CN116066977A - Method and device for training air conditioner operating parameter prediction model, electronic equipment, storage medium - Google Patents

Abstract

The present application relates to the field of air-conditioning technology, and discloses a method for training an air-conditioning operating parameter prediction model, including: inputting the first air-conditioning operating data sample into a preset neural network model for training to obtain a reference air-conditioning operating parameter prediction model; The air conditioner operation parameter prediction model acquires a plurality of second air conditioner operation data samples with second labels; wherein, the second air conditioner operation data samples with parameters representing abnormal operation of the air conditioner are more than the first sample with parameters representing normal operation of the air conditioner The second air conditioner operation data sample; use the second air conditioner operation data sample and the first air conditioner operation data sample to carry out incremental learning on the reference air conditioner operation parameter prediction model, and the obtained air conditioner operation parameter prediction model improves the air conditioner operation parameter prediction model to predict air conditioner operation parameter accuracy. The application also discloses a device for training an air conditioner operating parameter prediction model, electronic equipment, and a storage medium.

Description

Method and device for training air conditioner operating parameter prediction model, electronic equipment, storage medium

technical field

The present application relates to the technical field of air conditioners, for example, to a method and device for training an air conditioner operating parameter prediction model, electronic equipment, and a storage medium.

Background technique

At present, with the wide application of air conditioners in daily life and work, people pay more and more attention to the operation status of air conditioners. In order to avoid machine damage caused by long-term failure of air conditioners, it is necessary to know the operating parameters of air conditioners in real time. Compared with using measurement tools such as sensors to directly measure the operating parameters of the air conditioner, it is more convenient and faster to use the air conditioner operating parameter prediction model.

In the process of implementing the embodiments of the present disclosure, it is found that at least the following problems exist in related technologies:

In the prior art, when obtaining an air conditioner operating parameter prediction model, all sample data is input into a preset neural network model for training at one time to obtain an air conditioner operating parameter prediction model. However, during the operation of the air conditioner, the time in the normal operation state is much longer than the time in the abnormal operation state. As a result, in the actual training process of the model, the number of data samples when the air conditioner is faulty is much smaller than the number of data samples when the air conditioner is in normal operation. The model is difficult to accurately predict the parameters of the abnormal operation of the air conditioner.

Contents of the invention

In order to provide a basic understanding of some aspects of the disclosed embodiments, a brief summary is presented below. The summary is not intended to be an extensive overview nor to identify key/important elements or to delineate the scope of these embodiments, but rather serves as a prelude to the detailed description that follows.

Embodiments of the present disclosure provide a method and device for training an air conditioner operating parameter prediction model, electronic equipment, and a storage medium, so as to improve the accuracy of the air conditioner operating parameter prediction model in predicting the air conditioner operating parameters.

In some embodiments, the method for training an air conditioner operation parameter prediction model includes: acquiring a first air conditioner operation data sample with a first label, where the first label is a parameter used to characterize the abnormal operation of the air conditioner or used To characterize the parameters of the normal operation of the air conditioner; input the first air conditioner operation data sample into the preset neural network model for training, and obtain a reference air conditioner operation parameter prediction model; obtain multiple models with the first The second air conditioner operating data sample with two tags, the second tag is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner; wherein, the second air conditioner operating data with parameters characterizing the abnormal operation of the air conditioner There are more samples than the second air conditioner operation data samples with parameters representing the normal operation of the air conditioner; using the second air conditioner operation data samples and the first air conditioner operation data samples to perform incremental learning on the reference air conditioner operation parameter prediction model , to obtain the air conditioner operating parameter prediction model.

In some embodiments, the device for training an air conditioner operation parameter prediction model includes: a first acquisition module configured to acquire a first air conditioner operation data sample with a first label, the first label is for A parameter characterizing the abnormal operation of the air conditioner or a parameter characterizing the normal operation of the air conditioner; the training module is configured to input the first air conditioner operation data sample into a preset neural network model for training, and obtain a reference air conditioner operation parameter prediction model; Two acquisition modules, configured to acquire a plurality of second air conditioner operation data samples with a second label according to the reference air conditioner operation parameter prediction model, the second label is a parameter used to characterize the abnormal operation of the air conditioner or is used to characterize The parameters of the normal operation of the air conditioner; wherein, the second air conditioner operation data samples with the parameters representing the abnormal operation of the air conditioner are more than the second air conditioner operation data samples with the parameters representing the normal operation of the air conditioner; the incremental learning module is configured to use The second air conditioner operation data sample and the first air conditioner operation data sample perform incremental learning on the reference air conditioner operation parameter prediction model to obtain an air conditioner operation parameter prediction model.

In some embodiments, the device for training an air conditioner operating parameter prediction model includes: a processor and a memory storing program instructions, and the processor is configured to execute the above-mentioned method when executing the program instructions. A method for training air conditioner operating parameter prediction models.

In some embodiments, the electronic device includes the above-mentioned device for training an air conditioner operating parameter prediction model.

In some embodiments, the storage medium stores program instructions, and when the program instructions are run, execute the above-mentioned method for training an air conditioner operating parameter prediction model.

The method and device, electronic equipment, and storage medium for training the air conditioner operating parameter prediction model provided by the embodiments of the present disclosure can achieve the following technical effects: by inputting the first air conditioner operating data sample into the preset neural network model for training, the obtained Referring to the air conditioner operation parameter prediction model, according to the reference air conditioner operation parameter prediction model, a plurality of second air conditioner operation data samples with a second label are obtained, and the second label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner parameters; wherein, the second air conditioner operating data samples with parameters representing the abnormal operation of the air conditioner are more than the second air conditioner operating data samples with parameters representing the normal operation of the air conditioner; using the second air conditioner operating data samples and the first air conditioner operating data samples The sample performs incremental learning on the reference air conditioner operating parameter prediction model to obtain the air conditioner operating parameter prediction model. In this way, by inputting the first air conditioner operation data sample into the preset neural network model for training, a reference air conditioner operation parameter prediction model is obtained, and then the first model with more parameters representing the abnormal operation of the air conditioner than the parameters representing the normal operation of the air conditioner is used. Two air conditioner operation data samples carry out incremental learning on the reference air conditioner operation parameter prediction model, and the obtained air conditioner operation parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner, so that the use of the air conditioner operation parameter prediction model can improve the accuracy of the air conditioner operation parameters. The accuracy of making predictions.

The foregoing general description and the following description are exemplary and explanatory only and are not intended to limit the application.

Description of drawings

One or more embodiments are exemplified by the corresponding drawings, and these exemplifications and drawings do not constitute a limitation to the embodiments, and elements with the same reference numerals in the drawings are shown as similar elements, The drawings are not limited to scale and in which:

FIG. 1 is a schematic diagram of a method for training an air conditioner operating parameter prediction model provided by an embodiment of the present disclosure;

Fig. 2 is a schematic diagram of another method for training an air conditioner operating parameter prediction model provided by an embodiment of the present disclosure;

Fig. 3 is a schematic diagram of another method for training an air conditioner operating parameter prediction model provided by an embodiment of the present disclosure;

Fig. 4 is a schematic diagram of another method for training an air conditioner operating parameter prediction model provided by an embodiment of the present disclosure;

Fig. 5 is a schematic structural diagram of a device for training an air conditioner operating parameter prediction model provided by an embodiment of the present disclosure;

Fig. 6 is a schematic structural diagram of another device for training an air conditioner operating parameter prediction model provided by an embodiment of the present disclosure.

Detailed ways

In order to understand the characteristics and technical content of the embodiments of the present disclosure in more detail, the implementation of the embodiments of the present disclosure will be described in detail below in conjunction with the accompanying drawings. The attached drawings are only for reference and description, and are not intended to limit the embodiments of the present disclosure. In the following technical description, for purposes of explanation, numerous details are set forth in order to provide a thorough understanding of the disclosed embodiments. However, one or more embodiments may be practiced without these details. In other instances, well-known structures and devices may be shown simplified in order to simplify the drawings.

The terms "first", "second" and the like in the description and claims of the embodiments of the present disclosure and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific sequence or sequence. It should be understood that the parameters so used are interchangeable under appropriate circumstances so as to facilitate the embodiments of the disclosed embodiments described herein. Furthermore, the terms "comprising" and "having", as well as any variations thereof, are intended to cover a non-exclusive inclusion.

Unless stated otherwise, the term "plurality" means two or more.

In the embodiments of the present disclosure, the character "/" indicates that the preceding and following objects are an "or" relationship. For example, A/B means: A or B.

The term "and/or" is an associative relationship describing objects, indicating that there can be three relationships. For example, A and/or B means: A or B, or, A and B, these three relationships.

The term "correspondence" may refer to an association relationship or a binding relationship, and the correspondence between A and B means that there is an association relationship or a binding relationship between A and B.

As shown in FIG. 1 , an embodiment of the present disclosure provides a method for training an air conditioner operating parameter prediction model, including:

Step S101, acquiring a first air conditioner operation data sample with a first label, where the first label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner;

Step S102, inputting the first air conditioner operation data sample into a preset neural network model for training to obtain a reference air conditioner operation parameter prediction model;

Step S103, acquiring a plurality of second air conditioner operation data samples with a second label according to the reference air conditioner operation parameter prediction model, where the second label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner;

Step S104, using the second air conditioner operation data sample and the first air conditioner operation data sample to perform incremental learning on the reference air conditioner operation parameter prediction model to obtain the air conditioner operation parameter prediction model.

Using the method for training the air conditioner operating parameter prediction model provided by the embodiments of the present disclosure, by inputting the first air conditioner operating data sample into the preset neural network model for training, a reference air conditioner operating parameter prediction model is obtained, and using The operating parameters are more than the second air conditioner operating data sample with parameters representing the normal operation of the air conditioner. The reference air conditioner operation parameter prediction model is incrementally learned, and the obtained air conditioner operation parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner. Prediction makes it possible to improve the accuracy rate of air conditioner operation parameter prediction by using the air conditioner operation parameter prediction model.

Optionally, the neural network model includes: a convolutional neural network and the like.

In some embodiments, the first air conditioner operating data sample is the condenser temperature and condenser pressure of the air conditioner at the first preset time. The first label attached to the first air conditioner operation data sample is the pressure at the discharge port of the compressor, the discharge temperature of the compressor, the pressure at the suction port of the compressor, and the temperature of the chilled water corresponding to the first preset moment. time, if the air conditioner is in an abnormal operation state, the pressure at the compressor exhaust port, the compressor discharge temperature, the compressor suction port pressure and the chilled water temperature at the first preset time are the parameters that characterize the abnormal operation of the air conditioner; At the first preset moment, if the air conditioner is in normal operation, the pressure at the compressor discharge port, the compressor discharge temperature, the pressure at the compressor suction port and the chilled water temperature at the first preset moment represent the normal operation of the air conditioner. parameters. For example, the first preset time is 8:00, 8:01, 8:02, etc. At each first preset time, the condenser temperature and condenser pressure of the air conditioner are obtained, and the pressure, compression The discharge temperature of the machine, the pressure of the suction port of the compressor, and the chilled water temperature, that is, to obtain multiple first air conditioner operation data samples with the first label; input multiple first air conditioner operation data samples with the first label into the preset The neural network model is trained to obtain the reference air conditioner operating parameter prediction model.

Optionally, acquiring a plurality of second air-conditioning operating data samples with a second label according to the reference air-conditioning operating parameter prediction model includes: obtaining the accuracy rate of the reference air-conditioning operating parameter prediction model; when the accuracy rate is greater than a preset threshold , to acquire a plurality of second air conditioner operation data samples with a second label. In this way, when the accuracy rate is greater than the preset threshold, the training of the reference air conditioner operation parameter prediction model is completed, and a plurality of second air conditioner operation data samples with the second label are obtained, and then the second air conditioner operation data samples are used to compare the reference air conditioner operation parameters. The parameter prediction model is incrementally learned to obtain the air conditioner operation parameter prediction model, which improves the accuracy of the air conditioner operation parameter prediction model in predicting the air conditioner operation parameters.

Optionally, obtaining the accuracy rate of the reference air conditioner operation parameter prediction model includes: obtaining multiple air conditioner operation data; obtaining measurement results corresponding to each air conditioner operation data; inputting each air conditioner operation data into the reference air conditioner operation parameter prediction model to obtain each air conditioner operation The prediction results corresponding to the data; respectively compare the prediction results of each air conditioner operation data with the measurement results of each air conditioner operation data to obtain multiple comparison results; the comparison results are used to represent whether the difference between the prediction results and the measurement results is within the preset range According to each comparison result, the accuracy rate of the reference air conditioner operation parameter prediction model is obtained. In this way, the accuracy rate of the reference air conditioner operation parameter prediction model can be obtained more accurately according to the comparison result between the prediction results of each air conditioner operation data and the measurement results of each air conditioner operation data.

Optionally, the air conditioner operating data is the condenser temperature and condenser pressure of the air conditioner at the second preset time. The condenser temperature and condenser pressure of the air conditioner are respectively obtained at different second preset times, that is, a plurality of air conditioner operating data are obtained.

Optionally, the measurement results corresponding to the air conditioner operation data are: the actual measurement of the pressure at the discharge port of the compressor, the discharge temperature of the compressor, the pressure at the suction port of the compressor, and the temperature of the chilled water at the corresponding second preset time value.

Optionally, the prediction result corresponding to the air conditioner operation data is: the pressure at the compressor discharge port, the compressor discharge temperature, the pressure at the compressor suction port and Output value for chilled water temperature.

Optionally, the accuracy rate of the reference air-conditioning operation parameter prediction model is obtained according to each comparison result, including: when the comparison result is that the difference between the prediction result of the air-conditioning operation data and the measurement result of the corresponding air-conditioning operation data is within a preset range , it is determined that the reference air conditioner operating parameter prediction model is accurate; the accuracy of the reference air conditioner operating parameter prediction model is obtained by dividing the number of accurate predictions by the reference air conditioner operation parameter prediction model by the total number of reference air conditioner operation parameter prediction models.

In conjunction with what is shown in FIG. 2 , an embodiment of the present disclosure provides another method for training an air conditioner operating parameter prediction model, including:

Step S201, acquiring a first air conditioner operation data sample with a first label, where the first label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner;

Step S202, inputting the first air conditioner operation data sample into a preset neural network model for training to obtain a reference air conditioner operation parameter prediction model;

Step S203, acquiring a plurality of air conditioner operating data;

Step S204, acquiring measurement results corresponding to each air conditioner operating data;

Step S205, input the operation data of each air conditioner into the reference air conditioner operation parameter prediction model to obtain the prediction results corresponding to each air conditioner operation data;

Step S206, respectively comparing the prediction results of each air conditioner operation data with the measurement results of each air conditioner operation data to obtain multiple comparison results; the comparison results are used to indicate whether the difference between the prediction results and the measurement results is within a preset range;

Step S207, obtaining the accuracy rate of the reference air conditioner operation parameter prediction model according to each comparison result;

Step S208, when the accuracy rate is greater than the preset threshold, obtain a plurality of second air conditioner operation data samples with a second label; the second label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner parameter;

Step S209, using the second air conditioner operation data sample and the first air conditioner operation data sample to perform incremental learning on the reference air conditioner operation parameter prediction model to obtain the air conditioner operation parameter prediction model.

In this way, the reference air conditioner operation parameter prediction model is obtained according to the first air conditioner operation data sample, and the prediction results of a plurality of air conditioner operation data are obtained according to the reference air conditioner operation parameter prediction model, thereby obtaining the prediction results of each air conditioner operation data and the measurement of each air conditioner operation data As a result of the comparison of the results, a plurality of second air-conditioning operation data samples with the second label are obtained according to each comparison result, and the second air-conditioning operation data samples with more parameters representing the abnormal operation data of the air conditioner are used than those with parameters representing the normal operation of the air conditioner The data samples carry out incremental learning on the reference air conditioner operation parameter prediction model, and the obtained air conditioner operation parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner, so that it can improve the ability to use the air conditioner operation parameter prediction model to predict the air conditioner operation parameters. Accuracy.

Optionally, acquiring a plurality of second air-conditioning operation data samples with a second label includes: determining the total number of second air-conditioning operation data samples with a second label according to the accuracy of the reference air-conditioning operation parameter prediction model, Determine the number of samples of the second air-conditioning operation data with parameters representing the abnormal operation of the air conditioner; according to the number of samples of the second air-conditioning operation data with parameters representing the abnormal operation of the air conditioner and the total number of samples, obtain a plurality of second air-conditioning operation data with the second label Air conditioner operation data sample. Optionally, determining the total number of samples of the second air-conditioning operation data samples with the second label according to the accuracy rate of the reference air-conditioning operation parameter prediction model includes: obtaining the incremental ratio according to the accuracy rate of the reference air-conditioning operation parameter prediction model; The first preset algorithm uses the incremental ratio to perform calculations to obtain the total number of samples of the second air conditioner operating data samples with the second label. In this way, according to the accuracy rate of the reference air conditioner operating parameter prediction model to obtain the incremental ratio, according to the incremental ratio to obtain the total sample size of the second air conditioner operating data sample with the second label, the second air conditioner operating data using the total sample size The sample performs incremental learning on the reference air conditioner operation parameter prediction model, and the obtained air conditioner operation parameter prediction model can improve the accuracy of air conditioner operation parameter prediction using the air conditioner operation parameter prediction model.

Optionally, the accuracy rate of the reference air conditioner operating parameter prediction model is directly proportional to the increment ratio.

Optionally, by calculating F=E*(1-N), the total number of samples of the second air-conditioning operation data samples with the second label is obtained; wherein, F is the second air-conditioning operation data sample with the second label The total number of samples, E is the total number of samples of the first air-conditioning operation data sample with the first label, and N is the increment ratio.

Optionally, the accuracy rate includes a fault accuracy rate and a normal accuracy rate, and determining the number of second air conditioner operating data samples with parameters representing the abnormal operation of the air conditioner includes: obtaining the fault accuracy rate of the reference air conditioner operating parameter prediction model; The failure accuracy rate of the operating parameter prediction model is obtained from the failure failure rate of the reference air conditioner operation parameter prediction model; the second preset algorithm is used to calculate the failure rate of the reference air conditioner operation parameter prediction model, and the parameter with the abnormal operation of the air conditioner is obtained. The number of samples of the second air conditioner operation data. In this way, the failure rate of the reference air conditioner operation parameter prediction model is obtained according to the failure accuracy rate of the reference air conditioner operation parameter prediction model, and the number of second air conditioner operation data samples with parameters representing the abnormal operation of the air conditioner is obtained according to the failure failure rate. A large number of second air conditioner operating data samples with parameters representing the abnormal operation of the air conditioner carry out incremental learning on the reference air conditioner operating parameter prediction model, and the obtained air conditioner operating parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner, so that The accuracy rate of predicting the air conditioner operation parameters by using the air conditioner operation parameter prediction model can be improved.

Optionally, the difference between the measurement result corresponding to the air conditioner operation data and the prediction result corresponding to the air conditioner operation data is within a preset range, and the measurement result corresponding to the air conditioner operation data is measured when the air conditioner is running normally, that is, the air conditioner does not report fault, then refer to the air conditioner operation data prediction model to predict the normal operation parameters of the air conditioner to be correct.

Optionally, the difference between the measurement result corresponding to the air conditioner operation data and the prediction result corresponding to the air conditioner operation data is within a preset range, and the measurement result corresponding to the air conditioner operation data is measured when the air conditioner is running abnormally, that is, the air conditioner reports a fault , then the abnormal operation parameters of the air conditioner are predicted to be correct with reference to the air conditioner operation data prediction model.

Optionally, mark and predict the air conditioner failure for the prediction results within the preset parameter failure range; mark and measure the air conditioner failure for the measurement results when the air conditioner fails, and obtain the failure accuracy rate of the reference air conditioner operating parameter prediction model, including: obtaining the mark with The first number of prediction results for predicting air-conditioning failure; obtaining the difference between the prediction result and the measurement result is within a preset range, and marking the second number of prediction results for predicting air-conditioning failure; obtaining the measurement result marked for measuring air-conditioning failure The third quantity: Acquiring the fault accuracy rate of the reference air conditioner operation parameter prediction model according to the first quantity, the second quantity and the third quantity.

Optionally, by calculating M=N _b ÷ (N _a +N _c ), the fault accuracy rate of the reference air-conditioning operation parameter prediction model is obtained; wherein, M is the fault accuracy rate of the reference air-conditioning operation parameter prediction model, and N _a is the mark There is the first number of prediction results for predicting air-conditioning failure, N _b is the difference between the prediction result and the measurement result is within a preset range, and the second number of prediction results for predicting air-conditioning failure is marked, and N _c is marked for measuring air-conditioning The third number of faulty measurements.

Optionally, by calculating P=1-M, the failure failure rate of the reference air conditioner operating parameter prediction model is obtained; wherein, P is the failure failure rate of the reference air conditioner operation parameter prediction model, and M is the failure accuracy of the reference air conditioner operation parameter prediction model Rate.

Optionally, by calculating D ₂ =D ₁ *P, the number of samples of the second air conditioner operation data with parameters representing the abnormal operation of the air conditioner is obtained; wherein, D ₂ is the second air conditioner operation with parameters representing the abnormal operation of the air conditioner The number of data samples, _D1 is the number of first air conditioner operation data samples with parameters that characterize the abnormal operation of the air conditioner, and P is the failure rate of the reference air conditioner operation parameter prediction model.

Optionally, by calculating G=FD ₂ , the number of samples of the second air conditioner operation data with parameters representing the normal operation of the air conditioner is obtained, wherein G is the number of samples of the second air conditioner operation data with parameters representing the normal operation of the air conditioner, D ₂ is the number of second air conditioner operation data samples with parameters representing abnormal operation of the air conditioner, and F is the total number of samples of the second air conditioner operation data samples with the second label.

Optionally, using the second air conditioner operation data sample and the first air conditioner operation data sample to carry out incremental learning on the reference air conditioner operation parameter prediction model to obtain the air conditioner operation parameter prediction model, including: combining the second air conditioner operation data sample with the first air conditioner operation data sample The operating data samples are fused to generate incremental learning samples, and the incremental learning samples are input into the reference air conditioner operating parameter prediction model for incremental learning to obtain the air conditioner operating parameter prediction model. In this way, the air conditioner operation data samples for training the air conditioner operation parameter prediction model are increased, thereby improving the accuracy of the air conditioner operation parameter prediction model in predicting the air conditioner operation parameter.

In conjunction with what is shown in FIG. 3 , an embodiment of the present disclosure provides another method for training an air conditioner operating parameter prediction model, including:

Step S301, acquiring a first air conditioner operation data sample with a first label, where the first label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner;

Step S302, inputting the first air conditioner operation data sample into a preset neural network model for training to obtain a reference air conditioner operation parameter prediction model;

Step S303, obtaining the accuracy rate of the reference air conditioner operation parameter prediction model;

Step S304, if the accuracy rate is greater than the preset threshold, determine the total number of samples of the second air-conditioning operation data samples with the second label according to the accuracy rate;

Step S305, determining the number of second air conditioner operation data samples with parameters representing abnormal operation of the air conditioner;

Step S306, according to the number of samples of the second air conditioner operation data and the total number of samples with parameters representing the abnormal operation of the air conditioner, a plurality of second air conditioner operation data samples with a second label are obtained, and the second label is used to represent the abnormal operation of the air conditioner Parameters or parameters used to characterize the normal operation of the air conditioner;

Step S307, using the second air conditioner operation data sample and the first air conditioner operation data sample to perform incremental learning on the reference air conditioner operation parameter prediction model to obtain the air conditioner operation parameter prediction model.

In this way, the reference air conditioner operation parameter prediction model is obtained according to the first air conditioner operation data sample, and the prediction results of a plurality of air conditioner operation data are obtained according to the reference air conditioner operation parameter prediction model, thereby obtaining the accuracy rate of the reference air conditioner operation parameter prediction model. In the case of a preset threshold, the training of the reference air conditioner operating parameter prediction model is completed, a plurality of second air conditioner operating data samples with the second label are obtained, and the reference air conditioner operating parameter prediction model is incrementally learned by using the second air conditioner operating data samples , get the air conditioner operation parameter prediction model, improve the air conditioner operation parameter prediction model to predict the accuracy of air conditioner operation parameters.

Optionally, after obtaining a plurality of comparison results, it also includes: when the difference between the prediction result corresponding to the air-conditioning operation data and the measurement result corresponding to the air-conditioning operation data is not within a preset range, performing an operation on the prediction result corresponding to the air-conditioning operation data Correction: inputting the corrected forecast result and the air conditioner operation data corresponding to the corrected forecast result into a reference air conditioner operation parameter prediction model for training.

Optionally, correcting the prediction result corresponding to the air-conditioning operation data includes: replacing the prediction result corresponding to the air-conditioning operation data with the measurement result corresponding to the air-conditioning operation data. In this way, when the difference between the prediction result corresponding to the air-conditioning operation data and the measurement result corresponding to the air-conditioning operation data is not within the preset range, the prediction result corresponding to the air-conditioning operation data is replaced with the measurement result corresponding to the air-conditioning operation data, and the replaced The air-conditioning operation data corresponding to the prediction results and the replaced prediction results are input into the reference air-conditioning operation parameter prediction model for training again, which improves the accuracy of the reference air-conditioning prediction model.

In conjunction with what is shown in FIG. 4 , an embodiment of the present disclosure provides another method for training an air conditioner operating parameter prediction model, including:

Step S401, acquiring a first air conditioner operation data sample with a first label, where the first label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner;

Step S402, inputting the first air conditioner operation data sample into a preset neural network model for training to obtain a reference air conditioner operation parameter prediction model;

Step S403, acquiring a plurality of air conditioner operating data;

Step S404, obtaining the measurement results corresponding to the air conditioner operation data;

Step S405, input the operation data of each air conditioner into the reference air conditioner operation parameter prediction model to obtain the prediction results corresponding to each air conditioner operation data;

Step S406, judging whether the difference between the prediction result corresponding to the air conditioner operation data and the measurement result corresponding to the air conditioner operation data is within a preset range; if the difference between the prediction result corresponding to the air conditioner operation data and the measurement result corresponding to the air conditioner operation data is not within a preset range If the difference between the prediction result corresponding to the air-conditioning operation data and the measurement result corresponding to the air-conditioning operation data is within the preset range, execute step S409;

Step S407, replace the prediction result corresponding to the air conditioner operation data with the measurement result corresponding to the air conditioner operation data; then execute step 408;

Step S408, input the replaced prediction result and the air conditioner operation data corresponding to the replaced prediction result into the reference air conditioner operation parameter prediction model for training; then return to step S405;

Step S409, obtain the accuracy rate of the reference air conditioner operation parameter prediction model according to each comparison result; then execute step S410;

Step S410, when the accuracy rate is greater than the preset threshold, obtain a plurality of second air conditioner operation data samples with a second label, the second label is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner parameter;

Step S411 , using the second air conditioner operation data sample and the first air conditioner operation data sample to perform incremental learning on the reference air conditioner operation parameter prediction model to obtain the air conditioner operation parameter prediction model.

In this way, the reference air conditioner operation parameter prediction model is obtained according to the first air conditioner operation data sample, and a plurality of air conditioner operation data prediction results are obtained according to the reference air conditioner operation parameter prediction model. The prediction results corresponding to the air conditioner operation data and the measurement results corresponding to the air conditioner operation data If the difference is not within the preset range, the reference air conditioner operation parameter prediction model is retrained according to the measurement results corresponding to the air conditioner operation data; In the case of setting the range, the accuracy rate of the reference air conditioner operation parameter prediction model is obtained according to the comparison results, and a plurality of second air conditioner operation data samples with the second label are obtained according to the accuracy rate, and the abnormal operation data of the air conditioner is used. The second air conditioner operating data sample with more parameters than the parameters representing the normal operation of the air conditioner performs incremental learning on the reference air conditioner operation parameter prediction model, and the obtained air conditioner operation parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner, The accuracy rate of predicting the air-conditioning operating parameters by using the air-conditioning operating parameter predicting model can be improved.

In some embodiments, within the time period T, air conditioner operation data is collected several times for several air conditioners, the collected air conditioner operation data is stored in the database S, and the first preset number of abnormal operation of the air conditioners is screened out in the database State air conditioner operation data sample A and the second preset number of air conditioner normal operation state air conditioner operation data samples B, A and B are synthesized into a first air conditioner operation data sample C with a first label, and the first air conditioner operation data Sample C is input to the preset neural network model for training to obtain a reference air conditioner operating parameter prediction model. Obtain the accuracy rate of the reference air conditioner operating parameter prediction model; when the accuracy rate is greater than a preset threshold, such as 90%, obtain the incremental ratio according to the accuracy rate, and obtain the second label with the second label according to the incremental ratio. Second, the total number of samples of air-conditioning operation data samples, determine the number of second air-conditioning operation data samples with parameters that characterize the abnormal operation of the air-conditioner; obtain according to the number of samples of the second air-conditioning operation data and the total number of samples with parameters that characterize the abnormal operation of the air-conditioner A plurality of second air conditioner operation data samples D with a second label, using the second air conditioner operation data sample D and the first air conditioner operation data sample C to carry out incremental learning on the reference air conditioner operation parameter prediction model to obtain the air conditioner operation parameter prediction model . In this way, by referring to the accuracy of the air conditioner operating parameter prediction model to obtain the incremental ratio, according to the incremental ratio to obtain the total number of samples of the second air conditioner operating data samples with the second label, thereby obtaining the total number of samples with the second label The second air conditioner operating data sample, using the second air conditioner operating data sample and the first air conditioner operating data sample to carry out incremental learning on the reference air conditioner operating parameter prediction model, the obtained air conditioner operating parameter prediction model improves the air conditioner operating parameter prediction model. Accuracy of air conditioner operating parameters.

As shown in FIG. 5 , an embodiment of the present disclosure provides an apparatus for training an air conditioner operating parameter prediction model, including a first acquisition module 501 , a training module 502 , a second acquisition module 503 and an incremental learning module 504 . The first acquisition module 501 is configured to acquire a first air conditioner operating data sample with a first label, the first label is a parameter for characterizing abnormal operation of the air conditioner or a parameter for characterizing normal operation of the air conditioner; the training module 502 is configured as Input the first air conditioner operation data sample into a preset neural network model for training to obtain a reference air conditioner operation parameter prediction model; the second acquisition module 503 is configured to obtain a plurality of second tags with the second label according to the reference air conditioner operation parameter prediction model Two air conditioner operation data samples, the second tag is a parameter used to characterize the abnormal operation of the air conditioner or a parameter used to characterize the normal operation of the air conditioner; wherein, there are more second air conditioner operation data samples with parameters representing the abnormal operation of the air conditioner than those with The second air-conditioning operation data sample of the parameters of the normal operation of the air conditioner; the incremental learning module 504 is configured to use the second air-conditioning operation data sample and the first air-conditioning operation data sample to perform incremental learning on the reference air-conditioning operation parameter prediction model to obtain the air-conditioning operation Parameter prediction model.

Using the device for training the air conditioner operating parameter prediction model provided by the embodiment of the present disclosure, by inputting the first air conditioner operating data sample into the preset neural network model for training, a reference air conditioner operating parameter prediction model is obtained, and the model with the characteristic of air conditioner abnormality is used The operating parameters are more than the second air conditioner operating data sample with parameters representing the normal operation of the air conditioner. The reference air conditioner operation parameter prediction model is incrementally learned, and the obtained air conditioner operation parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner. Prediction makes it possible to improve the accuracy rate of air conditioner operation parameter prediction by using the air conditioner operation parameter prediction model.

Optionally, the third acquisition model is configured to acquire a plurality of second air-conditioner operation data samples with the second label according to the reference air-conditioner operation parameter prediction model in the following manner: obtain the accuracy rate of the reference air-conditioner operation parameter prediction model; If the accuracy rate is greater than the preset threshold, a plurality of second air conditioner operation data samples with the second label are acquired.

Optionally, the third acquisition model is configured to acquire the accuracy of the reference air conditioner operating parameter prediction model in the following manner: acquire a plurality of air conditioner operating data; acquire measurement results corresponding to each air conditioner operating data; input each air conditioner operating data into the reference air conditioner The operation parameter prediction model obtains the prediction results corresponding to the operation data of each air conditioner; respectively compares the prediction results of each air conditioner operation data with the measurement results of each air conditioner operation data to obtain multiple comparison results; the comparison results are used to represent the prediction results and measurement results Whether the difference is within the preset range; according to each comparison result, the accuracy rate of the reference air-conditioning operation parameter prediction model is obtained.

Optionally, the third acquisition model is configured to acquire a plurality of second air conditioner operating data samples with the second label in the following manner: determine the second air conditioner with the second label according to the accuracy rate of the reference air conditioner operation parameter prediction model The total number of samples of the operation data samples, determine the number of second air conditioner operation data samples with parameters representing the abnormal operation of the air conditioner; obtain multiple A second air conditioner operating data sample with a second label.

Optionally, the device for training an air conditioner operating parameter prediction model further includes a correction module. The correction model is configured to correct the prediction result corresponding to the air conditioner operation data when the difference between the prediction result corresponding to the air conditioner operation data and the measurement result corresponding to the air conditioner operation data is not within a preset range after obtaining multiple comparison results; The corrected prediction result and the air conditioner operation data corresponding to the corrected prediction result are input into the reference air conditioner operation parameter prediction model for training.

Optionally, the correction module is configured to correct the prediction result corresponding to the air-conditioning operation data by replacing the prediction result corresponding to the air-conditioning operation data with the measurement result corresponding to the air-conditioning operation data.

As shown in FIG. 6 , an embodiment of the present disclosure provides an apparatus for training an air conditioner operating parameter prediction model, including a processor (processor) 600 and a memory (memory) 601 . Optionally, the device may further include a communication interface (Communication Interface) 602 and a bus 603. Wherein, the processor 600 , the communication interface 602 , and the memory 601 can communicate with each other through the bus 603 . The communication interface 602 can be used for information transmission. The processor 600 can call the logic instructions in the memory 601 to execute the method for training the air conditioner operating parameter prediction model of the above-mentioned embodiment.

In addition, the above logic instructions in the memory 601 may be implemented in the form of software function units and when sold or used as an independent product, may be stored in a computer-readable storage medium.

The memory 601, as a computer-readable storage medium, can be used to store software programs and computer-executable programs, such as program instructions/modules corresponding to the methods in the embodiments of the present disclosure. The processor 100 executes the function application and parameter processing by running the program instructions/modules stored in the memory 601 , that is, implements the method for training the air conditioner operating parameter prediction model in the above embodiments.

The memory 601 may include a program storage area and a parameter storage area, wherein the program storage area may store an operating system and an application program required by at least one function; the parameter storage area may store parameters created according to the use of the terminal device, etc. In addition, the memory 601 may include a high-speed random access memory, and may also include a non-volatile memory.

Using the device for training the air conditioner operating parameter prediction model provided by the embodiments of the present disclosure, by inputting the first air conditioner operating data sample into the preset neural network model for training, a reference air conditioner operating parameter prediction model is obtained, and then using The second air conditioner operation data sample with more parameters representing the normal operation of the air conditioner than the second air conditioner operation data sample carries out incremental learning on the reference air conditioner operation parameter prediction model, and the obtained air conditioner operation parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner Prediction is performed, so that the accuracy rate of predicting the air conditioner operation parameters by using the air conditioner operation parameter prediction model can be improved.

An embodiment of the present disclosure provides an electronic device, including the above-mentioned apparatus for training an air conditioner operating parameter prediction model. The electronic device trains by inputting the first air conditioner operation data sample into a preset neural network model to obtain a reference air conditioner operation parameter prediction model, and then utilizes a model with more parameters representing abnormal operation of the air conditioner than parameters representing normal operation of the air conditioner. The second air conditioner operation data sample performs incremental learning on the reference air conditioner operation parameter prediction model, and the obtained air conditioner operation parameter prediction model can more accurately predict the abnormal operation parameters of the air conditioner, making it possible to improve the use of the air conditioner operation parameter prediction model for air conditioner operation. The prediction accuracy of parameters.

Optionally, the electronic device includes: a computer or a server.

An embodiment of the present disclosure provides a storage medium, which stores program instructions, and when the program instructions are run, execute the above-mentioned method for training an air conditioner operating parameter prediction model.

An embodiment of the present disclosure provides a computer program product, the computer program product includes a computer program stored on a computer-readable storage medium, the computer program includes program instructions, and when the program instructions are executed by a computer, the The computer executes the above method for training an air conditioner operating parameter prediction model.

The above-mentioned computer-readable storage medium may be a transitory computer-readable storage medium, or a non-transitory computer-readable storage medium.

The technical solutions of the embodiments of the present disclosure can be embodied in the form of software products, which are stored in a storage medium and include one or more instructions to enable a computer device (which may be a personal computer, a server, or a network equipment, etc.) to perform all or part of the steps of the method described in the embodiments of the present disclosure. The aforementioned storage medium can be a non-transitory storage medium, including: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc. A medium that can store program code, or a transitory storage medium.

The above description and drawings sufficiently illustrate the embodiments of the present disclosure to enable those skilled in the art to practice them. Other embodiments may incorporate structural, logical, electrical, procedural, and other changes. The examples merely represent possible variations. Individual components and functions are optional unless explicitly required, and the order of operations may vary. Portions and features of some embodiments may be included in or substituted for those of other embodiments. Also, the terms used in the present application are used to describe the embodiments only and are not used to limit the claims. As used in the examples and description of the claims, the singular forms "a", "an" and "the" are intended to include the plural forms as well unless the context clearly indicates otherwise . Similarly, the term "and/or" as used in this application is meant to include any and all possible combinations of one or more of the associated listed ones. Additionally, when used in this application, the term "comprise" and its variants "comprises" and/or comprising (comprising) etc. refer to stated features, integers, steps, operations, elements, and/or The presence of a component does not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and/or groupings of these. Without further limitations, an element defined by the statement "comprising a ..." does not exclude the presence of additional identical elements in the process, method or apparatus comprising said element. Herein, what each embodiment focuses on may be the difference from other embodiments, and the same and similar parts of the various embodiments may refer to each other. For the method, product, etc. disclosed in the embodiment, if it corresponds to the method part disclosed in the embodiment, then the relevant part can refer to the description of the method part.

Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software may depend on the specific application and design constraints of the technical solution. Said artisans may implement the described functions using different methods for each particular application, but such implementation should not be regarded as exceeding the scope of the disclosed embodiments. The skilled person can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the embodiments disclosed herein, the disclosed methods and products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units may only be a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined Or it can be integrated into another system, or some features can be ignored, or not implemented. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Some or all of the units can be selected according to actual needs to implement this embodiment. In addition, each functional unit in the embodiments of the present disclosure may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to embodiments of the disclosure. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or part of code that includes one or more Executable instructions. In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than that disclosed in the description, and sometimes there is no specific agreement between different operations or steps. order. For example, two consecutive operations or steps may, in fact, be performed substantially concurrently, or they may sometimes be performed in the reverse order, depending upon the functionality involved. Each block in the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by dedicated hardware implemented in combination with computer instructions.

Claims (10)

1. A method for training an air conditioner operating parameter prediction model, comprising:

acquiring a first air conditioner operation data sample with a first tag, wherein the first tag is a parameter for representing abnormal operation of an air conditioner or a parameter for representing normal operation of the air conditioner;

inputting the first air conditioner operation data sample into a preset neural network model for training to obtain a reference air conditioner operation parameter prediction model;

acquiring a plurality of second air conditioner operation data samples with second labels according to the reference air conditioner operation parameter prediction model, wherein the second labels are parameters used for representing abnormal operation of the air conditioner or parameters used for representing normal operation of the air conditioner; wherein, the second air conditioner operation data sample with the parameter for representing the abnormal operation of the air conditioner is more than the second air conditioner operation data sample with the parameter for representing the normal operation of the air conditioner;

and performing incremental learning on the reference air conditioner operation parameter prediction model by using the second air conditioner operation data sample and the first air conditioner operation data sample to obtain an air conditioner operation parameter prediction model.

2. The method of claim 1, wherein obtaining a plurality of second air conditioner operation data samples with second tags according to the reference air conditioner operation parameter prediction model comprises:

Acquiring the accuracy of the reference air conditioner operation parameter prediction model;

and under the condition that the accuracy rate is larger than a preset threshold value, acquiring a plurality of second air conditioner operation data samples with second labels.

3. The method of claim 2, wherein obtaining the accuracy of the reference air conditioner operating parameter prediction model comprises:

acquiring a plurality of air conditioner operation data; obtaining a measurement result corresponding to each air conditioner operation data;

inputting the air conditioner operation data into the reference air conditioner operation parameter prediction model to obtain a prediction result corresponding to the air conditioner operation data;

respectively comparing the predicted result of each air conditioner operation data with the measured result of each air conditioner operation data to obtain a plurality of comparison results; the comparison result is used for representing whether the difference value between the prediction result and the measurement result is within a preset range;

and obtaining the accuracy of the reference air conditioner operation parameter prediction model according to each comparison result.

4. The method of claim 2, wherein obtaining a plurality of second air conditioner operation data samples with second tags comprises:

determining the total sample number of second air conditioner operation data samples with second labels according to the accuracy rate, and determining the second air conditioner operation data sample number with parameters representing abnormal operation of the air conditioner;

And obtaining a plurality of second air conditioner operation data samples with second labels according to the second air conditioner operation data sample number with parameters representing abnormal operation of the air conditioner and the total sample number.

5. The method of claim 3, wherein after obtaining the plurality of comparison results, further comprising:

correcting the predicted result corresponding to the air conditioner operation data under the condition that the difference value between the predicted result corresponding to the air conditioner operation data and the measured result corresponding to the air conditioner operation data is not in a preset range;

and inputting the corrected predicted result and the air conditioner operation data corresponding to the corrected predicted result into the reference air conditioner operation parameter prediction model for training.

6. The method of claim 5, wherein correcting the predicted result corresponding to the air conditioner operation data comprises:

and replacing the predicted result corresponding to the air conditioner operation data with the measured result corresponding to the air conditioner operation data.

7. An apparatus for training an air conditioner operating parameter prediction model, comprising:

the first acquisition module is configured to acquire a first air conditioner operation data sample with a first label, wherein the first label is a parameter for representing abnormal operation of the air conditioner or a parameter for representing normal operation of the air conditioner;

The training module is configured to input the first air conditioner operation data sample into a preset neural network model for training to obtain a reference air conditioner operation parameter prediction model;

the second acquisition module is configured to acquire a plurality of second air conditioner operation data samples with second labels according to the reference air conditioner operation parameter prediction model, wherein the second labels are parameters used for representing abnormal operation of the air conditioner or parameters used for representing normal operation of the air conditioner; wherein, the second air conditioner operation data sample with the parameter for representing the abnormal operation of the air conditioner is more than the second air conditioner operation data sample with the parameter for representing the normal operation of the air conditioner;

and the increment learning module is configured to perform increment learning on the reference air conditioner operation parameter prediction model by using the second air conditioner operation data sample and the first air conditioner operation data sample to obtain an air conditioner operation parameter prediction model.

8. An apparatus for training an air conditioner operating parameter prediction model comprising a processor and a memory storing program instructions, wherein the processor is configured, when executing the program instructions, to perform the method for training an air conditioner operating parameter prediction model of any one of claims 1 to 6.

9. An electronic device comprising the apparatus for training an air conditioner operation parameter prediction model as claimed in claim 8.

10. A storage medium storing program instructions which, when executed, perform the method for training an air conditioner operating parameter prediction model of any one of claims 1 to 6.

Images