Movatterモバイル変換


[0]ホーム

URL:


CN117314391B - Operation and maintenance job management method and device, electronic equipment and storage medium - Google Patents

Operation and maintenance job management method and device, electronic equipment and storage medium
Download PDF

Info

Publication number
CN117314391B
CN117314391BCN202311274823.9ACN202311274823ACN117314391BCN 117314391 BCN117314391 BCN 117314391BCN 202311274823 ACN202311274823 ACN 202311274823ACN 117314391 BCN117314391 BCN 117314391B
Authority
CN
China
Prior art keywords
vibration
maintenance
dynamic information
similarity
characteristic
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202311274823.9A
Other languages
Chinese (zh)
Other versions
CN117314391A (en
Inventor
严可达
朱新桥
李震
向雪梅
单小芳
李泽胜
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Optical Valley Technology Co ltd
Original Assignee
Optical Valley Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Optical Valley Technology Co ltdfiledCriticalOptical Valley Technology Co ltd
Priority to CN202311274823.9ApriorityCriticalpatent/CN117314391B/en
Publication of CN117314391ApublicationCriticalpatent/CN117314391A/en
Application grantedgrantedCritical
Publication of CN117314391BpublicationCriticalpatent/CN117314391B/en
Activelegal-statusCriticalCurrent
Anticipated expirationlegal-statusCritical

Links

Classifications

Landscapes

Abstract

The application discloses an operation and maintenance job management method, an operation and maintenance job management device, electronic equipment and a storage medium, wherein the operation and maintenance job management method comprises the following steps: acquiring first dynamic information of an operation and maintenance working device and/or second dynamic information of operation and maintenance personnel; acquiring an operation vibration signal acquired by a vibration monitoring device arranged on a road side based on the first dynamic information and/or the second dynamic information; and inputting the operation vibration signal into a twin feature extraction network to obtain operation vibration features, and inputting the operation vibration features into a pre-trained operation and maintenance operation recognition model to obtain an operation and maintenance operation state recognition result. Through multiplexing underground facilities in the wisdom garden and monitoring, perhaps for perimeter security protection, arranged many passive vibration monitoring devices underground, utilize its collection scope wide, the signal dimension is low, the high characteristics of sensitivity, can be as far as possible and with the operation process of minimum resource consumption's monitoring fortune dimension operation, can more timely accurate carry out real-time supervision to fortune dimension operation state.

Description

Operation and maintenance job management method and device, electronic equipment and storage medium
Technical Field
The present application relates to the field of operation and maintenance management technologies, and in particular, to an operation and maintenance job management method, an apparatus, an electronic device, and a storage medium.
Background
The intelligent park is characterized in that various information technologies or innovative concepts are utilized to open and integrate systems and services of the park, so that the operation and maintenance efficiency of the park is improved, the management and services of the park are optimized, and the living quality of citizens is improved. In the operation and maintenance of a campus, for example, the operation process and the operation quality of the operation and maintenance operation of the campus, such as sanitation operation, road maintenance, greening maintenance, public facility maintenance, etc., are very important for the operation, management, etc. of the campus.
In the related art, the monitoring of the operation and maintenance operation process and the operation quality of a park is often performed during the operation process or after the operation is completed, the manual inspection is performed by a spot check or a special person, the efficiency is low, and different persons can have larger errors due to the fact that the operation process and the operation quality are judged in a subjective manner, so that the operation and maintenance operation cannot be performed efficiently and with high quality.
In some related technologies, video acquisition may be performed on the operation process, and the operation and maintenance operation is monitored through video analysis, however, the adopted camera is often a road side camera, which is limited by factors such as distance, weather conditions, light, and the like, and erroneous judgment may exist. If the camera is configured for each operation, the workload is increased, the operation cost is increased sharply, the configured camera cannot track and shoot in the whole process, the operation process is omitted, and the whole process is difficult to finish accurate monitoring.
Therefore, how to simply and accurately monitor the operation process and quality of the operation and maintenance operation is a technical problem to be solved.
Disclosure of Invention
The application provides an operation and maintenance operation management method, an operation and maintenance operation management device, electronic equipment and a storage medium, which at least solve the technical problem of how to simply and accurately monitor the operation process and quality of operation and maintenance operation in the related technology.
According to a first aspect of the present application, there is provided an operation and maintenance job management method, including: acquiring first dynamic information of an operation and maintenance working device and/or second dynamic information of operation and maintenance personnel; acquiring an operation vibration signal acquired by a vibration monitoring device arranged on a road side based on the first dynamic information and/or the second dynamic information; inputting the operation vibration signal into a twin feature extraction network to obtain operation vibration features, wherein the twin feature extraction network is obtained by training a sample pair constructed based on the operation vibration signal and the non-operation vibration signal; the operation vibration characteristics are input into a pre-trained operation and maintenance operation identification model to obtain an operation and maintenance operation state identification result, wherein the operation and maintenance operation identification model is obtained by training based on operation vibration characteristic samples and corresponding operation state labels, and the operation vibration characteristic samples comprise first operation vibration signals of an operation and maintenance operation device in different operation states and second operation vibration signals of an operation and maintenance worker in the operation states.
Optionally, the twin feature extraction network comprises: two neural network branches with shared weights, wherein each neural network branch comprises a convolution layer and an attention module; inputting the operation vibration signal into a twin feature extraction network, wherein obtaining the operation vibration feature comprises: extracting a plurality of dimensional vibration characteristic information through the convolution layer; and inputting the multidimensional vibration characteristic information into the attention module to perform self-adaptive weighted fusion on the multidimensional vibration characteristic information so as to obtain the operation vibration characteristic and the non-operation vibration characteristic.
Optionally, the twin feature extraction network further comprises: a first feature clustering module after the attention module is set; extracting operational vibration features in a first neural network branch and extracting non-operational vibration features in a second neural network; and inputting the operation vibration characteristics and the non-operation vibration characteristics into the first characteristic clustering module, and clustering the operation vibration characteristics and the non-operation vibration characteristics respectively so that the operation and maintenance operation identification model classifies the operation vibration characteristics and the non-operation vibration characteristics.
Optionally, the clustering the operational vibration feature and the non-operational vibration feature respectively includes: respectively calculating a first similarity between the operation vibration characteristics, a second similarity between the non-operation vibration characteristics and a third similarity between the operation vibration characteristics and the non-operation vibration characteristics; reducing a distance between operational vibration characteristics based on the first similarity and the third similarity to increase the first similarity while reducing the third similarity; reducing a distance between the non-operational vibration features based on the second similarity and the third similarity to increase the second similarity while reducing the third similarity; and increasing a distance between the operational vibration feature and the non-operational vibration feature based on the third similarity to reduce the third similarity.
Optionally, the first dynamic information includes first positioning information of the operation and maintenance working device and/or the second dynamic information includes second positioning information of the operation and maintenance personnel; the acquiring the operation vibration signal acquired by the vibration monitoring device arranged at the road side based on the first dynamic information and/or the second dynamic information comprises the following steps: acquiring the first positioning information and/or the second positioning information as vibration signals of a vibration source position; and carrying out local analysis and filtering on the vibration signal in the time domain and the frequency domain to obtain the operation vibration signal.
Optionally, the first dynamic information further includes a real-time running state of the operation and maintenance working device; the local analysis filtering of the time domain and the frequency domain is carried out on the vibration signal, and the operation vibration signal is obtained by the steps of: determining an operation vibration characteristic of the operation and maintenance working device based on the real-time operation state; and filtering the operation vibration signal based on the operation vibration characteristics to obtain an actual operation vibration signal of the operation and maintenance operation device.
Optionally, the operation and maintenance job management method further includes: acquiring image information acquired by an image acquisition device arranged on a road side based on the first dynamic information and/or the second dynamic information; extracting the operation image change characteristics of the operation and maintenance operation device and/or operation and maintenance operation personnel; performing time sequence and dimension alignment on the operation vibration characteristic and the operation image change characteristic; clustering and fusing the aligned operation vibration characteristics and the operation image change characteristics through a second characteristic clustering module to obtain fused operation characteristics; and updating parameters of the operation and maintenance operation identification model based on the fusion operation characteristics.
According to a second aspect of the present application, an embodiment of the present application provides an operation and maintenance job management apparatus, including: the first acquisition module is used for acquiring first dynamic information of the operation and maintenance working device and/or second dynamic information of operation and maintenance personnel; the second acquisition module is used for acquiring operation vibration signals acquired by the vibration monitoring device arranged on the road side based on the first dynamic information and/or the second dynamic information; the feature extraction module is used for inputting the operation vibration signal into a twin feature extraction network to obtain operation vibration features, and the twin feature extraction network is obtained by training a sample pair constructed based on the operation vibration signal and the non-operation vibration signal; the operation vibration characteristic sample comprises first operation vibration signals in different operation states of the operation and maintenance device and second operation vibration signals in the operation states of the operation and maintenance workers.
According to a third aspect of the present application, an embodiment of the present application further provides an electronic device, including: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the operation and maintenance job management method of any one of the first aspects above.
According to a third aspect of the present application, there is also provided a computer-readable storage medium storing a computer program which, when executed by a processor, implements the operation and maintenance job management method according to any one of the first aspects.
According to the operation and maintenance operation management method, the vibration signals collected by the road side vibration monitoring device are screened through the state information of the operation and maintenance operation device and the operation and maintenance operation personnel, operation vibration signals are obtained, operation vibration characteristics are extracted more pertinently through the twin characteristic extraction network, and the real-time operation state of operation and maintenance operation is identified through the operation and maintenance operation identification model identification vibration characteristics. Through multiplexing underground facilities in wisdom garden and monitoring, or for perimeter security protection, arranged many passive sensing equipment underground, for example, vibration cable, vibration optic fibre, signal such as ground wave sensor, utilize its collection scope wide, the signal dimension is low, the sensitivity is high characteristics, can monitor the operation process of operation and maintenance operation in the maximum scope, on the one hand can reduce the problem that the resource cost is big because of image data occupies memory and calculation power is too big, on the other hand, through multiplexing the real-time characteristic of the low dimension data of other sensors, can more timely accurate carry out real-time supervision to operation and maintenance operation state under the condition that need not too high calculation power and too much storage resource.
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 schematic flow chart of an operation and maintenance operation management method provided by the application;
FIG. 2 is a schematic diagram of an operation management device according to the present application;
fig. 3 is a schematic diagram of an electronic device provided by the present application.
Detailed Description
In order to more clearly illustrate the general inventive concept, a detailed description is given below by way of example with reference to the accompanying drawings.
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 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, system, 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.
Referring to the description of the background technology, the existing monitoring on the operation and maintenance operation of the park is usually performed by manual or road side cameras for visual monitoring, however, the real-time accurate monitoring is difficult to achieve whether the monitoring is manual or video monitoring.
The applicant has found that the construction of the smart park at present often requires a multi-source data acquisition end, various sensors, visual acquisition devices and the like, and a plurality of passive sensing devices such as vibration cables, vibration optical fibers, ground wave sensors and the like are arranged underground in underground pipelines or underground at the roadside of the park for monitoring underground facilities or for perimeter security. Many environmental monitoring sensors, such as noise detection sensors, etc., are disposed on the ground. The applicant has found that these sensors, when collecting data, tend to collect all the data within the detection range in which they are located, and therefore the operation and maintenance operations can be monitored using the data collected by the sensors arranged in combination with vision. Based on this, the present application proposes an operation and maintenance job management method, referring to fig. 1, the method may include the following steps:
S10, acquiring first dynamic information of the operation and maintenance working device and/or second dynamic information of operation and maintenance personnel. As an exemplary embodiment, the campus operation and maintenance operation may include sanitation operation, road maintenance, greening maintenance, public facility maintenance, etc., where the operation and maintenance operation is performed, an operation and maintenance operation device and an operation and maintenance operator are often required to participate simultaneously, and sometimes only the operation and maintenance operator may participate. In an embodiment, the first dynamic information and the second dynamic information may include first positioning information of the operation and maintenance operation device and second positioning information of an operation and maintenance operator, and the first dynamic information may further include operation state information of the operation and maintenance operation device, and may include, for example, operation state information of a vehicle, for example, a running state or a stopping state of the vehicle, and may also include operation state information of the vehicle, for example, whether the sweeper is in a cleaning state, whether the sprinkler is in a sprinkling state, and the like. The system can also comprise the operation state of maintenance equipment, the use information of the operation and maintenance facilities by staff, and the like.
S20, acquiring operation vibration signals acquired by vibration monitoring devices arranged on the road side based on the first dynamic information and/or the second dynamic information. As an exemplary embodiment, the first dynamic information includes first positioning information of the operation and maintenance working device and/or the second dynamic information includes second positioning information of the operation and maintenance person.
The first positioning information and/or the second positioning information can be firstly obtained as vibration signals of the vibration source position for the acquisition of the operation vibration signals; and carrying out local analysis and filtering on the vibration signal in the time domain and the frequency domain to obtain the operation vibration signal.
As an exemplary embodiment, the position of the target vibration source may be determined through the first positioning information and/or the second positioning information, the preliminary screening may be performed on all vibration signals through the relative positional relationship and distance between the target vibration source and the vibration monitoring device, and in addition, the type of the operation and maintenance device may be acquired, the fixed vibration characteristics may be determined, and the vibration signals may be further screened through the fixed vibration characteristics.
As an exemplary embodiment, the vibration signal may be subjected to local analysis filtering in the time domain and the frequency domain, the original signal may be subjected to wavelet transformation to obtain wavelet coefficients in a plurality of scales through wavelet analysis, and then the time domain and the frequency domain information may be simultaneously extracted through the change of the wavelet coefficients.
However, characteristics of signals generated by different objects tend to be different, and vibration signals generated by an operation and maintenance device and an operation and maintenance operator tend to be relatively fixed, so that before wavelet analysis, input signals of the wavelet analysis can be screened again based on machine learning. To train a vibration signal classification model. After training is completed, the initially screened signal set is input into a trained vibration signal classification model, and the signals in the signal set can be classified according to the states of the operation and maintenance device and the operation and maintenance worker. And carrying out wavelet analysis on the classification result, and carrying out local analysis and filtering on the vibration signal in the time domain and the frequency domain to obtain the operation vibration signal.
In order to avoid interference of the operation vibration signal on the operation vibration signal, in this embodiment, the real-time operation state of the operation and maintenance device may be collected by an operation state collection device installed on the vehicle or the device, for example, an ECU of the vehicle or the device may collect the operation state of the engine and the running state of the vehicle in real time as the real-time operation state, and determine the operation vibration characteristics of the operation vibration signal corresponding to the real-time operation state by the real-time operation state, and filter the operation vibration signal based on the operation vibration characteristics to obtain the actual operation vibration signal of the operation and maintenance device.
S30, inputting the operation vibration signals into a twin feature extraction network to obtain operation vibration features, wherein the twin feature extraction network is trained based on sample pairs constructed by the operation vibration signals and the non-operation vibration signals. As an exemplary embodiment, the operation vibration signal may be, for example, a vibration signal generated when an operator performs a cleaning operation on a street, the characteristics of the operation vibration signal may include a time-series market of cleaning, a force of cleaning, a speed of cleaning (i.e., an interval between cleaning operation vibration signals), etc., and the operation vibration signal for the operation and maintenance operation device may include, for example, a speed of cleaning by a cleaning vehicle, a force of cleaning, a range vibration characteristic of cleaning. The operation efficiency (interval duration, etc. of the corresponding operation vibration signal) of the road surface restoration device may also be included. The operation vibration characteristics can represent the current working state of the operation device and/or the operation personnel, the working duration, the working efficiency, the working mode and other working characteristics.
As an exemplary embodiment, the training for the twin feature extraction network may construct a feature sample pair by different types of vibration signals, for example, a working vibration signal and a non-working vibration signal may be constructed as a feature sample pair, wherein the non-working vibration signal may be a disturbance vibration signal generated during working, for example, a vibration signal generated when a vehicle is driven on a road surface, a vibration signal generated when a pedestrian passes, or the like.
The samples in the sample pair are different types of vibration signals, namely, the operation vibration signal and the non-operation vibration signal are constructed into a pair of sample pairs, one network branch extracts the operation vibration characteristics of the operation vibration signal, one network branch extracts the non-operation vibration characteristics of the non-operation vibration signal, and the parameters of the two network branches are shared, namely, the parameters of the two network branches are the same.
In this embodiment, the twin feature extraction network includes: two neural network branches of weight sharing, wherein each neural network branch includes a convolutional layer and an attention module.
Inputting the operation vibration signal into a twin feature extraction network, wherein obtaining the operation vibration feature comprises:
and extracting a plurality of dimensional vibration characteristic information through the convolution layer. In the present embodiment, the initial vibration characteristic information of the multiple dimensions, such as the frequency, amplitude, phase, and the like, of the vibration characteristics is extracted by the convolution layer. The multi-dimensional initial vibration characteristics have a coupling relation, and the different specific characteristics corresponding to different operation types are different, and the specific characteristics corresponding to the operation and maintenance operation device and the operation and maintenance operation personnel are also different.
Therefore, in this embodiment, the multi-dimensional vibration feature information is input to the attention module to perform adaptive weighted fusion on the multi-dimensional vibration feature information, so as to obtain the operation vibration feature and the non-operation vibration feature.
Specifically, the attention weight of the operation and maintenance operation device and the operation and maintenance operation personnel on the characteristics can be learned based on the attention module, so that the operation types, the operation and maintenance operation device and the operation and maintenance operation personnel can be more accurately identified and distinguished through the attention deviation on some specific characteristics. Therefore, the attention module is used for carrying out weighted fusion on the characteristics of multiple dimensions, and the fused vibration characteristics of the vibration characteristic information are obtained. After the fusion vibration characteristics of the operation vibration signals and the fusion vibration characteristics of the non-operation vibration signals are obtained respectively, the operation and maintenance operation identification model can be trained by taking the fusion vibration characteristics of the non-operation vibration signals as the fusion vibration characteristic negative sample of the operation vibration signals, so that the problem of inaccurate identification caused by introducing the non-operation vibration characteristics of the unfiltered non-operation vibration signals during identification is avoided.
In the present embodiment, the job vibration signal may also be learned by the countermeasure network, or the job vibration feature may be separated from the non-job vibration feature by the clustering network.
S40, inputting the operation vibration characteristics into a pre-trained operation and maintenance operation recognition model to obtain an operation and maintenance operation first state recognition result, wherein the operation and maintenance operation recognition model is obtained by training based on operation vibration characteristic samples and corresponding operation state labels, and the operation vibration characteristic samples comprise first operation vibration signals of an operation and maintenance operation device in different operation states and second operation vibration signals of an operation and maintenance worker in the operation states. As an exemplary embodiment, the operation and maintenance operation recognition model may be trained through marked operation vibration feature samples, and non-operation vibration feature samples may be introduced as negative samples to train the operation and maintenance operation recognition model together, so that the recognition model has a capability of distinguishing non-operation vibration features.
In this embodiment, vibration signals collected by the road side vibration monitoring device are screened through state information of the operation and maintenance operation device and operation and maintenance operation personnel to obtain operation vibration signals, operation vibration characteristics are extracted more specifically through the twin characteristic extraction network, and real-time operation states of operation and maintenance operations are identified through the operation and maintenance operation identification model to identify the vibration characteristics. Through multiplexing underground facilities in wisdom garden and monitoring, or for perimeter security protection, arranged many passive sensing equipment underground, for example, vibration cable, vibration optic fibre, signal such as ground wave sensor, utilize its collection scope wide, the signal dimension is low, the sensitivity is high characteristics, can monitor the operation process of operation and maintenance operation in the maximum scope, on the one hand can reduce the problem that the resource cost is big because of image data occupies memory and calculation power is too big, on the other hand, through multiplexing the real-time characteristic of the low dimension data of other sensors, can more timely accurate carry out real-time supervision to operation and maintenance operation state under the condition that need not too high calculation power and too much storage resource.
As an alternative embodiment, the distinction between the operational vibration feature and the non-operational vibration feature may be learned by clustering, and in this embodiment, the twin feature extraction network further includes: and setting a first feature clustering module after the attention module.
Extracting operational vibration features in a first neural network branch and extracting non-operational vibration features in a second neural network; and inputting the operation vibration characteristics and the non-operation vibration characteristics into the first characteristic clustering module, and clustering the operation vibration characteristics and the non-operation vibration characteristics respectively so that the operation and maintenance operation identification model classifies the operation vibration characteristics and the non-operation vibration characteristics.
As an exemplary embodiment, in the present embodiment, one or more cluster centers are determined based on semantic units of the plurality of vibration feature entities. The clusters (or clusters) generated by the clustering operation are a collection of data objects that are similar to each other and different from the objects in the other clusters. The cluster center is the most important one of the objects in the cluster, which is the most representative of the cluster and the most interpretable for the other objects in the cluster. In some embodiments, a cluster has only one cluster center. In some embodiments, the cluster center may be one or more vibration feature entities selected from a plurality of vibration feature entities, each cluster center serving as a reference object in calculating a similarity between the cluster center and other vibration feature entities of the plurality of vibration feature entities. After the cluster center is obtained, the vibration characteristic of the cluster center is taken as the specific vibration characteristic corresponding to the embodiment.
The first clustering module is used for respectively obtaining the operation vibration characteristics and the non-operation vibration characteristics so that the operation vibration characteristics are more similar, the operation vibration characteristics and the non-operation vibration characteristics are more dissimilar, and the identification model is easier to distinguish the operation vibration characteristics and the non-operation vibration characteristics.
In one embodiment, the clustering of the operational vibration features and the non-operational vibration features, respectively, includes: respectively calculating a first similarity between the operation vibration characteristics, a second similarity between the non-operation vibration characteristics and a third similarity between the operation vibration characteristics and the non-operation vibration characteristics; reducing a distance between operational vibration characteristics based on the first similarity and the third similarity to increase the first similarity while reducing the third similarity; reducing a distance between the non-operational vibration features based on the second similarity and the third similarity to increase the second similarity while reducing the third similarity; and increasing a distance between the operational vibration feature and the non-operational vibration feature based on the third similarity to reduce the third similarity. And by calculating the similarity and adjusting the similarity, the distance between the working vibration characteristics and the non-working vibration characteristics is increased, so that the distances between the characteristics of the same type are smaller, and the distances between the characteristics of different types are larger.
As an exemplary embodiment, when some image acquisition devices exist and the image acquisition devices can acquire a clearer image, the operation and maintenance operation process can be acquired in real time based on the image acquisition devices, the operation and maintenance operation process acquired in real time is compared with the operation and maintenance operation state identification result obtained through the operation and maintenance operation identification model, and parameters of the operation and maintenance operation identification model are corrected and updated by utilizing the comparison result, so that the operation and maintenance operation identification model is more accurate.
Specifically, extracting the operation image change characteristics of the operation and maintenance operation device and/or operation and maintenance operation personnel; performing time sequence and dimension alignment on the operation vibration characteristic and the operation image change characteristic; clustering and fusing the aligned operation vibration characteristics and the operation image change characteristics through a second characteristic clustering module to obtain fused operation characteristics; and updating parameters of the operation and maintenance operation identification model based on the fusion operation characteristics.
As an exemplary embodiment, job image information is input to an image feature extraction network to extract job image features, in this embodiment, job image features may be extracted through a convolution network and a time sequence feature extraction network that are cascaded, static features of a job image may be extracted through the convolution network, after the static features are obtained, a static feature sequence is constructed, and time sequence associated features in the static feature sequence, that is, dynamic features of the job image, may be extracted through the time sequence feature extraction network.
As an alternative embodiment, in the static feature extraction network, two-dimensional operations may be performed on the image information first. Firstly, inputting data into a convolution layer, and activating by using a LeakyReLU activation layer after batch standardization operation; then input into the convolution layer with the convolution kernel size of 1×1, and use LeakyReLU activation layer to make secondary activation, finally use the largest pooling layer with the pooling size of 2×2 to pool the feature map.
After the static features are obtained, time-series related features in the static feature sequence are extracted through a cyclic neural network (e.g., LSTM, GRU) to characterize the image change features, such as continuous dynamic features of the work action, and time-series cumulative features of the work action.
Since the operation vibration feature and the image change feature of the operation have a temporal relationship, the features can be aligned based on the temporal relationship, and the image change feature and the operation vibration feature are also different in feature dimension, in this embodiment, the two features are also required to be aligned in dimension, and in this embodiment, the operation vibration feature can be decomposed, for example, a wavelet packet decomposition tree is used to extract a plurality of frequency band signal components, the energy of the signals of each frequency band is obtained, the feature vector is constructed by using the energy of the signal of each frequency band as an element, and the vector is normalized to obtain the histogram of the feature vector. And similarly, binarizing the image change features, and obtaining a feature vector histogram of the image change features to further realize dimension alignment.
Inputting the aligned public facility signal characteristics and the image change characteristics into a second characteristic clustering module for clustering, and in the embodiment, fusing through an attention mechanism to obtain fused operation characteristics; splitting the aligned operation vibration characteristic and the image change characteristic into an operation vibration characteristic sequence and an image change characteristic sequence according to time sequence to obtain a characteristic pair sequence; calculating the feature similarity of each feature pair; determining an attention weight of each feature pair based on the feature similarity, the fusion weight being positively correlated with the feature similarity; and carrying out weighted fusion on the feature pairs based on the attention weight to obtain the fusion feature.
In this embodiment, the fusion weight may be based on the similarity of each feature pair in the aligned vibration feature sequence and image change feature sequence, and determine the attention weight of each feature pair based on the feature similarity, that is, the feature pair with high similarity has a larger corresponding feature weight.
As another alternative embodiment, the second feature clustering module may employ a depth cross attention module that may include a correlation layer, an attention layer, and a depth correlation layer. Forming a attention pattern according to the correlation of the operation vibration characteristic and the image change characteristic in time sequence, so as to emphasize the characteristic with strong correlation in the two data; wherein the correlation layer is dedicated to integrating the two features rather than simply making a series, summation. Obtaining a correlation matrix based on the time sequence correlation of the operation vibration characteristics and the image change characteristics in a correlation layer, and calculating a cross attention matrix corresponding to the correlation matrix in an attention layer; learning the cross attention matrix by using an attention layer to obtain a non-exclusive relation between the operation vibration characteristic and the image change characteristic; fusing the non-mutually exclusive relations by using an attention mechanism to obtain a relation characteristic diagram of the vibration characteristic of the operation and the image change characteristic; and in the depth correlation layer, performing feature fusion on the operation vibration feature and the image change feature by applying a mutual convolution operation on the relation feature map.
The fusion operation feature can more accurately represent the current operation state, so that parameters of the operation and maintenance operation identification model can be updated through the fusion operation feature, and the operation and maintenance operation identification model can deeply fuse the actual image feature and the actual vibration feature and then can identify the operation vibration feature according to the identification capability.
The embodiment of the application also provides an operation and maintenance operation management device, as shown in fig. 2, comprising:
a first obtaining module 21, configured to obtain first dynamic information of the operation and maintenance working device and/or second dynamic information of an operation and maintenance person;
a second acquisition module 22, configured to acquire an operation vibration signal acquired by a vibration monitoring device disposed on a road side based on the first dynamic information and/or the second dynamic information;
The feature extraction module 23 is configured to input the operation vibration signal into a twin feature extraction network to obtain an operation vibration feature, where the twin feature extraction network is obtained by training based on a sample pair constructed by the operation vibration signal and the non-operation vibration signal;
The identifying module 24 is configured to input the operation vibration characteristic into a pre-trained operation and maintenance operation identifying model to obtain a first state identifying result of the operation and maintenance operation, where the operation and maintenance operation identifying model is obtained by training based on an operation vibration characteristic sample and a corresponding operation state label, and the operation vibration characteristic sample includes first operation vibration signals in different operation states of the operation and maintenance operation device, and second operation vibration signals in an operation state of an operation and maintenance worker.
It should be noted that, the first obtaining module 21 in this embodiment may be used to perform the above-mentioned step S10, the second obtaining module 22 in this embodiment may be used to perform the above-mentioned step S20, the feature extraction module 23 in this embodiment may be used to perform the above-mentioned step S30, and the identification module 24 in this embodiment may be used to perform the above-mentioned step S40.
It should be noted that the above modules are the same as examples and application scenarios implemented by the corresponding steps, but are not limited to what is disclosed in the above embodiments. It should be noted that the above modules may be implemented in software or in hardware as part of the apparatus shown in fig. 3, where the hardware environment includes a network environment.
Thus, according to still another aspect of the embodiments of the present application, there is also provided an electronic device for implementing the above operation and maintenance job management method, which may be a server, a terminal, or a combination thereof.
Fig. 3 is a block diagram of an alternative electronic device, according to an embodiment of the application, as shown in fig. 3, comprising a processor 301, a communication interface 302, a memory 303 and a communication bus 304, wherein the processor 301, the communication interface 302 and the memory 303 perform communication with each other via the communication bus 304, wherein,
A memory 303 for storing a computer program;
the processor 301 is configured to execute the computer program stored in the memory 303, and implement the following steps:
acquiring first dynamic information of an operation and maintenance working device and/or second dynamic information of operation and maintenance personnel;
Acquiring an operation vibration signal acquired by a vibration monitoring device arranged on a road side based on the first dynamic information and/or the second dynamic information;
Inputting the operation vibration signal into a twin feature extraction network to obtain operation vibration features, wherein the twin feature extraction network is obtained by training a sample pair constructed based on the operation vibration signal and the non-operation vibration signal;
The operation vibration characteristics are input into a pre-trained operation and maintenance operation identification model to obtain an operation and maintenance operation state identification result, wherein the operation and maintenance operation identification model is obtained by training based on operation vibration characteristic samples and corresponding operation state labels, and the operation vibration characteristic samples comprise first operation vibration signals of an operation and maintenance operation device in different operation states and second operation vibration signals of an operation and maintenance worker in the operation states.
Alternatively, in the present embodiment, the above-described communication bus may be a PCI (PERIPHERAL COMPONENT INTERCONNECT, peripheral component interconnect standard) bus, or an EISA (Extended Industry Standard Architecture ) bus, or the like. The communication bus may be classified as an address bus, a data bus, a control bus, or the like. For ease of illustration, only one thick line is shown in fig. 3, but not only one bus or one type of bus.
The communication interface is used for communication between the electronic device and other devices.
The memory may include RAM or may include non-volatile memory (non-volatile memory), such as at least one disk memory. Optionally, the memory may also be at least one memory device located remotely from the aforementioned processor.
The processor may be a general purpose processor and may include, but is not limited to: CPU (Central Processing Unit ), NP (Network Processor, network processor), etc.; but also DSP (DIGITAL SIGNAL Processing), ASIC (Application SPECIFIC INTEGRATED Circuit), FPGA (Field-Programmable GateArray, field programmable gate array) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components.
Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments, and this embodiment is not described herein.
It will be appreciated by those skilled in the art that the structure shown in fig. 3 is only illustrative, and the device implementing the operation and maintenance operation management method may be a terminal device, and the terminal device may be a smart phone (such as an Android Mobile phone, an iOS Mobile phone, etc.), a tablet computer, a palm computer, a Mobile internet device (Mobile INTERNET DEVICES, MID), a PAD, etc. Fig. 3 is not limited to the structure of the electronic device. For example, the terminal device may also include more or fewer components (e.g., network interfaces, display devices, etc.) than shown in fig. 3, or have a different configuration than shown in fig. 3.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of the above embodiments may be implemented by a program for instructing a terminal device to execute in association with hardware, the program may be stored in a computer readable storage medium, and the storage medium may include: flash disk, ROM, RAM, magnetic or optical disk, etc.
According to yet another aspect of an embodiment of the present application, there is also provided a storage medium. Alternatively, in the present embodiment, the above-described storage medium may be used for executing the program code of the operation and maintenance job management method.
Alternatively, in this embodiment, the storage medium may be located on at least one network device of the plurality of network devices in the network shown in the above embodiment.
Alternatively, in the present embodiment, the storage medium is configured to store program code for performing the steps of:
acquiring first dynamic information of an operation and maintenance working device and/or second dynamic information of operation and maintenance personnel;
Acquiring an operation vibration signal acquired by a vibration monitoring device arranged on a road side based on the first dynamic information and/or the second dynamic information;
Inputting the operation vibration signal into a twin feature extraction network to obtain operation vibration features, wherein the twin feature extraction network is obtained by training a sample pair constructed based on the operation vibration signal and the non-operation vibration signal;
The operation vibration characteristics are input into a pre-trained operation and maintenance operation identification model to obtain an operation and maintenance operation state identification result, wherein the operation and maintenance operation identification model is obtained by training based on operation vibration characteristic samples and corresponding operation state labels, and the operation vibration characteristic samples comprise first operation vibration signals of an operation and maintenance operation device in different operation states and second operation vibration signals of an operation and maintenance worker in the operation states.
Alternatively, specific examples in the present embodiment may refer to examples described in the above embodiments, which are not described in detail in the present embodiment.
Alternatively, in the present embodiment, the storage medium may include, but is not limited to: various media capable of storing program codes, such as a U disk, ROM, RAM, a mobile hard disk, a magnetic disk or an optical disk.
The foregoing embodiment numbers of the present application are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
The integrated units in the above embodiments may be stored in the above-described computer-readable storage medium if implemented in the form of software functional units and sold or used as separate products. Based on such understanding, the technical solution of the present application may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing one or more computer devices (which may be personal computers, servers or network devices, etc.) to perform all or part of the steps of the method described in the embodiments of the present application.
In the foregoing embodiments of the present application, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In several embodiments provided by the present application, it should be understood that the disclosed client may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, such as the division of the units, is merely a logical function division, and may be implemented in another manner, for example, multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in the present embodiment.
In addition, each functional unit in the embodiments of the present application may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The application can be realized by adopting or referring to the prior art at the places which are not described in the application.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and variations of the present application will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. which come within the spirit and principles of the application are to be included in the scope of the claims of the present application.

Claims (10)

CN202311274823.9A2023-09-282023-09-28Operation and maintenance job management method and device, electronic equipment and storage mediumActiveCN117314391B (en)

Priority Applications (1)

Application NumberPriority DateFiling DateTitle
CN202311274823.9ACN117314391B (en)2023-09-282023-09-28Operation and maintenance job management method and device, electronic equipment and storage medium

Applications Claiming Priority (1)

Application NumberPriority DateFiling DateTitle
CN202311274823.9ACN117314391B (en)2023-09-282023-09-28Operation and maintenance job management method and device, electronic equipment and storage medium

Publications (2)

Publication NumberPublication Date
CN117314391A CN117314391A (en)2023-12-29
CN117314391Btrue CN117314391B (en)2024-05-28

Family

ID=89284372

Family Applications (1)

Application NumberTitlePriority DateFiling Date
CN202311274823.9AActiveCN117314391B (en)2023-09-282023-09-28Operation and maintenance job management method and device, electronic equipment and storage medium

Country Status (1)

CountryLink
CN (1)CN117314391B (en)

Citations (4)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
US10235882B1 (en)*2018-03-192019-03-19Derq Inc.Early warning and collision avoidance
CN111105047A (en)*2019-12-122020-05-05陕西瑞海工程智慧数据科技有限公司 An operation and maintenance monitoring method, device, electronic device and storage medium
CN111488828A (en)*2020-04-102020-08-04谭军华Entrepreneurship park personnel activity identification method and device and terminal equipment
CN115393817A (en)*2022-08-292022-11-25盐城工学院Multi-sensor information fusion agricultural machinery working environment identification method and device

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
WO2016074608A2 (en)*2014-11-112016-05-19冯春魁Methods and systems for vehicle operation monitoring and control, video monitoring, data processing, and overload monitoring and control
EP3735682A4 (en)*2019-03-222020-11-11SZ DJI Technology Co., Ltd.System and method for lane monitoring and providing lane departure warnings

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication numberPriority datePublication dateAssigneeTitle
US10235882B1 (en)*2018-03-192019-03-19Derq Inc.Early warning and collision avoidance
CN111105047A (en)*2019-12-122020-05-05陕西瑞海工程智慧数据科技有限公司 An operation and maintenance monitoring method, device, electronic device and storage medium
CN111488828A (en)*2020-04-102020-08-04谭军华Entrepreneurship park personnel activity identification method and device and terminal equipment
CN115393817A (en)*2022-08-292022-11-25盐城工学院Multi-sensor information fusion agricultural machinery working environment identification method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
地铁列车在线监测与网络化运维系统;潘文凯;赵勇;;技术与市场;20200929(第10期);全文*

Also Published As

Publication numberPublication date
CN117314391A (en)2023-12-29

Similar Documents

PublicationPublication DateTitle
CN110097037B (en)Intelligent monitoring method and device, storage medium and electronic equipment
Li et al.Knowledge-based power line detection for UAV surveillance and inspection systems
KR20200071799A (en)object recognition and counting method using deep learning artificial intelligence technology
US20140355839A1 (en)Mobile pothole detection system and method
CN109298785A (en) A man-machine joint control system and method for monitoring equipment
KR101884611B1 (en)Method of extracting objects-of-Interest out of CCTV video using metadate filtering of moving objects
CN110852236A (en)Target event determination method and device, storage medium and electronic device
CN112541393A (en)Transformer substation personnel detection method and device based on deep learning
CN111783718A (en)Target object state identification method and device, storage medium and electronic device
CN111753587A (en)Method and device for detecting falling to ground
CN113469137A (en)Abnormal behavior recognition method and device, storage medium and electronic device
CN113723176A (en)Target object determination method and device, storage medium and electronic device
CN117830053A (en)Perimeter security alarm system and method
CN111951505B (en)Fence vibration intrusion positioning and mode identification method based on distributed optical fiber system
KR20200123324A (en)A method for pig segmentation using connected component analysis and yolo algorithm
CN118968631B (en)Electric power operation site violation action recognition system based on multi-mode large model
CN102244769A (en)Object and key person monitoring system and method thereof
CN117314391B (en)Operation and maintenance job management method and device, electronic equipment and storage medium
CN119228133A (en) Method for determining the operation safety risk of operators and safety risk linkage system
CN118568621A (en)Intelligent park management system based on digital twinning
CN117312828B (en)Public facility monitoring method and system
CN110278197A (en)A kind of management system and its management method of Internet of Things
CN115424216A (en)Image processing method and processor
CN118115931A (en) A monitoring screen abnormality recognition method and device based on deep learning
CN114495028A (en)Vehicle fake plate identification method and device, electronic equipment and storage medium

Legal Events

DateCodeTitleDescription
PB01Publication
PB01Publication
SE01Entry into force of request for substantive examination
SE01Entry into force of request for substantive examination
GR01Patent grant
GR01Patent grant

[8]ページ先頭

©2009-2025 Movatter.jp