Air conditioner control system and air conditioner

Abstract

The invention discloses an air conditioner control system and an air conditioner, and the air conditioner control system comprises a parameter acquisition module which is configured to acquire the unit air conditioner area and indoor actual environment parameters in a window closing state; unit air conditioner area = indoor area / indoor air conditioner number; the model calling module is configured to input the unit air conditioner area obtained by the parameter obtaining module and the indoor actual environment parameters in the window closing state into a preset window opening prediction model, and the window opening prediction model outputs predicted indoor target environment parameters in the window opening state; and the control module is configured to control operation of the air conditioner according to the indoor target environment parameters output by the windowing prediction model. According to the air conditioner control system and the air conditioner, air flow during window opening ventilation can be simulated in the indoor window closing state, the natural wind function effect is improved, the technical problem that in the prior art, the natural wind function effect is poor is solved, it is guaranteed that the human feeling comfort degree is not changed, window opening is not needed, and potential safety hazards are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of air conditioning technology, in particular to an air conditioner control system and an air conditioner. BACKGROUND

[0002] In order to improve the air quality and environmental comfort of a closed space such as an office or a residential home, it is often necessary to open the window for ventilation. The natural air convection and the steady increase in oxygen content brought about by opening the window for ventilation give people a comfortable feeling that cannot be achieved by ordinary air conditioning products.

[0003] However, opening the window for ventilation faces many personal and property safety issues, such as the personal safety of children and the elderly, the impact of house dust and severe rain and snow weather on indoor property, and the increase in indoor safety hazards.

[0004] However, most current air conditioners only focus on changes in indoor temperature and humidity to achieve the purpose of indoor cooling or heating, and do not have a natural wind function. The focus is not on the flow, flow rate, and air temperature of the airflow, and people in the room are difficult to feel natural wind when only the air conditioner is turned on without opening the window. Even air conditioners with a natural wind function are based on a unified control mode and are difficult to effectively achieve air convection according to the current time and space size. Compared with traditional pure cooling and heating, although there is some improvement, it is still difficult to customize the simulation of indoor air circulation comfort according to the actual situation, and the natural wind function effect is poor, and there is still a distance from the healthy and comfortable feeling brought about by opening the window for ventilation. SUMMARY

[0005] The present application provides an air conditioner control system, which solves the technical problem of poor natural wind function effect in the prior art.

[0006] To achieve the above purpose, the present application adopts the following technical scheme:

[0007] The present application provides an air conditioner control system, which includes:

[0008] A parameter acquisition module configured to acquire indoor actual environment parameters under a closed window state per unit air conditioner area, wherein the unit air conditioner area = indoor area / number of indoor air conditioners;

[0009] A model calling module configured to input the indoor actual environment parameters under the closed window state per unit air conditioner area acquired by the parameter acquisition module into a preset open window prediction model, and the open window prediction model outputs predicted indoor target environment parameters under an open window state;

[0010] A control module configured to control the operation of the air conditioner according to the indoor target environment parameters output by the open window prediction model.

[0011] In some embodiments of the present application, the open window prediction model is provided with four, corresponding to four seasons respectively;

[0012] The model calling module is further configured to:

[0013] obtain date information, and determine a current season according to the date information;

[0014] select a corresponding window-opening prediction model according to the current season;

[0015] input the indoor actual environment parameter in the window-closed state per unit air conditioner area obtained by the parameter obtaining module into the corresponding window-opening prediction model.

[0016] In some embodiments of the present application, the indoor actual environment parameter in the window-closed state includes temperature, humidity, air volume, air speed, air direction, and oxygen content.

[0017] The indoor target environment parameter includes temperature, humidity, air volume, air speed, and air direction.

[0018] In some embodiments of the present application, the air conditioner control system further includes a model establishing module, which includes:

[0019] a sample obtaining unit, configured to obtain window-closed sample data and window-opened sample data; the window-closed sample data includes indoor environment parameters per unit air conditioner area and in a window-closed state; the window-opened sample data includes indoor environment parameters per unit air conditioner area and in a window-opened state;

[0020] a model training unit, configured to establish the window-opening prediction model and train the window-opening prediction model by using the window-closed sample data and the window-opened sample data.

[0021] In some embodiments of the present application, the model establishing module further includes a data preprocessing unit.

[0022] The data preprocessing unit is configured to perform normalization processing on the window-closed sample data and the window-opened sample data obtained by the sample obtaining unit.

[0023] The model training unit trains the window-opening prediction model by using the window-closed sample data and the window-opened sample data after normalization processing.

[0024] In some embodiments of the present application, the window-opening prediction model is an autoencoder model based on LSTM.

[0025] In some embodiments of the present application, the autoencoder model includes an encoder and a decoder.

[0026] The encoder receives the window-closed sample data as an input sequence, and encodes the input sequence into a hidden state.

[0027] The decoder receives the hidden state from the encoder as an initial state, inputs the windowed sample data, and outputs the predicted indoor environment parameter in the windowed state.

[0028] In some embodiments of the present application, the windowed prediction model is preset in the central controller.

[0029] In some embodiments of the present application, the air conditioner control system further comprises:

[0030] The intelligent terminal is configured to calculate the unit air conditioner area according to the indoor area and the number of indoor air conditioners input by the user, and send the calculated unit air conditioner area to the parameter acquisition module.

[0031] The present application provides an air conditioner comprising the air conditioner control system.

[0032] The technical solution of the present application has the following technical effects relative to the prior art: the air conditioner control system and the air conditioner of the present application, the parameter acquisition module acquires the unit air conditioner area and the actual indoor environment parameter in the window-closed state; the model calling module inputs the unit air conditioner area and the actual indoor environment parameter in the window-closed state acquired by the parameter acquisition module into the preset windowed prediction model, and the windowed prediction model outputs the predicted indoor target environment parameter in the window-opened state; the control module controls the operation of the air conditioner according to the indoor target environment parameter predicted by the windowed prediction model, so that the air conditioner simulates the air flow in the window-opened and ventilated state, and realizes the human-sense natural ventilation in the window-closed state. Therefore, the air conditioner control system and the air conditioner of the present application completely replace the window-opened ventilation by the prediction of the windowed prediction model and the air conditioner control, realize the natural wind function, do not need to open the window, reduce the safety hidden danger, and ensure the human-sense comfort degree unchanged; moreover, the windowed prediction model comprehensively considers the indoor area, the number of indoor air conditioners, and the actual indoor environment parameter, can predict the relatively accurate indoor target environment parameter, controls the operation of the air conditioner, so that the air conditioner simulates the air flow in the window-opened and ventilated state, improves the natural wind function effect of the air conditioner, and solves the technical problem of poor natural wind function effect of the air conditioner in the prior art.

[0033] Other features and advantages of the present application will become more apparent after reading the specific embodiments of the present application in combination with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0034] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below, and obviously, the drawings in the following description can be used to obtain other drawings without creative labor by those skilled in the art.

[0035] Figure 1 It is a structural block diagram of an embodiment of the air conditioner control system of the present application.

[0036] Figure 2 Flow chart of an embodiment of the model calling module performing steps;

[0037] Figure 3 Structural block diagram of another embodiment of the air conditioner control system of the present application;

[0038] Figure 4 Structural block diagram of an embodiment of the model building module;

[0039] Figure 5 Structural block diagram of an embodiment of the auto-encoder model;

[0040] Figure 6 Schematic diagram of the collection of the window-closed sample;

[0041] Figure 7 Schematic diagram of the collection of the window-opened sample;

[0042] Figure 8 Structural block diagram of an embodiment of the encoder and decoder;

[0043] Figure 9 Flow chart of an embodiment of the model training process;

[0044] Figure 10 Structural block diagram of an embodiment of the air conditioner of the present application. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0046] In the description of the present application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the purpose of facilitating the description of the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0047] The terms "first", "second", etc. are used only for the purpose of description and do not imply or indicate relative importance or a specific number of the technical features indicated. Thus, the features defined with "first", "second" can include one or more of the features explicitly or implicitly. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise specified.

[0048] In the description of the present application, it should be noted that unless otherwise explicitly specified and limited, the terms "mounting", "connecting", "connection" should be understood broadly, for example, it can be fixed connection, or detachable connection, or integrally connected; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium, or the communication inside two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0049] In the present application, unless otherwise explicitly specified and limited, the "upper" or "lower" of the first feature to the second feature can include that the first and second features are in direct contact, or that the first and second features are not in direct contact but are in contact through another feature between them. Moreover, the "upper", "above" and "on" of the first feature to the second feature includes that the first feature is directly above and obliquely above the second feature, or only means that the horizontal height of the first feature is higher than that of the second feature. The "under", "below" and "under" of the first feature to the second feature includes that the first feature is directly below and obliquely below the second feature, or only means that the horizontal height of the first feature is less than that of the second feature.

[0050] The following disclosure provides many different embodiments or examples for implementing different structures of the present application. In order to simplify the disclosure of the present application, the components and settings of specific examples are described below. Of course, they are only examples, and the purpose is not to limit the present application. In addition, the present application can repeatedly refer to numbers and / or letters in different examples, and such repetition is for the purpose of simplification and clarity, which itself does not indicate the relationship between the various embodiments and / or settings discussed. In addition, the present application provides examples of various specific processes and materials, but those skilled in the art can realize the application of other processes and / or the use of other materials.

[0051] The air conditioner performs the refrigeration cycle and the heating cycle of the air conditioner by using the compressor, the condenser, the expansion valve and the evaporator, and the controller performs control to achieve the flow control of the refrigerant and the opening control of the expansion valve, etc. The refrigeration cycle and the heating cycle include a series of processes involving compression, condensation, expansion and evaporation, and supply refrigerant to the air that has been adjusted and heat exchanged.

[0052] The compressor compresses refrigerant gas in a high-temperature and high-pressure state and discharges the compressed refrigerant gas. The discharged refrigerant gas flows into the condenser. The condenser condenses the compressed refrigerant into a liquid phase, and heat is released to the surrounding environment through the condensation process.

[0053] The expansion valve expands the liquid-phase refrigerant in a high-temperature and high-pressure state condensed in the condenser into a low-pressure liquid-phase refrigerant. The evaporator evaporates the refrigerant expanded in the expansion valve and returns the refrigerant gas in a low-temperature and low-pressure state to the compressor. The evaporator can achieve a refrigeration effect by exchanging heat with a material to be cooled using latent heat of evaporation of the refrigerant. Throughout the cycle, the air conditioner can adjust the temperature of the indoor space.

[0054] The air conditioner outdoor unit refers to the part of the refrigeration cycle including the compressor and the outdoor heat exchanger, the air conditioner indoor unit includes the indoor heat exchanger, and the expansion valve can be provided in the air conditioner outdoor unit or the indoor unit.

[0055] The indoor heat exchanger and the outdoor heat exchanger are used as a condenser or an evaporator. When the indoor heat exchanger is used as a condenser, the air conditioner functions as a heater in a heating mode, and when the indoor heat exchanger is used as an evaporator, the air conditioner functions as a cooler in a cooling mode.

[0056] Embodiment one,

[0057] The air conditioner control system of the embodiment includes a parameter acquisition module, a model calling module, a control module, etc., as shown in Figure 1 .

[0058] The parameter acquisition module is configured to acquire the unit air conditioner area and the actual indoor environmental parameters in the window closed state.

[0059] The unit air conditioner area = indoor area / indoor air conditioner number. For example, the indoor area is 100 square meters, and there are 2 air conditioners in the indoor, then the unit air conditioner area = 100 / 2 = 50 square meters.

[0060] The unit air conditioner area is collected by the intelligent terminal. The actual indoor environmental parameters are collected by the sensors arranged on the air conditioner.

[0061] The parameter acquisition module receives the unit air conditioner area sent by the intelligent terminal and the actual indoor environmental parameters (including temperature, humidity, air volume, air speed, air direction, oxygen content) collected by the sensor.

[0062] The model calling module is configured to input the unit air conditioner area and the actual indoor environmental parameters in the window closed state acquired by the parameter acquisition module into the preset window opening prediction model, and the window opening prediction model outputs the predicted indoor target environmental parameters in the window opening state.

[0063] The preset window opening prediction model is a window opening prediction model that is pre-trained, and the unit air conditioner area and the actual indoor environment parameter in the window closed state are input into the window opening prediction model, so that the predicted indoor target environment parameter in the window opened state is output.

[0064] The control module is configured to control the operation of the air conditioner according to the predicted indoor target environment parameter output by the window opening prediction model.

[0065] The control module generates a corresponding control instruction according to the received indoor target environment parameter, and controls the operation of the air conditioner.

[0066] The input of the window opening prediction model is the unit air conditioner area and the actual indoor environment parameter in the window closed state, and the output is the predicted indoor target environment parameter in the window opened state. The control module controls the operation of the air conditioner according to the predicted indoor target environment parameter predicted by the window opening prediction model, so that the air conditioner simulates the air flow in the window opened ventilation state, realizes the natural wind function, realizes the human sense natural ventilation in the window closed indoor state, and realizes the human sense comfort unchanged in the window opened ventilation state.

[0067] The air conditioner control system of the embodiment, the design parameter acquisition module acquires the unit air conditioner area and the actual indoor environment parameter in the window closed state; the design model calling module inputs the unit air conditioner area and the actual indoor environment parameter in the window closed state acquired by the parameter acquisition module into the preset window opening prediction model, and the window opening prediction model outputs the predicted indoor target environment parameter in the window opened state; the control module controls the operation of the air conditioner according to the predicted indoor target environment parameter predicted by the window opening prediction model, so that the air conditioner simulates the air flow in the window opened ventilation state, realizes the human sense natural ventilation in the window closed indoor state. Therefore, the air conditioner control system of the embodiment completely replaces the window opened ventilation by the prediction of the window opening prediction model and the control of the air conditioner, realizes the natural wind function, does not need to open the window, reduces the safety hidden danger, and guarantees the human sense comfort unchanged; and the window opening prediction model comprehensively considers the indoor area, the indoor air conditioner quantity and the actual indoor environment parameter, can predict more accurate indoor target environment parameters to control the operation of the air conditioner, so that the air conditioner can simulate the air flow in the window opened ventilation state in the window closed indoor state, improves the natural wind function effect of the air conditioner, and solves the technical problem of poor natural wind function effect of the air conditioner in the prior art.

[0068] The air conditioner control system of the embodiment can simulate the real state of the indoor environment in the window opened ventilation state, considers the local indoor air conditioner quantity and the indoor area of the user, obtains more accurate and more personalized results, and realizes personalized setting.

[0069] In some embodiments of the application, in order to further improve the prediction accuracy, four window opening prediction models are provided, which correspond to four seasons respectively. Of course, the four window opening prediction models are trained separately according to the seasons.

[0070] The model calling module is further configured to: obtain date information, determine a current season according to the date information, select a corresponding open-window prediction model according to the current season, and input the unit air conditioner area and the indoor actual environment parameter in the closed-window state obtained by the parameter obtaining module into the corresponding open-window prediction model.

[0071] Therefore, the model calling module specifically performs the following steps, as shown in Figure 2

[0072] Step S11: Obtain date information.

[0073] The model calling module accesses a cloud platform to obtain current date information.

[0074] Step S12: Determine a current season according to the date information.

[0075] Step S13: Select a corresponding open-window prediction model according to the current season.

[0076] A corresponding table of seasons and open-window prediction models is established in advance. By looking up the table, the corresponding open-window prediction model of the current season can be obtained.

[0077] Step S14: Input the unit air conditioner area and the indoor actual environment parameter in the closed-window state obtained by the parameter obtaining module into the corresponding open-window prediction model. The open-window prediction model outputs a predicted indoor target environment parameter in the open-window state.

[0078] By designing steps S11-S14, the corresponding open-window prediction model is determined according to the current season, so that a more accurate indoor target environment parameter in the open-window state is predicted.

[0079] In some embodiments of the present application, in order to more accurately control the operation of the air conditioner and achieve more accurate natural wind simulation, the indoor actual environment parameter in the closed-window state includes temperature, humidity, air volume, air speed, air direction, and oxygen content. The indoor target environment parameter in the open-window state predicted by the open-window prediction model includes temperature, humidity, air volume, air speed, and air direction.

[0080] Temperature, humidity, air volume, air speed, and air direction can comprehensively represent the indoor target environment, and accordingly generate a control instruction to control the operation of the air conditioner, so that the air conditioner can simulate more accurate air flow and improve user experience.

[0081] In some embodiments of the present application, the air conditioner control system further includes a model establishing module, as shown in Figure 3

[0082] The model establishing module includes a sample obtaining unit and a model training unit, as shown in Figure 4

[0083] ​​​The sample acquisition unit is configured to acquire closed-window sample data and open-window sample data, wherein the closed-window sample data comprises indoor environmental parameters in a closed-window state per unit air conditioner area, and the open-window sample data comprises indoor environmental parameters in an open-window state per unit air conditioner area.

[0084] The model training unit is configured to establish an open-window prediction model and train the open-window prediction model by using the closed-window sample data and the open-window sample data.

[0085] The open-window prediction model is trained by using the closed-window sample data and the open-window sample data, so that a qualified open-window prediction model is trained, and the trained open-window prediction model can accurately predict the indoor target environmental parameters in the open-window state.

[0086] The indoor environmental parameters in the closed-window state acquired by the sample acquisition unit comprise temperature, humidity, wind speed, wind direction, wind volume and oxygen content.

[0087] The indoor environmental parameters in the open-window state acquired by the sample acquisition unit comprise temperature, humidity, wind speed, wind direction, wind volume and oxygen content.

[0088] In some embodiments of the present application, the model establishing module further comprises a data preprocessing unit.

[0089] The data preprocessing unit is configured to perform normalization processing on the closed-window sample data and the open-window sample data acquired by the sample acquisition unit.

[0090] The model training unit trains the open-window prediction model by using the normalized closed-window sample data and the normalized open-window sample data.

[0091] By setting the data preprocessing unit, the data is normalized to ensure the consistency of the data distribution, and then the normalized sample data is used to train the open-window prediction model, so as to ensure that an accurate model is trained and accurate prediction is achieved.

[0092] In some embodiments of the present application, the open-window prediction model is a self-encoder model based on LSTM (Long Short-Term Memory).

[0093] Long Short-Term Memory (LSTM) is a kind of time recurrent neural network.

[0094] The Seq2Seq (Sequence-to-Sequence) self-encoder model uses LSTM as a component to better capture long-term dependencies.

[0095] The LST M layer is used as a core functional unit to build a self-encoder model, allowing variable length output. The self-encoder model is an unsupervised learning model designed to learn an effective representation of data.

[0096] The self-encoder model based on LSTM is convenient for training, and the normalized closed window sample data and open window sample data are used to train the self-encoder model. After training, the self-encoder model can achieve accurate prediction.

[0097] The self-encoder model includes an encoder and a decoder, which are used to reconstruct the input data into output data.

[0098] In some embodiments of the present application, the self-encoder model includes an encoder and a decoder, as shown in Figure 5

[0099] The encoder receives the closed window sample data as the input sequence and encodes the input sequence into the hidden state.

[0100] The decoder receives the hidden state from the encoder as the initial state and inputs the open window sample data to output the predicted indoor environmental parameters in the open window state.

[0101] The closed window sample data includes unit air conditioner area, temperature, humidity, wind speed, wind direction, air volume, and oxygen content.

[0102] The open window sample data includes unit air conditioner area, temperature, humidity, wind speed, wind direction, air volume, and oxygen content.

[0103] The predicted indoor environmental parameters in the open window state output by the decoder include temperature, humidity, air volume, wind speed, and wind direction.

[0104] By designing the above-mentioned encoder and decoder, it is convenient to train using closed window sample data and open window sample data. After training, a qualified self-encoder model can be obtained to achieve accurate prediction.

[0105] Next, data collection, data preprocessing, model building and training are described in detail.

[0106] The temperature, humidity, wind speed, wind direction, air volume, and oxygen content in the indoor area (the indoor area = unit air conditioner area when data collection) are collected by sensors. A single sample is collected at a time step of 24 hours, and the closed window and open window states are collected as sample data for model training.

[0107] ​The window-closed sample data includes temperature, humidity, wind speed, wind direction, air volume, oxygen content, and indoor area in the closed window state. The temperature, humidity, wind speed, wind direction, air volume, oxygen content, and indoor area are collected once an hour, and 24 times a day, as shown in Figure 6 . .

[0108] The window-open sample data includes temperature, humidity, wind speed, wind direction, air volume, oxygen content, and indoor area in the open window state. The temperature, humidity, wind speed, wind direction, air volume, oxygen content, and indoor area are collected once an hour, and 24 times a day, as shown in Figure 7 . .

[0109] Before training the model using the sample data, the window-open sample data and the window-closed sample data need to be normalized to ensure consistency in the distribution of the data.

[0110] For example, linear normalization is performed using the following formula to linearly transform the original data x to the range of [0, 1].

[0111] The normalization formula is x' = (x-min(x)) / (max(x)-min(x)).

[0112] Where x' is the normalized data, x is the original data (collected feature value), min(x) is the minimum value of the feature x at the same location in the sample set, and max(x) is the maximum value of the feature x at the same location in the sample set.

[0113] The window-closed sample data, as the input sequence of the encoder, contains seven features: temperature, humidity, wind speed, wind direction, air volume, oxygen content, and unit air conditioner area, with a total of 24 time steps, and the shape of a single sample is (24, 7).

[0114] The window-closed sample data is reshaped to (sample number, time step length, feature) before being input into the model. A total of 365 sample data is planned to be collected throughout the year. The sample data is sequentially input into the encoder and the decoder for training.

[0115] Referring to Figure 8 , the encoder receives the window-closed sample data as the input sequence and outputs a fixed-length hidden state. The decoder receives the hidden state output by the encoder as the initial state and inputs the window-open sample data at each time step.

[0116] The LSTM layer of the decoder outputs intermediate feature parameters, and the fully connected layer of the decoder converts the output of the LSTM layer to the target dimension, i.e., simulates the temperature, humidity, air volume, wind speed, and wind direction when the window is open.

[0117] Throughout the training process, the encoder and the decoder are alternately trained, so that the final decoder can accurately generate the temperature, humidity, air volume, air speed, and air direction when the window is opened.

[0118] The specific steps of training the autoencoder model using the window-closed sample data and the window-open sample data are described with reference to FIG. 8. Figure 9

[0119] Step S21: The window-open sample data is taken as the input sequence of the encoder, and the shape is (sample number, time step, feature).

[0120] Step S22: The LST M layer of the encoder encodes the input sequence into a hidden state and outputs it to the decoder.

[0121] Step S23: The decoder receives the hidden state from the encoder as the initial state and inputs the window-open sample data at each time step. The encoder generates intermediate feature parameters (time series) through the LSTM layer, and the fully connected layer of the decoder converts the intermediate feature parameters to the target dimension, i.e., the temperature, humidity, air volume, air speed, and air direction when the window is opened.

[0122] After the window-open prediction model is trained, it can be deployed on a local device (such as a central controller) or on a cloud platform.

[0123] In some embodiments of the present application, in order to achieve faster and safer prediction, the window-open prediction model is pre-installed in the central controller.

[0124] By setting the window-open prediction model on the central controller, the data is more secure and the operation is faster during model training and use.

[0125] The model calling module is also integrated into the central controller, which facilitates the calling of the window-open prediction model.

[0126] In some embodiments of the present application, the control module is integrated into the central controller, which facilitates centralized control of multiple indoor units.

[0127] In some other embodiments of the present application, the control module can also be integrated into the indoor unit to facilitate the control of the indoor unit.

[0128] The model establishment module is also integrated into the central controller, and the establishment and training of the window-open prediction model can be completed before the air conditioner is shipped.

[0129] The four window-open prediction models trained according to the seasons are pre-installed in the central controller chip, and the central controller can locally store two hours of collected data of indoor temperature, humidity, air speed, air volume, air direction, and oxygen content.

[0130] ​In some embodiments of the present application, in order to facilitate user input of parameters and control of the air conditioner, the air conditioner control system further comprises a smart terminal.

[0131] The smart terminal is configured to receive user input of indoor area and indoor air conditioner quantity, and calculate unit air conditioner area according to the user input of indoor area and indoor air conditioner quantity, and send the calculated unit air conditioner area to the parameter acquisition module.

[0132] The parameter acquisition module is also integrated on the central controller for facilitating communication with the smart terminal.

[0133] The smart terminal communicates with the central controller through a cloud platform, as shown in Figure 10

[0134] The smart terminal is installed with an APP for user input of parameters and instructions, etc. to control the air conditioner.

[0135] The smart terminal can be a smart phone, a PAD, etc.

[0136] Of course, the smart terminal can also be a wire control device, a remote control device, etc.

[0137] When the user wants to start the natural ventilation function of the air conditioner, the indoor area and the air conditioner quantity are input on the APP of the smart terminal, the APP of the smart terminal calculates the unit air conditioner area, and sends the calculated unit air conditioner area to the central controller through the cloud platform. After receiving the unit air conditioner area, the central controller imports the window opening prediction model together with the locally stored indoor environment data (temperature, humidity, air volume, air speed, air direction, oxygen content), obtains the indoor temperature, humidity, air volume, air speed and air direction in the natural window opening state through the window opening prediction model, and then sends them to the indoor unit. The indoor unit operates according to the parameters sent by the central controller to realize natural comfort in the closed window state.

[0138] The indoor unit sets the operating parameters according to the indoor environment parameters predicted by the window opening prediction model to realize the human-sensing natural ventilation in the closed window condition.

[0139] The air conditioner control system of the present embodiment realizes that the air conditioner can replace the window ventilation in function and human-sensing comfort in the closed window state by simulating the air flow change in the indoor environment opening ventilation on the edge (air conditioner indoor unit) through the built-in self-encoder model on the air conditioner central controller, thereby enhancing the safety of houses, personnel and properties, and improving the indoor environment air quality.

[0140] ​The air conditioner control system of the embodiment is based on LSTM to build a self-encoder, uses an LSTM layer for processing time sequence in the self-encoder model, and is deployed on an air conditioner central controller to enable the air conditioner to learn airflow direction, temperature change, humidity change, wind speed, wind direction and the like in the open window natural ventilation state at the local air conditioner, to meet different needs based on personalized input parameters (indoor area, air conditioner quantity), to realize exclusive natural ventilation control for different time points and different indoor buildings, to replace open window ventilation without affecting human comfort, to improve indoor safety, and to enable the air conditioner to simulate human feeling and air flow in an open window ventilation environment for a commercial or residential building with high closedness.

[0141] The air conditioner control system of the embodiment adopts a Seq2Seq self-encoder model, can simulate the real state of the indoor environment in the open window ventilation state at each time point, considers the local air conditioner quantity and indoor area of the user to obtain more accurate and personalized results. Meanwhile, the model runs at the local end (central controller) of the user to realize target prediction more quickly and more safely. The air conditioner control system of the embodiment is suitable for air conditioner control of large buildings and high-rise residential buildings, completely replaces open window ventilation through model simulation and air conditioner control, reduces safety hazards, and ensures constant human comfort.

[0142] Embodiment two,

[0143] Based on the design of the air conditioner control system of embodiment one, embodiment two proposes an air conditioner including the air conditioner control system.

[0144] By designing the air conditioner control system in the air conditioner, the air flow in the open window ventilation state can be simulated in the closed window state of the indoor environment, the natural wind function effect of the air conditioner is improved, the technical problem of poor natural wind function effect of the air conditioner in the prior art is solved, the safety of houses, personnel and property is improved, and the air quality of the indoor environment is improved.

[0145] The air conditioner includes an outdoor unit, an indoor unit, a central controller, an air conditioner control system and the like, as shown in Figure 10 .

[0146] In some embodiments of the application, the air conditioner control system is arranged on the central controller, which facilitates the establishment, training and use of the open window prediction model, and can ensure data security and running speed.

[0147] 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.

[0148] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. An air conditioning control system, characterized in that, include: The parameter acquisition module is configured to acquire indoor environmental parameters per unit air-conditioned area under closed window conditions; the unit air-conditioned area = indoor area / number of indoor air-conditioned units; The model calling module is configured to input the unit air-conditioned area and the actual indoor environmental parameters under closed window conditions obtained by the parameter acquisition module into a preset window opening prediction model, and the window opening prediction model outputs the predicted indoor target environmental parameters under open window conditions. The control module is configured to control the operation of the air conditioner based on the indoor target environmental parameters output by the window opening prediction model.

2. The air conditioning control system according to claim 1, characterized in that: The window opening prediction model is set up with four models, each corresponding to one of the four seasons; The model invocation module is also configured as follows: Obtain date information and determine the current season based on the date information; Select the corresponding window opening prediction model based on the current season; The parameters obtained by the parameter acquisition module, such as the unit air-conditioned area and the actual indoor environmental parameters under closed window conditions, are input into the corresponding window opening prediction model.

3. The air conditioning control system according to claim 1, characterized in that: The actual indoor environmental parameters under the closed window condition include temperature, humidity, air volume, wind speed, wind direction, and oxygen content; The indoor target environmental parameters include temperature, humidity, air volume, wind speed, and wind direction.

4. The air conditioning control system according to claim 1, characterized in that: The air conditioning control system also includes a model building module, which includes: A sample acquisition unit is used to acquire closed window sample data and open window sample data; the closed window sample data includes indoor environmental parameters per unit air-conditioned area and when the windows are closed; the open window sample data includes indoor environmental parameters per unit air-conditioned area and when the windows are open. The model training unit is used to establish the windowing prediction model and train the windowing prediction model using the closed window sample data and the open window sample data.

5. The air conditioning control system according to claim 4, characterized in that: The model building module also includes a data preprocessing unit; The data preprocessing unit is used to normalize the closed-window sample data and open-window sample data acquired by the sample acquisition unit. The model training unit uses normalized closed-window sample data and open-window sample data to train the open-window prediction model.

6. The air conditioning control system according to claim 4, characterized in that: The windowing prediction model is an autoencoder model based on LSTM.

7. The air conditioning control system according to claim 6, characterized in that: The autoencoder model includes an encoder and a decoder; The encoder receives closed-window sample data as an input sequence and encodes the input sequence into a hidden state. The decoder receives the hidden state from the encoder as the initial state, inputs open window sample data, and outputs the predicted indoor environmental parameters under the open window state.

8. The air conditioning control system according to claim 1, characterized in that: The windowing prediction model is pre-set in the central controller.

9. The air conditioning control system according to any one of claims 1 to 8, characterized in that: The air conditioning control system also includes: The intelligent terminal is used to calculate the unit air-conditioning area based on the indoor area and the number of indoor air conditioners input by the user, and send the calculated unit air-conditioning area to the parameter acquisition module.

10. An air conditioner, characterized in that, The air conditioning control system includes any one of claims 1 to 9.

Images