Training method for prediction model, prediction method and apparatus for optimal operating parameters of air conditioner, and air conditioner

Abstract

A training method for a prediction model, a prediction method and apparatus for optimal operating parameters of an air conditioner, and an air conditioner. The training method comprises: collecting hardware parameters of an air conditioner under different heat exchanger dirt blockage states and different refrigerant quantities in an air conditioner enthalpy difference laboratory; determining optimal operating parameters of the air conditioner under different heat exchanger dirt blockage states and different refrigerant quantities, and taking the determined optimal operating parameters of the air conditioner as tag data; constructing a training data set on the basis of the heat exchanger dirt blockage states, the refrigerant quantities, the tag data, and the hardware parameters of the air conditioner, wherein the training data set comprises a plurality of data units, and each data unit corresponds to one heat exchanger dirt blockage state and one refrigerant quantity; and sequentially inputting the data units into an initial prediction model for training to obtain a target prediction model capable of predicting optimal operating parameters of an air conditioner.

Description

Training method of prediction model, prediction method and device of optimal operation parameter of air conditioner, air conditioner

[0001] The present application is based on and claims priority to Chinese Patent Application No. 202410886464.0, filed on July 3, 2024, the entire contents of which are hereby incorporated by reference into the present application. TECHNICAL FIELD

[0002] The present disclosure relates to the technical field of air conditioner control, for example to a training method of prediction model, a prediction method and device of optimal operation parameter of air conditioner, and an air conditioner. BACKGROUND

[0003] The optimal operation parameter of an air conditioner under various working conditions is an important factor for the air conditioner to achieve energy saving. The optimal operation parameter of the air conditioner is set under the standard working condition of the enthalpy difference room when the air conditioner is newly manufactured, and under the condition that there is no heat exchanger blockage and the refrigerant quantity is 100%. Setting accurate optimal operation parameters can ensure that the air conditioner operates at the optimal energy consumption ratio in the latest state.

[0004] However, as the use time of the air conditioner increases, the air conditioner will have soft faults such as heat exchanger blockage and refrigerant leakage, which will cause the degradation of the heating or cooling performance of the air conditioner, and the optimal control parameter of the air conditioner will deviate, which will cause the previously set optimal operation parameter under the standard state to be unable to make the air conditioner operate at the true optimal state, resulting in energy waste.

[0005] It should be noted that the information disclosed in the above background section is only used to strengthen the understanding of the background of the present disclosure, and therefore can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] In order to have a basic understanding of some aspects of the disclosed embodiments, a brief summary is given below. The summary is not a general review, nor is it intended to determine key / important components or delineate the scope of protection of these embodiments, but as a prelude to the detailed description below.

[0007] The present disclosure provides a training method of prediction model, a prediction method and device of optimal operation parameter of air conditioner, and an air conditioner, which can train a prediction model for predicting optimal operation parameters that meet the current actual conditions of the air conditioner, so that the air conditioner can operate at the optimal state based on the optimal operation parameters predicted by the prediction model, reducing energy waste.

[0008] According to a first aspect of the present disclosure, a training method of a prediction model for predicting optimal operation parameters of an air conditioner is provided, comprising:

[0009] Collect air conditioner hardware parameters under different heat exchanger dirty blocking states and different refrigerant amounts in an air conditioner enthalpy difference room;

[0010] Determine the optimal operation parameters of the air conditioner under different heat exchanger dirty blocking states and different refrigerant amounts, and determine the optimal operation parameters of the air conditioner as label data;

[0011] Based on the heat exchanger dirty blocking state, the refrigerant amount, the label data, and the air conditioner hardware parameters, a training data set is constructed, wherein the training data set includes a plurality of data units, each data unit includes the heat exchanger dirty blocking state and the refrigerant amount, and the air conditioner hardware parameters and the label data under the heat exchanger dirty blocking state and the refrigerant amount;

[0012] Each data unit is sequentially input into an initial prediction model for training to obtain a target prediction model capable of predicting the optimal operation parameters of the air conditioner.

[0013] According to a second aspect of the present disclosure, a method for predicting optimal operation parameters of an air conditioner is provided, comprising:

[0014] Collecting current air conditioner hardware parameters of the air conditioner;

[0015] Determining the current heat exchanger dirty blocking state and the refrigerant amount of the air conditioner;

[0016] Inputting the air conditioner hardware parameters, the heat exchanger dirty blocking state, and the refrigerant amount into a target prediction model, wherein the target prediction model is obtained by using the training method for predicting the optimal operation parameters of the air conditioner provided by the second aspect of the present disclosure;

[0017] Using the target prediction model to predict the current optimal operation parameters of the air conditioner.

[0018] In some embodiments, inputting the air conditioner hardware parameters into the target prediction model comprises:

[0019] Based on the air conditioner hardware parameters, the heat exchanger dirty blocking state, and the refrigerant amount, n sequences are constructed;

[0020] Inputting sequence 1 to sequence (n-1) into a corresponding second network unit, and inputting sequence n into the first network unit.

[0021] According to a third aspect of the present disclosure, a training device for a prediction model for predicting optimal operation parameters of an air conditioner is provided, comprising a processor and a memory storing program instructions, the processor being configured to execute the training method for predicting the optimal operation parameters of the air conditioner provided by the first aspect of the present disclosure when the program instructions are executed.

[0022] According to a fourth aspect of the present disclosure, a prediction device for optimal operation parameters of an air conditioner is provided, comprising a processor and a memory storing program instructions, the processor being configured to execute the prediction method for optimal operation parameters of an air conditioner provided by the second aspect of the present disclosure when the program instructions are executed.

[0023] According to a fifth aspect of the present disclosure, an air conditioner is provided, comprising:

[0024] an air conditioner body;

[0025] the training device for the prediction model for predicting optimal operation parameters of an air conditioner provided by the third aspect of the present disclosure is arranged in the air conditioner body; and / or

[0026] the prediction device for optimal operation parameters of an air conditioner provided by the fourth aspect of the present disclosure is arranged in the air conditioner body.

[0027] The foregoing general description and the following description are merely exemplary and explanatory, and are not intended to limit the present disclosure. BRIEF DESCRIPTION OF DRAWINGS

[0028] One or more embodiments are exemplarily illustrated by the accompanying drawings corresponding thereto, which do not constitute a limitation on the embodiments, elements with the same reference numerals in the drawings are shown as similar elements, the drawings do not constitute a proportional limitation, and wherein:

[0029] FIG. 1 is a flow diagram of a training method for a prediction model for predicting optimal operation parameters of an air conditioner according to an embodiment of the present disclosure;

[0030] FIG. 2 is a structural diagram of a prediction model according to an embodiment of the present disclosure;

[0031] FIG. 3 is a flow diagram of a prediction method for optimal operation parameters of an air conditioner according to an embodiment of the present disclosure;

[0032] FIG. 4 is a structural diagram of a training device for a prediction model for predicting optimal operation parameters of an air conditioner according to an embodiment of the present disclosure;

[0033] FIG. 5 is a structural diagram of a prediction device for optimal operation parameters of an air conditioner according to an embodiment of the present disclosure;

[0034] FIG. 6 is a structural diagram of an air conditioner according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0035] In order to enable a person skilled in the art to more fully understand the features and technical contents of the embodiments of the present disclosure, the implementation of the embodiments of the present disclosure is described in detail below with reference to the accompanying drawings, which are only used for reference and do not limit the embodiments of the present disclosure. In the following technical description, in order to facilitate explanation, a plurality of details are provided to provide a full understanding of the disclosed embodiments. However, one or more embodiments can still be implemented without these details. In other cases, well-known structures and devices can be simplified to facilitate the drawings.

[0036] Any reference in the specification to any prior art is not, and should not be taken as, an acknowledgment or any form of suggestion that this prior art forms part of the common general knowledge in the field of the application or in any other field, or that this prior art could be practiced or be a prior publication under the applicable jurisdiction.

[0037] 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 do not necessarily have to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented. In addition, the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusion.

[0038] Unless otherwise specified, the term "a plurality of" means two or more.

[0039] In the embodiments of the present disclosure, the character " / " represents an "or" relationship between the objects before and after it. For example, A / B represents: A or B.

[0040] The term "and / or" is a description of the association relationship between objects, which means that there can be three relationships. For example, A and / or B means: A or B, or, A and B, the three relationships.

[0041] The term "corresponding" can refer to an association relationship or a binding relationship. A and B correspond to each other means that there is an association relationship or a binding relationship between A and B.

[0042] The embodiments of the present disclosure provide a training device for training a prediction model for predicting optimal operating parameters of an air conditioner. The training device is a device with computing capability, for example, the training device can be a personal computer, a server, or a network device, etc. The electronic device can be used to execute the training method for training a prediction model for predicting optimal operating parameters of an air conditioner provided by the embodiments of the present disclosure.

[0043] In combination with the training device provided by the embodiments of the present disclosure, the embodiments of the present disclosure provide a training method for training a prediction model for predicting optimal operating parameters of an air conditioner. In combination with FIG. 1, the training method for training a prediction model for predicting optimal operating parameters of an air conditioner includes:

[0044] S101, collecting air conditioner hardware parameters under different heat exchanger dirty blocking states and different refrigerant amounts in an air conditioner enthalpy difference room;

[0045] S102, determining air conditioner optimal operation parameters under different heat exchanger dirty blocking states and different refrigerant amounts, and taking the determined air conditioner optimal operation parameters as label data;

[0046] S103, constructing a training data set based on the heat exchanger dirty blocking state, the refrigerant amount, the label data, and the air conditioner hardware parameters;

[0047] S104, sequentially inputting each data unit into an initial prediction model for training to obtain a target prediction model capable of predicting air conditioner optimal operation parameters.

[0048] The air conditioner enthalpy difference room is a professional laboratory mainly used for testing the performance of air conditioners, heat pumps and other refrigeration equipment, especially their key indicators such as refrigerating capacity, heating capacity, energy efficiency ratio under different working conditions. In the present embodiment, the heat exchanger dirty blocking state and the refrigerant amount of the hole field can be changed in the air conditioner enthalpy difference room, and the air conditioner hardware parameters can be collected under different heat exchanger dirty blocking states and different refrigerant amounts.

[0049] In the present embodiment, the air conditioner hardware parameters include exhaust temperature, suction temperature, defrost sensor temperature, outer ring temperature, subcooler bypass inlet temperature, total current, outer valve opening, outer coil temperature, auxiliary valve opening, plate heat exchanger inlet temperature, plate heat exchanger outlet temperature, oil temperature, compressor frequency, expansion valve opening, inner fan speed and outer fan speed, etc.

[0050] Regarding S103, in the present embodiment, the training data set includes multiple data units, each data unit containing the heat exchanger dirty blocking state and the refrigerant amount, and the air conditioner hardware parameters and the label data under the heat exchanger dirty blocking state and the refrigerant amount.

[0051] The present embodiment provides a training method of a prediction model for predicting air conditioner optimal operation parameters. The training data set is constructed based on the air conditioner hardware parameters and the air conditioner optimal operation parameters under different heat exchanger dirty blocking states and different refrigerant amounts, and the heat exchanger dirty blocking state and the refrigerant amount corresponding to the air conditioner hardware parameters and the air conditioner optimal operation parameters. The prediction model is trained by using the training data set, so that the prediction model has the ability to predict the air conditioner optimal operation parameters. By inputting the current heat exchanger dirty blocking state, the refrigerant amount and the air conditioner hardware parameters into the prediction model, the optimal operation parameters conforming to the current actual situation of the air conditioner can be obtained by using the prediction model, so that the air conditioner can operate in the optimal state based on the predicted optimal operation parameters, achieve a higher energy consumption ratio, and reduce energy waste.

[0052] In some embodiments, the training data set is constructed based on the heat exchanger dirty block state, the refrigerant amount, the label data, and the air conditioner hardware parameters, including: arranging temperature points and pressure points of the air conditioner enthalpy difference room as input features of a preset algorithm; preprocessing the collected air conditioner hardware parameters by using the preset algorithm, wherein the preprocessing includes rejecting non-steady state data and feature extraction; dividing the heat exchanger dirty block state and the refrigerant amount, and the air conditioner hardware parameters and the label data under the heat exchanger dirty block state and the refrigerant amount into a data unit, to obtain a plurality of data units.

[0053] In the embodiments of the present disclosure, the preset algorithm can be a SHAP (SHapley Additive exPlanations) algorithm. The SHAP algorithm is an algorithm for explaining the prediction results of a machine learning model, which provides the explainability of the model by calculating the contribution of each feature to the model output. Here, the supercooler bypass inlet temperature is the most important modeling input variable, showing a Shapley value of about 0.8. This indicates that the opening degree of the supercooler bypass valve has an important influence on the heat exchanger dirty block. The Shapley values of the supercooler bypass inlet temperature, the condenser coil temperature a, the high-pressure dry pipe temperature, the fan supply air temperature, and the evaporator high-pressure branch pipe temperature are all more than 0.2, which are the top five input variables. According to relevant professional knowledge, the condensing temperature, the opening degree of the valve, and the heat exchange temperature difference are important variables affecting the degree of heat exchanger dirty block, and the features directly or indirectly affecting these variables are very consistent with the feature importance ranking results, which means that the developed prediction model can be explained according to the domain knowledge.

[0054] In the embodiments of the present disclosure, in order to make the prediction model converge quickly during training, the air conditioner hardware parameters can be normalized. For example, the maximum value of the compressor frequency is 116 Hz, the maximum value of the expansion valve opening degree is 480, the maximum value of the indoor fan speed is 1200 RPM, and the maximum value of the outdoor fan speed is 1200 RPM. The compressor frequency, the expansion valve opening degree, the indoor fan speed, and the outdoor fan speed are divided by their maximum values, respectively, so as to normalize the range of the compressor frequency, the expansion valve opening degree, the indoor fan speed, and the outdoor fan speed to 0 to 1, which can improve the convergence speed during model training.

[0055] Through analysis of the relationship between the running state of the air conditioner and the hardware parameters of the air conditioner, there is a certain sequence and coupling between the running parameters of the air conditioner. Specifically, in the embodiments of the present disclosure, the number of optimal running parameters of the air conditioner is multiple, and the optimal running parameters of the air conditioner have sequence and coupling. Based on this, the prediction model of the embodiments of the present disclosure uses an improved recurrent neural network. The recurrent neural network is a kind of weight sharing artificial neural network, that is, the current output of a sequence is also related to the previous output. The specific form of expression is that the network will memorize the previous information and apply it to the calculation of the current output, that is, the nodes between the hidden layers are no longer unconnected but connected, and the input of the hidden layer includes not only the output of the input layer but also the output of the hidden layer at the previous moment. In combination with FIG. 2, here, the initial prediction model using the improved recurrent neural network includes a first network unit corresponding to each optimal running parameter of the air conditioner, and each first network unit corresponding to an optimal running parameter of the air conditioner is used to predict the optimal running parameter of the air conditioner. The first network units corresponding to each optimal running parameter of the air conditioner are connected in series according to the sequence of the optimal running parameters of the air conditioner, and the output result of the subsequent first network unit is related to the output result of the previous first network unit.

[0056] In the embodiments of the present disclosure, the optimal running parameters of the air conditioner include an optimal compressor frequency, an optimal expansion valve opening degree, an optimal indoor fan speed and an optimal outdoor fan speed. The sequence of the optimal compressor frequency, the optimal expansion valve opening degree, the optimal indoor fan speed and the optimal outdoor fan speed increases in turn, and the first network unit corresponding to the optimal compressor frequency, the first network unit corresponding to the optimal expansion valve opening degree, the first network unit corresponding to the optimal indoor fan speed and the first network unit corresponding to the optimal outdoor fan speed are connected in series.

[0057] The output result of the first network unit corresponding to the optimal expansion valve opening degree is related to the output result of the first network unit corresponding to the optimal compressor frequency, the output result of the first network unit corresponding to the optimal indoor fan speed is related to the output result of the first network unit corresponding to the optimal expansion valve opening degree, and the output result of the first network unit corresponding to the optimal outdoor fan speed is related to the output result of the first network unit corresponding to the optimal indoor fan speed.

[0058] In the embodiments of the present disclosure, the initial prediction model further includes a plurality of second network units, and the plurality of second network units are connected in series, and the last second network unit is connected with the first network unit at the front. The output result of the subsequent second network unit is related to the output result of the previous second network unit, and the output result of the first network unit at the front is related to the output result of the last second network unit. In combination with FIG. 2, the output result of the first network unit corresponding to the optimal compressor frequency is related to the output result of the last second network unit.

[0059] In the embodiments of the present disclosure, the data units are sequentially input into the initial prediction model for training, including: sequentially inputting each data unit to perform the following steps: constructing n sequences based on the data unit, wherein the number of the second network units is (n-1); inputting sequence 1 to sequence (n-1) into a corresponding second network unit respectively, and inputting sequence n into the first network unit respectively; training the initial prediction model based on the input sequences.

[0060] The embodiments of the present disclosure provide a prediction device for optimal operation parameters of an air conditioner. The prediction device is a device with computing capability, for example, the prediction device can be a personal computer, a server, or a network device, etc. The electronic device can be used to execute the prediction method for optimal operation parameters of an air conditioner provided by the embodiments of the present disclosure.

[0061] In combination with the prediction device provided by the embodiments of the present disclosure, the embodiments of the present disclosure provide a prediction method for optimal operation parameters of an air conditioner. In combination with FIG. 3, the prediction method includes:

[0062] S301, collecting current air conditioner hardware parameters of the air conditioner;

[0063] S302, determining a dirty block state of a heat exchanger and a refrigerant amount of the air conditioner;

[0064] S303, inputting the air conditioner hardware parameters, the dirty block state of the heat exchanger, and the refrigerant amount into a target prediction model;

[0065] S304, predicting the current optimal operation parameters of the air conditioner by using the target prediction model.

[0066] Regarding S303, in the embodiments of the present disclosure, the target prediction model is obtained by the training method of the prediction model for predicting the optimal operation parameters of the air conditioner provided in the above embodiments.

[0067] In some embodiments, inputting the air conditioner hardware parameters into the target prediction model includes: constructing n sequences based on the air conditioner hardware parameters, the dirty block state of the heat exchanger, and the refrigerant amount; inputting sequence 1 to sequence (n-1) into a corresponding second network unit respectively, and inputting sequence n into the first network unit respectively.

[0068] In combination with FIG. 4, the disclosure embodiment provides a training device 400 for a prediction model for predicting optimal operating parameters of an air conditioner, the training device 400 comprising a processor 401 and a memory 402. Optionally, the training device 400 can further comprise a communication interface 403 and a bus 404. Wherein the processor 401, the communication interface 403, the memory 402 can complete the communication among each other through the bus 404. The communication interface 403 can be used for information transmission. The processor 401 can call the logical instructions in the memory 402 to execute the training method of the prediction model for predicting optimal operating parameters of an air conditioner of the above-mentioned embodiments.

[0069] In addition, the logical instructions in the memory 402 described above can be implemented in the form of a software functional unit and sold or used as an independent product when used, which can be stored in a computer readable storage medium.

[0070] The memory 402 as a kind of computer readable storage medium can be used to store software programs, computer executable programs, such as program instructions / modules corresponding to the method in the disclosure embodiment. The processor 401 executes the program instructions / modules stored in the memory 402, thereby executing functional applications and data processing, i.e. implementing the training method of the prediction model for predicting optimal operating parameters of an air conditioner in the above-mentioned embodiments.

[0071] The memory 402 can include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required by a function; The data storage area can store data created according to the use of the terminal device and the like. In addition, the memory 402 can include a high-speed random access memory, and can also include a non-volatile memory.

[0072] In combination with FIG. 5, the disclosure embodiment provides a prediction device 500 for optimal operating parameters of an air conditioner, the prediction device 500 comprising a processor 501 and a memory 502. Optionally, the prediction device 500 can further comprise a communication interface 503 and a bus 504. Wherein the processor 501, the communication interface 503, the memory 502 can complete the communication among each other through the bus 504. The communication interface 503 can be used for information transmission. The processor 501 can call the logical instructions in the memory 502 to execute the prediction method of optimal operating parameters of an air conditioner of the above-mentioned embodiments.

[0073] In addition, the logic instructions in the memory 502 described above can be implemented in the form of a software function unit and sold or used as an independent product, which can be stored in a computer readable storage medium.

[0074] The memory 502 as a computer readable storage medium can be used to store software programs, computer executable programs, such as program instructions / modules corresponding to the method in the embodiments of the present disclosure. The processor 501 executes the function application and data processing by running the program instructions / modules stored in the memory 502, that is, realizes the prediction method of the optimal operating parameter of the air conditioner in the above-mentioned embodiments.

[0075] The memory 502 can include a program storage area and a data storage area, wherein the program storage area can store an operating system and application programs required by at least one function; the data storage area can store data created according to the use of the terminal device, etc. In addition, the memory 502 can include a high-speed random access memory, and can also include a non-volatile memory.

[0076] In combination with FIG. 6, the embodiments of the present disclosure provide an air conditioner 600, which includes an air conditioner body 601, and further includes the training device 400 for predicting the prediction model of the optimal operating parameter of the air conditioner and / or the prediction device 500 for predicting the optimal operating parameter of the air conditioner provided in the above-mentioned embodiments. In the case where the air conditioner 600 includes the training device 400, the training device 400 is arranged in the air conditioner body 601. In the case where the air conditioner 600 includes the prediction device 500, the prediction device 500 is arranged in the air conditioner body 601.

[0077] The embodiments of the present disclosure provide a computer readable non-transitory storage medium, which stores program instructions, and the program instructions execute the following steps when running:

[0078] Collecting air conditioner hardware parameters under different heat exchanger dirty blocking states and different refrigerant amounts in an air conditioner enthalpy difference room;

[0079] Determining the optimal operating parameter of the air conditioner under different heat exchanger dirty blocking states and different refrigerant amounts, and taking the determined optimal operating parameter of the air conditioner as label data;

[0080] Constructing a training data set based on the heat exchanger dirty blocking state, the refrigerant amount, the label data and the air conditioner hardware parameters, wherein the training data set includes a plurality of data units, each data unit contains the heat exchanger dirty blocking state and the refrigerant amount, and the air conditioner hardware parameters and the label data under the heat exchanger dirty blocking state and the refrigerant amount;

[0081] The data units are sequentially input into an initial prediction model for training, and a target prediction model capable of predicting optimal operation parameters of the air conditioner is obtained.

[0082] The embodiments of the present disclosure provide a computer-readable non-transitory storage medium storing program instructions, and the program instructions perform the following steps when executed:

[0083] Collecting current air conditioner hardware parameters of the air conditioner;

[0084] Determining a current heat exchanger dirty state and a refrigerant amount of the air conditioner;

[0085] Inputting the air conditioner hardware parameters, the heat exchanger dirty state and the refrigerant amount into the target prediction model;

[0086] Predicting the current optimal operation parameters of the air conditioner by using the target prediction model.

[0087] The embodiments of the present disclosure provide a computer program, which, when executed by a computer, causes the computer to implement the training method of the prediction model for predicting optimal operation parameters of an air conditioner or the prediction method of optimal operation parameters of an air conditioner.

[0088] The embodiments of the present disclosure provide a computer program product, which includes computer instructions stored on a computer-readable storage medium, and the program instructions, when executed by a computer, cause the computer to implement the training method of the prediction model for predicting optimal operation parameters of an air conditioner or the prediction method of optimal operation parameters of an air conditioner.

[0089] The technical solutions of the embodiments of the present disclosure can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes one or more instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute 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, such as a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.

[0090] The training method of the prediction model, the prediction method of optimal operation parameters of an air conditioner, the device and the air conditioner provided by the embodiments of the present disclosure can achieve the following technical effects:

[0091] The training method of the prediction model for predicting optimal operation parameters of an air conditioner provided by the embodiments of the present disclosure comprises the following steps: constructing a training data set based on the air conditioner hardware parameters and the optimal operation parameters of the air conditioner under different heat exchanger dirty blocking states and different refrigerant amounts, and the heat exchanger dirty blocking states and the refrigerant amounts corresponding to the air conditioner hardware parameters and the optimal operation parameters of the air conditioner; training the prediction model by using the training data set, so that the prediction model has the ability to predict the optimal operation parameters of the air conditioner. By inputting the current heat exchanger dirty blocking state, the refrigerant amount and the air conditioner hardware parameters into the prediction model, the optimal operation parameters conforming to the current actual situation of the air conditioner can be obtained by using the prediction model, so that the air conditioner can be operated in an optimal state based on the predicted optimal operation parameters, a higher energy consumption ratio can be achieved, and energy waste can be reduced.

[0092] The above description and drawings sufficiently illustrate the embodiments of the present disclosure to enable one skilled in the art to practice them. Other embodiments can include structural, logical, electrical, process, and other changes. The embodiments are merely representative of possible variations. Individual components and functions are optional unless explicitly required, and the order of operations can be varied. Parts and features of some embodiments can be included or substituted for parts and features of other embodiments. Also, the words used in this application are for describing the embodiments and are not intended to limit the claims. As used in the description of the embodiments and 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 refers to any and all possible combinations of one or more associated listed items. In addition, when used in this application, the term "comprise" and its variations "comprises" and / or "comprising" and the like mean the presence of the stated feature, integer, step, operation, element, and / or component, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. Without more limitations, the element defined by the phrase "comprising a" does not exclude the presence of additional identical elements in the process, method, or device including the stated element. In this document, each embodiment focuses on the differences from other embodiments, and the same or similar parts between various embodiments can be referred to each other. For the method, product, etc. disclosed by the embodiments, if it corresponds to the method part disclosed by the embodiments, the relevant part can be referred to the description of the method part.

[0093] Those skilled in the art can understand that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. The skilled person can use different methods for each specific application to realize the described functions, but such implementation should not be considered beyond the scope of the embodiments of the present disclosure. The skilled person can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.

[0094] In the embodiments disclosed herein, the disclosed methods, products (including but not limited to devices, equipment, etc.) can be implemented in other ways. For example, the above-described device embodiments are only schematic, for example, the division of the units can only be a logical function division, and actual implementation can have another division manner, for example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interface, device or unit, and can be electrical, mechanical or other forms. The units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to implement the embodiments. In addition, each functional unit in the embodiments of the present disclosure can be integrated in one processing unit, or each unit can be a physically independent unit, or two or more units can be integrated in one unit.

[0095] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing device to cause a series of operational steps to be performed on the computer, other programmable apparatus or other processing device to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions / acts specified in the flowchart and / or block diagram block or blocks.

Claims

1. A training method of a prediction model for predicting optimal operating parameters of an air conditioner, comprising: collecting air conditioner hardware parameters under different heat exchanger fouling states and different refrigerant amounts in an air conditioner enthalpy difference chamber; determining optimal operating parameters of the air conditioner under different heat exchanger fouling states and different refrigerant amounts, and taking the determined optimal operating parameters of the air conditioner as label data; constructing a training data set based on the heat exchanger fouling state, the refrigerant amount, the label data, and the air conditioner hardware parameters, wherein the training data set comprises a plurality of data units, each data unit containing the heat exchanger fouling state and the refrigerant amount, and the air conditioner hardware parameters and the label data under the heat exchanger fouling state and the refrigerant amount; sequentially inputting each data unit into an initial prediction model for training to obtain a target prediction model capable of predicting the optimal operating parameters of the air conditioner.

2. The method of claim 1, wherein, Constructing a training data set based on the heat exchanger fouling state, the refrigerant amount, the label data, and the air conditioner hardware parameters comprises: arranging temperature points and pressure points of the air conditioner enthalpy difference chamber as input features of a preset algorithm; preprocessing the collected air conditioner hardware parameters using the preset algorithm, wherein the preprocessing includes removing non-steady-state data and feature extraction; dividing the heat exchanger fouling state and the refrigerant amount, and the air conditioner hardware parameters and the label data under the heat exchanger fouling state and the refrigerant amount into one data unit to obtain a plurality of data units.

3. The method of claim 1, wherein, The number of optimal operating parameters of the air conditioner is a plurality, and the optimal operating parameters of the air conditioner have sequence and coupling; The initial prediction model comprises first network units corresponding to each optimal operating parameter of the air conditioner, and each first network unit corresponding to an optimal operating parameter of the air conditioner is used to predict the optimal operating parameter; The first network units corresponding to each optimal operating parameter of the air conditioner are connected in series according to the sequence of the optimal operating parameters of the air conditioner, and the output result of a subsequent first network unit is related to the output result of a previous first network unit.

4. The method of claim 3, wherein, The optimal operating parameters of the air conditioner include optimal compressor frequency, optimal expansion valve opening, optimal indoor fan speed, and optimal outdoor fan speed; The sequence of the optimal compressor frequency, the optimal expansion valve opening, the optimal indoor fan speed, and the optimal outdoor fan speed increases in turn, and the first network unit corresponding to the optimal compressor frequency, the first network unit corresponding to the optimal expansion valve opening, the first network unit corresponding to the optimal indoor fan speed, and the first network unit corresponding to the optimal outdoor fan speed are connected in series.

5. The method of claim 3, wherein, The initial prediction model further comprises a plurality of second network units, the plurality of second network units are connected in series, and the last second network unit is connected to the first network unit at the front; The output result of a subsequent second network unit is related to the output result of a previous second network unit, and the output result of the first network unit at the front is related to the output result of the last second network unit.

6. The method of claim 5, wherein, Sequentially inputting each data unit into the initial prediction model for training comprises: sequentially inputting each data unit into the initial prediction model for training, comprising: performing the following steps on each data unit: constructing n sequences based on the data unit, wherein the number of second network units is (n-1); inputting sequence 1 to sequence (n-1) into a corresponding second network unit, and inputting sequence n into a first network unit; The initial prediction model is trained based on the inputted sequences.

7. A method for predicting optimal operating parameters of an air conditioner, comprising: collecting current air conditioner hardware parameters of the air conditioner; determining current heat exchanger fouling state and refrigerant amount of the air conditioner; inputting the air conditioner hardware parameters, the heat exchanger fouling state and the refrigerant amount into a target prediction model, wherein the target prediction model is obtained using the training method for the prediction model for predicting optimal operating parameters of an air conditioner according to any one of claims 1 to 6; predicting current optimal operating parameters of the air conditioner using the target prediction model.

8. A training device for a prediction model for predicting optimal operating parameters of an air conditioner, comprising a processor and a memory storing program instructions, wherein the processor is configured to execute the training method for the prediction model for predicting optimal operating parameters of an air conditioner according to any one of claims 1 to 6 when running the program instructions.

9. A prediction device for optimal operating parameters of an air conditioner, comprising a processor and a memory storing program instructions, wherein the processor is configured to execute the prediction method for optimal operating parameters of an air conditioner according to claim 7 when running the program instructions.

10. An air conditioner, comprising: an air conditioner body; the training device for a prediction model for predicting optimal operating parameters of an air conditioner according to claim 8, wherein the training device is arranged in the air conditioner body; and / or, the prediction device for optimal operating parameters of an air conditioner according to claim 9, wherein the prediction device is arranged in the air conditioner body.

11. A computer-readable non-transitory storage medium storing program instructions, wherein the program instructions, when executed, perform the following steps: collecting air conditioner hardware parameters under different heat exchanger fouling states and different refrigerant amounts in an air conditioner enthalpy difference room; determining optimal operating parameters of the air conditioner under different heat exchanger fouling states and different refrigerant amounts, and taking the determined optimal operating parameters of the air conditioner as label data; constructing a training data set based on the heat exchanger fouling state, the refrigerant amount, the label data and the air conditioner hardware parameters, wherein the training data set comprises a plurality of data units, each data unit comprising the heat exchanger fouling state and the refrigerant amount, and the air conditioner hardware parameters and the label data under the heat exchanger fouling state and the refrigerant amount; sequentially inputting each data unit into an initial prediction model for training to obtain a target prediction model capable of predicting optimal operating parameters of the air conditioner.

12. A computer-readable non-transitory storage medium storing program instructions, wherein the program instructions, when executed, perform the following steps: collecting current air conditioner hardware parameters of the air conditioner; determining current heat exchanger fouling state and refrigerant amount of the air conditioner; The air conditioner hardware parameters, heat exchanger dirty state and refrigerant amount are input to the target prediction model, wherein, the target prediction model is obtained using the following training method: collecting air conditioner hardware parameters under different heat exchanger fouling states and different refrigerant amounts in an air conditioner enthalpy difference room; determining optimal operating parameters of the air conditioner under different heat exchanger fouling states and different refrigerant amounts, and taking the determined optimal operating parameters of the air conditioner as label data; The training data set is constructed based on a heat exchanger dirty state, a refrigerant amount, label data, and air conditioner hardware parameters, wherein the training data set includes multiple data units, each data unit containing the heat exchanger dirty state and the refrigerant amount, and air conditioner hardware parameters and label data under the heat exchanger dirty state and the refrigerant amount; Each data unit is sequentially input into an initial prediction model for training, to obtain a target prediction model capable of predicting optimal air conditioner operating parameters; The target prediction model is used to predict current optimal air conditioner operating parameters.

13. A computer program, which, when executed by a computer, causes the computer to implement a training method of a prediction model for predicting optimal air conditioner operating parameters according to any one of claims 1 to 6.

14. A computer program product, which comprises computer instructions stored on a computer readable storage medium, which, when executed by a computer, causes the computer to implement a training method of a prediction model for predicting optimal air conditioner operating parameters according to any one of claims 1 to 6.

Images