CN120135094A - Data-driven wind-induced vibration active control method and device, storage medium, and electronic device - Google Patents

Abstract

The invention provides a wind vibration active control method and device based on data driving, a storage medium and an electronic device, wherein the method comprises the steps of obtaining a running speed of a vehicle, obtaining initial rear window opening data of the vehicle and configuring a target noise reduction position in the vehicle, calculating initial wind vibration noise pleasure degree in the vehicle based on the running speed, obtaining a wind vibration active control big model corresponding to the target noise reduction position, calculating target front window opening data combined and linked with the initial rear window opening data by utilizing the wind vibration active control big model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, and controlling a window of the vehicle according to the target front window opening data. According to the embodiment of the invention, the data-driven large model is adopted, and the pleasure degree of wind vibration noise is calculated through the pleasure degree model combined with the subjective and objective, so that the accuracy of the pleasure degree of wind vibration noise is improved.

Description

Wind vibration active control method and device based on data driving, storage medium and electronic device

Technical Field

The invention relates to the technical field of NVH (Noise, vibration, harshness, noise, vibration and harshness), in particular to a wind vibration active control method and device based on data driving of a vehicle, a storage medium and an electronic device.

Background

In the related art, with the continuous development and progress of the automobile industry, the riding comfort of the automobile has become one of the most important performance indexes, and the NVH performance of the automobile greatly influences the riding experience of passengers in the automobile, so that the good NVH performance can greatly improve the purchasing desire of consumers. In recent years, with the increasing popularity of new energy automobiles, the influence of wind noise is increasingly remarkable, and the wind noise becomes a research hot spot in a new period, and the wind vibration of windowing is a key problem of complaining of wind noise by current users.

Wind noise is the aerodynamic acoustic response of the air in the vehicle to the external transient airflow. Wind vibration noise can be divided into skylight wind vibration and side window wind vibration, but the side window wind vibration is relatively difficult to control, and particularly rear window wind vibration is not a good control method all the time. However, when the vehicle is driven daily, the window opening requirement of the rear passengers is continuous, no matter the passengers are self-driven to enjoy the beauty, the children desire fresh air, or the old can open the window to avoid carsickness, and the rear window is opened along with the rising of the vehicle speed, so that the passengers in the vehicle can be uncomfortable to a great extent due to the strong ear pressing sense.

In the related art, wind vibration noise is regulated by automatically opening and closing the window by a user, the regulation efficiency is low, the user cannot grasp the optimal window regulating position and opening degree, even the wind vibration noise after regulation is larger and larger, each seat can generally only regulate the windows beside the seat, and if the wind vibration noise at a specific position is required to be reduced, the windows of a plurality of seats need to be tried to be regulated, and the operation is complicated.

Disclosure of Invention

The invention provides a wind vibration active control method and device based on data driving, a storage medium and an electronic device, and aims to solve the technical problems in the related art.

According to one embodiment of the invention, a data-driven wind vibration active control method is provided, and comprises the steps of obtaining a running speed of a vehicle, obtaining initial rear window opening data of the vehicle and configuring a target noise reduction position in the vehicle, calculating initial wind vibration noise pleasure degree of the interior of the vehicle based on the running speed, obtaining a wind vibration active control large model corresponding to the target noise reduction position, calculating target front window opening data combined and linked with the initial rear window opening data by utilizing the wind vibration active control large model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, wherein the wind vibration active control large model is used for outputting front window opening data with optimal wind vibration noise pleasure degree of the target position on the premise of the current running speed and the initial rear window opening data, and controlling windows of the vehicle according to the target front window opening data.

Optionally, calculating the initial wind vibration noise pleasure degree of the vehicle interior based on the running speed comprises determining a speed interval in which the running speed is located, generating a pleasure degree weight coefficient of a noise sound quality parameter according to the speed interval, configuring a wind vibration noise pleasure degree model based on the pleasure degree weight coefficient, and calculating the initial wind vibration noise pleasure degree of the vehicle interior by adopting the wind vibration noise pleasure degree model.

Optionally, calculating the initial wind vibration noise pleasure degree of the vehicle interior by using the wind vibration noise pleasure degree model comprises obtaining a noise sound quality parameter collected by a microphone of the vehicle at the target noise reduction position, and calculating the initial wind vibration noise pleasure degree PD of the vehicle interior by using a formula of PD=x₁/SPL +x₂/R+x₃/L +x₄/S, wherein SPL represents a sound pressure level, R represents roughness, L represents loudness, S represents sharpness, x1, x2, x3 and x4 respectively represent weight coefficients of the sound pressure level, roughness, loudness and sharpness, and SPL, R, L, S is the noise sound quality parameter.

Optionally, the noise sound quality parameters comprise a low-frequency noise parameter and a high-frequency noise parameter, the generating of the pleasure degree weight coefficient of the noise sound quality parameter according to the speed interval comprises judging whether the maximum value of the speed interval is smaller than a first threshold value, generating a first pleasure degree weight coefficient if the maximum value of the speed interval is smaller than the first threshold value, judging whether the maximum value of the speed interval is smaller than a second threshold value if the maximum value of the speed interval is larger than or equal to the first threshold value, wherein the second threshold value is larger than the first threshold value, the weight of the low-frequency noise parameter in the first pleasure degree weight coefficient is larger than the weight of the high-frequency noise parameter, generating a second pleasure degree weight coefficient if the maximum value of the speed interval is smaller than a second threshold value, and generating a third pleasure degree weight coefficient if the maximum value of the speed interval is larger than or equal to the second threshold value, wherein the weight of the low-frequency noise parameter in the second pleasure degree weight coefficient is equal to the weight of the high-frequency noise parameter, and the weight of the low-frequency noise parameter in the third pleasure degree weight coefficient is smaller than the pleasure degree parameter.

Optionally, after the window of the vehicle is controlled according to the target front window opening data, the method further comprises collecting, at the target noise reduction position, a wind vibration noise pleasure degree after the window of the vehicle is controlled according to the target front window opening data, judging whether the wind vibration noise pleasure degree is greater than a pleasure degree threshold, updating the wind vibration active control large model according to the running speed, the initial rear window opening data, the target front window opening data and the wind vibration noise pleasure degree if the wind vibration noise pleasure degree is less than or equal to the pleasure degree threshold, outputting target front window opening data with the wind vibration noise pleasure degree greater than the pleasure degree threshold according to the updated target front window opening data, and determining that the wind vibration mode is completed if the wind vibration noise pleasure degree is greater than the pleasure degree threshold.

Optionally, updating the wind vibration active control large model by using the running speed, the initial rear window opening data, the target front window opening data and the wind vibration noise pleasure degree comprises the steps of iteratively executing the steps of acquiring the wind vibration noise pleasure degree after the window adjustment of the previous adjustment period until the wind vibration noise pleasure degree acquired by the target noise reduction position is larger than a pleasure degree threshold value, configuring the running speed, the initial rear window opening data and the wind vibration noise pleasure degree after the window adjustment of the previous adjustment period as input sample data of the current adjustment period, configuring the target front window opening data acquired by the previous adjustment period as output sample data of the current adjustment period, optimizing the wind vibration active control large model by using the input sample data and the output sample data, inputting the running speed and the initial rear window opening data of the current adjustment period into the wind vibration active control large model to acquire the target front window opening data of the current adjustment period, acquiring the wind vibration noise pleasure degree after the window adjustment of the current adjustment period, configuring the target front window opening data after the window adjustment of the current adjustment period into the current adjustment period, and judging whether the wind vibration noise pleasure degree of the vehicle is larger than the threshold value or not.

Optionally, after calculating the initial wind vibration noise pleasure degree of the vehicle interior based on the running speed, judging whether the initial wind vibration noise pleasure degree is smaller than a pleasure degree threshold value, and if the initial wind vibration noise pleasure degree is smaller than the pleasure degree threshold value, calling a wind vibration active control large model matched with the target noise reduction position to calculate target front window opening data linked with the initial rear window opening data combination.

The method comprises the steps of acquiring a plurality of groups of wind vibration sample data of a sample vehicle, wherein each group of wind vibration sample data comprises a running speed and windowing combination data, wind vibration noise pleasure degree acquired at a target noise reduction position, acquiring whole vehicle size data of the sample vehicle, taking the whole vehicle size data as a fixed parameter, taking the running speed and windowing combination data in the wind vibration sample data as independent variables and the wind vibration noise pleasure degree as dependent variables, and training an initial large model to obtain the wind vibration active control large model corresponding to the target noise reduction position.

Optionally, acquiring the whole vehicle size data of the sample vehicle comprises acquiring the vehicle type information of the sample vehicle, and searching vehicle size data and window size parameters matched with the vehicle type information in a preset database, wherein the vehicle size data comprises the length, width and height of the vehicle, and the window size parameters comprise the upper side length of window glass, the lower side length of the window glass, the total height of the window glass and the thickness of the window frame.

Optionally, acquiring initial back window opening data of the vehicle comprises acquiring initial back window opening positions of the vehicle, wherein the initial back window opening positions comprise at least one of left back window and right back window, and acquiring initial back window opening of the vehicle, wherein the initial back window opening data comprise the initial back window opening positions and the initial opening.

Optionally, according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, calculating target front window opening data combined and linked with the initial rear window opening data by using a wind vibration active control big model matched with the target noise reduction position comprises the steps of calling the wind vibration active control big model matched with the target noise reduction position, inputting the running speed, the initial wind vibration noise pleasure degree and the initial rear window opening data into the wind vibration active control big model, and outputting to obtain first target front window opening data of the vehicle, wherein the first target front window opening data comprises a first target front window position and a window opening degree thereof, and the wind vibration noise pleasure degree is used for representing the subjective bearing degree of passengers in the vehicle on wind vibration noise.

The method comprises the steps of acquiring a windowing combination table matched with the whole vehicle size data, wherein each table item of the windowing combination table comprises a running speed, windowing combination data and corresponding wind vibration noise pleasure degrees, the windowing combination data comprise opening data of a plurality of windowing positions, the running speed and the initial rear window opening data are used as keywords, and searching second target front window opening data with highest wind vibration noise pleasure degrees in the windowing combination data of the windowing combination table, wherein the second target front window opening data comprise a second target front window position and the windowing opening degree thereof, and the wind vibration noise pleasure degrees are used for representing subjective bearing degrees of passengers in the vehicle on wind vibration noise.

Optionally, before calculating the target front window opening data linked with the initial rear window opening data combination by using the wind vibration active control big model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, judging whether the vehicle starts a wind vibration removal mode or not, if the vehicle starts the wind vibration removal mode, determining that the target front window opening data linked with the initial rear window opening data combination is calculated by using the wind vibration active control big model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree.

Optionally, judging whether the vehicle starts the wind vibration removal mode comprises the steps of obtaining rainfall data of a rainfall sensor of the vehicle, wherein the wind vibration noise pleasure degree is used for representing subjective bearing degree of passengers in the vehicle on wind vibration noise, and judging whether the vehicle starts the wind vibration removal mode based on the rainfall data, the initial wind vibration noise pleasure degree, the running speed and the initial rear window opening data.

Optionally, judging whether the vehicle starts a wind vibration removing mode based on the rainfall data, the initial wind vibration noise pleasure degree, the running vehicle speed and the initial rear window opening data comprises judging whether the rainfall data is smaller than a rainfall threshold value, judging whether the initial wind vibration noise pleasure degree is smaller than a pleasure degree threshold value, judging whether the running vehicle speed is larger than a vehicle speed threshold value and judging whether the initial rear window opening data is larger than a vehicle window opening threshold value, and determining that the vehicle starts the wind vibration removing mode if the rainfall data is smaller than the rainfall threshold value, the initial wind vibration noise pleasure degree is smaller than a pleasure degree threshold value, the running vehicle speed is larger than a vehicle speed threshold value and the initial rear window opening data is larger than the vehicle window opening threshold value.

Optionally, controlling the window of the vehicle according to the target front window opening data comprises analyzing the target front window and the target opening degree in the target front window opening data, sending a windowing instruction to an opening degree adjustment actuator of the vehicle, wherein the windowing instruction is used for instructing to adjust the target front window to the target opening degree under the condition of maintaining the opening degree and the opening degree of the initial rear window opening data, and controlling the opening degree adjustment actuator to adjust the target front window to the target opening degree.

According to another embodiment of the invention, a data-driven wind vibration active control device is provided, and comprises an acquisition module, a calculation module and a control module, wherein the acquisition module is used for acquiring the running speed of a vehicle, acquiring initial rear window opening data of the vehicle, calculating initial wind vibration noise pleasure degree in the vehicle, acquiring whole vehicle size data of the vehicle and configuring target noise reduction positions in the vehicle, the calculation module is used for calculating a plurality of groups of target front window opening data combined and linked with the initial rear window opening data by using a wind vibration active control large model matched with the target noise reduction positions according to the running speed, the initial rear window opening data, the initial wind vibration noise pleasure degree and the whole vehicle size data, and the control module is used for selecting one target front window opening data from the plurality of groups of target front window opening data to control the window of the vehicle.

Optionally, the first calculation module comprises a determination unit for determining a speed interval in which the running speed is located, a generation unit for generating a pleasure degree weight coefficient of a noise sound quality parameter according to the speed interval, a configuration unit for configuring a wind vibration noise pleasure degree model based on the pleasure degree weight coefficient, and a calculation unit for calculating initial wind vibration noise pleasure degree in the vehicle by adopting the wind vibration noise pleasure degree model.

Optionally, the noise quality parameters comprise low-frequency noise parameters and high-frequency noise parameters, the generating unit comprises a judging subunit, a first generating subunit and a second generating subunit, wherein the judging subunit is used for judging whether the maximum value of the speed interval is smaller than a first threshold value, the first generating subunit is used for generating a first pleasure degree weight coefficient if the maximum value of the speed interval is smaller than the first threshold value, the maximum value of the speed interval is larger than or equal to the first threshold value, judging whether the maximum value of the speed interval is smaller than a second threshold value, the second threshold value is larger than the first threshold value, the weight of the low-frequency noise parameters in the first pleasure degree weight coefficient is larger than the weight of the high-frequency noise parameters, the second generating subunit is used for generating a second pleasure degree weight coefficient if the maximum value of the speed interval is larger than or equal to the second threshold value, and the weight of the low-frequency noise parameters in the second pleasure degree weight coefficient is smaller than the high-frequency noise parameters.

According to another aspect of the embodiments of the present application, there is also provided a storage medium including a stored program that performs the above steps when running.

According to another aspect of the embodiment of the application, there is also provided an electronic device, including a processor, a communication interface, a memory and a communication bus, where the processor, the communication interface, and the memory complete communication with each other through the communication bus, where the memory is used to store a computer program, and the processor is used to execute the steps in the above method by running the program stored on the memory.

Embodiments of the present application also provide a computer program product comprising instructions which, when run on a computer, cause the computer to perform the steps of the above method.

The invention has the beneficial effects that:

1. The front window opening position and the window opening degree for eliminating wind vibration noise of the target noise reduction position can be predicted through the wind vibration active control large model, so that the optimal wind vibration control effect on the target noise reduction position is realized;

2. The wind vibration active control large model can be optimized automatically according to real vehicle wind vibration data, a large number of repeated wind vibration control bottoming tests are not needed, and development time and cost are saved;

3. The pleasure degree weight coefficient of the noise sound quality parameter is configured according to the actual running speed of the vehicle, so that the feeling of the low-frequency noise parameter and the high-frequency noise parameter on the human ear at different speeds can be more accurately reflected, and the accuracy and objectivity of the pleasure degree of wind vibration noise are improved;

4. The pleasure degree of wind vibration noise is calculated through the noise quality parameters of the specific combination, and a wind vibration removal mode for starting a wind vibration active control function is triggered, so that the accuracy of the pleasure degree of the wind vibration noise is improved, the wind vibration removal mode can be entered at a more accurate time, and the wind vibration of a vehicle is eliminated;

5. The wind vibration problem of the vehicle window can be effectively solved through the wind vibration active control system based on data driving, and the reduction of the frequency amplitude sound pressure level of the wind vibration problem by more than 20dB is realized.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:

FIG. 1 is a block diagram of a hardware architecture of an automobile in accordance with an embodiment of the present invention;

FIG. 2 is a flow chart of a data-driven wind-vibration-based active control method according to an embodiment of the present invention;

FIG. 3 is a flow chart of the establishment of a wind noise pleasure model in an embodiment of the present invention;

FIG. 4 is a schematic diagram of the wind vibration of an automobile in an embodiment of the invention;

FIG. 5 is a schematic illustration of a vehicle glazing in an embodiment of the invention;

FIG. 6 is a flow chart of the establishment of a large model for active control of wind vibrations in an embodiment of the present invention;

FIG. 7 is a graph showing a comparison of spectra before and after a vehicle is turned on in a wind-vibration mode according to an embodiment of the present invention;

FIG. 8 is a flow chart of an active control method of wind vibration based on data driving in an embodiment of the invention;

FIG. 9 is a flow chart of wind vibration control large model optimization based on feedback control in an embodiment of the invention;

fig. 10 is a block diagram of a wind vibration active control device based on data driving according to an embodiment of the present invention.

Detailed Description

In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Example 1

The method embodiment provided by the first embodiment of the application can be executed in an automobile, a server, a processor, a vehicle window controller or similar processing device. Taking an example of running on a car, fig. 1 is a block diagram of a hardware structure of a car according to an embodiment of the present application. As shown in fig. 1, the car may include one or more (only one is shown in fig. 1) processors 10 (the processors 10 may include, but are not limited to, a microprocessor MCU or a processing means such as a programmable logic device FPGA) and a memory 11 for storing data, and optionally, a transmission device 12 for communication functions and an input-output device 13. It will be appreciated by those skilled in the art that the configuration shown in fig. 1 is merely illustrative and is not intended to limit the configuration of the vehicle described above. For example, the automobile may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

The memory 11 may be used to store a car program, for example, a software program of an application software and a module, such as a car program corresponding to a data-driven wind vibration active control method of a car according to an embodiment of the present invention, and the processor 10 executes various functional applications and data processing by running the car program stored in the memory 11, that is, implements the method described above. Memory 11 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 11 may further include memory located remotely from processor 10, which may be connected to the car through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

The transmission device 12 is used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of an automobile. In one example, the transmission device 12 includes a network adapter (Network Interface Controller, simply referred to as a NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 12 may be a Radio Frequency (RF) module for communicating with the internet wirelessly.

In this embodiment, a method for actively controlling wind vibration based on data driving is provided, and fig. 2 is a flowchart of a method for actively controlling wind vibration based on data driving according to an embodiment of the present invention, as shown in fig. 2, where the flowchart includes the following steps:

step S200, acquiring the running speed of a vehicle, acquiring initial rear window opening data of the vehicle and configuring a target noise reduction position in the vehicle;

Alternatively, the running speed of the vehicle and the initial rear window opening data may be acquired from a communication line (e.g., CAN bus) of the vehicle, a controller, a vehicle machine, or the like.

The initial rear window opening data is current window opening data of the vehicle, such as manual window opening data of passengers or drivers, including window opening positions and opening degrees of one or more rear windows, and the like.

Optionally, acquiring initial back window opening data of the vehicle comprises acquiring initial back window opening positions of the vehicle, wherein the initial back window opening positions comprise at least one of left back window and right back window, and acquiring initial back window opening of the vehicle, wherein the initial back window opening data comprise the initial back window opening positions and the initial opening.

Step S201 of calculating an initial wind-vibration noise pleasure degree of the inside of the vehicle based on the running vehicle speed;

The initial wind noise pleasure degree is the wind noise pleasure degree (simply called pleasure degree) when the vehicle is currently opened with the rear window, and is used for representing the subjective bearing degree of passengers in the vehicle on the wind noise.

At different driving speeds, the influence of the wind vibration noise pleasure degree on passengers in the vehicle is different.

Step S202, acquiring a wind vibration active control large model corresponding to the target noise reduction position;

Alternatively, the target noise reduction position may be a main driving position, a sub driving position, a second row of left rear seat positions, a second row of right rear seat positions, a third row of left rear seat positions, a third row of right rear seat positions, or the like. The target noise reduction position may be selected by a user, or may be determined by the load condition of each seat or the priority of each seat, if only the main driving position has a passenger, the target noise reduction position is the main driving position, the model 1 corresponding to the main driving position is obtained, and if the priority of the auxiliary driving position is the highest, the target noise reduction position is the auxiliary driving position, and the model 2 corresponding to the auxiliary driving position is obtained.

Step S203, calculating target front window opening data combined and linked with the initial rear window opening data by using a wind vibration active control large model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, wherein the wind vibration active control large model is used for outputting front window opening data with optimal wind vibration noise pleasure degree of the target noise reduction position on the premise of the current running speed and the initial rear window opening data;

In this embodiment, the wind vibration active control large model includes a plurality of models or includes a plurality of sets of model parameters, each model or each set of model parameters corresponds to a target noise reduction position, and is configured to output front window opening data with the optimal (maximum) wind vibration noise pleasure degree at the corresponding target noise reduction position on the premise of maintaining the current running speed and the current rear window opening data (initial rear window opening data), so as to realize linkage opening of the front window following the rear window, and eliminate wind vibration noise in the vehicle.

Optionally, the target front window opening data output by the wind vibration active control large model includes a plurality of groups, the plurality of groups of target front window opening data includes one or more groups of target front window opening data, each group of target front window opening data corresponds to one target pleasure degree parameter, and the target pleasure degree parameter is an influence factor of subjective bearing degree of wind vibration noise by a user, such as sound pressure level, roughness, loudness and sharpness, for example, comprehensive optimization (simultaneously considering sound pressure level, roughness, loudness and sharpness) corresponds to one front window opening data, sound pressure level optimization corresponds to one front window opening data, roughness optimization corresponds to one front window opening data, and the like. In another case, considering that the ventilation quantity and the windowing area of multiple types of ventilation quantity and windowing area can all achieve the situation that the pleasure degree of wind vibration noise is optimal, on the basis that the pleasure degree of wind vibration noise is optimal, the ventilation quantity and windowing area can be further considered to be optimal, and each group of target front window opening data corresponds to one group of ventilation quantity and/or windowing area, for example, the ventilation quantity is maximum, the ventilation quantity is minimum, the windowing area is maximum, the windowing area is minimum, and the like.

The self-oscillation formed by the airflow caused by the opening of the rear window and the resonance of the inner cavity of the vehicle lead to wind vibration, and the frequency of the self-oscillation can be changed by changing the airflow in the vehicle, so that the natural frequency of the inner cavity of the vehicle is avoided, wind vibration noise is eliminated, and the linkage opening of the front window can enable the airflow to enter and exit in the vehicle when the rear window is opened, and the frequency of the self-oscillation is reduced by reducing the pressure difference between the inside and outside of the vehicle, so that wind vibration noise is effectively restrained.

The initial rear window opening data and the target front window opening data of the embodiment are combined linkage windowing data, and include at least two windows combined in linkage and corresponding opening degrees. The target front window opening data includes a window opening position, an opening degree, and the like of one or more front windows, for example, an opening degree of a left front window, an opening degree of a right front window. The linkage of the front window and the rear window is realized, and wind vibration noise caused by opening the rear window only is reduced.

Step S204, controlling the window of the vehicle according to the target front window opening data;

Initial rear window opening data of the vehicle is fixed while controlling front window opening based on the target front window opening data. If the initial back window opening data for the next adjustment cycle, then the target front window opening data needs to be recalculated.

There are various requirements for opening rear windows in the driving scene of an automobile, such as smoke exhausting habit of adults, air convection is needed to be formed by opening windows to exhaust smoke, the ability of adjusting physiological functions of old people is reduced, children are easy to feel carsickness when riding the automobile, discomfort of the body is relieved by means of ventilation of windows, the children are in a position of riding the automobile, and the children like repeatedly lifting the windows to explore things outside the windows or eat pungent taste foods, so that people need to open windows and loose taste, and the automobile is driven to seaside or forest by themselves, so that people want to blow sea wind, breathe fresh air, enjoy beauty and have window opening requirements. When the rear window is opened to generate wind vibration, the front window is opened in a linkage mode to achieve elimination, and the scene of front-rear window linkage adjustment in the embodiment comprises a single-opening left rear window linkage front window, a single-opening right rear window linkage front window and a double-rear window linkage front window, wherein the positions (left front window and right front window), the quantity and the opening degree of the linkage opening front windows are calculated through the running speed, the initial rear window opening data and the whole vehicle size data.

The method comprises the steps of obtaining the running speed of a vehicle, obtaining initial rear window opening data of the vehicle and configuring a target noise reduction position in the vehicle, calculating initial wind vibration noise pleasure degree in the vehicle based on the running speed, obtaining a wind vibration active control large model corresponding to the target noise reduction position, calculating target front window opening data linked with the initial rear window opening data by utilizing the wind vibration active control large model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, wherein the wind vibration active control large model is used for outputting front window opening data with optimal wind vibration noise pleasure degree of the target noise reduction position on the premise of the current running speed and the initial rear window opening data, controlling vehicle windows of the vehicle according to the target front window opening data, solving the technical problem that the vehicle cannot actively control wind vibration, and eliminating or weakening wind vibration phenomena of the vehicle at the target noise reduction position in related technologies, and improving comfort degree of users.

In one implementation of the present embodiment, calculating the initial wind noise pleasure level in the vehicle based on the running vehicle speed includes determining a speed section in which the running vehicle speed is located, generating a pleasure level weight coefficient of a noise sound quality parameter according to the speed section, configuring a wind noise pleasure level model based on the pleasure level weight coefficient, and calculating the initial wind noise pleasure level in the vehicle using the wind noise pleasure level model.

Optionally, the noise sound quality parameters comprise a low-frequency noise parameter and a high-frequency noise parameter, the generating of the pleasure degree weight coefficient of the noise sound quality parameter according to the speed interval comprises judging whether the maximum value of the speed interval is smaller than a first threshold value, generating a first pleasure degree weight coefficient if the maximum value of the speed interval is smaller than the first threshold value, judging whether the maximum value of the speed interval is smaller than a second threshold value if the maximum value of the speed interval is larger than or equal to the first threshold value, wherein the second threshold value is larger than the first threshold value, the weight of the low-frequency noise parameter in the first pleasure degree weight coefficient is larger than the weight of the high-frequency noise parameter, generating a second pleasure degree weight coefficient if the maximum value of the speed interval is smaller than a second threshold value, and generating a third pleasure degree weight coefficient if the maximum value of the speed interval is larger than or equal to the second threshold value, wherein the weight of the low-frequency noise parameter in the second pleasure degree weight coefficient is equal to the weight of the high-frequency noise parameter, and the weight of the low-frequency noise parameter in the third pleasure degree weight coefficient is smaller than the pleasure degree parameter.

Optionally, the noise quality parameters include sound pressure level, roughness, loudness, sharpness, wherein the low-frequency noise parameters are noise quality parameters that bring about low-frequency wind vibration, including sound pressure level and loudness, and the high-frequency noise parameters are noise quality parameters that bring about high-frequency wind vibration, including roughness and sharpness.

In the embodiment, the fact that the low-frequency wind vibration is more prominent than the high-frequency wind noise is considered in the low-speed state of the vehicle, and the human ear feel on the low-frequency wind vibration is more obvious. In the high-speed state, the influence of the low-frequency wind vibration is reduced and the influence of the high-frequency wind noise is increased due to the influence of the masking effect as the high-frequency wind noise component is increased. The influence of noise with different frequencies on the sound quality parameters is greatly different, the sound pressure level and the loudness are greatly influenced by low-frequency noise, and the roughness and the sharpness are more sensitive to high-frequency noise. Therefore, in order to make the pleasure degree model more applicable, the wind vibration noise pleasure degree model is established and configured according to the speed interval where the running vehicle speed is located.

For example, three speed sections are set, namely, a low speed section of 0-70 km/h, a medium speed section of 70-90 km/h, the high-speed interval is 90km/h to +.infinity, the first threshold is 70km/h, and the second threshold is 90km/h.

In one implementation scenario, the vehicle runs at a low speed in an urban area, wind vibration noise appears after the rear window is opened, the low-frequency wind vibration is more prominent than the high-frequency wind noise, and the feeling of the human ear on the low-frequency wind vibration is more obvious. The running speed is 40km/h, the influence of low-frequency noise parameters (including sound pressure level and loudness) on passengers in the vehicle is larger in a low-speed interval of 0-70 km/h, the weight ratio of the low-frequency noise parameters (including roughness and sharpness) is configured to be 35% and 35%, the influence of the high-frequency noise parameters on the passengers in the vehicle is smaller, and the weight ratio of the high-frequency noise parameters is configured to be 15% and is smaller than the weight of the low-frequency noise parameters.

In one implementation scene, the vehicle runs at a medium speed in national roads, wind vibration noise appears after a rear window is opened, low-frequency wind vibration and high-frequency wind vibration appear simultaneously, a human ear can obviously sense the low-frequency wind vibration and the high-frequency wind vibration simultaneously, the running speed is 80km/h, the low-frequency noise parameters (including sound pressure level and loudness) and the high-frequency wind noise parameters (including roughness and sharpness) are jointly influenced in a medium speed interval of 70 km/h-90 km/h, the weights are equivalent, the weight ratio of each parameter of the low-frequency noise parameters is 25% and 25%, the weight ratio of each parameter of the high-frequency noise parameters is 25% and 25%, and the high-frequency noise parameters are equal to the weight of the low-frequency noise parameters.

In one implementation scene, the vehicle runs at a high speed on an expressway, wind vibration noise appears after a rear window is opened, the high-frequency wind vibration is more prominent than the low-frequency wind noise, the feeling of human ears on the high-frequency wind vibration is more obvious, the running speed is 100km/h, the running speed is 90km/h to + -infinity in a high-speed interval, the low-frequency noise parameters (including sound pressure level and loudness) have small influence on passengers in the vehicle, the weight ratio is 15% and 15%, the high-frequency noise parameters (including roughness and sharpness) have large influence on the passengers in the vehicle, the weight ratio is 35% and 35%, and the weight ratio is larger than that of the low-frequency noise parameters.

The vehicle speed is divided into three types, the vehicle speed is 70km/h and below, the pleasure degree model under the vehicle speed mainly considers the influence caused by low-frequency wind vibration, the sound pressure level and the loudness of the low-frequency wind vibration are the most main influence parameters of wind vibration noise pleasure degree, the weight coefficient is larger than the coefficient under other vehicle speeds, the roughness and the sharpness occupy smaller proportion, the noise data of the vehicle speed is divided into a second type, the pleasure degree model under the vehicle speed needs to consider the comprehensive influence caused by the low-frequency wind vibration and the high-frequency wind noise, the wind vibration noise pleasure degree is influenced by the sound pressure level, the loudness, the roughness and the sharpness, the noise data under the vehicle speed is the third type, the wind vibration noise pleasure degree under the vehicle speed is more considered the influence of a high-frequency wind noise part, the roughness and the sharpness weight occupying proportion in the vibration noise pleasure degree model is increased, and the sound pressure level and the loudness occupying ratio are reduced.

By adopting the scheme of the embodiment, the pleasure degree weight coefficient of the noise sound quality parameter is configured based on the actual running speed of the vehicle, so that more accurate wind vibration noise pleasure degree can be calculated.

In one example, calculating the initial wind noise pleasure of the interior of the vehicle using the wind noise pleasure model includes obtaining a noise sound quality parameter collected by a microphone of the vehicle at the target noise reduction location, and calculating the initial wind noise pleasure of the interior of the vehicle using the following formula pd=x₁/SPL +x₂/R+x₃/L +x₄/S, where SPL represents sound pressure level, R represents roughness, L represents loudness, S represents sharpness, x₁、x₂、x₃、x₄ represents weight coefficients of sound pressure level, roughness, loudness, sharpness, and SPL, R, L, S is the noise sound quality parameter.

It should be noted that the formula for solving PD in the above example is not unique (that is, the combination of formulas capable of reflecting the deviation between PD and SPL, R, L, S satisfies the inventive concept), the formula for solving PD may be modified, for example, pd=a/(x₁*SPL +x₂*R+x₃*L +x₄ ×s), where a is a constant ,PD=x₁/SPL*k₁+x₂/R*k₂+x₃/L*k₃+x₄/S*k₄,k₁、k、k₃、k₄, and represents the fitting coefficients of sound pressure level, roughness, loudness, and sharpness, respectively.

In this embodiment, four key psychoacoustic parameters including sound pressure level, roughness, loudness and sharpness are selected to describe the perceptual characteristics of the auditory system of the human ear on the sound, so that the subjective feeling of the human ear on the wind vibration noise can be reflected better compared with other parameters and combinations thereof, in consideration of the characteristics of the wind vibration noise, such as a specific frequency range (usually lower than 30 Hz), a specific amplitude (the peak sound pressure level is higher than 100 dB) and the sharp peak frequency.

Fig. 3 is a flowchart for establishing a pleasure degree model of wind vibration noise in the embodiment of the present invention, where sound quality parameters such as sound pressure level, roughness, loudness, sharpness, etc. are combined with subjective evaluation tests in advance to establish an objective calculation formula for pleasure degree, and the flowchart includes that microphones are firstly arranged at positions of respective seat headrests or corresponding ceiling positions above the seat headrests, and are connected with a data acquisition device. And respectively acquiring a plurality of groups of wind vibration noise in the vehicle under different vehicle types, different vehicle speeds, different vehicle window combinations and different opening degrees, and calculating sound quality parameters such as the sound pressure level, roughness, loudness, sharpness and the like of the wind vibration noise. And then, carrying out subjective evaluation on the sound quality of the wind vibration noises in the groups by using a grade grading method to obtain subjective pleasure degree scores, wherein the higher the subjective pleasure degree score is, the better the pleasure degree of the user on the noise sample is represented, and conversely, the lower the score is, the more objectionable the user is. And finally, fitting the calculated sound quality parameter and the subjective pleasure degree score to obtain a fitting formula between the wind vibration noise sound quality parameter and the subjective pleasure degree, namely an objective pleasure degree formula, namely a formula capable of reflecting the wind vibration noise sound quality, namely formula (3). The subjective feeling of the human ear on the automobile wind vibration noise can be reflected by utilizing the objective pleasure degree under the condition of not organizing subjective evaluation tests.

PD=x₁/SPL +x₂/R+x₃/L +x₄/S (3)

In the above formula, PD represents pleasure, SPL represents sound pressure level, R represents roughness, L represents loudness, S represents sharpness, x₁、x₂、x₃、x₄ represents weight coefficients of sound pressure level, roughness, loudness and sharpness, respectively, and the weight of the weight coefficients can be flexibly adjusted based on the actual running speed of the vehicle or the preference of a user to a certain noise sound quality parameter.

The subjective evaluation test is carried out by establishing an acoustic quality evaluation model about windowed wind vibration noise and utilizing a level evaluation method, and a calculation model about pleasure degree is established based on acoustic quality parameters such as sound pressure level, loudness, roughness, sharpness and the like, so that the problem that subjective feeling of human ears on wind vibration noise cannot be comprehensively and accurately reflected by simply depending on sound pressure level or noise amplitude in the existing wind vibration control evaluation system is solved.

In one implementation scenario of the embodiment, acquiring a wind vibration active control large model corresponding to the target noise reduction position comprises the steps of acquiring multiple groups of wind vibration sample data of a sample vehicle, wherein each group of wind vibration sample data comprises a running speed and windowing combination data, wind vibration noise pleasure degree acquired at the target noise reduction position, the windowing combination data comprises rear window opening data and front window opening data, acquiring whole vehicle size data of the sample vehicle, taking the whole vehicle size data as fixed parameters, taking the running speed and windowing combination data in the wind vibration sample data as independent variables and the wind vibration noise pleasure degree as dependent variables, and training an initial large model to obtain the wind vibration active control large model corresponding to the target noise reduction position.

Optionally, acquiring the whole vehicle size data of the sample vehicle comprises acquiring the vehicle type information of the sample vehicle, and searching vehicle size data and window size parameters matched with the vehicle type information in a preset database, wherein the vehicle size data comprises the length, width and height of the vehicle, and the window size parameters comprise the upper side length of window glass, the lower side length of the window glass, the total height of the window glass and the thickness of the window frame.

The applicant finds that wind vibration of a side window (such as a rear window) is caused by Helmholtz resonance between pressure pulsation generated in a vehicle and occurrence of a cavity in the vehicle due to self-excitation of a free shear layer, when the vehicle runs at a certain speed, the speed difference of air flow inside and outside the vehicle exceeds a critical value, the pressure difference is large, an unstable shear layer is easily formed on the front edge of the skylight or the side window, periodic falling off is generated to form a shear vortex, pressure change and disturbance waves are generated in the process of forming, developing and crushing the shear vortex, the shear vortex propagates upwards in an acoustic mode according to sound velocity, the acoustic mode acts on the front edge of a cavity opening, the vortex falling off of a new round is excited, a periodic feedback mechanism is formed, the pressure oscillation with a certain specific frequency is generated in the cavity area, the periodic excitation of the vortex is called self-excitation, noise caused by wind vibration in the vehicle is wind vibration noise, and the wind vibration frequency is calculated as shown in formula (1), and fig. 4 is a schematic diagram of wind vibration formation of the vehicle in the embodiment of the invention. For the geometric cavity structure in the automobile, when the automobile window is opened, the automobile window and the passenger cabin form a Helmholtz resonant cavity, and the natural frequency of the resonant cavity is determined by the geometric shape of the cavity and the air property in the cavity, and the natural frequency empirical formula of the cavity is shown as a formula (2). When the frequency of self-oscillation approaches or equals to the natural frequency of the Helmholtz resonator formed by the vehicle window and the passenger cabin, resonance phenomenon is excited, and wind vibration phenomenon is most intense.

(1)

(2)

In the above-mentioned method, the step of,In order for the incoming flow rate to be high,Is the length of the open cavity, n is the number of falling vortex modes, c is the sound velocity,For the window opening area of the vehicle window,Is the volume of the cavity in the vehicle,Is the thickness of the window frame, i.e., the thickness of the B pillar of the vehicle.

Due to the incoming flow velocityAnd the speed of the vehicleDirectly related to open cavity lengthThe analogy is window length, which is related to window opening area. FIG. 5 is a schematic view of a vehicle window opening area in an embodiment of the inventionIs related with the descending height m of the window (the ratio of m to the total height h of the glass is the opening degree p of the window), the upper side length a of the window glass, the lower side length b of the window glass and the total height h of the window glass, the volume of the cavity in the vehicleThe length L, the width W and the height H of the whole vehicle are strongly related; in addition, the front window and the rear window have different windowing positions, and the wind vibration degree has larger difference.

Therefore, the influence factor of the windowing wind vibration is mainly the speed of the vehicleThe vehicle size data and the window size parameters comprise a vehicle window opening degree p, a vehicle window glass upper side length a, a vehicle window glass lower side length b, a vehicle window glass total height H, a vehicle window frame thickness d, a vehicle whole vehicle length L, a vehicle width W, a vehicle height H and a window opening position M when the vehicle window is opened.

In one implementation manner of the embodiment, according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, calculating the target front window opening data combined and linked with the initial rear window opening data by using the wind vibration active control big model matched with the target noise reduction position comprises the steps of calling the wind vibration active control big model matched with the target noise reduction position, inputting the running speed, the initial wind vibration noise pleasure degree and the initial rear window opening data into the wind vibration active control big model, and outputting to obtain first target front window opening data of the vehicle, wherein the first target front window opening data comprises a first target front window position and a window opening degree thereof, and the wind vibration noise pleasure degree is used for representing subjective bearing degree of passengers in the vehicle on wind vibration noise.

In the embodiment, the wind vibration active control large model is BP (Back Propagation) neural network model, and training is carried out by adopting sample data in the early stage to obtain a pre-trained wind vibration active control large model.

Based on the description of the embodiment, when the wind vibration active control large model is trained, the independent variables comprise vehicle size data and window size parameters, vehicle speed, window opening combination data (window opening positions of front and rear windows and opening degrees thereof), dependent variables are wind vibration noise pleasure degrees, the vehicle size data and the window size parameters are fixed parameters under the condition that the vehicle is unchanged, and the change parameters of the independent variables are the vehicle speed and window opening combination data. And training an initial model through sample data consisting of a plurality of groups of independent variables and dependent variables to obtain the wind vibration active control large model.

The neural network model has self-adaption and learning capability for the complex uncertainty problem, and provides a new idea and method for solving the windowing wind vibration problem. The neural network model can learn and store a large number of mappings between input vectors and output vectors without the need to build a mathematical model expressing the mappings in advance. The learning rule is to use various quick descent methods to continuously adjust the node contact weight and node output threshold value in the network, so as to minimize the mean square error between the network output result and the known result.

FIG. 6 is a flow chart for establishing a large wind vibration active control model in the embodiment of the invention, wherein the wind vibration noise data in a plurality of groups of vehicles under different working conditions are collected by utilizing equipment such as a data acquisition front end, a microphone and the like in a wind tunnel laboratory environment. Noise data test positions comprise four positions of a main driving headrest left ear (FLL), a secondary driving headrest right ear (FRR), a left rear seat headrest left ear (RLL) and a right rear seat headrest right ear (RRR), and wind vibration active control large models of four target noise reduction positions of a main driving, a secondary driving, a left rear seat and a right rear seat can be respectively obtained through training.

The test working conditions comprise different window opening combinations, different window opening degrees and different vehicle speeds, and the specific test working conditions are divided into a plurality of types according to the different window opening combinations and can be divided into a left single-opening rear window 2, a right single-opening rear window 3, a left rear window and a right front window 4 simultaneously and a left rear window and a right rear window 5 simultaneously.

The speed of the vehicle is divided into 9 gears from 50km/h to 130km/h according to 10km/h as one gear. The opening of the vehicle window is first gear according to the proportion of 10 percent, and the vehicle window is totally divided into 10 gears from 10 percent to 100 percent. And carrying out a windowing wind vibration test in the environment of the wind tunnel laboratory, and sequentially testing wind vibration noise data under different working conditions.

And calculating sound quality parameters such as sound pressure level, loudness, roughness, sharpness and the like of the wind vibration sample data under different working conditions, and obtaining pleasure values corresponding to different vehicle speeds, different vehicle window opening combinations and different opening of the vehicle model by utilizing the sound quality model. And selecting the lower value of the wind vibration noise pleasure degree value at the headrest position of the seat closest to the windowing position and the pleasure degree value at the main driving position to represent the wind vibration noise level in the vehicle, and calculating the wind vibration noise pleasure degree value at the target noise reduction position as the wind vibration noise pleasure degree value in the vehicle under the condition that the right rear window is opened.

As shown in fig. 5, for a certain vehicle model, when the window glass is in a right trapezoid and the window opening is p, the calculation formula of the opening area S is obtained by deriving from a calculation formula of the trapezoid area, and is as follows:

(4)

Wherein a represents the upper side length of the window glass, b represents the lower side length of the window glass, h represents the total height of the window glass, and p represents the percentage of the opening of the window glass.

From equation (4), it is known that the vehicle window opening area is strongly correlated with the window glass upper side length a, the window glass lower side length b, and the window glass total height h, indicating the window opening percentage p. And (3) respectively selecting different car types of the car and the SUV under different car speeds, different windowing positions and different car window openings, obtaining an optimal windowing combination and the opening thereof by taking pleasure in the wind vibration sound quality evaluation model in the step (2) as a judgment standard, and collecting and sorting samples, wherein the number of the samples is more than 1000. Samples were randomly scrambled, taking 70% as training samples and the other 30% as test samples.

Optionally, the wind vibration active control large model is a BP neural network model, the wind vibration noise pleasure degree data corresponding to the opening degree of the vehicle window combination is input into the neural network model, and the structure of the neural network model comprises an input layer, an output layer and an hidden layer. In the wind vibration active control large model after training is completed, initial rear window opening data, vehicle speed v, whole vehicle size data (whole vehicle length L, width W, height H, vehicle window frame thickness d and the like), wind vibration noise pleasure degree PD are used as input layers, optimal front window opening data (position and opening degree) are used as output layers, hyperbolic tangent functions Tansig are selected as activation functions of the hidden layers, and the activation functions of the output layers are linear functions Purelin. And (3) continuously and repeatedly training and optimizing the BP neural network model, performing performance evaluation by adopting a mean square error criterion, and finding out a wind vibration active control model structure with the optimal fitting effect.

In the process of training the BP neural network model, the BP neural network model is optimized by utilizing a genetic algorithm. In order to optimize the weight and the threshold value of the BP neural network model, firstly, the initial weight value of the BP neural network model is encoded and initialized to form a population, then, the fitness value of each individual is calculated, then, the operations of selection, intersection, variation and the like are carried out according to the fitness value of the individual to generate a new generation population, and a plurality of iterations are carried out until the optimal individual, namely the weight and the threshold value of the BP neural network model, are obtained, and finally, the wind vibration active control large model is built.

The speed, the length, the width and the height of the vehicle, the opening degree of the vehicle window, the opening thickness and the windowing position in the test sample are used as input values to be imported into a model structure, a windowing combination with optimal predicted pleasure and the opening degree of the windowing combination are obtained, the windowing combination is compared with the actual windowing combination and the opening degree of the test sample, the prediction function of the large model is self-checked, the confirmation error is controllable, and the wind vibration active control large model structure after training is guaranteed to have a good prediction function.

After the wind vibration removal control mode is started, the system inputs the information of the current speed, the initial wind vibration noise pleasure degree, the window opening position, the corresponding opening degree and the like into the wind vibration active control large model, and the controller starts to calculate the window combination with the optimal pleasure degree and the opening degree under the conditions of the current speed, the window opening position and the window opening degree based on the wind vibration active control large model.

In another implementation manner of the embodiment, the method comprises the steps of calling a windowing combination table matched with the whole vehicle size data, wherein each table item of the windowing combination table comprises a running speed, windowing combination data and corresponding wind vibration noise pleasure degrees, the windowing combination data comprise opening data of a plurality of windowing positions, the running speed and the initial rear window opening data are used as keywords, and searching second target front window opening data with highest wind vibration noise pleasure degrees in the windowing combination data of the windowing combination table, wherein the second target front window opening data comprise a second target front window position and the windowing opening degree thereof, and the wind vibration noise pleasure degrees are used for representing subjective bearing degrees of passengers in the vehicle on wind vibration noise.

The windowing combination table of this embodiment includes a plurality of table entries, and is stored in rows or columns, where the windowing combination data includes opening data of at least two windowing positions, as shown in table 1, which illustrates the windowing position combinations of RL (rear left) & RR (rear right), RL & FR (front right), RR & FL (front left), where the windowing combination data of the 9 th row table entry is 10% of window opening at the RL position, and 10% of window opening at the RR position:

TABLE 1

In one implementation mode of the embodiment, after calculating the initial wind vibration noise pleasure degree of the vehicle interior based on the running speed, judging whether the initial wind vibration noise pleasure degree is smaller than a pleasure degree threshold value, and if the initial wind vibration noise pleasure degree is smaller than the pleasure degree threshold value, calling a wind vibration active control large model matched with the target noise reduction position to calculate target front window opening data combined and linked with the initial rear window opening data.

In this embodiment, if the rear window of the vehicle is opened, the initial wind vibration noise pleasure degree detected and calculated in real time is smaller than the pleasure degree threshold value, the wind vibration active control large model matched with the target noise reduction position is triggered and invoked, the target front window opening data is output, and the front window opening is executed.

In one implementation mode of the embodiment, before calculating the target front window opening data combined and linked with the initial rear window opening data by using the wind vibration active control big model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, judging whether the vehicle starts a wind vibration removal mode or not, and determining that a plurality of groups of target front window opening data combined and linked with the initial rear window opening data are calculated by using the wind vibration active control big model matched with the target noise reduction position according to the running speed, the initial rear window opening data, the initial wind vibration noise pleasure degree and the whole vehicle size data if the vehicle starts the wind vibration removal mode.

In one example, judging whether the vehicle starts a wind vibration removal mode comprises obtaining rainfall data of a rainfall sensor of the vehicle, wherein the wind vibration noise pleasure degree is used for representing subjective bearing degree of passengers in the vehicle on wind vibration noise, and judging whether the vehicle starts the wind vibration removal mode based on the rainfall data, the initial wind vibration noise pleasure degree, the running speed and the initial rear window opening data.

In one example, determining whether the vehicle is to turn on a wind removal mode based on the rain data, the wind noise pleasure level, the travel vehicle speed, and the initial rear window opening data includes determining whether the rain data is less than a rain threshold, determining whether the wind noise pleasure level is less than a pleasure level threshold, determining whether the travel vehicle speed is greater than a vehicle speed threshold, and determining whether the initial rear window opening data is greater than a vehicle window opening threshold, and determining the vehicle on a wind removal mode if the rain data is less than a rain threshold, the wind noise pleasure level is less than a pleasure level threshold, the travel vehicle speed is greater than a vehicle speed threshold, and the initial rear window opening data is greater than a vehicle window opening threshold.

The vehicle speed threshold value v₀ is set according to the vehicle speed at which wind vibration noise just begins to appear in the wind tunnel test process, for example, is set to 50km/h, and the vehicle window opening threshold value P₀ is set according to the minimum opening at which wind vibration noise appears in the test process, and is set to 20%. The pleasure degree threshold T₀ is set according to the average value of the uncomfortable situations of the subjective feeling of the user in the subjective evaluation test process, and the rainfall threshold F₀ is set to be a rainfall value corresponding to the 2-gear windscreen wiper speed, namely, the windscreen wiper speed is smaller than 2-gear windscreen wiper speed is met, and then the wind vibration control system can be started.

And judging wind vibration active control conditions by utilizing the vehicle speed, the vehicle window opening, the pleasure degree and the rainfall in the current state, when the vehicle speed is greater than a vehicle speed threshold, the vehicle window opening is greater than a vehicle window opening threshold, the rainfall is less than a rainfall threshold, and the pleasure degree is less than a pleasure degree threshold, automatically starting a wind vibration removing mode, or inquiring whether a passenger needs to start the wind vibration removing mode through a loudspeaker playing voice, and starting the wind vibration removing mode if the user feedback agrees.

In one implementation mode of the embodiment, the method for controlling the window of the vehicle according to the target front window opening data comprises the steps of analyzing target front window and target opening of the target front window opening data, sending an opening command to an opening adjustment executor of the vehicle, wherein the opening command is used for indicating that the target front window is adjusted to the target opening under the condition that the opening position and the opening of the initial rear window opening data are maintained, and controlling the opening adjustment executor to adjust the target front window to the target opening.

Optionally, if the target front window opening data includes multiple sets of data, a target pleasure degree parameter selected by a user may be detected, where the target pleasure degree parameter is used to represent a pleasure degree parameter corresponding to user preference, and one target front window opening data is selected from the multiple sets of target front window opening data based on the target pleasure degree parameter.

The processor outputs the optimal window combination and the opening thereof to a window opening adjusting actuator, and the actuator adjusts the corresponding window to the required opening.

The plurality of sets of target front window opening data in this embodiment include one or more sets of target front window opening data, where each set of target front window opening data corresponds to a target pleasure parameter, and the target pleasure parameter is an influence factor of subjective tolerance of a user to wind vibration noise, for example, sound pressure level, roughness, loudness, sharpness, for example, comprehensive optimization (simultaneously considering sound pressure level, roughness, loudness, sharpness, for example, weight coefficients are a, b, c, d respectively) corresponds to one front window opening data, sound pressure level optimal corresponds to one front window opening data, roughness optimal corresponds to one front window opening data, and so on.

Optionally, the target pleasure degree parameter may be null, for example, factory setting is null, and when the target pleasure degree parameter may be null, the vehicle defaults to output a set of target front window opening data with the optimal combination (that is, the maximum wind vibration noise pleasure degree calculated according to pd=x₁/SPL +x₂/R+x₃/L +x₄/S calculated according to the running speed).

The user can set the target pleasure degree parameter based on the personalized needs and preference conditions, if the user selects the target pleasure degree parameter to be corresponding to the comprehensive optimum, one front window opening data corresponding to the comprehensive optimum (meanwhile, the sound pressure level, the roughness, the loudness and the sharpness are considered), and if the user preference is particularly sensitive to the roughness, the selected target pleasure degree parameter is corresponding to the roughness to be optimal, one front window opening data corresponding to the roughness to be optimal is selected. In addition, a target ventilation quantity parameter and/or a target windowing area parameter selected by a user can be detected, wherein the target ventilation quantity parameter is used for representing a ventilation quantity parameter (such as maximum ventilation quantity and minimum ventilation quantity) corresponding to user preference and a windowing area parameter (such as maximum windowing area and minimum windowing area) corresponding to user preference, and further, one target front window opening data is selected from the plurality of groups of target front window opening data based on the target pleasure degree parameter, so that the optimal ventilation quantity or the optimal windowing area is realized under the condition of ensuring the maximum pleasure degree of wind vibration noise.

Fig. 7 is a spectrum comparison diagram before and after a wind vibration removal mode is started in a vehicle in an embodiment of the invention, as shown in fig. 7, the spectrum diagram of FLL (front driving headrest left ear) before and after the wind vibration active control system is started in a full-open state of a left rear window, wherein a solid line in the diagram represents an original state, and a dotted line represents a state after the wind vibration active control system is started, it can be seen that an obvious low-frequency peak exists before the wind vibration control system is started, and the peak is a main cause of discomfort to human ears. After the wind vibration control system is started, the peak value disappears, subjective feeling is obviously improved, and the control target is achieved.

FIG. 8 is a flow chart of a method for actively controlling wind vibration based on data driving in an embodiment of the present invention, including:

information such as a vehicle speed v, a vehicle window opening p, a windowing position M, a rainfall f and the like is read from a CAN bus in the vehicle. The upper side length a of the window glass, the lower side length b of the window glass, the total height H of the window glass, the length L, the width W and the height H of the whole vehicle and the opening thickness d of the window when the window is opened can be directly obtained according to specific vehicle types;

obtaining pleasure values corresponding to the vehicle model under the combined conditions of different vehicle speeds, different window opening positions, different window opening degrees and different window opening degrees according to the wind vibration noise sound quality evaluation model;

Since the opening wind vibration is strongly related to the vehicle speed and the opening of the vehicle window, a vehicle speed threshold v₀ and an opening threshold P₀ are set according to the vehicle model in the early stage in order to reduce unnecessary vehicle window linkage opening actions. In addition, in a rainy day, water enters the vehicle to affect the driving experience, so that the rainy day needs to detect the rainfall and set a rainfall threshold F₀.

And judging by utilizing the vehicle speed, the vehicle window opening, the pleasure degree and the rainfall, and when the vehicle speed is greater than a vehicle speed threshold v₀, the vehicle window opening is greater than a vehicle window opening threshold P₀, the rainfall is less than a rainfall threshold F₀ and the pleasure degree is less than a pleasure degree threshold T₀, playing a voice by a loudspeaker to inquire whether a passenger needs to start a wind-removing vibration mode.

After the passenger confirms that the wind vibration removal mode needs to be started, based on the established wind vibration active control large model, the controller starts to calculate the window opening position and the opening of the vehicle model with optimal pleasure under the current vehicle speed, the window opening position and the opening of the vehicle window;

The processor outputs the optimal windowing position and the opening thereof to the actuator, and the actuator adjusts the linkage of the corresponding vehicle window to the required opening.

In one implementation of the embodiment, after the window of the vehicle is controlled according to the target front window opening data, the method further comprises collecting, at the target noise reduction position, a wind vibration noise pleasure degree after the window of the vehicle is controlled according to the target front window opening data, judging whether the wind vibration noise pleasure degree is greater than a pleasure degree threshold, updating the wind vibration active control large model according to the running speed, the initial rear window opening data, the target front window opening data and the wind vibration noise pleasure degree if the wind vibration noise pleasure degree is less than or equal to the pleasure degree threshold, outputting target front window opening data with the wind vibration noise pleasure degree greater than the pleasure degree threshold according to the updated target front window opening data, and determining that the wind vibration removal mode is executed if the wind vibration noise pleasure degree is greater than the pleasure degree threshold.

In an example of model optimization, updating the wind vibration active control large model by using the running speed, the initial rear window opening data, the target front window opening data and the wind vibration noise pleasure degree comprises the steps of iteratively executing the steps of acquiring wind vibration noise pleasure degree after finishing window adjustment of a previous adjustment period until the wind vibration noise pleasure degree acquired by the target noise reduction position is larger than a pleasure degree threshold value, configuring the running speed, the initial rear window opening data and the wind vibration noise pleasure degree after the window adjustment of the previous adjustment period as input sample data of the current adjustment period, configuring target front window opening data obtained by the previous adjustment period as output sample data of the current adjustment period, optimizing the wind vibration active control large model by using the input sample data and the output sample data, inputting the running speed and the initial rear window opening data of the current adjustment period into the wind vibration active control large model to acquire the target front window opening data of the current adjustment period, acquiring the wind vibration noise pleasure degree after controlling the window opening data of the vehicle according to the updated target front window of the current adjustment period, and judging whether the wind vibration noise pleasure degree is larger than the threshold value or not.

And judging the wind vibration control effect, if the control target is achieved, ending the wind vibration control flow, if the control target is not achieved, continuing to optimize the wind vibration active control large model until the window opening position and the opening degree of the wind vibration active control large model output after optimization reach the target, namely, the pleasure degree of wind vibration noise collected by the target noise reduction position is greater than a pleasure degree threshold value. FIG. 9 is a flow chart of wind vibration control large model optimization based on feedback control in an embodiment of the invention, the process comprising:

And (3) continuously acquiring the wind vibration noise data in the vehicle after the opening of the vehicle window is adjusted by using the microphone, performing real-time signal processing, calculating the pleasure degree of the wind vibration noise data, judging whether the pleasure degree of the current wind vibration noise is larger than a threshold value, and ending the control program if the pleasure degree of the current wind vibration noise is larger than the threshold value level. If the speed is smaller than the threshold value, taking the opening positions of the front window and the rear window, the opening degree of the vehicle window and the current pleasure degree as new sample input, continuing training the wind vibration active control large model, reusing the optimized wind vibration active control large model to output the current speed, the initial rear window opening position and the corresponding target front window position and opening degree under the opening degree of the vehicle window, and giving the target front window position and the opening degree to an actuator to control the vehicle window and the corresponding opening degree.

By adopting the scheme of the embodiment, the large model is utilized to optimize the windowing position and the opening of the vehicle window, thereby realizing the control of the windowing wind vibration, eliminating the wind vibration phenomenon of the vehicle and improving the comfort level of the user. The wind vibration active control large model can be optimized automatically according to real vehicle wind vibration data, and even if the current output target front window opening data of the active control large model cannot meet the pleasure degree requirement (smaller than a pleasure degree threshold value) in a vehicle, new target front window opening data can be output after the model is optimized. On the other hand, the model parameters can be adaptively updated on the vehicle to obtain the wind vibration active control large model matched with the vehicle type (vehicle size data and window size parameters) without repeated wind vibration noise bottoming test, and development time and cost are saved.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method according to the embodiments of the present invention.

Example 2

The embodiment also provides a wind vibration active control device based on data driving for the vehicle, which is used for realizing the embodiment and the preferred implementation mode, and the description is omitted. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the means described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware, is also possible and contemplated.

Fig. 10 is a block diagram of a data-driven wind vibration active control device for a vehicle according to an embodiment of the present invention, as shown in fig. 10, the device includes:

a first acquisition module 100 for acquiring a running speed of a vehicle, acquiring initial rear window opening data of the vehicle, and configuring a target noise reduction position in the vehicle;

A first calculation module 101 for calculating an initial wind vibration noise pleasure degree of the vehicle interior based on the running vehicle speed;

The second obtaining module 102 is configured to obtain a wind vibration active control large model corresponding to the target noise reduction position;

A second calculation module 103, configured to calculate, according to the running speed, the initial rear window opening data, and the initial wind vibration noise pleasure degree, target front window opening data combined and linked with the initial rear window opening data by using a wind vibration active control large model matched with the target noise reduction position, where the wind vibration active control large model is configured to output front window opening data with an optimal wind vibration noise pleasure degree of the target noise reduction position on the premise of the current running speed and the initial rear window opening data;

And the control module 104 is used for controlling the window of the vehicle according to the target front window opening data.

Optionally, the first calculation module comprises a determination unit for determining a speed interval in which the running speed is located, a generation unit for generating a pleasure degree weight coefficient of a noise sound quality parameter according to the speed interval, a configuration unit for configuring a wind vibration noise pleasure degree model based on the pleasure degree weight coefficient, and a calculation unit for calculating initial wind vibration noise pleasure degree in the vehicle by adopting the wind vibration noise pleasure degree model.

Optionally, the noise quality parameters comprise low-frequency noise parameters and high-frequency noise parameters, the generating unit comprises a judging subunit, a first generating subunit and a second generating subunit, wherein the judging subunit is used for judging whether the maximum value of the speed interval is smaller than a first threshold value, the first generating subunit is used for generating a first pleasure degree weight coefficient if the maximum value of the speed interval is smaller than the first threshold value, the maximum value of the speed interval is larger than or equal to the first threshold value, judging whether the maximum value of the speed interval is smaller than a second threshold value, the second threshold value is larger than the first threshold value, the weight of the low-frequency noise parameters in the first pleasure degree weight coefficient is larger than the weight of the high-frequency noise parameters, the second generating subunit is used for generating a second pleasure degree weight coefficient if the maximum value of the speed interval is larger than or equal to the second threshold value, and the weight of the low-frequency noise parameters in the second pleasure degree weight coefficient is smaller than the high-frequency noise parameters.

It should be noted that each of the above modules may be implemented by software or hardware, and the latter may be implemented by, but not limited to, the above modules all being located in the same processor, or each of the above modules being located in different processors in any combination.

Example 3

An embodiment of the invention also provides a storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.

Alternatively, in the present embodiment, the above-described storage medium may be configured to store a computer program for performing the steps of:

s1, acquiring a running speed of a vehicle, acquiring initial rear window opening data of the vehicle, and configuring a target noise reduction position in the vehicle;

S2, calculating initial wind vibration noise pleasure degree of the interior of the vehicle based on the running speed;

s3, acquiring a wind vibration active control large model corresponding to the target noise reduction position;

s4, calculating target front window opening data combined and linked with the initial rear window opening data by utilizing a wind vibration active control large model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, wherein the wind vibration active control large model is used for outputting front window opening data with optimal wind vibration noise pleasure degree of the target noise reduction position on the premise of the current running speed and the initial rear window opening data;

and S5, controlling the window of the vehicle according to the target front window opening data.

Alternatively, in the present embodiment, the storage medium may include, but is not limited to, a USB flash disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a removable hard disk, a magnetic disk, or an optical disk, etc. various media in which a computer program may be stored.

An embodiment of the invention also provides an electronic device comprising a memory having stored therein a computer program and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.

Optionally, the electronic apparatus may further include a transmission device and an input/output device, where the transmission device is connected to the processor, and the input/output device is connected to the processor.

Alternatively, in the present embodiment, the above-described processor may be configured to execute the following steps by a computer program:

s1, acquiring a running speed of a vehicle, acquiring initial rear window opening data of the vehicle, and configuring a target noise reduction position in the vehicle;

S2, calculating initial wind vibration noise pleasure degree of the interior of the vehicle based on the running speed;

s3, acquiring a wind vibration active control large model corresponding to the target noise reduction position;

s4, calculating target front window opening data combined and linked with the initial rear window opening data by utilizing a wind vibration active control large model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, wherein the wind vibration active control large model is used for outputting front window opening data with optimal wind vibration noise pleasure degree of the target noise reduction position on the premise of the current running speed and the initial rear window opening data;

and S5, controlling the window of the vehicle according to the target front window opening data.

Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments and optional implementations, and this embodiment is not described herein.

The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

From the above description of embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus a general purpose hardware platform, or may be implemented by hardware. Based on such understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the related art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform the method described in the respective embodiments or some parts of the embodiments.

It is to be understood that the terminology used herein is for the purpose of describing particular example embodiments only, and is not intended to be limiting. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise. The terms "comprises," "comprising," "includes," "including," and "having" are inclusive and therefore specify the presence of stated features, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, elements, components, and/or groups thereof. The method steps, processes, and operations described herein are not to be construed as necessarily requiring their performance in the particular order described or illustrated, unless an order of performance is explicitly stated. It should also be appreciated that additional or alternative steps may be used.

The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (21)

1. The wind vibration active control method based on data driving is characterized by comprising the following steps of:

acquiring a running speed of a vehicle, acquiring initial rear window opening data of the vehicle, and configuring a target noise reduction position in the vehicle;

calculating an initial wind-vibration noise pleasure degree of the interior of the vehicle based on the running vehicle speed;

Acquiring a wind vibration active control large model corresponding to the target noise reduction position;

Calculating target front window opening data combined and linked with the initial rear window opening data by utilizing a wind vibration active control large model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, wherein the wind vibration active control large model is used for outputting front window opening data with optimal wind vibration noise pleasure degree of the target noise reduction position on the premise of the current running speed and the initial rear window opening data;

and controlling the window of the vehicle according to the target front window opening data.

2. The method of claim 1, wherein calculating an initial wind-vibration noise pleasure of the vehicle interior based on the travel vehicle speed comprises:

determining a speed interval in which the running vehicle speed is located;

generating a pleasure degree weight coefficient of the noise sound quality parameter according to the speed interval;

Configuring a wind vibration noise pleasure degree model based on the pleasure degree weight coefficient;

and calculating the initial wind vibration noise pleasure degree of the vehicle interior by adopting the wind vibration noise pleasure degree model.

3. The method of claim 2, wherein calculating an initial wind noise pleasure for the vehicle interior using the wind noise pleasure model comprises:

Acquiring noise quality parameters acquired by a microphone of the vehicle at the target noise reduction position;

The initial wind-vibration noise pleasure PD of the vehicle interior is calculated using the following formula:

Pd=x₁/SPL + x₂/R+x₃/L + x₄/S, where SPL represents sound pressure level, R represents roughness, L represents loudness, S represents sharpness, x₁、x₂、x₃、x₄ represents weight coefficients of sound pressure level, roughness, loudness, sharpness, respectively, and SPL, R, L, S is a noise quality parameter.

4. The method of claim 2, wherein the noise figure of merit parameters include a low frequency noise parameter and a high frequency noise parameter, and generating a pleasure weight coefficient for the noise figure of merit from the speed interval comprises:

Judging whether the maximum value of the speed interval is smaller than a first threshold value or not;

If the maximum value of the speed interval is greater than or equal to a first threshold value, judging whether the maximum value of the speed interval is smaller than a second threshold value, wherein the second threshold value is greater than the first threshold value, and the weight of a low-frequency noise parameter in the first pleasure degree weight coefficient is greater than the weight of a high-frequency noise parameter;

And if the maximum value of the speed interval is greater than or equal to the second threshold value, generating a third pleasure degree weight coefficient, wherein the weight of the low-frequency noise parameter in the second pleasure degree weight coefficient is equal to the weight of the high-frequency noise parameter, and the weight of the low-frequency noise parameter in the third pleasure degree weight coefficient is smaller than the weight of the high-frequency noise parameter.

5. The method of claim 1, wherein after controlling the window of the vehicle in accordance with the target front window opening data, the method further comprises:

collecting the pleasure degree of wind vibration noise after the window of the vehicle is controlled according to the target front window opening data at the target noise reduction position;

judging whether the pleasure degree of the wind vibration noise is larger than a pleasure degree threshold value or not;

And if the wind vibration noise pleasure degree is smaller than or equal to a pleasure degree threshold value, updating the wind vibration active control large model by using the running speed, the initial rear window opening data, the target front window opening data and the wind vibration noise pleasure degree, outputting target front window opening data with the wind vibration noise pleasure degree larger than the pleasure degree threshold value by using the updated wind vibration active control large model, and controlling the window of the vehicle according to the updated target front window opening data, and if the wind vibration noise pleasure degree is larger than the pleasure degree threshold value, determining that executing the wind vibration removing mode is completed.

6. The method of claim 5, wherein updating the wind-vibration active control large model with the travel vehicle speed, the initial rear window opening data, the target front window opening data, wind-vibration noise pleasure comprises:

iteratively executing the following steps until the pleasure degree of wind vibration noise acquired by the target noise reduction position is greater than a pleasure degree threshold value:

The method comprises the steps of obtaining wind vibration noise pleasure degree after window adjustment of a previous adjustment period is completed, configuring the running speed of the previous adjustment period, initial rear window opening data and the wind vibration noise pleasure degree after adjustment as input sample data of a current adjustment period, configuring target front window opening data obtained in the previous adjustment period as output sample data of the current adjustment period, optimizing a wind vibration active control large model by adopting the input sample data and the output sample data, inputting the running speed of the current adjustment period and the initial rear window opening data into the wind vibration active control large model to obtain target front window opening data of the current adjustment period, collecting the wind vibration noise pleasure degree after the window of the vehicle is controlled according to the target front window opening data updated in the current adjustment period at the target noise reduction position, and judging whether the wind vibration noise pleasure degree is larger than a pleasure degree threshold.

7. The method according to claim 1, characterized in that after calculating an initial wind-vibration noise pleasure of the vehicle interior based on the running vehicle speed, the method further comprises:

Judging whether the pleasure degree of the initial wind vibration noise is smaller than a pleasure degree threshold value or not;

and if the pleasure degree of the initial wind vibration noise is smaller than a pleasure degree threshold value, calling a wind vibration active control large model matched with the target noise reduction position to calculate target front window opening data combined and linked with the initial rear window opening data.

8. The method of claim 1, wherein obtaining a wind vibration active control large model corresponding to the target noise reduction location comprises:

collecting multiple groups of wind vibration sample data of a sample vehicle, wherein each group of wind vibration sample data comprises a running speed and windowing combination data, and wind vibration noise pleasure degree is collected at the target noise reduction position, and the windowing combination data comprises rear window opening data and front window opening data;

acquiring whole vehicle size data of the sample vehicle;

And training an initial large model by taking the whole vehicle size data as a fixed parameter, taking the running speed and the windowing combination data in the wind vibration sample data as independent variables and the wind vibration noise pleasure degree as dependent variables to obtain a wind vibration active control large model corresponding to the target noise reduction position.

9. The method of claim 8, wherein obtaining the sample vehicle full vehicle size data comprises:

Acquiring the model information of the sample vehicle;

And searching vehicle size data and window size parameters matched with the vehicle type information in a preset database, wherein the vehicle size data comprise the length, width and height of the vehicle, and the window size parameters comprise the upper side length of the window glass, the lower side length of the window glass, the total height of the window glass and the thickness of the window frame.

10. The method of claim 1, wherein obtaining initial rear window opening data of the vehicle comprises:

acquiring an initial rear window opening position of a vehicle, wherein the initial rear window opening position comprises at least one of a left rear window and a right rear window;

And acquiring initial windowing opening of the vehicle, wherein the initial rear window opening data comprise the initial rear window windowing position and the initial windowing opening.

11. The method of claim 1, wherein calculating target front window opening data combined and linked with the initial rear window opening data using the wind vibration active control large model matched with the target noise reduction position according to the running vehicle speed, the initial rear window opening data and the initial wind vibration noise pleasure degree comprises:

A wind vibration active control large model matched with the target noise reduction position is called;

And inputting the driving speed, the initial wind vibration noise pleasure degree and the initial rear window opening data into the wind vibration active control large model, and outputting to obtain first target front window opening data of the vehicle, wherein the first target front window opening data comprises a first target front window position and a window opening degree thereof, and the wind vibration noise pleasure degree is used for representing the subjective bearing degree of passengers in the vehicle on wind vibration noise.

12. The method according to claim 1, wherein the method further comprises:

A windowing combination table matched with the whole vehicle size data is called, wherein each table item of the windowing combination table comprises a running speed, windowing combination data and corresponding wind vibration noise pleasure, and the windowing combination data comprises opening data of a plurality of windowing positions;

And searching second target front window opening data with highest wind vibration noise pleasure degree in the window opening combination data of the window opening combination table by taking the running speed and the initial rear window opening data as keywords, wherein the second target front window opening data comprises a second target front window position and the window opening degree thereof, and the wind vibration noise pleasure degree is used for representing the subjective bearing degree of passengers in the vehicle on wind vibration noise.

13. The method of claim 1, wherein prior to calculating target front window opening data in combined linkage with the initial rear window opening data using the wind vibration active control large model matched to the target noise reduction position based on the travel vehicle speed, the initial rear window opening data, the initial wind vibration noise pleasure, the method further comprises:

judging whether the vehicle starts a wind vibration removal mode or not;

and if the vehicle is in a wind vibration removal mode, determining the noise pleasure degree of the initial wind vibration according to the running speed, the initial rear window opening data and the initial wind vibration, and calculating the target front window opening data combined and linked with the initial rear window opening data by using a wind vibration active control large model matched with the target noise reduction position.

14. The method of claim 13, wherein determining whether the vehicle is turning on a wind-vibration removal mode comprises:

acquiring rainfall data of a rainfall sensor of the vehicle, wherein the wind vibration noise pleasure degree is used for representing subjective bearing degree of passengers in the vehicle on wind vibration noise;

And judging whether the vehicle starts a wind vibration removal mode or not based on the rainfall data, the initial wind vibration noise pleasure degree, the running vehicle speed and the initial rear window opening data.

15. The method of claim 14, wherein determining whether the vehicle is to turn on a wind-out mode based on the rainfall data, the initial wind-noise pleasure, the travel speed, the initial rear window opening data comprises:

Judging whether the rainfall data is smaller than a rainfall threshold, judging whether the pleasure degree of the initial wind vibration noise is smaller than a pleasure degree threshold, judging whether the running vehicle speed is larger than a vehicle speed threshold, and judging whether the initial rear window opening data is larger than a vehicle window opening threshold;

and if the rainfall data is smaller than a rainfall threshold, the initial wind vibration noise pleasure degree is smaller than a pleasure degree threshold, the running speed is larger than a vehicle speed threshold, and the initial rear window opening data is larger than a vehicle window opening threshold, determining that the vehicle starts a wind vibration removing mode.

16. The method of claim 1, wherein controlling the window of the vehicle in accordance with the target front window opening data comprises:

analyzing the target front window and the target opening degree in the target front window opening data;

transmitting a windowing instruction to an opening adjustment executor of the vehicle, wherein the windowing instruction is used for instructing that the target front window is adjusted to the target windowing opening in a combined way under the condition of maintaining the windowing position and the opening of the initial rear window opening data;

and controlling the opening degree adjusting actuator to adjust the target front window to the target opening degree.

17. The utility model provides a wind vibration initiative control device based on data drive which characterized in that includes:

The first acquisition module is used for acquiring the running speed of the vehicle, acquiring the initial rear window opening data of the vehicle and configuring the target noise reduction position in the vehicle;

A first calculation module for calculating an initial wind vibration noise pleasure level of the vehicle interior based on the running vehicle speed;

The second acquisition module is used for acquiring a wind vibration active control large model corresponding to the target noise reduction position;

The second calculation module is used for calculating target front window opening data combined and linked with the initial rear window opening data by utilizing a wind vibration active control large model matched with the target noise reduction position according to the running speed, the initial rear window opening data and the initial wind vibration noise pleasure degree, wherein the wind vibration active control large model is used for outputting front window opening data with optimal wind vibration noise pleasure degree of the target noise reduction position on the premise of the current running speed and the initial rear window opening data;

and the control module is used for controlling the window of the vehicle according to the target front window opening data.

18. The apparatus of claim 17, wherein the first computing module comprises:

a determining unit, configured to determine a speed interval in which the running vehicle speed is located;

the generating unit is used for generating a pleasure degree weight coefficient of the noise sound quality parameter according to the speed interval;

A configuration unit, configured to configure a wind vibration noise pleasure degree model based on the pleasure degree weight coefficient;

And the calculating unit is used for calculating the initial wind vibration noise pleasure degree of the vehicle interior by adopting the wind vibration noise pleasure degree model.

19. The apparatus of claim 18, wherein the noise quality parameters include a low frequency noise parameter and a high frequency noise parameter, the generating unit comprising:

A judging subunit, configured to judge whether a maximum value of the speed interval is smaller than a first threshold;

If the maximum value of the speed interval is greater than or equal to a first threshold value, judging whether the maximum value of the speed interval is less than a second threshold value, wherein the second threshold value is greater than the first threshold value, and the weight of a low-frequency noise parameter in the first pleasure degree weight coefficient is greater than the weight of a high-frequency noise parameter;

And the second generation subunit is used for generating a second pleasure degree weight coefficient if the maximum value of the speed interval is smaller than a second threshold value, and generating a third pleasure degree weight coefficient if the maximum value of the speed interval is larger than or equal to the second threshold value, wherein the weight of the low-frequency noise parameter in the second pleasure degree weight coefficient is equal to the weight of the high-frequency noise parameter, and the weight of the low-frequency noise parameter in the third pleasure degree weight coefficient is smaller than the weight of the high-frequency noise parameter.

20. A storage medium having a computer program stored therein, wherein the computer program is arranged to perform the method of any of claims 1 to 16 when run.

21. An electronic device comprising a memory and a processor, characterized in that the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the method of any of the claims 1 to 16.