Temperature regulation and control method, device and equipment and storage medium

Abstract

The invention discloses a temperature regulation and control method and device, equipment and a storage medium, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining historical temperature data of a preset temperature field, and classifying the historical temperature data based on a preset label, so as to construct training data based on the classified historical temperature data; training the initial binary tree model and the initial deep neural network model based on the training data to obtain a target binary tree model and a target deep neural network model; obtaining current target temperature data of the preset temperature field, and inputting the target temperature data into the target binary tree model and the target deep neural network model to generate an initial classification result of the target temperature data; and determining a target temperature control score of the target temperature data based on the initial classification result, and triggering a temperature regulation operation corresponding to the preset temperature field according to the target temperature control score. Therefore, the change of the water temperature set value of the indirect air cooling tower can be responded in real time, and the safety of the indirect air cooling tower is improved.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to a temperature control method, device, equipment and storage medium. Background Art

[0002] On the original indirect air cooling tower intelligent control platform, the water temperature setting value of the traditional temperature control needs to be set manually. In a changing environment, people cannot respond quickly to modify the setting value, which causes the temperature control to deviate from the optimal value, and there is a general risk of fins being out of control and freezing the pipes.

[0003] In summary, how to respond in real time to changes in the water temperature set value of the indirect air cooling tower is a technical problem that needs to be solved urgently. Summary of the invention

[0004] In view of this, the object of the present invention is to provide a temperature control method, device, equipment and storage medium, which can respond to the change of the water temperature setting value of the indirect air cooling tower in real time. The specific scheme is as follows:

[0005] In a first aspect, the present application provides a temperature control method, comprising:

[0006] Acquire historical temperature data of a preset temperature field, and classify the historical temperature data based on preset labels, so as to construct training data based on the classified historical temperature data;

[0007] Training an initial binary tree model and an initial deep neural network model based on the training data to obtain a target binary tree model and a target deep neural network model;

[0008] Acquire current target temperature data of the preset temperature field, and input the target temperature data into the target binary tree model and the target deep neural network model respectively to generate an initial classification result of the target temperature data;

[0009] A target temperature control score of the target temperature data is determined based on the initial classification result, and a temperature control operation corresponding to the preset temperature field is triggered according to the target temperature control score.

[0010] Optionally, acquiring historical temperature data of a preset temperature field and classifying the historical temperature data based on preset tags includes:

[0011] Acquire the historical temperature data of the preset temperature field, and clean the historical temperature data according to a preset data cleaning rule to obtain cleaned temperature data;

[0012] The post-cleaning temperature data is classified based on a preset classification label; wherein the preset classification label is a frozen tube label or a safety label.

[0013] Optionally, the inputting the target temperature data into the target binary tree model and the target deep neural network model respectively to generate an initial classification result of the target temperature data includes:

[0014] The target temperature data is respectively input into the target binary tree model and the target deep neural network model to obtain a first classification result output by the target binary tree model and a second classification result output by the target deep neural network model; the first classification result and the second classification result include classification labels and classification probabilities.

[0015] Optionally, determining a target temperature control score of the target temperature data based on the initial classification result includes:

[0016] Determining preset score weights of the target binary tree model and the target deep neural network model;

[0017] The target temperature control score of the target temperature data is determined based on the first classification result, the second classification result and the preset score weight.

[0018] Optionally, the temperature control method further includes:

[0019] Determine a target classification result corresponding to the target temperature data based on the target temperature control score, and obtain a proofreading result corresponding to the target classification result;

[0020] The accuracy of the target classification result is determined according to the proofreading result, and the preset score weight is updated based on the accuracy, so as to determine the target temperature control score of the target temperature data based on the updated preset score weight.

[0021] Optionally, determining a target temperature control score of the target temperature data based on the initial classification result, and triggering a temperature control operation corresponding to the preset temperature field according to the target temperature control score, includes:

[0022] Determine the target temperature control score of the target temperature data within a preset time period based on the initial classification result, and construct a temperature change curve of the preset temperature field within the preset time period according to the target temperature control score;

[0023] Filtering target frozen pipe data from the historical temperature data based on preset data screening rules, and generating a corresponding frozen pipe temperature curve according to the target frozen pipe data using the target binary tree model and the target deep neural network model;

[0024] The curve similarity between the temperature change curve and the freezing pipe temperature curve is determined, and the temperature control operation of the preset temperature field is triggered when the curve similarity is greater than a preset similarity threshold.

[0025] Optionally, the process of training the initial binary tree model and the initial deep neural network model based on the training data to obtain the target binary tree model and the target deep neural network model further includes:

[0026] Obtaining a predicted temperature control score corresponding to the training data output by the initial deep neural network model, and determining a target temperature control score corresponding to the training data based on the initial deep neural network model and the initial binary tree model;

[0027] Determine a deviation rate between the predicted temperature control score and the target temperature control score corresponding to the training data, and update a preset initialization weight matrix in the initial neural network model based on the deviation rate to obtain the target deep neural network model.

[0028] In a second aspect, the present application provides a temperature control device, comprising:

[0029] A training data construction module, used for acquiring historical temperature data of a preset temperature field and classifying the historical temperature data based on preset labels, so as to construct training data based on the classified historical temperature data;

[0030] A model training module, used for training an initial binary tree model and an initial deep neural network model based on the training data to obtain a target binary tree model and a target deep neural network model;

[0031] A data classification module, used to obtain the current target temperature data of the preset temperature field, and input the target temperature data into the target binary tree model and the target deep neural network model respectively, so as to generate an initial classification result of the target temperature data;

[0032] A temperature control module is used to determine a target temperature control score of the target temperature data based on the initial classification result, and trigger a temperature control operation corresponding to the preset temperature field according to the target temperature control score.

[0033] In a third aspect, the present application provides an electronic device, including:

[0034] Memory, used to store computer programs;

[0035] The processor is used to execute the computer program to implement the aforementioned temperature control method.

[0036] In a fourth aspect, the present application provides a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the aforementioned temperature control method is implemented.

[0037] In the present application, the historical temperature data of the preset temperature field is first obtained, and the historical temperature data is classified based on the preset labels, so as to construct training data based on the classified historical temperature data; then the initial binary tree model and the initial deep neural network model are trained based on the training data to obtain the target binary tree model and the target deep neural network model; then the current target temperature data of the preset temperature field is obtained, and the target temperature data is respectively input into the target binary tree model and the target deep neural network model to generate the initial classification result of the target temperature data; finally, the target temperature control score of the target temperature data is determined based on the initial classification result, and the corresponding temperature control operation of the preset temperature field is triggered according to the target temperature control score. As can be seen from the above, in the present application, the target binary tree model and the target deep neural network model are obtained by training the historical temperature data of the preset temperature field, and the initial classification result of the target temperature data is generated by using the target binary tree model and the target deep neural network model, and then the target temperature control score of the target temperature data is obtained based on the initial classification result, so as to trigger the corresponding temperature control operation of the preset temperature field based on the target temperature control score. In this way, the present application uses a large amount of real historical temperature data to train the target binary tree model and the target deep neural network model, determines the target temperature control score through the output results of the target binary tree model and the target deep neural network model, and regulates the water temperature of the temperature field according to the target temperature control score. In this way, the present application can respond more quickly to the optimization setting of the water temperature setting value, so that the economic benefits and safety are improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0039] Figure 1 A flow chart of a temperature control method provided in this application;

[0040] Figure 2 A coefficient diagram of a specific temperature field data provided for this application;

[0041] Figure 3 A schematic diagram of a specific temperature field heat dissipation surface provided in this application;

[0042] Figure 4 A schematic diagram of calculating the single-point temperature control score of a specific temperature field heat dissipation surface provided in this application;

[0043] Figure 5 A specific flow chart of the temperature control method provided in this application;

[0044] Figure 6 A schematic diagram of the structure of a temperature control device provided in this application;

[0045] Figure 7 A structural diagram of an electronic device provided for this application. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] On the original indirect air cooling tower's indirect cooling intelligent control platform, the water temperature setting value of the traditional temperature control needs to be set manually. In a changing environment, people cannot respond quickly to modify the setting value, which causes the temperature control to deviate from the optimal value, and there is a general risk of the fins being out of control and freezing the pipes. To this end, the present application provides a temperature control solution that can respond to changes in the water temperature setting value of the indirect air cooling tower in real time.

[0048] See also Figure 1 As shown, the embodiment of the present invention discloses a temperature control method, which may include:

[0049] Step S11, acquiring historical temperature data of a preset temperature field, and classifying the historical temperature data based on preset labels, so as to construct training data based on the classified historical temperature data.

[0050] In this embodiment, the acquisition of historical temperature data of a preset temperature field and classification of the historical temperature data based on preset labels may include: acquiring the historical temperature data of the preset temperature field, and cleaning the historical temperature data according to preset data cleaning rules to obtain cleaned temperature data; classifying the cleaned temperature data based on preset classification labels; wherein the preset classification labels are frozen pipe labels or safety labels. Specifically, it is first necessary to select several thousand valid historical temperature data, and the historical temperature data include: frozen pipes, temperature data with a high risk of frozen pipes, and absolutely safe temperature data. Thereafter, the historical temperature data is cleaned for data validity according to the business and thermal balance, and outliers in the historical temperature data are cleaned, data completed, and ring points filled. Manual verification and labeling are performed to calibrate the state of a heat dissipation surface of the preset temperature field, 0-frozen pipe, 1-safe. Thereafter, the data similarity between the historical temperature data is determined, and the data with a data similarity greater than or equal to that between the historical temperature data are deleted from the historical temperature data. The above temperature data is retained, and the remaining post-cleaning temperature data is classified using semi-supervised machine learning to determine the binary labels of the post-cleaning temperature data, which are frozen tube labels or safety labels. Then the classified historical temperature data is split to obtain two sets of training data of equal quantity. Figure 2 As shown, in this embodiment, the correlation between the sector cold water main pipe temperature, the circulating water and hot water main pipe temperature, the ambient temperature, the power generation load, the shutter opening, etc. is calculated and the coefficient is manually adjusted.

[0051] Step S12: training an initial binary tree model and an initial deep neural network model based on the training data to obtain a target binary tree model and a target deep neural network model.

[0052] In this embodiment, two sets of training data are input into the initial binary tree model and the initial deep neural network model respectively. When training the initial binary tree model, one set of training data is split into N groups, and binary tree classification training is performed based on the idea of ​​Bagging (i.e. Bootstrap Aggregating) to obtain the target binary tree model.

[0053] Understandably, see Figure 3 and Figure 4As shown, in order to optimize the neural network model, the process of training the initial binary tree model and the initial deep neural network model based on the training data to obtain the target binary tree model and the target deep neural network model may also include: obtaining the predicted temperature control score corresponding to the training data output by the initial deep neural network model, and determining the target temperature control score corresponding to the training data based on the initial deep neural network model and the initial binary tree model; determining the deviation rate between the predicted temperature control score and the target temperature control score corresponding to the training data, and updating the preset initialization weight matrix in the initial neural network model based on the deviation rate to obtain the target deep neural network model. Specifically, according to the change in the trend of the temperature data of the preset temperature field, a multi-layer neural network model of LeNET based on time series change is built, and the temperature field is completely safe as 100, and the temperature data of the completely frozen pipe is recorded as 0. The evaluation is performed for a short period of time to achieve a set of data representing the trend fluctuation of the temperature field data. It should be noted that the preset temperature field is fixed to a data layout of M layers × 3 columns. The temperature data at each moment is recorded, and the fluctuation score is calculated every 30 seconds. The temperature data at the corresponding position is directly identified by a vector within a period of time, and the vector calculation is performed in a way that the continuous time between the same layers affects each other. In a specific implementation, M is 5 layers, and the vector of a single point on the heat dissipation surface of the preset temperature field after time calculation is , the training weight parameter size is The weight parameters of the initialization weight matrix are obtained by training the multi-layer neural network model of LeNet based on time-varying labels as frozen pipes or safe binary classification. From this, it can be concluded that the final learning process is Time series temperature data The weight parameters are flattened to obtain a 15-digit vector matrix. Through algorithm regression training, the obtained value is recorded as the predicted temperature control score of the temperature field. The training data is input into the initial deep neural network model to obtain the predicted temperature control score corresponding to the training data output by the initial deep neural network model. Then, the target temperature control score corresponding to the training data is determined based on the classification results output by the initial deep neural network model and the initial binary tree model. The deviation rate between the predicted temperature control score and the target temperature control score is calculated, and the weight parameters in the preset initialization weight matrix in the initial neural network model are updated based on the deviation rate, so that the target deep neural network model can be obtained.

[0054] Step S13, obtaining the current target temperature data of the preset temperature field, and inputting the target temperature data into the target binary tree model and the target deep neural network model respectively to generate an initial classification result of the target temperature data.

[0055] In this embodiment, the target temperature data is input into the target binary tree model and the target deep neural network model respectively to generate the initial classification result of the target temperature data, which may include: inputting the target temperature data into the target binary tree model and the target deep neural network model respectively to obtain a first classification result output by the target binary tree model and a second classification result output by the target deep neural network model; the first classification result and the second classification result include classification labels and classification probabilities. Specifically, The target temperature data are respectively input into the target binary tree model and the target deep neural network model to obtain the classification of the two types of models output by the target binary tree model and the target deep neural network model and the probability of the categories.

[0056] Step S14: determining a target temperature control score of the target temperature data based on the initial classification result, and triggering a temperature control operation corresponding to the preset temperature field according to the target temperature control score.

[0057] In this embodiment, the target temperature control score of the target temperature data determined based on the initial classification result may include: determining the preset score weights of the target binary tree model and the target deep neural network model; determining the target temperature control score of the target temperature data based on the first classification result, the second classification result and the preset score weights. In a specific implementation, the preset score weights of the target binary tree model and the target deep neural network model are set to , the classification probability corresponding to the first classification result output by the target binary tree model is multiplied and added to the classification as K1, the classification probability corresponding to the second classification result output by the target deep neural network model is multiplied and added to the classification as K2, and the target temperature control score of the target temperature data is determined as .

[0058] It is understandable that in order to improve the accuracy of the target temperature control score, after determining the temperature control score of the target temperature data based on the target classification result, it may also include: determining the target classification result corresponding to the target temperature data based on the target temperature control score, and obtaining the proofreading result corresponding to the target classification result; determining the accuracy of the target classification result according to the proofreading result, and updating the preset score weight based on the accuracy, so as to determine the target temperature control score of the target temperature data based on the updated preset score weight. Specifically, after obtaining the target temperature control score, the target classification result corresponding to the target temperature data can be determined, and then the target classification result is manually proofread to obtain the corresponding proofreading result, and the accuracy of the target classification result is calculated based on the proofreading result, and the preset score weight is optimized according to the accuracy. The basic step size of optimization is 0.1. For example, if the prediction accuracy is 0.8, the score weight of the target binary tree model is adjusted from 0.5 to 0.4, and the score weight of the target deep neural network model is adjusted from 0.5 to 0.6. If the prediction accuracy is 0.915, the next optimization iteration = accuracy change * last weight first-order gradient. , the score weight of the target binary tree model is adjusted to 0.25, and the score weight of the target deep neural network model is adjusted to 0.75. Otherwise, the score weight is reduced until the accuracy reaches a stable range. After obtaining the target temperature control score, the temperature control operation corresponding to the preset temperature field can be triggered according to the target temperature control score.

[0059] As can be seen from the above, in this embodiment, the historical temperature data of the preset temperature field is first obtained, and the historical temperature data is classified based on the preset labels, so as to construct training data based on the classified historical temperature data; then, the initial binary tree model and the initial deep neural network model are trained based on the training data to obtain the target binary tree model and the target deep neural network model; then, the current target temperature data of the preset temperature field is obtained, and the target temperature data is respectively input into the target binary tree model and the target deep neural network model to generate the initial classification result of the target temperature data; finally, the target temperature control score of the target temperature data is determined based on the initial classification result, and the corresponding temperature control operation of the preset temperature field is triggered according to the target temperature control score. As can be seen from the above, in this embodiment, the target binary tree model and the target deep neural network model are obtained by training the historical temperature data of the preset temperature field, and the initial classification result of the target temperature data is generated by using the target binary tree model and the target deep neural network model, and then the target temperature control score of the target temperature data is obtained based on the initial classification result, so as to trigger the corresponding temperature control operation of the preset temperature field based on the target temperature control score. In this way, in this embodiment, a large amount of real historical temperature data is used to train the target binary tree model and the target deep neural network model, the target temperature control score is determined by the output results of the target binary tree model and the target deep neural network model, and the water temperature of the temperature field is regulated according to the target temperature control score. In this way, this embodiment can respond to the optimization setting of the water temperature setting value more quickly, so that economic benefits and safety are improved.

[0060] Based on the previous embodiment, it can be known that the present application can determine the target temperature control score of the preset temperature field. Next, this embodiment will elaborate on how to determine whether to trigger the corresponding temperature control operation of the preset temperature field according to the target temperature control score. Figure 5 As shown, the embodiment of the present invention further discloses a temperature control method, which may include:

[0061] Step S21, determining a target temperature control score of the target temperature data within a preset time period based on the initial classification result, and constructing a temperature change curve of a preset temperature field within the preset time period according to the target temperature control score.

[0062] In a specific implementation, a heat dissipation surface of a preset temperature field is taken as a unit, and a moment is The data matrix is ​​used as a unit surface. After calculation by the target binary tree model and the target deep neural network model, a classification state with classification probability is obtained, such as 0:0.4, 1:0.6. The frozen pipe score can be recorded as 40 points and the safety score as 60. After a period of time, many scores will be obtained, such as: 40, 29, 27, 42, ..., 48, forming a temperature change curve, which is recorded as Q1. After a large amount of data analysis and comparison with frozen pipe data, the antifreeze data duration is at least 10 minutes.

[0063] Step S22: Filter out target frozen pipe data from the historical temperature data based on preset data screening rules, and generate a corresponding frozen pipe temperature curve according to the target frozen pipe data using a target binary tree model and a target deep neural network model.

[0064] In this embodiment, about one hundred pieces of target frozen pipe data are manually selected from the historical temperature data, and the target frozen pipe data are input into the target binary tree model and the target deep neural network model, and a corresponding frozen pipe temperature curve is generated, which is recorded as Q2.

[0065] Step S23, determining the curve similarity between the temperature change curve and the freezing pipe temperature curve, and triggering the temperature control operation of the preset temperature field when the curve similarity is greater than a preset similarity threshold.

[0066] In one embodiment, the frozen tube fraction is set to , the safety score is set to The limit is determined by the probability of the two classifications trained by the target binary tree model and the target deep neural network model. For example, the highest probability of data classification as frozen pipe label 0 is 0.65, and the probability of data classification as safety label 1 is above 0.65, so the frozen pipe limit is set to 65. The temperature change curve of the first N minutes of the target temperature data is used to calculate the maximum fitting linear trend. If the temperature change curve Q1 for N consecutive minutes is highly similar to the trend of the first N minutes of the frozen pipe temperature curve Q2 under the premise of the safety score, the maximum fitting linear trend is calculated. , it means that the pipe freezing phenomenon will occur in the preset temperature field in the near future, and the protection needs to be triggered in advance until the similarity is lower than The low temperature protection is released when the temperature is below 1000 °C. Among them, N is variable, and the subsequent data analysis shows that the antifreeze data duration is at least 10 minutes.

[0067] In order to determine the curve similarity coefficient, hundreds of N-minute curves were obtained from the data of hundreds of frozen pipes through calculations by the target binary tree model and the target deep neural network model. After being summarized into a graph, the coefficient of similarity was calculated. The curve at the edge of the range, where As standard deviation, calculate the similarity with the freezing pipe temperature curve Q2. Assuming it is 0.9, when updating the coefficient every year, the historical data is unified and recalculated to update the similarity coefficient until the continuous coefficient volatility is less than 0.1.

[0068] It is understandable that the score limit is updated: the frozen pipe score is 64, and the safety score is continuously above 65, which is the normal control range. During the control process, there will be some temperature control scores outside the predicted data of previous years. For example, the score fluctuates around 64 during a period of time, but because it does not meet the continuous low temperature for N minutes or the similarity with the frozen pipe curve is higher than 0.9, the low temperature protection will not be triggered. The lower limit of the iterative update curve similarity can be recorded and used as the data learning basis for antifreeze safety control. This data is automatically calibrated as a safe state. In a new batch of data in the second year, the safety line of the temperature field is retrained, and the frozen pipe score and duration are replanned. In a specific implementation method, the score curve of the temperature field before the frozen pipe in previous years is first analyzed; then the old safety line is 65. In the new round of temperature data, it is found that starting from 52, there are multiple scores above 65 during the period of N minutes, and the similarity with the curve starting from the frozen pipe score of 65 and lasting for about M minutes is 0.8, indicating that in the N minutes before 65, there are still After the new model is trained twice, the new safety line ; Finally, if after a certain experiment, the similarity between the change curve of the temperature field risk score and the frozen pipe curve is If the score is above the safety line, you can stop updating the score.

[0069] As can be seen from the above, in this embodiment, the temperature change curve of the preset temperature field is calculated, and the frozen pipe temperature curve corresponding to the target frozen pipe data in the historical temperature data is calculated, and then the curve similarity between the temperature change curve and the frozen pipe temperature curve is calculated, and finally the temperature control operation of the preset temperature field is triggered when the curve similarity is greater than the preset similarity threshold. In this way, this embodiment can respond to the optimization setting of the water temperature setting value more quickly.

[0070] Accordingly, see Figure 6 As shown, the embodiment of the present application also provides a temperature control device, which may include:

[0071] A training data construction module 11 is used to obtain historical temperature data of a preset temperature field and classify the historical temperature data based on preset labels so as to construct training data based on the classified historical temperature data;

[0072] A model training module 12, used to train an initial binary tree model and an initial deep neural network model based on the training data to obtain a target binary tree model and a target deep neural network model;

[0073] A data classification module 13 is used to obtain the current target temperature data of the preset temperature field, and input the target temperature data into the target binary tree model and the target deep neural network model respectively to generate an initial classification result of the target temperature data;

[0074] The temperature control module 14 is used to determine a target temperature control score of the target temperature data based on the initial classification result, and trigger a temperature control operation corresponding to the preset temperature field according to the target temperature control score.

[0075] As can be seen from the above, in this application, the historical temperature data of the preset temperature field is first obtained, and the historical temperature data is classified based on the preset labels, so as to construct training data based on the classified historical temperature data; then, the initial binary tree model and the initial deep neural network model are trained based on the training data to obtain the target binary tree model and the target deep neural network model; then, the current target temperature data of the preset temperature field is obtained, and the target temperature data is respectively input into the target binary tree model and the target deep neural network model to generate the initial classification result of the target temperature data; finally, the target temperature control score of the target temperature data is determined based on the initial classification result, and the corresponding temperature control operation of the preset temperature field is triggered according to the target temperature control score. As can be seen from the above, in this application, the target binary tree model and the target deep neural network model are obtained by training the historical temperature data of the preset temperature field, and the initial classification result of the target temperature data is generated by using the target binary tree model and the target deep neural network model, and then the target temperature control score of the target temperature data is obtained based on the initial classification result, so as to trigger the corresponding temperature control operation of the preset temperature field based on the target temperature control score. In this way, the present application uses a large amount of real historical temperature data to train the target binary tree model and the target deep neural network model, determines the target temperature control score through the output results of the target binary tree model and the target deep neural network model, and regulates the water temperature of the temperature field according to the target temperature control score. In this way, the present application can respond more quickly to the optimization setting of the water temperature setting value, so that the economic benefits and safety are improved.

[0076] In some specific implementations, the training data construction module 11 may include:

[0077] A data cleaning unit, used for acquiring the historical temperature data of the preset temperature field, and cleaning the historical temperature data according to a preset data cleaning rule to obtain cleaned temperature data;

[0078] The first data classification unit is used to classify the post-cleaning temperature data based on a preset classification label; wherein the preset classification label is a freezing tube label or a safety label.

[0079] In some specific implementations, the data classification module 13 may include:

[0080] A second data classification unit is used to input the target temperature data into the target binary tree model and the target deep neural network model respectively to obtain a first classification result output by the target binary tree model and a second classification result output by the target deep neural network model; the first classification result and the second classification result include classification labels and classification probabilities.

[0081] In some specific embodiments, the temperature control module 14 may include:

[0082] A weight determination unit, used to determine preset score weights of the target binary tree model and the target deep neural network model;

[0083] A target temperature control score determining unit is used to determine the target temperature control score of the target temperature data based on the first classification result, the second classification result and the preset score weight.

[0084] In some specific embodiments, the temperature control device may further include:

[0085] A proofreading result acquisition module, used to determine a target classification result corresponding to the target temperature data based on the target temperature control score, and to acquire a proofreading result corresponding to the target classification result;

[0086] An accuracy determination module is used to determine the accuracy of the target classification result according to the proofreading result, and to update the preset score weight based on the accuracy, so as to determine the target temperature control score of the target temperature data based on the updated preset score weight.

[0087] In some specific embodiments, the temperature control module 14 may include:

[0088] a temperature change curve construction unit, configured to determine the target temperature control score of the target temperature data within a preset time period based on the initial classification result, and to construct a temperature change curve of the preset temperature field within the preset time period according to the target temperature control score;

[0089] A frozen pipe temperature curve construction unit, used for filtering target frozen pipe data from the historical temperature data based on preset data screening rules, and generating a corresponding frozen pipe temperature curve according to the target frozen pipe data by using the target binary tree model and the target deep neural network model;

[0090] The temperature control unit is used to determine the curve similarity between the temperature change curve and the freezing pipe temperature curve, and trigger the temperature control operation of the preset temperature field when the curve similarity is greater than a preset similarity threshold.

[0091] In some specific implementations, the model training module 12 may include:

[0092] A predicted temperature control score acquisition unit, used to acquire a predicted temperature control score corresponding to the training data output by the initial deep neural network model, and determine a target temperature control score corresponding to the training data based on the initial deep neural network model and the initial binary tree model;

[0093] A target deep neural network model determination unit is used to determine the deviation rate between the predicted temperature control score and the target temperature control score corresponding to the training data, and to update the preset initialization weight matrix in the initial neural network model based on the deviation rate to obtain the target deep neural network model.

[0094] Furthermore, the present application also discloses an electronic device. Figure 7 It is a structural diagram of an electronic device 20 shown according to an exemplary embodiment, and the content in the figure cannot be regarded as any limitation on the scope of use of this application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a power supply 23, a communication interface 24, an input and output interface 25 and a communication bus 26. Among them, the memory 22 is used to store a computer program, and the computer program is loaded and executed by the processor 21 to implement the relevant steps in the temperature control method disclosed in any of the aforementioned embodiments. In addition, the electronic device 20 in this embodiment can specifically be an electronic computer.

[0095] In this embodiment, the power supply 23 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 24 can create a data transmission channel between the electronic device 20 and the external device, and the communication protocol it follows is any communication protocol that can be applied to the technical solution of the present application, and is not specifically limited here; the input and output interface 25 is used to obtain external input data or output data to the outside world, and its specific interface type can be selected according to specific application needs and is not specifically limited here.

[0096] In addition, the memory 22, as a carrier for storing resources, can be a read-only memory, a random access memory, a disk or an optical disk, etc. The resources stored thereon can include an operating system 221, a computer program 222, etc., and the storage method can be temporary storage or permanent storage.

[0097] The operating system 221 is used to manage and control the hardware devices and computer program 222 on the electronic device 20, which can be Windows Server, Netware, Unix, Linux, etc. In addition to including a computer program that can be used to complete the temperature control method performed by the electronic device 20 disclosed in any of the aforementioned embodiments, the computer program 222 can further include a computer program that can be used to complete other specific tasks.

[0098] Furthermore, the present application also discloses a computer-readable storage medium for storing a computer program; wherein the computer program, when executed by a processor, implements the temperature control method disclosed above. The specific steps of the method can refer to the corresponding contents disclosed in the above embodiments, and will not be repeated here.

[0099] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0100] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0101] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented directly using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0102] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0103] The technical solution provided by the present application is introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea. At the same time, for general technicians in this field, according to the idea of ​​the present application, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present application.

Claims

1. A temperature control method, characterized in that: include: Acquire historical temperature data of a preset temperature field, and classify the historical temperature data based on preset labels, so as to construct training data based on the classified historical temperature data; Training an initial binary tree model and an initial deep neural network model based on the training data to obtain a target binary tree model and a target deep neural network model; Acquire current target temperature data of the preset temperature field, and input the target temperature data into the target binary tree model and the target deep neural network model respectively to generate an initial classification result of the target temperature data; A target temperature control score of the target temperature data is determined based on the initial classification result, and a temperature control operation corresponding to the preset temperature field is triggered according to the target temperature control score.

2. The temperature control method according to claim 1, characterized in that: The acquiring of historical temperature data of a preset temperature field and classifying the historical temperature data based on preset labels includes: Acquire the historical temperature data of the preset temperature field, and clean the historical temperature data according to a preset data cleaning rule to obtain cleaned temperature data; The post-cleaning temperature data is classified based on a preset classification label; wherein the preset classification label is a frozen tube label or a safety label.

3. The temperature control method according to claim 1, characterized in that: The inputting the target temperature data into the target binary tree model and the target deep neural network model respectively to generate an initial classification result of the target temperature data includes: The target temperature data is respectively input into the target binary tree model and the target deep neural network model to obtain a first classification result output by the target binary tree model and a second classification result output by the target deep neural network model; the first classification result and the second classification result include classification labels and classification probabilities.

4. The temperature control method according to claim 3, characterized in that: The determining a target temperature control score of the target temperature data based on the initial classification result includes: Determining preset score weights of the target binary tree model and the target deep neural network model; The target temperature control score of the target temperature data is determined based on the first classification result, the second classification result and the preset score weight.

5. The temperature control method according to claim 4, characterized in that: Also includes: Determine a target classification result corresponding to the target temperature data based on the target temperature control score, and obtain a proofreading result corresponding to the target classification result; The accuracy of the target classification result is determined according to the proofreading result, and the preset score weight is updated based on the accuracy, so as to determine the target temperature control score of the target temperature data based on the updated preset score weight.

6. The temperature control method according to claim 1, characterized in that: The determining of the target temperature control score of the target temperature data based on the initial classification result, and triggering the temperature control operation corresponding to the preset temperature field according to the target temperature control score, includes: Determine the target temperature control score of the target temperature data within a preset time period based on the initial classification result, and construct a temperature change curve of the preset temperature field within the preset time period according to the target temperature control score; Filtering target frozen pipe data from the historical temperature data based on preset data screening rules, and generating a corresponding frozen pipe temperature curve according to the target frozen pipe data using the target binary tree model and the target deep neural network model; The curve similarity between the temperature change curve and the freezing pipe temperature curve is determined, and the temperature control operation of the preset temperature field is triggered when the curve similarity is greater than a preset similarity threshold.

7. The temperature control method according to any one of claims 1 to 6, characterized in that: The process of training the initial binary tree model and the initial deep neural network model based on the training data to obtain the target binary tree model and the target deep neural network model also includes: Obtaining a predicted temperature control score corresponding to the training data output by the initial deep neural network model, and determining a target temperature control score corresponding to the training data based on the initial deep neural network model and the initial binary tree model; Determine a deviation rate between the predicted temperature control score and the target temperature control score corresponding to the training data, and update a preset initialization weight matrix in the initial neural network model based on the deviation rate to obtain the target deep neural network model.

8. A temperature control device, characterized in that: include: A training data construction module, used for acquiring historical temperature data of a preset temperature field and classifying the historical temperature data based on preset labels, so as to construct training data based on the classified historical temperature data; A model training module, used for training an initial binary tree model and an initial deep neural network model based on the training data to obtain a target binary tree model and a target deep neural network model; A data classification module, used to obtain the current target temperature data of the preset temperature field, and input the target temperature data into the target binary tree model and the target deep neural network model respectively, so as to generate an initial classification result of the target temperature data; A temperature control module is used to determine a target temperature control score of the target temperature data based on the initial classification result, and trigger a temperature control operation corresponding to the preset temperature field according to the target temperature control score.

9. An electronic device, characterized in that: The electronic device includes a processor and a memory; wherein the memory is used to store a computer program, and the computer program is loaded and executed by the processor to implement the temperature control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store a computer program, which, when executed by a processor, implements the temperature control method according to any one of claims 1 to 7.

Images