Control method of air conditioner, air conditioner, and storage medium

Abstract

The application discloses a control method of an air conditioner, the air conditioner and a storage medium, and relates to the field of heat pump systems. The method comprises the following steps: acquiring actual working condition data of the air conditioner; determining a temperature variation load and a temperature maintenance load of the air conditioner according to the actual working condition data, wherein the temperature variation load is a heat exchange amount required by an indoor space of the air conditioner when the indoor space changes temperature in a target direction, and the temperature maintenance load is a heat exchange amount required by the indoor space for maintaining temperature; and controlling the air conditioner to operate according to the temperature variation load and the temperature maintenance load. The application aims to improve the comfort and save energy consumption of air conditioner control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of heat pump systems, in particular to a control method of an air conditioner, an air conditioner and a storage medium. BACKGROUND

[0002] During the operation of the air conditioner, the indoor heat exchanger is in a heat exchange state through the cooperation of the operation of the compressor, the fan, the electronic expansion valve and other components, and the air and the indoor heat exchanger exchange heat to achieve the adjustment of the indoor environment.

[0003] In the related art, each device in the air conditioner generally characterizes the indoor load based on the indoor temperature or humidity, and controls the operation of each device according to the indoor temperature or humidity. However, the indoor temperature or humidity cannot accurately characterize the actual load of the air conditioner, and there are problems of discomfort and energy consumption. SUMMARY

[0004] The main purpose of the present application is to provide a control method of an air conditioner, an air conditioner and a storage medium, which aims to improve the comfort and energy saving of the air conditioner control.

[0005] To achieve the above purpose, the present application provides a control method of an air conditioner, which comprises:

[0006] obtaining actual working condition data of the air conditioner;

[0007] determining a temperature changing load and a temperature maintaining load of the air conditioner according to the actual working condition data, the temperature changing load being a heat exchange amount required when the indoor space adjusted by the air conditioner changes temperature in a target direction, and the temperature maintaining load being a heat exchange amount required for the indoor space to maintain temperature;

[0008] controlling the operation of the air conditioner according to the temperature changing load and the temperature maintaining load.

[0009] In an embodiment, the step of controlling the operation of the air conditioner according to the temperature changing load and the temperature maintaining load comprises:

[0010] controlling the operation of the air conditioner according to the sum of the temperature changing load and the temperature maintaining load.

[0011] In an embodiment, the step of determining the temperature changing load and the temperature maintaining load of the air conditioner according to the actual working condition data comprises:

[0012] inputting the actual working condition data into a load prediction model to obtain an output result of the load prediction model, the output result comprising the temperature changing load and the temperature maintaining load;

[0013] wherein the load prediction model is a model generated based on deep learning.

[0014] In an embodiment, the load prediction model comprises an input layer, a feature extraction layer, a hidden state layer, and an output layer connected in sequence, the input layer is configured to obtain the input operating condition data, the feature extraction layer is configured to extract the space-time features in the input operating condition data, the hidden state layer comprises a first latent variable representing the building parameters of the building where the air conditioner is located and a second latent variable representing the heat transfer parameters of the indoor space and the external environment, and the output layer is configured to calculate the variable temperature load based on the first latent variable and the input operating condition data, and the output layer is configured to calculate the constant temperature load based on the first latent variable, the second latent variable, and the input operating condition data.

[0015] In an embodiment, the building parameters comprise the length, width, and height of the indoor space.

[0016] In an embodiment, the heat transfer parameters comprise the thermal resistance of the envelope of the indoor space, the heat transfer coefficient of the envelope, and the temperature difference coefficient between the indoor space and the external space in different environmental states.

[0017] In an embodiment, the temperature difference coefficient comprises a first temperature difference coefficient, a second temperature difference coefficient, and a third temperature difference coefficient, the first temperature difference coefficient is the temperature difference coefficient when the external space corresponding to the wall of the indoor space is outdoor air, the second temperature difference coefficient is the temperature difference coefficient when the external space corresponding to the wall of the indoor space is an indoor environment without air conditioning, and the third temperature difference coefficient is the temperature difference coefficient when the external space corresponding to the wall of the indoor space is an indoor environment with air conditioning.

[0018] In an embodiment, before the step of obtaining the actual operating condition data of the air conditioner, the method further comprises:

[0019] obtaining historical operating condition data of the air conditioner and corresponding output capacity of the air conditioner, the parameter types of the historical operating condition data are correspondingly set with the parameter types of the actual operating condition data;

[0020] training the model parameters in the preset deep learning model according to the historical operating condition data and the output capacity;

[0021] using the preset deep learning model with the determined model parameters as the load prediction model.

[0022] In an embodiment, the control method of the air conditioner further comprises:

[0023] constructing a first loss function corresponding to the output layer as a constraint condition that the sum of the variable temperature load and the constant temperature load output by the output layer is equal to the output capacity of the corresponding air conditioner;

[0024] The step of training model parameters in the preset deep learning model according to the historical working condition data and the output capacity comprises:

[0025] The model parameters in the preset deep learning model are iteratively trained according to the historical working condition data, the output capacity, and the first loss function.

[0026] In an embodiment, the control method of the air conditioner further comprises:

[0027] A second loss function corresponding to the hidden state layer is constructed with consistency of the building parameters and / or the heat conduction parameters corresponding to different time points as a constraint condition.

[0028] The step of iteratively training the model parameters in the preset deep learning model according to the historical working condition data, the output capacity, and the first loss function comprises:

[0029] The model parameters in the preset deep learning model are iteratively trained according to the historical working condition data, the output capacity, the first loss function, and the second loss function.

[0030] In an embodiment, the step of constructing the second loss function corresponding to the hidden state layer with consistency of the building parameters and / or the heat conduction parameters corresponding to different time points as a constraint condition comprises:

[0031] The second loss function is constructed with consistency of the building parameters, the thermal resistance, and the heat transfer coefficient corresponding to different time points and consistency of the temperature difference coefficient within the same time period as constraint conditions.

[0032] In addition, to achieve the above object, the present application further provides an air conditioner, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the computer program is configured to implement the steps of the control method of the air conditioner as described above.

[0033] In addition, to achieve the above object, the present application further provides a storage medium, which is a computer readable storage medium, and the storage medium stores a computer program, and the computer program is executed by a processor to implement the steps of the control method of the air conditioner as described above.

[0034] The one or more technical solutions provided in the application have at least the following technical effects: in the solution, the variable temperature load and the temperature maintenance load of the air conditioner are determined based on actual working condition data, the air conditioner is controlled to operate according to the variable temperature load and the temperature maintenance load, and therefore, the operation of the air conditioner is no longer regulated based on the indoor temperature or humidity. The variable temperature load and the temperature maintenance load can accurately represent the actual output demand of the air conditioner, so as to realize accurate regulation of the operation of the air conditioner and effectively improve the comfort and energy saving of the air conditioner control. BRIEF DESCRIPTION OF DRAWINGS

[0035] The accompanying drawings, which are incorporated into and form part of the specification, illustrate embodiments consistent with the present application and, together with the specification, serve to explain the principles of the application.

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced as follows. Obviously, those skilled in the art can obtain other drawings from these drawings without creative effort.

[0037] Figure 1 The device structure schematic diagram of the hardware running environment involved in the control method of the air conditioner in the embodiments of the present application is shown in the figure.

[0038] Figure 2 The flowchart provided by the control method of the air conditioner in the first embodiment of the present application is shown in the figure.

[0039] Figure 3 The model framework schematic diagram of the load prediction model involved in the control method of the air conditioner in the embodiments of the present application is shown in the figure.

[0040] The purpose of the present application, functional features and advantages will be further explained with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0041] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application, and are not used to limit the present application.

[0042] In order to better understand the technical solutions of the present application, the specific embodiments will be described in detail with reference to the accompanying drawings and the specific embodiments.

[0043] The main solution of the embodiments of the present application is: obtaining actual working condition data of the air conditioner; determining variable temperature load and temperature maintenance load of the air conditioner according to the actual working condition data, the variable temperature load being heat exchange amount required by indoor space adjusted by the air conditioner to change temperature in a target direction, and the temperature maintenance load being heat exchange amount required by the indoor space to maintain temperature; and controlling the air conditioner to operate according to the variable temperature load and the temperature maintenance load.

[0044] In this embodiment, for ease of description, the following is described with an air conditioner as the execution subject.

[0045] In the related art, each device in the air conditioner generally represents the indoor load based on the indoor temperature or humidity, and each device is operated and controlled according to the indoor temperature or humidity. However, the indoor temperature or humidity cannot accurately represent the actual load of the air conditioner, and there are problems of discomfort and energy consumption.

[0046] The present application provides the above-mentioned solution, determines the temperature variation load and temperature maintenance load of the air conditioner based on actual working condition data, controls the operation of the air conditioner according to the temperature variation load and temperature maintenance load, and based on this, no longer controls the operation of the air conditioner based on the indoor temperature or humidity. The temperature variation load and temperature maintenance load can accurately represent the actual output demand of the air conditioner, thereby realizing accurate control of the operation of the air conditioner and effectively improving the comfort and energy saving of the air conditioner control.

[0047] The present application provides the above-mentioned solution, determines the temperature variation load and temperature maintenance load of the air conditioner based on actual working condition data, controls the operation of the air conditioner according to the temperature variation load and temperature maintenance load, and based on this, no longer controls the operation of the air conditioner based on the indoor temperature or humidity. The temperature variation load and temperature maintenance load can accurately represent the actual output demand of the air conditioner, thereby realizing accurate control of the operation of the air conditioner and effectively improving the comfort and energy saving of the air conditioner control.

[0048] In this embodiment, refer to Figure 1 , the air conditioner comprises a compressor 1, a reversing assembly 2 (such as a four-way valve, etc.), and an indoor heat exchanger, a throttling device 3 and an outdoor heat exchanger connected in sequence.

[0049] The indoor heat exchanger is correspondingly provided with an indoor fan 4, and the outdoor heat exchanger is correspondingly provided with an outdoor fan 5.

[0050] The exhaust port of the compressor 1, the gas return port of the compressor 1, the indoor heat exchanger and the outdoor heat exchanger are connected with the reversing assembly 2, and the reversing assembly 2 has a first operating state and a second operating state:

[0051] When the reversing assembly 2 operates in the first operating state, the exhaust port of the compressor 1 is communicated with the outdoor heat exchanger and the gas return port of the compressor 1 is communicated with the indoor heat exchanger, and when the compressor 1 is turned on, the refrigerant discharged from the compressor 1 flows through the outdoor heat exchanger, the throttling device 3 and the indoor heat exchanger in sequence and then returns to the compressor 1, and the indoor heat exchanger is in an evaporation state;

[0052] When the reversing assembly 2 operates in the second operating state, the exhaust port of the compressor 1 is communicated with the indoor heat exchanger and the gas return port of the compressor 1 is communicated with the outdoor heat exchanger, and when the compressor 1 is turned on, the refrigerant discharged from the compressor 1 flows through the indoor heat exchanger, the throttling device 3 and the outdoor heat exchanger in sequence and then returns to the compressor 1, and the indoor heat exchanger is in a condensation state.

[0053] The air conditioner further comprises an environment detection module 6 to detect an environment state parameter of an environment where the air conditioner is located. The environment can include an indoor environment and an outdoor environment, and the environment state parameter can include at least one of an environment temperature, an environment humidity, an environment enthalpy, an environment moisture content, etc.

[0054] With reference to Figure 1 The air conditioner further comprises a control device 100, and the compressor 1, the throttling device 3, the indoor fan 4, the outdoor fan 5, the reversing assembly 2, and the environment detection module 6 are all in communication connection with the control device 100.

[0055] The control device 100 comprises at least one processor 1001, and a memory 1002 and a timer 1003 in communication connection with the at least one processor 1001; the memory 1002 stores instructions executable by the at least one processor 1001, and the instructions are executed by the at least one processor 1001 to enable the at least one processor 1001 to execute the control method of the air conditioner in the following embodiments.

[0056] With reference to Figure 1 Fig. 1 shows a structural schematic diagram of a control device 100 suitable for implementing the embodiments of the present application. The control device 100 in the embodiments of the present application can include, but is not limited to, mobile terminals such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), and the like, and fixed terminals such as digital TVs, desktop computers, and the like. Figure 1 The control device 100 shown is only an example, and should not bring any limitation to the functions and use range of the embodiments of the present application.

[0057] As Figure 1As shown, the control device 100 can include a processor 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to programs stored in a memory 1002, which can be programs in a read only memory (ROM) or programs loaded from a storage device into a random access memory (RAM). In the RAM, various programs and data required for the operation of the control device 100 are also stored. The processor 1001, the memory 1002 (ROM and RAM), and the like are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus. Generally, the following systems can be connected to the I / O interface: input devices including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, and the like; output devices including, for example, a liquid crystal display (LCD), a speaker, a vibrator, and the like; storage devices including, for example, a magnetic tape, a hard disk, and the like; and communication devices. The communication devices can allow the control device 100 to communicate with other devices wirelessly or by wire to exchange data. Although the control device 100 having various systems is shown in the drawing, it should be understood that all of the systems shown are not required to be implemented or possessed. More or less systems can be alternatively implemented or possessed.

[0058] In particular, according to embodiments of the present disclosure, the method flow described in the following embodiments can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for executing the method shown in the flow chart. In such embodiments, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device. When the computer program is executed by the processor 1001, the above-mentioned functions defined in the control method of the air conditioner of embodiments of the present disclosure are performed.

[0059] The air conditioner provided by the present application adopts the control method of the air conditioner in the following embodiments, which can solve the technical problem of how to improve the comfort and energy saving of air conditioner control. Compared with the prior art, the air conditioner provided by the present application has the same beneficial effects as the control method of the air conditioner provided by the following embodiments, and other technical features in the air conditioner are the same as the features disclosed in the following embodiments, which are not described here.

[0060] It should be noted that the execution subject of the embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone or the like, or an electronic device, an air conditioner or the like capable of realizing the above functions. The following takes the air conditioner as an example to describe the embodiment and the following embodiments.

[0061] Based on this, the application provides an air conditioner control method, which refers to Figure 2 , Figure 2 The flowchart of the first embodiment of the air conditioner control method of the application is shown in the figure.

[0062] In the embodiment, the air conditioner control method comprises steps S10-S30:

[0063] Step S10, obtaining actual working condition data of the air conditioner;

[0064] The actual working condition data is data related to the load of the air conditioner in the current running process of the air conditioner. The actual working condition data can include data at one time or at multiple different times.

[0065] The actual working condition data can include state data of the air conditioner itself and environment data of the environment where the air conditioner is located.

[0066] In the embodiment, the actual working condition data includes but is not limited to the following data: return air temperature (current indoor temperature), indoor heat exchanger temperature, outdoor temperature, indoor temperature when the air conditioner is not turned on, set temperature, humidity, fan speed, compressor running frequency, initial temperature difference, evaporation temperature, condensation temperature, etc.

[0067] Step S10 is performed in the process of running the air conditioner in a preset mode, and the indoor heat exchanger is in a condensation state or an evaporation state in the preset mode, for example, the preset mode can be a heating mode or a cooling mode or a dehumidification mode.

[0068] Step S20, determining the temperature change load and the temperature maintenance load of the air conditioner according to the actual working condition data, the temperature change load being the heat exchange amount required by the indoor space adjusted by the air conditioner to change the temperature in a target direction, and the temperature maintenance load being the heat exchange amount required by the indoor space to maintain the temperature;

[0069] The target direction is the target direction of the temperature change of the indoor environment required by the preset mode currently running by the air conditioner. When the preset mode is a cooling mode or a dehumidification mode, the target direction is downward; when the preset mode is a heating mode, the target direction is upward.

[0070] In the embodiment, maintaining the temperature means maintaining the temperature unchanged. In other implementations, maintaining the temperature can also mean that the temperature fluctuation value of the maintained temperature is less than a preset amplitude.

[0071] The corresponding relationship between the actual working condition data and the variable temperature load and the maintaining temperature load can be established in advance, and the corresponding relationship can include a relational expression or an algorithm model, etc. The variable temperature load and the maintaining temperature load corresponding to the current actual working condition data can be determined based on the corresponding relationship.

[0072] In an implementation, the variable temperature load and the maintaining temperature load can be respectively set with a corresponding relationship with the working condition data. A first relationship between the working condition data and the variable temperature load is established in advance, a second relationship between the working condition data and the maintaining temperature load is established in advance, a first data corresponding to the variable temperature load in the actual working condition data is determined, a second data corresponding to the maintaining temperature load in the actual working condition data is determined, the variable temperature load corresponding to the first data is determined based on the first relationship, and the maintaining temperature load corresponding to the second data is determined based on the second relationship. The first relationship can include a function relationship expression or an algorithm model, etc. The second relationship can include a function relationship expression or an algorithm model, etc.

[0073] In another implementation, the variable temperature load and the maintaining temperature load are combined to set a corresponding relationship with the working condition data. A third relationship among the working condition data, the variable temperature load and the maintaining temperature load is established in advance, and the variable temperature load and the maintaining temperature load corresponding to the current actual working condition data can be determined based on the third relationship. The third relationship can include a function relationship expression or an algorithm model, etc.

[0074] In the embodiment, the variable temperature load and the maintaining temperature load are determined based on the actual working condition data. In other implementations, the variable temperature load can be determined based on the actual working condition data in the variable temperature stage, and the maintaining temperature load can be determined based on the actual working condition data in the maintaining temperature stage.

[0075] Step S30, controlling the air conditioner to operate according to the variable temperature load and the maintaining temperature load.

[0076] The compressor adjusts the operating frequency, the fan adjusts the operating speed, and the electronic expansion valve adjusts the opening degree in the air conditioner according to the variable temperature load and the maintaining temperature load.

[0077] In the embodiment, the air conditioner is controlled to operate according to the sum of the variable temperature load and the maintaining temperature load. When the sum of the variable temperature load and the maintaining temperature load is greater than the current output capacity of the air conditioner, the compressor is controlled to increase the operating frequency and the fan is controlled to increase the operating speed. When the sum of the variable temperature load and the maintaining temperature load is less than or equal to the current output capacity of the air conditioner, the compressor and the fan are controlled to maintain the current state, or the compressor is controlled to reduce the operating frequency and the fan is controlled to reduce the operating speed. Controlling the air conditioner to operate based on the sum of the variable temperature load and the maintaining temperature load can realize that the variable temperature load is mainly used to control the output capacity of the air conditioner when the indoor space is in the variable temperature stage, and the maintaining temperature load is mainly used to control the output capacity of the air conditioner when the indoor space is in the maintaining temperature stage.

[0078] In some other implementations, when the air conditioner is controlled to operate according to the variable temperature load and the temperature maintenance load, the first weight value corresponding to the variable temperature load and the second weight value corresponding to the temperature maintenance load can be determined according to a temperature difference between the temperature of the indoor space adjusted by the air conditioner and the set temperature, the target load can be calculated by weighted average of the variable temperature load and the temperature maintenance load according to the first weight value and the second weight value, and the air conditioner is controlled to operate according to the target load. When the target load is greater than the current output capacity of the air conditioner, the compressor is controlled to increase the operating frequency and the fan is controlled to increase the operating speed; when the target load is less than or equal to the current output capacity of the air conditioner, the compressor and the fan are controlled to maintain the current state, or the compressor is controlled to reduce the operating frequency and the fan is controlled to reduce the operating speed.

[0079] The embodiment provides a control method of an air conditioner. The variable temperature load and the temperature maintenance load of the air conditioner are determined based on actual working condition data, and the air conditioner is controlled to operate according to the variable temperature load and the temperature maintenance load. Therefore, the operation of the air conditioner is no longer regulated based on the indoor temperature or humidity, the variable temperature load and the temperature maintenance load can accurately represent the actual output demand of the air conditioner, and thus accurate regulation of the operation of the air conditioner is realized, and the comfort and energy saving of the air conditioner control are effectively improved.

[0080] Based on any of the above embodiments, in the second embodiment of the present application, the content consistent with or similar to the above embodiments can be referred to the above description, and will not be described in detail. On this basis, step S20 comprises:

[0081] The actual working condition data is input into a load prediction model to obtain an output result of the load prediction model, and the output result comprises the variable temperature load and the temperature maintenance load. The load prediction model is a model generated based on deep learning.

[0082] The input parameters of the load prediction model are working condition parameters, and the output parameters of the load prediction model are the variable temperature load and the temperature maintenance load.

[0083] The deep learning algorithm applied by the load prediction model can include at least one of the following: a recurrent neural network algorithm (RNN), a time convolution network algorithm (TCN), a long short-term memory network algorithm (LSTM), a transformer network algorithm (Transformer), a graph network algorithm (GNN), and the like.

[0084] The different load prediction models can be correspondingly used when the air conditioner is currently operated in different preset modes.

[0085] In the embodiment, the variable temperature load and the temperature maintenance load are determined based on the load prediction model, which is beneficial to accurately detect the actual load of the air conditioner and further improve the accuracy of the air conditioner control, thereby further improving the comfort and saving energy.

[0086] In an implementable embodiment, in combination Figure 3 , the load prediction model comprises an input layer, a feature extraction layer, a hidden state layer, and an output layer connected in sequence, the input layer is configured to obtain input working condition data, the feature extraction layer is configured to extract spatio-temporal features in the input working condition data, the hidden state layer comprises a first latent variable representing building parameters of a building where the air conditioner is located and a second latent variable representing heat transfer parameters of the indoor space and the external environment, and the output layer is configured to calculate a variable temperature load based on the first latent variable and the input working condition data, and the output layer is configured to calculate a constant temperature load based on the first latent variable, the second latent variable, and the input working condition data.

[0087] The feature extraction layer captures spatio-temporal features in the input data through a time-series network (such as RNN, TCN, LSTM, Transformer, etc.) or a graph network (GNN). In combination Figure 3 , the functions of the modules in the feature extraction layer are as follows:

[0088] Dilated Causal Convolution (Dilated Causal Conv): mainly uses convolution kernels to extract spatio-temporal features of input data in the input layer, wherein the dependency between features is captured through full connection, the short-term time series relationship of the features is strengthened through the causal layer, and the long-term time series relationship is captured through the dilated layer to obtain feature data;

[0089] WeightNorm: performs a unified normalization operation on the convolution kernel weights, so that the normalized convolution kernel can be used for spatio-temporal feature extraction in the dilated causal convolution process, which can reduce gradient explosion or disappearance in the model training process and improve the training efficiency of the model;

[0090] Rectified Linear Unit (ReLU): a commonly used activation function, which performs non-negative processing on the above feature data, and sets the value to 0 if it is less than or equal to 0, and the value remains unchanged if it is greater than 0. Through this activation function, the non-linear relationship between the input data and the output data can be extracted;

[0091] Dropout: randomly sets the feature values of the data processed by the rectified linear activation to zero, reduces the mutual dependence between model nodes, and thus reduces the overfitting risk of the model and the structural risk of the model design.

[0092] The data obtained by the dropout processing can be processed again in the above manner, and the data obtained by the dropout processing can be input into the hidden state layer for further processing after a certain number of repetitions.

[0093] The building parameters can include size parameters of the indoor space, which can include length, width, height, etc. of the indoor space. Correspondingly, the first hidden variable can include a first sub-variable corresponding to the length, a second sub-variable corresponding to the width, and a third sub-variable corresponding to the height. In some implementations, the height of the indoor space can be a given value (e.g., 2.8 meters, etc., different types of buildings can be set to different values), and the building parameters represented by the first hidden variable can include the length and width of the indoor space.

[0094] The heat conduction parameters can include the thermal resistance of the envelope of the indoor space and / or the heat transfer coefficient of the envelope and / or the temperature difference coefficient between the indoor space and the external space in different environmental states, etc.

[0095] The indoor space can have multiple walls, and the environmental states of the external space corresponding to different walls can be different, so the indoor-outdoor temperature difference values between the external space and the indoor space can be different. Based on this, the temperature difference coefficient can be a weighted coefficient of the indoor-outdoor temperature difference value of the corresponding external space, and the indoor-outdoor temperature difference representing the overall temperature difference between the indoor space and all external spaces can be calculated by weighted average according to the indoor-outdoor temperature difference values of the external space in different environmental states and the corresponding temperature difference coefficients.

[0096] The temperature difference coefficient can include a first temperature difference coefficient, a second temperature difference coefficient, and a third temperature difference coefficient, or the temperature difference coefficient can include a first temperature difference coefficient and a second temperature difference coefficient. When the external space corresponding to the wall is outdoor air, the corresponding temperature difference coefficient can be the first temperature difference coefficient; when the external space corresponding to the wall is an indoor environment without air conditioning, the corresponding temperature difference coefficient can be the second temperature difference coefficient; and when the external space corresponding to the wall is an indoor environment with air conditioning, the corresponding temperature difference coefficient can be the third temperature difference coefficient. Wherein, the sum of the first temperature difference coefficient, the second temperature difference coefficient and the third temperature difference coefficient is 1, then the third temperature difference coefficient can be determined based on the first temperature difference coefficient and the second temperature difference coefficient.

[0097] In this embodiment, the heat conduction parameters include the ratio of the heat transfer coefficient to the thermal resistance of the envelope, the first temperature difference coefficient, and the second temperature difference coefficient, which can be defined as the heat conduction coefficient. Correspondingly, the second hidden variable can include a fourth sub-variable corresponding to the heat conduction coefficient, a fifth sub-variable corresponding to the first temperature difference coefficient, and a sixth sub-variable corresponding to the second temperature difference coefficient.

[0098] The spatio-temporal features extracted by the feature extraction layer are input to the hidden state layer, which does not perform feature extraction. Each hidden variable only performs dimension alignment and splicing with the corresponding feature in the input spatio-temporal feature. The first hidden variable and the second hidden variable in the hidden state layer can be continuously corrected in the process of model iterative training, so as to continuously approach the actual state value of the indoor space.

[0099] The output layer can calculate the temperature-changing load according to the following formula (1) by using the current value of the first hidden variable and the corresponding first sub-data in the input working condition data:

[0100] (1)

[0101] wherein, is the temperature-changing load, is the specific heat capacity, is the air density, is the length of the indoor space, is the width of the indoor space, is the height of the indoor space, and m is the air mass of the indoor space, is the temperature change value of the indoor space, wherein, , are both known constants, , , are the current values of the first hidden variable, is the parameter value determined based on the input working condition data.

[0102] The output layer can calculate the temperature-maintaining load according to the following formula (2) and (3) by using the current value of the second hidden variable and the corresponding second sub-data in the input working condition data: :

[0103] (2)

[0104] (3)

[0105] wherein, the heat transfer coefficient K and the thermal resistance R are both related to heat conduction, and thus the heat conduction coefficient Kr=K / R is defined, is the length of the indoor space, is the width of the indoor space, is the height of the indoor space, is the indoor-outdoor temperature difference representing the overall temperature difference between the indoor space and all external spaces, is the temperature of the outdoor air, is the temperature of the indoor space, is the temperature of the indoor environment without air conditioning, is the temperature of the indoor environment with air conditioning, is the temperature difference between the external space and the indoor space when the external space is the outdoor air, is the first temperature difference coefficient, is the temperature difference between the external space and the indoor space when the external space is the indoor environment without air conditioning, is the second temperature difference coefficient, a temperature difference value between the external space and the indoor space when the external space is an indoor environment with an open air conditioner, a third temperature difference coefficient, an area of the indoor space. Wherein, , , are current values of the first hidden variables, K / R, , are current values of the second hidden variables, , , , are data in the current input working condition data.

[0106] In the embodiment, the load prediction model is set in the above manner, and the physical parameters such as the building parameters and the heat conduction parameters required for determining the variable temperature load and the temperature maintenance load are no longer fixed parameters set in advance, but can be autonomously perceived and learned by the air conditioner through the load prediction model, which is conducive to improving the accuracy of the determined variable temperature load and temperature maintenance load, and effectively improving the comfort and energy saving of air conditioning control.

[0107] In a feasible implementation, before the step of acquiring the actual working condition data of the air conditioner, the method further includes: acquiring historical working condition data of the air conditioner and corresponding output capacity of the air conditioner, the parameter types of the historical working condition data are set corresponding to the parameter types of the actual working condition data; training model parameters in a preset deep learning model according to the historical working condition data and the output capacity; and taking the preset deep learning model with the determined model parameters as the load prediction model.

[0108] Before the current time, in the process of running the air conditioner in a preset mode, the working condition data and the output capacity of the air conditioner are collected every set time length, the working condition data at multiple different times are obtained, and after abnormal processing, smoothing data and aggregation processing according to a preset time length, the historical working condition data are obtained. The collected working condition data include but are not limited to the following data: return air temperature, indoor heat exchanger temperature, outdoor temperature, indoor temperature when the air conditioner is not turned on, initial temperature difference, evaporation temperature, condensation temperature, set temperature, humidity, fan speed, compressor operating frequency, indoor temperature change value, temperature difference value between the indoor space and different external spaces, etc.

[0109] The output capacity can include actual refrigerating capacity or actual heating capacity of the air conditioner, etc.

[0110] The preset deep learning model consists of an input layer, a feature extraction layer, a hidden state layer, and an output layer, connected sequentially as described above. Model parameters include neural network weights and trainable constants. The trainable constants include the first and second hidden variables in the hidden state layer. All model parameters of the preset deep learning model are initial, untrained values. When the model parameters of the preset deep learning model are iteratively trained until the training termination condition is met (e.g., the deviation between the predicted load and the actual load is less than a preset deviation), the model parameters can be considered determined, and the preset deep learning model at this point can be used as a load prediction model.

[0111] In this embodiment, combined with Figure 3 The control method for the air conditioner further includes: constructing a first loss function corresponding to the output layer, with the sum of the variable-temperature load and the constant-temperature load output by the output layer equal to the corresponding output capacity of the air conditioner as a constraint; the step of training the model parameters in the preset deep learning model based on the historical operating data and the output capacity includes: iteratively training the model parameters in the preset deep learning model based on the historical operating data, the output capacity, and the first loss function. Here, the trained model parameters are neural network weights. The corresponding output capacity refers to the output capacity corresponding to the operating data input to the current input layer. Based on this, the accuracy of the variable-temperature load and constant-temperature load output by the load prediction model can be guaranteed through the capacity conservation constraint.

[0112] In this embodiment, combined with Figure 3 The control method for the air conditioner further includes: constructing a second loss function corresponding to the hidden state layer with the consistency of the building parameters and / or the heat conduction parameters corresponding to different times as a constraint; the step of iteratively training the model parameters in the preset deep learning model according to the historical operating data, the output capacity, and the first loss function includes: iteratively training the model parameters in the preset deep learning model according to the historical operating data, the output capacity, the first loss function, and the second loss function.

[0113] The consistency of building parameters and / or heat conduction parameters at different times here means that the building parameters and / or heat conduction parameters do not change over time. The constructed second loss function includes the following loss function relationship (4):

[0114] (4);

[0115] Where N is the number of samples during the training process. For the first i The variable in the first... j The values ​​in each sample For the first iThe mean of a variable in all samples, the first i The variable refers to one of length, width, height, first temperature difference coefficient, second temperature difference coefficient, and heat transfer coefficient, where 6 represents the total number of variables.

[0116] The loss value corresponding to the current training sample set can be calculated through the second loss function, and the numerical value of the hidden variable in the hidden state layer can be adjusted based on the loss value.

[0117] Since the area of the same room remains unchanged at any time during operation, the length, width, and height are consistent; once the building is completed, the building material will not change, and the thermal resistance and heat transfer coefficient are consistent; the algebraic relationship between the indoor and outdoor temperature difference and the external space temperature and other variables has relative stability. Based on this, the consistency constraint is added to the hidden variable in the hidden state layer in the above manner, thereby effectively improving the accuracy of the model prediction load.

[0118] Further, the building parameters, the thermal resistance, and the heat transfer coefficient consistency corresponding to different time points are used as constraint conditions, and the consistency of the temperature difference coefficient in the same period is used as a constraint condition to construct a second loss function. In this process, the temperature difference coefficients corresponding to different time periods can be different. Based on the constraint condition that the temperature difference coefficients in the same period have consistency and the temperature difference coefficients in different periods can be different, the following loss function relationship (5) is constructed:

[0119] (5);

[0120] wherein, represents a sample training batch (data in the same period can be divided into multiple batches), i represents the sample of the current batch, h represents the period, represents the seed distribution of the current sample and all samples before the current time in the h period, represents the seed distribution of the current sample and all samples before the current time in the h period, represents the Jensen-Shannon divergence (used to evaluate the consistency of the distribution), represents the distribution of the current sample in the h period, represents the distribution of the current sample in the h period.

[0121] The second loss function comprises the loss function relationship (4) and the loss function relationship (5) described above, wherein the first loss value corresponding to the building parameters (such as length, width, height, etc.) and the heat conduction coefficient can be calculated by the loss function relationship (4) described above, the total number of variables in the loss function relationship (4) can be 4 at this time, and the value of the hidden variable corresponding to the building parameters (such as length, width, height, etc.) and the heat conduction coefficient in the hidden state layer can be adjusted based on the first loss value, the second loss value corresponding to the first temperature difference coefficient and the second temperature difference coefficient in the current training sample set can be calculated by the loss function relationship (5) described above, and the value of the hidden variable corresponding to the first temperature difference coefficient and the second temperature difference coefficient in the hidden state layer can be adjusted based on the second loss value.

[0122] In the embodiment, the first temperature difference coefficient and the second temperature difference coefficient may be different in different time periods in different use scenarios (for example, the other side of the wall is a neighbor, but the air conditioner is not turned on during the day and is turned on at night, so the first temperature difference coefficient and the second temperature difference coefficient are different during the day and at night), but are consistent in the same scenario (for example, the other side of the wall is a neighbor, and if the air conditioner is turned on at 10 o'clock at night, it is highly probable that the air conditioner will be turned on at 10 o'clock every day in summer), based on which the loss function relationship (5) described above is constructed, so as to ensure that the load prediction matches the actual scenario, thereby improving the accuracy of the load prediction based on the model.

[0123] In the embodiment, the load prediction model is trained by the above method, which is beneficial to accurately predict the temperature-changing load and the temperature-maintaining load, so as to improve the comfort of the air conditioner control while saving energy consumption.

[0124] In other embodiments, when the historical temperature-changing load and the historical temperature-maintaining load corresponding to the known historical working condition data are known, the preset deep learning model can also be trained according to the historical working condition data, the historical temperature-changing load, and the historical temperature-maintaining load.

[0125] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the control method of the air conditioner of the present application, and more forms of simple transformation based on this technical concept are within the protection scope of the present application.

[0126] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, the computer-readable program instructions being used to execute the control method of the air conditioner in the above embodiments.

[0127] The computer readable storage medium provided in the present application may, for example, be a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more conductive wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read only memory (ROM), an erasable programmable read only memory (EPROM or flash memory), an optical fiber, a portable compact disk read only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted in any suitable medium, including but not limited to electrical wire, optical cable, RF (Radio Frequency), etc., or any suitable combination of the above.

[0128] The above computer readable storage medium can be contained in the air conditioner; or can exist separately without being assembled into the air conditioner.

[0129] The above computer readable storage medium carries one or more programs, which, when executed by the processor, cause the processor to execute the flow in the control method embodiment of the air conditioner.

[0130] Computer program code for carrying out operations of the present application can be written in one or more programming languages or combinations of languages including object oriented programming languages such as Java, Smalltalk, C++ or conventional procedural programming languages such as "C" or similar programming languages. The program code can execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the latter scenario, the remote computer can be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or the connection can be made to an external computer (for example, through the Internet using an Internet Service Provider).

[0131] The readable storage medium provided by the application is a computer readable storage medium, which stores computer readable program instructions (i.e. computer programs) for executing the control method of the air conditioner, and can solve the technical problem of how to improve the comfort and energy saving of air conditioner control. Compared with the prior art, the beneficial effects of the computer readable storage medium provided by the application are the same as those of the control method of the air conditioner provided by the above-mentioned embodiments, and will not be repeated here.

[0132] The flowcharts and block diagrams in the drawings illustrate the possible implementation architecture, function and operation of the system, method and computer program product according to various embodiments of the application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different order than that shown in the figure. For example, two blocks that are shown in succession can actually be executed substantially in parallel, and they can also be executed in reverse order, depending on the involved functions. It should also be noted that each block in the block diagram and / or flowchart and the combination of blocks in the block diagram and / or flowchart can be implemented by a dedicated hardware-based system that performs the specified function or operation, or can be implemented by a combination of special-purpose hardware and computer instructions.

[0133] It should be understood that parts of the present application can be implemented by hardware, software, firmware or their combination. The modules described in the embodiments of the present application can be implemented by software or hardware. Among them, the name of the module does not constitute a limitation of the unit itself in some cases. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0134] The above only describes some embodiments of the application, and does not limit the patent scope of the application. Any equivalent structural transformation made by using the content of the application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the application. Therefore, the protection scope of the application should be subject to the protection scope of the claims.

Claims

1. A control method for an air conditioner, characterized in that, The control method for the air conditioner includes: Obtain the actual operating condition data of the air conditioner; The actual operating condition data is input into the load prediction model to obtain the output results of the load prediction model. The output results include the variable temperature load and the temperature maintenance load of the air conditioner. The variable temperature load is the heat exchange required when the indoor space is regulated by the air conditioner to change the temperature in the target direction. The temperature maintenance load is the heat exchange required to maintain the temperature of the indoor space. The air conditioner is controlled to operate based on the sum of the variable temperature load and the constant temperature load. The load prediction model is a deep learning-based model, comprising an input layer, a feature extraction layer, a hidden state layer, and an output layer connected in sequence. The input layer is used to acquire input operating data. The feature extraction layer is used to extract spatiotemporal features from the input operating data. The hidden state layer includes a first latent variable representing the building parameters of the building where the air conditioner is located and a second latent variable representing the heat conduction parameters between the indoor space and the external environment. The output layer is used to calculate the variable temperature load based on the first latent variable and the input operating data, and the constant temperature load is calculated based on the first latent variable, the second latent variable, and the input operating data.

2. The control method for an air conditioner as described in claim 1, characterized in that, The architectural parameters include the length, width, and height of the interior space; And / or, the heat conduction parameters include the thermal resistance of the interior space's enclosure structure, the heat transfer coefficient of the enclosure structure, and the temperature difference coefficient between the interior space and the exterior space under different environmental conditions.

3. The control method for an air conditioner as described in claim 2, characterized in that, The temperature difference coefficient includes a first temperature difference coefficient, a second temperature difference coefficient, and a third temperature difference coefficient. The first temperature difference coefficient is the temperature difference coefficient when the external space corresponding to the wall of the indoor space is outdoor air. The second temperature difference coefficient is the temperature difference coefficient when the external space corresponding to the wall of the indoor space is an indoor environment without air conditioning. The third temperature difference coefficient is the temperature difference coefficient when the external space corresponding to the wall of the indoor space is an indoor environment with air conditioning.

4. The control method for an air conditioner as described in claim 1, characterized in that, Before the step of obtaining the actual operating condition data of the air conditioner, the method further includes: Acquire historical operating condition data of the air conditioner and the corresponding output capacity of the air conditioner, wherein the parameter types of the historical operating condition data are set to correspond to the parameter types of the actual operating condition data; The model parameters in the preset deep learning model are trained based on the historical operating data and the output capability. The preset deep learning model with determined model parameters is used as the load prediction model.

5. The control method for an air conditioner as described in claim 4, characterized in that, The control method for the air conditioner also includes: Using the constraint that the sum of the variable temperature load and the constant temperature load output by the output layer equals the output capacity of the corresponding air conditioner, a first loss function corresponding to the output layer is constructed. The step of training the model parameters in the preset deep learning model based on the historical working condition data and the output capability includes: The model parameters in the preset deep learning model are iteratively trained based on the historical operating data, the output capability, and the first loss function.

6. The control method for an air conditioner as described in claim 5, characterized in that, The control method for the air conditioner also includes: Using the consistency of the building parameters and / or the heat conduction parameters at different times as constraints, a second loss function corresponding to the hidden state layer is constructed; The step of iteratively training the model parameters in the preset deep learning model based on the historical operating data, the output capability, and the first loss function includes: The model parameters in the preset deep learning model are iteratively trained based on the historical operating data, the output capability, the first loss function, and the second loss function.

7. The control method for an air conditioner as described in claim 6, characterized in that, The heat conduction parameters include the thermal resistance of the building envelope of the indoor space, the heat transfer coefficient of the building envelope, and the temperature difference coefficient between the indoor space and the external space under different environmental conditions. The step of constructing the second loss function corresponding to the hidden state layer, using the consistency of the building parameters and / or the heat conduction parameters at different times as constraints, includes: The second loss function is constructed by taking the consistency of the building parameters, thermal resistance, and heat transfer coefficient at different times as constraints, and the consistency of the temperature difference coefficient within the same time period as constraints.

8. An air conditioner, characterized in that, The air conditioner includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the control method for the air conditioner as described in any one of claims 1 to 7.

9. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the control method for the air conditioner as described in any one of claims 1 to 7.

Images