Air conditioner control method, device, system and air conditioner

Abstract

The present invention discloses an air conditioner control method, device, system, and air conditioner. The method comprises: obtaining the cooling capacity of the air conditioner and obtaining outdoor environmental parameters; inputting the cooling capacity and environmental parameters into a trained neural network model to predict a net heat load; dynamically generating adjustment instructions for controlling the operating state of the air conditioner based on the net heat load; and controlling corresponding components of the air conditioner according to the adjustment instructions. The present invention utilizes the predicted net heat load to generate adjustment instructions for controlling the operating state of the air conditioner, and then controls corresponding components of the air conditioner according to the adjustment instructions, thereby improving the response speed of the air conditioner and achieving real-time control.

Description

Technical Field

[0001] The present invention relates to the technical field of air conditioning, and in particular to an air conditioner control method, device, system and air conditioner. Background Art

[0002] Air conditioners, as common temperature and humidity control devices, are widely used in residential and work environments, and their performance and intelligence are attracting increasing attention. However, most current home air conditioners still employ fixed control strategies that fail to fully account for the diverse usage scenarios of different users, such as room size, building envelope, and heat load. This results in a lack of effective correlation between the air conditioner's cooling capacity and the rate of change of space temperature in specific application scenarios.

[0003] Based on this, existing air conditioning control strategies primarily rely on a feedback mechanism. This involves indirectly estimating the difference between the air conditioner's cooling capacity and the room's heat load based on the rate of change of room temperature after the air conditioner has been running for a period of time. Operating parameters such as compressor frequency and wind speed are adjusted accordingly. However, this feedback control method exhibits significant lag, making it impossible to quickly match optimal control parameters during the air conditioner's initial operation. Furthermore, it is slow to respond to rapid changes in the external environment, making it difficult to optimize operating parameters in a timely manner. This impacts the air conditioner's temperature control accuracy and the user's comfort experience. Summary of the Invention

[0004] The embodiments of the present invention provide an air conditioner control method, device, system and air conditioner, aiming to solve the problem of slow response speed of air conditioners in the prior art.

[0005] In a first aspect, an embodiment of the present invention provides an air conditioner control method, comprising:

[0006] Obtain the cooling capacity of the air conditioner and the outdoor environmental parameters;

[0007] Inputting the cooling capacity and environmental parameters into a trained neural network model to predict the net heat load;

[0008] dynamically generating an adjustment instruction for controlling the operating state of the air conditioner based on the net heat load;

[0009] The corresponding components of the air conditioner are controlled according to the adjustment instruction.

[0010] In a second aspect, an embodiment of the present invention provides an air conditioner control device, comprising:

[0011] A data acquisition unit, used to obtain the cooling capacity of the air conditioner and outdoor environmental parameters;

[0012] A data prediction unit inputs the cooling capacity and environmental parameters into a trained neural network model to predict a net heat load;

[0013] an instruction generating unit, configured to dynamically generate an adjustment instruction for controlling an operating state of the air conditioner based on the net heat load;

[0014] The instruction application unit is used to control the corresponding components of the air conditioner according to the adjustment instruction.

[0015] In a third aspect, an embodiment of the present invention provides an air conditioner control system, comprising a cloud server, a user terminal, and the air conditioner control device of the second aspect, wherein the cloud server is connected to the user terminal and the air conditioner control device respectively.

[0016] In a fourth aspect, an embodiment of the present invention provides an air conditioner, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the air conditioner control method of the first aspect when executing the computer program.

[0017] An embodiment of the present invention provides an air conditioner control method, comprising: obtaining the cooling capacity of the air conditioner and obtaining outdoor environmental parameters; inputting the cooling capacity and environmental parameters into a trained neural network model to predict a net heat load; dynamically generating adjustment instructions for controlling the operating state of the air conditioner based on the net heat load; and controlling corresponding components of the air conditioner according to the adjustment instructions. The present invention utilizes the predicted net heat load to generate adjustment instructions for controlling the operating state of the air conditioner, and then controls corresponding components of the air conditioner according to the adjustment instructions, thereby improving the response speed of the air conditioner and achieving real-time control.

[0018] The embodiments of the present invention further provide an air conditioner control device, a system and an air conditioner, which also have the above-mentioned beneficial effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 A schematic flow chart of an air conditioner control method provided by an embodiment of the present invention;

[0021] Figure 2 A structural diagram of a multi-layer neuron model provided by an embodiment of the present invention;

[0022] Figure 3A schematic block diagram of an air conditioner control device provided by an embodiment of the present invention;

[0023] Figure 4 This is an architectural diagram of an air conditioner control system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0025] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.

[0026] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the present invention. As used in the specification and appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0027] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.

[0028] See below Figure 1 , Figure 1 A flowchart of an air conditioner control method provided by an embodiment of the present invention specifically includes steps S101 to S104.

[0029] S101, obtaining the cooling capacity of the air conditioner and obtaining outdoor environmental parameters;

[0030] S102, inputting the cooling capacity and environmental parameters into a trained neural network model to predict a net heat load;

[0031] S103, dynamically generating an adjustment instruction for controlling the operating state of the air conditioner based on the net heat load;

[0032] S104: Control corresponding components of the air conditioner according to the adjustment instruction.

[0033] In step S101, the air conditioner's real-time cooling capacity and outdoor environmental parameters are obtained. Simultaneously, the cloud server (smart home system) can call the weather API based on the user's address information to obtain real-time weather information, such as temperature, humidity, and wind speed. The air conditioner connects to the home network via its built-in Wi-Fi module and communicates with the cloud server, transmitting data including the air conditioner's on / off status, operating mode, and ambient temperature, thereby obtaining outdoor environmental parameters.

[0034] In one embodiment, step S101 includes:

[0035] respectively collecting the air inlet temperature and the air outlet temperature of the indoor unit of the air conditioner;

[0036] Calculate the difference between the air inlet temperature and the air outlet temperature to obtain the indoor ambient temperature difference;

[0037] Collect the air volume information of the indoor unit and the specific heat capacity of the air in the indoor environment;

[0038] The ambient temperature difference, air volume information and air specific heat capacity are input into a cooling capacity calculation model to calculate the cooling capacity of the air conditioner.

[0039] In this embodiment, the temperature data of the air inlet and outlet of the indoor unit of the air conditioner are collected separately. The indoor unit is equipped with multiple high-precision temperature sensors, which are respectively arranged at the air inlet and outlet positions to collect the current air inlet temperature in real time. and air outlet temperature Among them, the inlet air temperature can be equivalent to the indoor ambient temperature , the two can replace each other depending on the specific situation.

[0040] Furthermore, the difference between the collected inlet and outlet temperatures is calculated to obtain the temperature change of the indoor air during the operation of the air conditioner, that is, the ambient temperature difference ΔT = - . Then collect the current air volume information of the air conditioner indoor unit Air volume information can be measured using the air conditioner's built-in wind speed sensor or fan operating parameters, representing the volumetric flow rate of air handled by the air conditioner per unit time. Furthermore, based on the physical properties of actual indoor air, the specific heat capacity (C) of air in the current environment is collected or preset. This parameter represents the amount of heat required to change the temperature of a unit mass of air and is a constant term in cooling capacity calculations.

[0041] Finally, the above ambient temperature difference, air volume information and air specific heat capacity are used as input parameters and imported into the cooling capacity calculation model preset by the air conditioner to calculate the real-time cooling capacity of the air conditioner under the current working conditions. The cooling capacity calculation formula is as follows:

[0042] ;

[0043] Among them, Q(t) is the heat exchange amount of the air conditioner to the air per unit time (i.e., cooling capacity), expressed in kilowatts (kW) or joules / second (J / s), which can be used to predict the net heat load in the subsequent neural network model.

[0044] In step S102, the acquired cooling capacity and environmental parameters are input into a pre-trained neural network model. During the training phase, the neural network model has established a nonlinear functional relationship between cooling capacity, environmental parameters, and net heat load. After inputting the parameters, the neural network model can output a predicted net heat load at the corresponding moment. Net heat load refers to the total heat load of the space (room) where the air conditioner is located minus the air conditioner's cooling capacity. Traditional control methods estimate the net heat load by subtracting the total heat load from the air conditioner's cooling capacity based on the room temperature change rate after the air conditioner has been running for a certain period of time. However, this embodiment uses a neural network model to predict the net heat load in real time, replacing the traditional estimation method. This method then performs subsequent dynamic adjustments, reducing control lag.

[0045] Combine Figure 2 As shown, in one embodiment, step S102 includes:

[0046] Obtaining training set data, and inputting the training set data into a multi-layer neuron model for training;

[0047] By adjusting the connection weight parameters in the multi-layer neuron model, the error between the predicted net heat load and the actually measured net heat load is minimized, so as to complete the fitting training of the multi-layer neuron model;

[0048] The cooling capacity and environmental parameters obtained in real time are input into the trained multi-layer neuron model for prediction to obtain the net heat load.

[0049] In this embodiment, during the model training phase, training set data is obtained to construct a multi-layer neuron model. The training set data includes parameters such as cooling capacity, indoor ambient temperature, external ambient temperature, and external ambient humidity corresponding to multiple time points. In addition, it also includes the actual net heat load corresponding to the time point obtained by experimental or simulation methods. The above-mentioned training data can be derived from the historical operation records of the laboratory test environment or the target application area. The above-mentioned training set data is input into the constructed multi-layer neuron model for fitting training. The model consists of an input layer, one or more hidden layers, and an output layer. The number of nodes in the input layer is consistent with the number of input parameters, and the output layer is used to output the predicted net heat load. Specifically, a suitable activation function is selected, such as sigmoid and tanh, and during the training process, the model output value is calculated through forward propagation, and the predicted result is compared with the actual net heat load. The prediction error is measured using a preset loss function (such as mean square error). The loss function is set to measure the error between the predicted value and the actual value. In the subsequent training process, it is continuously iterated and optimized until the loss function reaches a preset threshold or converges.

[0050] Furthermore, the backpropagation algorithm is combined with an optimizer (such as gradient descent) to iteratively adjust the connection weight parameters in the model, thereby continuously reducing the error between the predicted value and the actual value. This process continues until the loss function converges to a set threshold or the maximum number of training rounds is reached, thus completing the fitting training of the neural network model.

[0051] After the model training is completed, the model application phase begins. During the actual operation of the air conditioner, the current cooling capacity is obtained in real time. and environmental parameters (such as ambient temperature ,humidity , etc.), and imported into the trained multi-layer neuron model as input variables for forward prediction, outputting the net heat load corresponding to the current moment. , the specific calculation formula is as follows:

[0052] .

[0053] It is also possible to incorporate more diverse data into the model training process to simulate various actual situations, including different building structures, climatic conditions, and usage habits, to improve the model's generalization ability.

[0054] In step S103, based on the net heat load value predicted by the neural network model and its changing trend, control instructions for adjusting the operating state of the air conditioner are dynamically generated. The control instructions may include adjustments to parameters such as compressor frequency, fan speed, and electronic expansion valve opening.

[0055] In one embodiment, step S103 includes:

[0056] comparing the net heat load to a load threshold;

[0057] If the net heat load is greater than or equal to the load threshold and has an increasing trend, generating an adjustment instruction to increase the compressor frequency by a first preset amplitude;

[0058] If the net heat load is less than the load threshold and has no increasing trend, an adjustment instruction is generated to maintain the compressor frequency unchanged or to increase the compressor frequency according to a second preset amplitude; wherein the first preset amplitude is greater than the second preset amplitude.

[0059] In this embodiment, the net heat load output by the neural network model is obtained and compared with a preset load threshold. The load threshold can be pre-set based on factors such as the user's target temperature, indoor space size, and typical heat load characteristics to define different control ranges for the air conditioner's operation. A determination is then made as to whether the net heat load is greater than or equal to the load threshold and whether it is showing a trend of continuous increase. If the determination is "yes," meaning the net heat load is greater than or equal to the load threshold and is showing an increasing trend, an adjustment instruction is generated to increase the air conditioner's operating intensity. This instruction includes increasing the compressor operating frequency by a first preset amount to rapidly enhance cooling capacity to meet current and upcoming load demands. If the determination is "no," meaning the net heat load is less than the load threshold and is not showing an increasing trend, another adjustment instruction is generated to maintain the current compressor frequency unchanged or to slightly increase it by a second preset amount. This strategy aims to reduce energy consumption, extend equipment life, and avoid unnecessary frequent adjustments while ensuring effective temperature control.

[0060] It's important to note that the first preset amplitude is larger than the second preset amplitude. The former is used to cope with high-intensity load changes, while the latter is suitable for light load adjustments, thereby achieving graded control of the compressor frequency under different operating conditions. Through the above logical judgment and strategy setting, the air conditioner operating parameters can be flexibly adjusted according to the changing trend of the net heat load, achieving precise temperature control, energy saving and high efficiency.

[0061] In step S104, the air conditioner executes control operations on the corresponding operating components based on the generated adjustment instructions. Through this closed-loop control process, the air conditioner can quickly respond to current environmental changes and intelligently adjust its operating status without waiting for temperature feedback lag, thereby effectively improving temperature control efficiency and user comfort.

[0062] In one embodiment, after step S104, the following steps are included:

[0063] Collecting indoor ambient temperature data before and after a preset time interval, and calculating the indoor temperature change rate before and after the preset time interval based on the indoor ambient temperature data;

[0064] The cooling capacity, environmental parameters and indoor temperature change rate are input into the neural network model for training to optimize the neural network model.

[0065] In this embodiment, during the continuous operation of the air conditioner, the ambient temperature sensor installed inside the indoor unit collects indoor temperature data before and after the preset time interval. That is, the corresponding indoor ambient temperature is collected at time t and time t+Δt respectively. and , used to calculate the indoor temperature change rate ΔT / Δt within the time interval Δt. This temperature change rate can reflect the actual effect of the air conditioner on indoor temperature regulation within a certain period of time and is an important dynamic parameter for measuring the operating status of the air conditioner. The calculation formula is as follows:

[0066] ;

[0067] The calculated indoor temperature change rate over a time period, along with the air conditioner cooling capacity and environmental parameters during the same time period, is input into the neural network model for incremental training or online optimization. This optimization process incorporates newly added actual operating data into the training set and fine-tunes the model parameters, gradually establishing a more accurate mapping between input and output. By introducing the indoor temperature change rate as an auxiliary input variable, the neural network model can more comprehensively capture the dynamic characteristics of the environment, improving the accuracy of net heat load prediction in complex usage scenarios and further enhancing the model's adaptability to different building structures, user habits, and external climatic conditions. This optimization mechanism continues over long-term user usage, helping to build a highly adaptive and continuously learning air conditioning control system, effectively improving the intelligence level and control performance of the air conditioner. It is important to note that model training may be affected by other factors such as ventilation and human activity, requiring the user to manually select an appropriate training period.

[0068] In one embodiment, the air conditioner control method further includes:

[0069] Collecting user operation record data during the operation of the air conditioner;

[0070] Building a user usage behavior dataset based on the operation record data, and inputting the user usage behavior dataset into a personalized behavior recognition model to extract the user's usage preference characteristics in different time periods;

[0071] Based on the usage preference characteristics, calculating the probability of correlation between the user operation parameter and the net heating load or the cooling capacity in a corresponding time period;

[0072] Generate personalized operation suggestions based on the correlation probability, and push the personalized operation suggestions to the user terminal.

[0073] In this embodiment, user operation log data is collected in real time during air conditioner operation, including information such as temperature setting values, fan speed selections, operating modes, and operating time periods. This information is automatically recorded and stored locally or uploaded to a cloud server by the air conditioner's built-in control module. Based on the collected operation log data, a user usage behavior dataset is constructed and input into a pre-trained personalized behavior recognition model. This personalized behavior recognition model is capable of classifying and extracting features from user operation behaviors over different time periods, identifying specific user preferences in different environmental conditions, such as morning and evening, and high or low temperatures. For example, it can identify user preferences such as a tendency to increase the set temperature and reduce the fan speed at night, or to frequently set the temperature to high fan speed during summer midday.

[0074] Next, based on the extracted usage preference features, the system further calculates the probability of correlation between the user's operating parameters (such as temperature setpoint and fan speed level) and the net heat load or cooling capacity during the air conditioner's operation within the corresponding time period. Based on this probability, personalized operation recommendations can be generated, such as "automatically increase the set temperature to 26°C and reduce the fan speed at 10 pm" or "automatically enable rapid cooling mode during hot periods." These personalized operation recommendations are pushed to the user's terminal (such as a mobile app or air conditioner remote control interface). Users can choose to accept or adjust the recommendations based on their actual needs, thus achieving a deep match between the air conditioner's operating mode and user habits, further improving the air conditioner's intelligent control capabilities and user satisfaction.

[0075] In one embodiment, the air conditioner control method further includes:

[0076] Receive a user's confirmation instruction for the personalized operation suggestion, and modify the weight parameters of the neural network model according to the confirmation instruction.

[0077] In this embodiment, after generating personalized operational recommendations and pushing them to the user terminal, the user can interactively confirm these recommendations through a smart terminal interface (such as a mobile application or the air conditioner's remote control panel). When the user approves the recommendations, the system receives and records the user's confirmation. Based on this confirmation, the system modifies the weight parameters of the neural network model. Specifically, the operating parameters corresponding to the personalized operational recommendations (such as set temperature, fan speed, and operating time period) that the user confirms and adopts are used as positive feedback signals and marked as the user's preferred parameter configuration. The system then recalculates the weights of the relevant input features in the model based on historical data such as net heating load or cooling capacity during the corresponding time period. This modification process can be achieved by introducing a fine-tuning mechanism based on the existing neural network model. For example, while retaining the original model structure and basic parameters, a learning rate is set to update the weights of specific parameters, thereby ensuring that the model output more closely reflects the user's actual usage preferences. This modification strategy can be implemented locally on the device or by distributing the training tasks to the cloud to reduce the burden on the air conditioner's local computing resources.

[0078] Through the above method, the system can continuously optimize the model performance based on user participation and confirmation, so that the neural network can not only adapt to different physical environmental conditions, but also gradually learn and reflect the usage preferences of different users, achieving a more personalized air conditioner intelligent control effect.

[0079] Combine Figure 3 As shown, the embodiment of the present invention further provides an air conditioner control device, the air conditioner control device 300, comprising:

[0080] The data acquisition unit 301 is used to obtain the cooling capacity of the air conditioner and the outdoor environmental parameters;

[0081] The data prediction unit 302 is used to input the cooling capacity and environmental parameters into the trained neural network model to predict the net heat load;

[0082] An instruction generating unit 303 is configured to dynamically generate an adjustment instruction for controlling the operating state of the air conditioner based on the net heat load;

[0083] The instruction application unit 304 is used to control corresponding components of the air conditioner according to the adjustment instruction.

[0084] In this embodiment, the data acquisition unit 301 obtains the cooling capacity of the air conditioner and the outdoor environmental parameters; the data prediction unit 302 inputs the cooling capacity and environmental parameters into the trained neural network model to predict the net heat load; the instruction generation unit 303 dynamically generates adjustment instructions for controlling the operating status of the air conditioner based on the net heat load; the instruction application unit 304 controls the corresponding components of the air conditioner according to the adjustment instructions.

[0085] In one embodiment, the air conditioner control device 300 is further configured to:

[0086] Collecting indoor ambient temperature data before and after a preset time interval, and calculating the indoor temperature change rate before and after the preset time interval based on the indoor ambient temperature data;

[0087] The cooling capacity, environmental parameters and indoor temperature change rate are input into the neural network model for training to optimize the neural network model.

[0088] In one embodiment, the air conditioner control device 300 is further configured to:

[0089] Collecting user operation record data during the operation of the air conditioner;

[0090] Building a user usage behavior dataset based on the operation record data, and inputting the user usage behavior dataset into a personalized behavior recognition model to extract the user's usage preference characteristics in different time periods;

[0091] Based on the usage preference characteristics, calculating the probability of correlation between the user operation parameter and the net heating load or the cooling capacity in a corresponding time period;

[0092] Generate personalized operation suggestions based on the correlation probability, and push the personalized operation suggestions to the user terminal.

[0093] In one embodiment, the air conditioner control device 300 is further configured to:

[0094] Receive a user's confirmation instruction for the personalized operation suggestion, and modify the weight parameters of the neural network model according to the confirmation instruction.

[0095] In one embodiment, the data acquisition unit 301 is specifically configured to:

[0096] respectively collecting the air inlet temperature and the air outlet temperature of the indoor unit of the air conditioner;

[0097] Calculate the difference between the air inlet temperature and the air outlet temperature to obtain the indoor ambient temperature difference;

[0098] Collect the air volume information of the indoor unit and the specific heat capacity of the air in the indoor environment;

[0099] The ambient temperature difference, air volume information and air specific heat capacity are input into a cooling capacity calculation model to calculate the cooling capacity of the air conditioner.

[0100] In one embodiment, the data prediction unit 302 is specifically configured to:

[0101] Obtaining training set data, and inputting the training set data into a multi-layer neuron model for training;

[0102] By adjusting the connection weight parameters in the multi-layer neuron model, the error between the predicted net heat load and the actually measured net heat load is minimized, so as to complete the fitting training of the multi-layer neuron model;

[0103] The cooling capacity and environmental parameters obtained in real time are input into the trained multi-layer neuron model for prediction to obtain the net heat load.

[0104] In one embodiment, the instruction generation unit 303 is specifically configured to:

[0105] comparing the net heat load to a load threshold;

[0106] If the net heat load is greater than or equal to the load threshold and has an increasing trend, generating an adjustment instruction to increase the compressor frequency by a first preset amplitude;

[0107] If the net heat load is less than the load threshold and has no increasing trend, an adjustment instruction is generated to maintain the compressor frequency unchanged or to increase the compressor frequency according to a second preset amplitude; wherein the first preset amplitude is greater than the second preset amplitude.

[0108] Since the embodiments of the apparatus part correspond to the embodiments of the method part, please refer to the description of the embodiments of the method part for the embodiments of the apparatus part, and they will not be repeated here.

[0109] Combine Figure 4 As shown, an embodiment of the present invention further provides an air conditioner control system, comprising a cloud server, a user terminal, and the air conditioner control device as described above, wherein the cloud server is connected to the user terminal and the air conditioner control device respectively.

[0110] In this embodiment, the air conditioner control device is set inside the air conditioner to implement the air conditioner control method. Figure 4The control module in the air conditioner control device is not included, but some functions of the air conditioner control device can be placed in the cloud server. The indoor air inlet temperature sensor is used to collect the air inlet temperature of the indoor unit, the indoor air outlet temperature sensor is used to collect the air outlet temperature of the indoor unit, and the indoor ambient temperature sensor is used to collect the indoor ambient temperature. The air inlet temperature and indoor ambient temperature can be interchanged. The networking module is used to interact with the cloud server.

[0111] The cloud server is responsible for high-performance computing and data storage. It receives operational data, environmental information, and user operation records from the air conditioner control unit and centrally trains, optimizes, and updates the neural network model. The cloud server also features a user behavior analysis module that generates personalized operational recommendations based on historical user usage data. The optimized model parameters are then distributed to the local air conditioner unit, enabling online optimization and adaptive adjustment of the system model.

[0112] User terminals, typically smartphones, tablets, or home control hubs, enable human-computer interaction. Users can use these terminals to view the air conditioner's operating status in real time, receive and confirm personalized operating recommendations, and manually adjust operating parameters or set control preferences. These terminals communicate with the cloud server over the network, receiving push notifications and sending user instructions and feedback to the cloud server.

[0113] Through the above-mentioned structural configuration, this system can realize data interaction and functional coordination between the air conditioner, cloud server and user terminal. On the basis of ensuring efficient prediction and precise control, it further improves the intelligence level of the system and user experience, and has good practicality and scalability.

[0114] The present invention also provides an air conditioner that may include a memory and a processor. The memory stores a computer program. When the processor calls the computer program in the memory, the steps provided in the above embodiment can be implemented. Of course, the air conditioner may also include various network interfaces, a power supply, and other components.

[0115] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of this application, several improvements and modifications can be made to this application, and these improvements and modifications also fall within the scope of protection of the claims of this application.

[0116] It should also be noted that, in this specification, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

Claims

1. A method for controlling an air conditioner, characterized in that: include: Obtain the cooling capacity of the air conditioner and the outdoor environmental parameters; Inputting the cooling capacity and environmental parameters into a trained neural network model to predict the net heat load; dynamically generating an adjustment instruction for controlling the operating state of the air conditioner based on the net heat load; controlling corresponding components of the air conditioner according to the adjustment instructions; The method of inputting the cooling capacity and environmental parameters into a trained neural network model to predict the net heat load includes: obtaining training set data and inputting the training set data into a multi-layer neural model for training; minimizing the error between the predicted net heat load and the actually measured net heat load by adjusting the connection weight parameters in the multi-layer neural model to complete the fitting training of the multi-layer neural model; and inputting the cooling capacity and environmental parameters obtained in real time into the trained multi-layer neural model for prediction to obtain the net heat load.

2. The air conditioner control method according to claim 1, wherein: After controlling the corresponding components of the air conditioner according to the adjustment instruction, the method includes: Collecting indoor ambient temperature data before and after a preset time interval, and calculating the indoor temperature change rate before and after the preset time interval based on the indoor ambient temperature data; The cooling capacity, environmental parameters and indoor temperature change rate are input into the neural network model for training to optimize the neural network model.

3. The air conditioner control method according to claim 1, wherein: Also includes: Collecting user operation record data during the operation of the air conditioner; Building a user usage behavior dataset based on the operation record data, and inputting the user usage behavior dataset into a personalized behavior recognition model to extract the user's usage preference characteristics in different time periods; Based on the usage preference characteristics, calculating the probability of correlation between the user operation parameter and the net heating load or the cooling capacity in a corresponding time period; Generate personalized operation suggestions based on the correlation probability, and push the personalized operation suggestions to the user terminal.

4. The air conditioner control method according to claim 3, wherein: Also includes: Receive a user's confirmation instruction for the personalized operation suggestion, and modify the weight parameters of the neural network model according to the confirmation instruction.

5. The air conditioner control method according to claim 1, wherein: The obtaining of the cooling capacity of the air conditioner includes: respectively collecting the air inlet temperature and the air outlet temperature of the indoor unit of the air conditioner; Calculate the difference between the air inlet temperature and the air outlet temperature to obtain the indoor ambient temperature difference; Collect the air volume information of the indoor unit and the specific heat capacity of the air in the indoor environment; The ambient temperature difference, air volume information and air specific heat capacity are input into a cooling capacity calculation model to calculate the cooling capacity of the air conditioner.

6. The air conditioner control method according to claim 1, wherein: The dynamically generating an adjustment instruction for controlling the operating state of the air conditioner based on the net heat load includes: comparing the net heat load to a load threshold; If the net heat load is greater than or equal to the load threshold and has an increasing trend, generating an adjustment instruction to increase the compressor frequency by a first preset amplitude; If the net heat load is less than the load threshold and has no increasing trend, an adjustment instruction is generated to maintain the compressor frequency unchanged or to increase the compressor frequency according to a second preset amplitude; wherein the first preset amplitude is greater than the second preset amplitude.

7. An air conditioner control device, characterized in that: include: A data acquisition unit, used to obtain the cooling capacity of the air conditioner and outdoor environmental parameters; A data prediction unit, configured to input the cooling capacity and environmental parameters into a trained neural network model to predict a net heat load; an instruction generating unit, configured to dynamically generate an adjustment instruction for controlling an operating state of the air conditioner based on the net heat load; an instruction application unit, configured to control corresponding components of the air conditioner according to the adjustment instruction; The data prediction unit is specifically used to obtain training set data and input the training set data into a multi-layer neuron model for training; by adjusting the connection weight parameters in the multi-layer neuron model, the error between the predicted net heat load and the actually measured net heat load is minimized to complete the fitting training of the multi-layer neuron model; the cooling capacity and environmental parameters obtained in real time are input into the trained multi-layer neuron model for prediction to obtain the net heat load.

8. An air conditioner control system, characterized in that: It comprises a cloud server, a user terminal, and the air conditioner control device as claimed in claim 7, wherein the cloud server is connected to the user terminal and the air conditioner control device respectively.

9. An air conditioner, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the air conditioner control method according to any one of claims 1 to 6 when executing the computer program.

Images