Disclosure of Invention
The invention aims to provide an intelligent monitoring method and system for a state of temperature control equipment. The method comprises the steps of collecting current monitoring data, environment sensing data and historical data of temperature control equipment, processing the current display monitoring data, the environment sensing data and the historical data in the current monitoring data by utilizing an environment influence weight prediction model to obtain current environment influence weights, obtaining display state monitoring values according to the current environment influence weights and the current display monitoring data, evaluating the current control monitoring data in the current monitoring data by utilizing a control monitoring standard value according to a threshold judgment result of the display state monitoring values to obtain control state monitoring values, inputting current position information and current state monitoring information into the temperature control equipment position prediction model according to a threshold judgment result of the control state monitoring values, and optimizing by utilizing a particle swarm algorithm to obtain updated position information.
In order to achieve the above purpose, the present invention provides the following technical solutions:
a temperature control equipment state intelligent monitoring method comprises the following steps:
collecting current monitoring data, environment sensing data, historical display monitoring data and historical environment sensing data of temperature control equipment;
training an environmental impact weight prediction model by utilizing the historical display monitoring data and the historical environmental sensing data to obtain a historical environmental impact weight;
inputting the current display monitoring data, the environment sensing data and the historical environment influence weight in the current monitoring data into the pre-trained environment influence weight prediction model to obtain the current environment influence weight;
comparing the display state monitoring value with a threshold value, if the display state monitoring value is lower than the display threshold value, setting the display state monitoring value to 0 and outputting display abnormality information, otherwise, evaluating the current control monitoring data in the current monitoring data by using a control monitoring standard value to obtain a control state monitoring value;
And comparing the control state monitoring value with a threshold value, setting the control state monitoring value to 0 and outputting control abnormal information if the control state monitoring value is lower than the control threshold value, otherwise, inputting the current position information and the current state monitoring information into a temperature control equipment position prediction model and optimizing by utilizing a particle swarm algorithm to obtain updated position information.
Further, the current monitoring data comprises current display monitoring data and current control monitoring data, the environment sensing data comprises environment illumination sensing data and environment temperature sensing data, and the historical environment sensing data comprises historical environment illumination sensing data and historical environment temperature sensing data.
Further, current display monitoring data, the environment sensing data and the historical environment influence weight in the current monitoring data are input into the pre-trained environment influence weight prediction model to obtain the current environment influence weight; the specific implementation process for obtaining the display state monitoring value by weighting the current display monitoring data by using the current environmental impact weight comprises the following steps:
Inputting current display monitoring data, environment sensing data and historical environment influence weights into a pre-trained environment influence weight prediction model, and optimizing updating parameters to obtain current environment influence weights;
The current environment influence weight comprises an influence temperature display weight, an influence humidity display weight and an influence power display weight;
Obtaining current temperature display data, current humidity display data and current power display data from the current display monitoring data;
Carrying out difference value calculation on the current temperature display data, the current humidity display data and the current power display data and the actual temperature data, the actual humidity data and the actual power data respectively to obtain a current temperature display deviation value, a current humidity display deviation value and a current power display deviation value;
and evaluating the current temperature display deviation value, the current humidity display deviation value and the current power display deviation value by using the current environmental impact weight to obtain a display state monitoring value.
Further, the specific implementation process of evaluating the current control monitoring data in the current monitoring data by using the control monitoring standard value to obtain the control state monitoring value comprises the following steps:
acquiring current control monitoring data and a control monitoring standard value;
The current control monitoring data comprise current control mode data, current control abnormal data, current control delay data and current control temperature data;
obtaining a control mode coefficient and a control anomaly coefficient according to the current control mode data and the current control anomaly data;
And evaluating the control mode coefficient, the control abnormal coefficient, the current control delay data and the current control temperature data to obtain a control state monitoring value.
Further, the specific implementation process of inputting the current position information and the current state monitoring information into the temperature control equipment position prediction model and optimizing by utilizing the particle swarm algorithm to obtain updated position information comprises the following steps:
acquiring historical position information, historical state monitoring information, current position information and current state monitoring information;
The current state monitoring information comprises a current display state monitoring value and a current control state monitoring value;
training a temperature control equipment position prediction model by utilizing the historical position information and the historical state monitoring information to obtain a pre-training equipment position prediction model;
Inputting the current position information and the current state monitoring information into the pre-training equipment position prediction model to obtain predicted position information;
and setting the predicted position information as an initialized particle position, taking the maximized display state monitoring value and the maximized control state monitoring value as optimization targets, and iterating by using a particle swarm algorithm to obtain updated position information.
Further, the specific implementation process of iterating by using the particle swarm algorithm to obtain the updated position information comprises the following steps:
setting an optimization target of a particle swarm algorithm to be a maximized display state monitoring value and a maximized control state monitoring value;
Randomly generating a group of particles, setting an initial position and an initial speed for each particle to form an initial particle group, and simultaneously setting a global optimal position and an optimal fitness, wherein the initial position is set as predicted position information output by a pre-training equipment position prediction model;
updating the speed and position of the particles by using a particle swarm optimization algorithm according to the current position, the current speed, the historical optimal position and the global optimal position of the particles;
And then, judging whether the current iteration number exceeds the preset maximum iteration number, if so, terminating the optimization process, and outputting an adjustment variable value corresponding to the global optimal position, namely updating the position information.
The intelligent monitoring system for the state of the temperature control equipment comprises a system control module, a data acquisition module, a display state monitoring module, a control state monitoring module, a temperature control equipment position prediction module and an output module;
the system control module is used for controlling the starting, suspending and stopping of the system;
The data acquisition module is used for acquiring current monitoring data, environment sensing data, historical display monitoring data and historical environment sensing data of the temperature control equipment;
The display state monitoring module is used for weighting the current display monitoring data by utilizing the current environmental impact weight output by the environmental impact weight prediction model to obtain a display state monitoring value and comparing display threshold values;
the control state monitoring module is used for evaluating the current control monitoring data by utilizing a control monitoring standard value to obtain a control state monitoring value and comparing control threshold values;
The temperature control equipment position prediction module is used for performing data processing by using a temperature control equipment position prediction model and a particle swarm algorithm to obtain updated position information;
The output module is used for outputting the abnormal information and the updated position information.
Further, the specific implementation process of the display state monitoring module for weighting the current display monitoring data by using the current environmental impact weight output by the environmental impact weight prediction model to obtain the display state monitoring value includes:
training the environmental impact weight prediction model by utilizing the historical display monitoring data and the historical environmental sensing data to obtain a historical environmental impact weight;
inputting current display monitoring data, environment sensing data and the historical environment influence weight into the environment influence weight prediction model, and optimizing updating parameters to obtain the current environment influence weight;
Carrying out difference value calculation on the current temperature display data, the current humidity display data and the current power display data and the actual temperature data, the actual humidity data and the actual power data respectively to obtain a current temperature display deviation value, a current humidity display deviation value and a current power display deviation value;
and evaluating the current temperature display deviation value, the current humidity display deviation value and the current power display deviation value by using the current environmental impact weight to obtain a display state monitoring value.
Further, the control state monitoring module is configured to evaluate current control monitoring data by using a control monitoring standard value, and a specific implementation process for obtaining a control state monitoring value includes:
acquiring current control monitoring data and a control monitoring standard value;
The current control monitoring data comprise current control mode data, current control abnormal data, current control delay data and current control temperature data;
obtaining a control mode coefficient and a control anomaly coefficient according to the current control mode data and the current control anomaly data;
And evaluating the control mode coefficient, the control abnormal coefficient, the current control delay data and the current control temperature data to obtain a control state monitoring value.
Further, the temperature control device position prediction module is configured to perform data processing by using a temperature control device position prediction model and a particle swarm algorithm, and the specific implementation process for obtaining updated position information includes:
acquiring historical position information, historical state monitoring information, current position information and current state monitoring information;
The current state monitoring information comprises a current display state monitoring value and a current control state monitoring value;
training a temperature control equipment position prediction model by utilizing the historical position information and the historical state monitoring information to obtain a pre-training equipment position prediction model;
Inputting the current position information and the current state monitoring information into the pre-training equipment position prediction model to obtain predicted position information;
and setting the predicted position information as an initialized particle position, taking the maximized display state monitoring value and the maximized control state monitoring value as optimization targets, and iterating by using a particle swarm algorithm to obtain updated position information.
Compared with the prior art, the invention has the beneficial effects that:
1. The invention provides a display state monitoring function for monitoring the display state of temperature control equipment, which is characterized in that firstly, historical display monitoring data and historical environment sensing data are used for training an environment influence weight prediction model to obtain historical environment influence weights, then, the current environment influence weights output by the environment influence weight prediction model are used for weighting the current display monitoring data to obtain a display state monitoring value, and the display state monitoring value can reflect the interference degree of environment illumination and temperature change on the display state of the temperature control equipment, so that the accuracy and reliability of intelligent monitoring of the state of the temperature control equipment can be effectively improved.
2. The invention provides a control state monitoring function which is used for monitoring the control state of temperature control equipment, the function obtains a control mode coefficient and a control abnormality coefficient through collected current control mode data and current control abnormality data, then the control mode coefficient and the control abnormality coefficient are combined to evaluate current control delay data and current control temperature data to obtain a control state monitoring value, and the function can effectively monitor each dimension controlled by the temperature control equipment according to the control state monitoring value, so that the accuracy and the reliability of intelligent monitoring of the state of the temperature control equipment are effectively improved.
3. The temperature control equipment position prediction function is used for adjusting the temperature control equipment position according to real-time monitoring information, the temperature control equipment position is predicted by using a pre-training equipment position prediction model obtained by training historical position information and historical state monitoring information to obtain predicted position information, then the function uses a particle swarm algorithm to iteratively optimize the predicted position information to obtain updated position information, and the position information obtained by the function is less in environmental interference and can realize stable control performance, so that the accuracy and reliability of intelligent monitoring of the temperature control equipment state are further improved.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
The intelligent temperature controller integrates the functions of measurement, display, control, alarm, recording and the like, is provided with functions of WiFi, OT communication protocols, multiple communication interfaces and the like, integrates the functions of AI and voice, can realize accurate temperature measurement and accurate on-site and remote control of household equipment temperature and working states and mobile phones, and brings great challenges to monitoring of temperature control equipment states due to complex design and functions.
In order to improve the user experience of an intelligent temperature controller, a manufacturer of the intelligent temperature controller introduces the intelligent monitoring method for the state of the temperature control equipment, and the flow can refer to fig. 1, and the specific implementation mode is as follows:
firstly, collecting current monitoring data, environment sensing data, historical display monitoring data and historical environment sensing data of temperature control equipment;
Further, the current monitoring data comprises current display monitoring data and current control monitoring data, the environment sensing data comprises environment illumination sensing data and environment temperature sensing data, and the historical environment sensing data comprises historical environment illumination sensing data and historical environment temperature sensing data.
Further, the parameter values of the current display monitor data and the environmental sensor data are shown in table 1.
TABLE 1 parameter data currently displaying monitoring data and environmental sensing data
In the embodiment, the historical environment sensing data is used for training an environment influence weight prediction model to predict the change trend of the display data caused by the past environment factors, and the current monitoring data and the environment sensing data provide a powerful data basis for the display state monitoring and the control state monitoring.
Training an environmental impact weight prediction model by utilizing the historical display monitoring data and the historical environmental sensing data to obtain a historical environmental impact weight;
inputting the current display monitoring data, the environment sensing data and the historical environment influence weight in the current monitoring data into the pre-trained environment influence weight prediction model to obtain the current environment influence weight;
further, the specific implementation process of the display state monitoring value comprises the following steps:
Inputting current display monitoring data, environment sensing data and historical environment influence weights into a pre-trained environment influence weight prediction model, and optimizing updating parameters to obtain current environment influence weights;
The current environment influence weight comprises an influence temperature display weight, an influence humidity display weight and an influence power display weight;
The structure of the environmental impact weight prediction model in this embodiment may refer to fig. 2, which includes an input layer, a CNN-LSTM feature extraction layer, a feature merging layer, an environmental impact weight prediction layer, and an output layer, where the input layer is configured to map current display monitor data, environmental sensor data, and historical environmental impact weight into a feature space for further processing, the CNN-LSTM feature extraction layer is configured to extract a temporal feature of LSTM and a spatial feature of CNN, the feature merging layer is configured to perform channel merging on the temporal feature and the spatial feature, the environmental impact weight prediction layer uses a multi-layer perceptron to obtain a weight prediction result, and the output layer outputs the current environmental impact weight through a linear activation function.
Obtaining current temperature display data, current humidity display data and current power display data from the current display monitoring data;
Carrying out difference value calculation on the current temperature display data, the current humidity display data and the current power display data and the actual temperature data, the actual humidity data and the actual power data respectively to obtain a current temperature display deviation value, a current humidity display deviation value and a current power display deviation value;
and evaluating the current temperature display deviation value, the current humidity display deviation value and the current power display deviation value by using the current environmental impact weight to obtain a display state monitoring value.
Further, the calculation formula of the display state monitoring value is as follows:
;
Wherein XSJC is a display state monitoring value, the value is in the range of [0,1] and the larger the value is, the more normal the temperature control equipment is displayed; displaying weights for influencing temperature; displaying data for the current temperature acquired for the ith time; Is actual temperature data; displaying weights for influencing humidity; Displaying data for the current humidity acquired for the ith time; Is the actual humidity data; display weights for influencing power; Displaying data for the current power acquired for the ith time; Is the actual power data.
Further, the temperature display weight is affectedInfluence humidity display weightAnd influencing power display weightAnd the weight value is between 0 and 1, which is obtained by outputting the environmental impact weight prediction model.
Further, the four sets of current display data collected for 8-11 times obtain the display state monitoring result of the temperature control device by using a calculation formula of the display state monitoring value, as shown in the following table 2, and the display threshold is set to be 0.9.
Table 2, temperature control device display status monitoring results
The embodiment provides a display state monitoring function for monitoring the display state of temperature control equipment, which comprises the steps of firstly training an environmental impact weight prediction model by using historical display monitoring data and historical environment sensing data to obtain historical environmental impact weights, and then weighting current display monitoring data by using the current environmental impact weights output by the environmental impact weight prediction model to obtain a display state monitoring value, wherein the display state monitoring value can reflect the interference degree of environmental illumination and temperature change on the display state of the temperature control equipment, so that the accuracy and reliability of intelligent monitoring of the state of the temperature control equipment can be effectively improved.
Comparing the display state monitoring value with a threshold value, if the display state monitoring value is lower than the display threshold value, setting the display state monitoring value to 0 and outputting display abnormality information, otherwise, evaluating the current control monitoring data in the current monitoring data by using a control monitoring standard value to obtain a control state monitoring value;
Further, the specific implementation process of evaluating the current control monitoring data in the current monitoring data by using the control monitoring standard value to obtain the control state monitoring value comprises the following steps:
acquiring current control monitoring data and a control monitoring standard value;
the current control monitoring data comprises current control mode data, current control abnormal data, current control delay data and current control temperature data, wherein the current control mode data is represented as indoor temperature data in different control modes, and the current control abnormal data is represented as an abnormal deviation value of target temperature and actual temperature;
obtaining a control mode coefficient and a control anomaly coefficient according to the current control mode data and the current control anomaly data;
And evaluating the control mode coefficient, the control abnormal coefficient, the current control delay data and the current control temperature data to obtain a control state monitoring value.
Further, the control pattern coefficient and the control abnormality coefficient are expressed as:
Wherein, the method comprises the steps of,In order to control the coefficient of the mode,For the indoor temperature change value when the temperature control device is turned on and off,The temperature is set for the target to be at,In order to control the coefficient of anomaly,In order to be able to achieve the actual temperature,The control mode coefficient represents the influence degree of the temperature control equipment on indoor temperature change under different working modes, and the control abnormal coefficient represents the temperature control performance of the temperature control equipment under abnormal working states.
Further, the calculation formula of the control state monitoring value is:
;
Wherein KZJC is a control state monitoring value, the value is in the range of [0,1] and the larger the value is, the more normal the temperature control equipment is controlled; in order to control the coefficient of the mode,KZPG is a control state evaluation value, M is the number of times of data acquisition; To control delay weights; current control delay data acquired for the j-th time; to control the delay threshold; to control the temperature weight; The current control temperature data acquired for the j-th time; to control the temperature threshold.
In the embodiment, the control delay weight and the control temperature weight are both set to 0.5, and the control delay threshold and the control temperature threshold are related to the model and the performance of the temperature control equipment and can be adjusted according to actual conditions.
Further, the four sets of current control monitoring data collected for 23-26 times obtain the control state monitoring result of the temperature control equipment by using a calculation formula of the control state monitoring value, as shown in the following table 3, and the control threshold is set to be 0.85.
Table 3, temperature control device display status monitoring results
The embodiment provides a control state monitoring function for monitoring the control state of temperature control equipment, the function obtains a control mode coefficient and a control abnormality coefficient through collected current control mode data and current control abnormality data, then evaluates current control delay data and current control temperature data by combining the control mode coefficient and the control abnormality coefficient to obtain a control state monitoring value, and the function can effectively monitor each dimension controlled by the temperature control equipment according to the control state monitoring value, so that the accuracy and the reliability of intelligent monitoring of the state of the temperature control equipment are effectively improved.
And comparing the control state monitoring value with a threshold value, setting the control state monitoring value to 0 and outputting control abnormal information if the control state monitoring value is lower than the control threshold value, otherwise, inputting the current position information and the current state monitoring information into a temperature control equipment position prediction model and optimizing by utilizing a particle swarm algorithm to obtain updated position information.
Further, the specific implementation process of inputting the current position information and the current state monitoring information into the temperature control equipment position prediction model and optimizing by utilizing the particle swarm algorithm to obtain updated position information comprises the following steps:
acquiring historical position information, historical state monitoring information, current position information and current state monitoring information;
The current state monitoring information comprises a current display state monitoring value and a current control state monitoring value;
training a temperature control equipment position prediction model by utilizing the historical position information and the historical state monitoring information to obtain a pre-training equipment position prediction model;
Inputting the current position information and the current state monitoring information into the pre-training equipment position prediction model to obtain predicted position information;
And setting the predicted position information as an initialized particle position, taking the maximized state monitoring information as an optimization target, and iterating by using a particle swarm algorithm to obtain updated position information.
The temperature control equipment position prediction function is used for adjusting the temperature control equipment position according to real-time monitoring information, predicting the temperature control equipment position by using a pre-training equipment position prediction model obtained by training historical position information and historical state monitoring information to obtain predicted position information, and then iteratively optimizing the predicted position information by using a particle swarm algorithm to obtain updated position information, wherein the position information obtained by the function is less in environmental interference and can realize stable control performance, so that the accuracy and reliability of intelligent monitoring of the temperature control equipment state are further improved.
In this embodiment, an intelligent monitoring method for a state of a temperature control device is provided. The method comprises the steps of obtaining a current environmental impact weight, obtaining a display state monitoring value by using the current display monitoring data and a control state monitoring value obtained by evaluating according to the current control monitoring data, effectively shortening the monitoring time of the equipment in different states, improving the accuracy and reliability of intelligent monitoring of the state of the temperature control equipment, and obtaining updated position information by using the prediction result of a temperature control equipment position prediction model and iterative optimization of a particle swarm algorithm according to the threshold judgment result of the control state monitoring value, so that the stability of the working state of the temperature control equipment can be improved, and user experience is improved.
Example two
In this embodiment, an intelligent monitoring system for a state of a temperature control device is provided, a system structure is shown in fig. 3, and a specific implementation manner is as follows:
the system control module is used for controlling the starting, suspending and stopping of the system;
The data acquisition module is used for acquiring current monitoring data, environment sensing data, historical display monitoring data and historical environment sensing data of the temperature control equipment;
the display state monitoring module is used for weighting the current display monitoring data by utilizing the current environmental impact weight output by the environmental impact weight prediction model to obtain a display state monitoring value and comparing display threshold values;
Further, the specific implementation process of the display state monitoring unit for weighting the current display monitoring data by using the current environmental impact weight output by the environmental impact weight prediction model to obtain the display state monitoring value includes:
training the environmental impact weight prediction model by utilizing the historical display monitoring data and the historical environmental sensing data to obtain a historical environmental impact weight;
inputting current display monitoring data, environment sensing data and the historical environment influence weight into the environment influence weight prediction model, and optimizing updating parameters to obtain the current environment influence weight;
Carrying out difference value calculation on the current temperature display data, the current humidity display data and the current power display data and the actual temperature data, the actual humidity data and the actual power data respectively to obtain a current temperature display deviation value, a current humidity display deviation value and a current power display deviation value;
and evaluating the current temperature display deviation value, the current humidity display deviation value and the current power display deviation value by using the current environmental impact weight to obtain a display state monitoring value.
Further, the display state monitoring unit obtains the monitoring result of the display state monitoring unit by using the four groups of current display data collected from 42 th to 45 th times and combining the calculation formula of the display state monitoring value, as shown in the following table 4.
Table 4, display status monitoring unit monitoring results
Further, the judging unit is used for comparing the display state monitoring value with a display threshold value.
The control state monitoring module is used for evaluating the current control monitoring data by utilizing a control monitoring standard value to obtain a control state monitoring value and comparing a control threshold value;
further, the control state monitoring unit evaluates the current control monitoring data by using a control monitoring standard value, and the specific implementation process for obtaining the control state monitoring value comprises the following steps:
acquiring current control monitoring data and a control monitoring standard value;
The current control monitoring data comprise current control mode data, current control abnormal data, current control delay data and current control temperature data;
obtaining a control mode coefficient and a control anomaly coefficient according to the current control mode data and the current control anomaly data;
And evaluating the control mode coefficient, the control abnormal coefficient, the current control delay data and the current control temperature data to obtain a control state monitoring value.
Further, the control state monitoring unit obtains the monitoring result of the control state monitoring unit by using the four groups of current control monitoring data collected from 84 th to 87 th times and combining with a calculation formula of the control state monitoring value, as shown in the following table 5.
Table 5, monitoring results by the control status monitoring unit
Further, the judging unit is used for comparing the control state monitoring value with a control threshold value.
The temperature control equipment position prediction module is used for performing data processing by utilizing a temperature control equipment position prediction model and a particle swarm algorithm to obtain updated position information;
Further, the specific implementation process of the temperature control device position prediction unit for obtaining the predicted position information by using the temperature control device position prediction model includes:
acquiring historical position information, historical state monitoring information, current position information and current state monitoring information;
The current state monitoring information comprises a current display state monitoring value and a current control state monitoring value;
training a temperature control equipment position prediction model by utilizing the historical position information and the historical state monitoring information to obtain a pre-training equipment position prediction model;
And inputting the current position information and the current state monitoring information into the pre-training equipment position prediction model to obtain predicted position information.
Further, the process of the position optimization unit iterating by using the particle swarm algorithm to obtain updated position information includes:
setting an optimization target of a particle swarm algorithm to be a maximized display state monitoring value and a maximized control state monitoring value;
Randomly generating a group of particles, giving an initial position and an initial speed to each particle to form an initial particle group, and simultaneously setting a global optimal position and an optimal fitness, wherein the initial position is set as predicted position information output by a pre-training equipment position prediction model;
updating the speed and position of the particles by using a particle swarm optimization algorithm according to the current position, the current speed, the historical optimal position and the global optimal position of the particles;
And then, judging whether the current iteration number exceeds the preset maximum iteration number, if so, terminating the optimization process, and outputting an adjustment variable value corresponding to the global optimal position, namely updating the position information.
And the output module is used for outputting the abnormal information and updating the position information.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.