An adaptive control method and system for an air conditioner

Abstract

The application provides an adaptive control method and system of an air conditioner, the method comprising: obtaining device state time series data of the air conditioner, obtaining time series environment data and user habit history data; inputting the device state time series data, the time series environment data and the user habit history data into a prediction model trained in advance, performing future working condition prediction based on the prediction model to obtain a future working condition prediction result; inputting the future working condition prediction result into a control model trained in advance, determining an optimal control parameter set corresponding to a maximum value of a target function of the control model under the future working condition; and sending the optimal control parameter set to an edge control end, so that the edge control end drives the air conditioner to operate based on the optimal control parameter set. The application can dynamically adapt to scene changes and optimize control parameters, thereby reducing air conditioner energy consumption, achieving precise energy saving, and improving indoor environment comfort.

Description

Technical Field

[0001] This invention relates to the field of air conditioning technology, and in particular to an adaptive control method and system for air conditioners. Background Technology

[0002] Against the backdrop of global energy shortages and the growing popularity of low-carbon and environmentally friendly concepts, energy efficiency optimization has become a core development direction for air conditioners, which are high-energy-consuming home appliances. Existing air conditioner control technology typically adopts a fixed parameter control mode, which can only start / stop or adjust the frequency based on simple temperature thresholds.

[0003] Existing air conditioner control technology has many limitations. The control parameters are fixed and cannot adapt to dynamic changing scenarios. When faced with dynamic scenarios such as outdoor temperature fluctuations and indoor heat load changes, it is easy to have "overcooling / heating" or "undercooling / heating" phenomena, resulting in energy waste and insufficient comfort. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide an adaptive control method and system for an air conditioner, which can calculate the optimal parameter combination under different operating conditions, dynamically adapt to scene changes and optimize control parameters, thereby reducing air conditioner energy consumption, achieving precise energy saving and improving indoor environmental comfort.

[0005] According to an embodiment of the present invention, an adaptive control method for an air conditioner is provided, applied to an adaptive control system of an air conditioner. The adaptive control system of the air conditioner includes a control cloud and an edge control terminal, wherein the control cloud and the edge control terminal are communicatively connected, and the edge control terminal is disposed in the air conditioner. The adaptive control method for the air conditioner includes: Acquire the device status time-series data of the air conditioner, acquire time-series environmental data and user habit history data; The device status time-series data, the time-series environmental data, and the user habit history data are input into a pre-trained prediction model. Based on the prediction model, future operating conditions are predicted to obtain future operating condition prediction results. The future operating condition prediction results include future indoor environmental conditions, expected heat load of the air conditioner, future equipment performance degradation prediction results, and future equipment operating power prediction results. The predicted results of the future operating conditions are input into the pre-trained control model to determine the optimal set of control parameters corresponding to the maximum value of the objective function of the control model under the future operating conditions. The optimal control parameter set is sent to the edge control terminal so that the edge control terminal drives the air conditioner to operate based on the optimal control parameter set.

[0006] By adopting the above technical solution, the equipment status time-series data, time-series environmental data, and user habit history data are input into the prediction model to predict future operating conditions. This allows for accurate perception of multi-dimensional data such as environment, equipment, and user habits. By inputting the prediction results of future operating conditions into the control model, the optimal parameter combination under different operating conditions can be calculated. This enables dynamic adaptation to scene changes and optimization of control parameters, thereby reducing air conditioning energy consumption, achieving precise energy saving, and improving indoor environmental comfort.

[0007] Preferably, the control model includes a neural network model, and the training steps of the control model include: Obtain the training set of the control model; wherein the training set of the control model includes multiple operating condition data and the corresponding set of optimal control parameters under each operating condition data; The training set of the control model is input into the control model for model training until the objective function of the control model reaches the optimal value, thus obtaining the trained control model.

[0008] Preferably, the objective function of the control model is related to the energy consumption parameters, comfort parameters, and equipment loss parameters of the air conditioner.

[0009] By adopting the above technical solution, the control model takes into account multi-dimensional data, can filter environmental noise and accidental operational interference, has high control strategy stability, and improves the reliability of air conditioner control.

[0010] Preferably, the objective function of the control model is calculated as follows:

[0011] in, To optimize the objective, Energy consumption parameters For comfort parameters, These are equipment loss parameters. , and These are weighting coefficients, respectively. The energy consumption parameter is related to the actual power of the air conditioner in the current cycle, the comfort parameter is related to the indoor ambient temperature, indoor humidity and air conditioner noise, and the equipment loss parameter is related to the vibration data of the current cycle, the actual power and the continuous operation time of the equipment.

[0012] Preferably, before the step of inputting the future operating condition prediction result into the pre-trained control model, the method further includes: After the air conditioner operates based on the optimal control parameter set of the previous cycle, the actual power of the air conditioner in the current cycle is obtained, and the actual power of the current cycle is input into the control model to update the objective function of the control model.

[0013] By adopting the above technical solution, the control model is updated based on the actual operating parameters of the air conditioner, enabling the control model to have continuous learning capabilities. As the operating time increases, the control strategy is continuously optimized, achieving "the more it is used, the smarter it becomes".

[0014] Preferably, the optimal set of control parameters includes compressor frequency, fan speed, and set temperature; The device status time-series data includes the cumulative running time of the air conditioner's compressor and the trend of filter pressure difference; The time-series environmental data includes outdoor ambient temperature, humidity, and light intensity data for a future preset duration. The user habit history data includes historical temperature settings and historical wind speed settings set by the user.

[0015] According to an embodiment of the present invention, another aspect provides an adaptive control system for an air conditioner, comprising: a control cloud and an edge control terminal, wherein the control cloud is communicatively connected to the edge control terminal, and the edge control terminal is disposed in the air conditioner; The control cloud includes a computer-readable storage medium storing a computer program and a processor, the computer program being read and executed by the processor to implement the method as described in any of the first aspects.

[0016] Preferably, the edge control terminal includes a data acquisition module; The data acquisition module is used to acquire the time-series data of the air conditioner's device status, time-series environmental data, and historical user habit data.

[0017] Preferably, the adaptive control system of the air conditioner further includes: a communication link; the control cloud includes a cloud database; The edge control terminal is used to encrypt and send the collected device status timing data, timing environment data, and user habit history data to the cloud database through the communication link.

[0018] Preferably, the edge control terminal further includes: an edge computing module; The control cloud is also used to send the optimal control parameter set to the edge computing module via the communication link; The edge computing module is also used to parse the optimal control parameter set sent by the control cloud, and drive the execution module to adjust the parameters based on the optimal control parameter set.

[0019] The present invention has the following beneficial effects: by inputting equipment status time-series data, time-series environmental data, and user habit history data into the prediction model to predict future operating conditions, it can accurately perceive multi-dimensional data such as environment, equipment, and user habits. By inputting the prediction results of future operating conditions into the control model, the optimal parameter combination under different operating conditions can be calculated, which can dynamically adapt to scene changes and optimize control parameters, thereby reducing air conditioning energy consumption, achieving precise energy saving, and improving indoor environmental comfort.

[0020] Other features and advantages of the embodiments of the present invention will be set forth in the following description, or some features and advantages may be inferred from the description or determined without doubt, or may be learned by practicing the techniques described above in the embodiments of the present invention.

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0022] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0023] The structures, proportions, sizes, etc. illustrated in this specification are only for the purpose of assisting those skilled in the art in understanding and reading the content disclosed herein, and are not intended to limit the conditions under which the present invention can be implemented. Therefore, they have no substantial technical significance. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0024] Figure 1 A flowchart of an adaptive control method for an air conditioner provided by the present invention; Figure 2 A flowchart for training a prediction model provided by the present invention; Figure 3 The present invention provides an adaptive control flowchart for an air conditioner. Detailed Implementation

[0025] The following specific embodiments illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0027] This embodiment provides an adaptive control method for an air conditioner. This method can be applied to an adaptive control system for an air conditioner. The adaptive control system includes a control cloud and an edge control terminal. The control cloud and the edge control terminal are communicatively connected. The edge control terminal is installed in the air conditioner. (See example...) Figure 1 The flowchart of the adaptive control method for the air conditioner shown is shown. The method mainly includes the following steps S102 to S106: Step S102: Obtain the device status time-series data of the air conditioner, obtain the time-series environmental data and user habit history data; The hardware of the aforementioned edge control terminal includes a data acquisition module, an edge computing module (embedded processor), an air conditioner execution module (compressor, fan, and valve control unit, etc.), and a communication module (such as a Wi-Fi / BLE / 5G module). The aforementioned data acquisition module includes a temperature sensor, a humidity sensor, an air quality sensor, a power sensor, an equipment status monitoring unit, and a user command receiving unit. The hardware of the control cloud includes a server cluster, a cloud database (also known as a database storage unit), and a model training unit.

[0028] The edge control terminal collects device status time-series data, time-series environmental data, and user habit history data through the data acquisition module. The edge control terminal then sends the collected device status time-series data, time-series environmental data, and user habit history data to the control cloud through the communication module. The control cloud stores the received device status time-series data, time-series environmental data, and user habit history data in the cloud database.

[0029] Various sensors are deployed in the indoor and outdoor units of the air conditioner and at the user interface. Temperature / humidity sensors collect real-time indoor and outdoor temperature and humidity (accuracy ±0.1℃ / ±1%RH), air quality sensors monitor indoor PM2.5, formaldehyde and other indicators, power sensors monitor the real-time power consumption of the compressor and fan, equipment status monitoring unit collects status data such as compressor cumulative running time, filter pressure difference and fault codes, and user command receiving module records operation data such as set temperature, start-up and shutdown time and operating mode.

[0030] In one embodiment, the aforementioned device status time-series data includes the cumulative running time of the air conditioner's compressor and the trend of filter pressure difference; the time-series environmental data includes outdoor ambient temperature, humidity, and light intensity data for a future preset duration (weather forecast); and the user habit history data includes historical set temperature data and historical set wind speed data set by the user.

[0031] Step S104: Input the equipment status time series data, time series environment data and user habit history data into the pre-trained prediction model, and make future operating condition predictions based on the prediction model to obtain the future operating condition prediction results. After receiving device status time-series data, time-series environmental data, and historical user habit data, the cloud database analyzes and learns based on existing historical operation datasets, device model characteristics, group user behavior models, and environmental prediction information, using a deep neural network intelligent control algorithm built into the cloud.

[0032] The aforementioned prediction model can be a deep neural network. This model is used to proactively perceive and quantify the situation of the air conditioning system. It is trained on a dataset of labeled future operating conditions, including time-series data of equipment status, time-series environmental data, and historical user habit data. The training objective of the prediction model is to accurately predict and output the system's demand and status forecasts for a future period (such as the next cycle) based on the input data. The prediction results can include predictions of room temperature changes, expected heat load, equipment performance degradation (component life prediction), and equipment operating power. The prediction model provides a forward-looking basis for the control model, enabling the system to anticipate changes rather than simply respond to them.

[0033] See also Figure 2 The flowchart shown illustrates the training process of the prediction model. After normalizing the dataset samples, they are input into the prediction model (such as a Transformer model or a deep neural network model) for forward propagation. Then, loss is calculated, and backpropagation is performed to optimize the weight parameters. The output of the prediction model is obtained, and the accuracy is calculated based on the output. The training is repeated multiple times until the preset number of training iterations is reached, resulting in the trained control model.

[0034] The control cloud inputs device status time-series data, time-series environmental data, and user habit history data into a pre-trained prediction model, enabling the prediction model to output future operating condition prediction results. These future operating condition prediction results include future indoor environmental conditions, expected heat load of the air conditioner, future equipment performance degradation prediction results, and future equipment operating power prediction results. These future operating condition prediction results can provide the control model with a basis for generating control parameters.

[0035] Step S106: Input the future operating condition prediction results into the pre-trained control model to determine the optimal set of control parameters corresponding to the maximum value of the objective function of the control model under the future operating conditions. The control model in the cloud can identify the optimal combination of operating control parameters under different operating conditions, and then generate an optimal control parameter set, which may include compressor frequency, fan speed and set temperature; so that the air conditioner can adjust the control parameters based on the optimal control parameter set.

[0036] Step S108: Send the optimal control parameter set to the edge control terminal so that the edge control terminal can drive the air conditioner to operate based on the optimal control parameter set.

[0037] The cloud-based control system sends the generated optimal combination of operating control parameters to the edge control unit, enabling the air conditioner to automatically adjust its control parameters based on the optimal set of control parameters. This improves the air conditioner's energy efficiency, optimizes comfort, and extends the equipment's lifespan.

[0038] After the air conditioner operates based on the optimal set of control parameters, the edge control terminal can also collect the actual operating data (including equipment status time-series data and actual user habit data) and actual time-series environmental data of the air conditioner after operating based on the optimal set of control parameters and send them to the control cloud. This allows the control cloud to continuously perform online incremental training and iterative optimization of the prediction model and control model based on the actual effect data after execution, thereby continuously improving the prediction accuracy and the effectiveness of the control strategy. This enables the entire system to have self-evolution capabilities, forming a closed-loop optimization system of "monitoring-learning-adjustment". At the same time, the intelligent algorithm built into the control cloud will continuously learn and iterate based on new user data.

[0039] The adaptive control method for air conditioners provided in this embodiment predicts future operating conditions by inputting equipment status time-series data, time-series environmental data, and historical user habit data into a prediction model. It can accurately perceive multi-dimensional data such as environment, equipment, and user habits. By inputting the prediction results of future operating conditions into the control model, the optimal parameter combination under different operating conditions can be calculated. It can dynamically adapt to scene changes and optimize control parameters, thereby reducing air conditioning energy consumption, achieving precise energy saving, and improving indoor environmental comfort.

[0040] In one implementation, the control model includes a neural network model. This embodiment provides training steps for the control model, including: Obtain the training set of the control model; wherein, the training set of the control model includes multiple operating condition data and the corresponding set of optimal control parameters under each operating condition data; The training set of the control model is input into the control model for model training until the objective function of the control model reaches the optimal value, thus obtaining the trained control model.

[0041] After normalizing the training set, it is input into the control model (deep neural network model) for forward propagation, then loss calculation is performed, and backpropagation is performed to optimize the weight parameters to obtain the output of the control model. The optimization target score is calculated based on the output, and multiple iterations of training are performed until the value of the objective function reaches the optimal value, thus obtaining the trained control model.

[0042] The control model is the core of generating optimal control strategies in the cloud, see, for example... Figure 3 The adaptive control flowchart of the air conditioner shown illustrates that the control model receives future operating condition predictions from the prediction model and performs multi-objective optimization calculations based on the current set targets (i.e., the current set temperature, set fan speed, and compressor frequency). The goal is to find an optimal set of control parameters that maximizes the defined comprehensive utility function while satisfying the predicted scenario and user constraints. The control model sends the generated optimal control parameter set to the edge control terminal located within the air conditioner. The edge control terminal drives the air conditioner based on this optimal control parameter set. After the air conditioner operates based on the optimal control parameter set, the edge control terminal can also send the actual operating data of the air conditioner to the prediction and control models to update and optimize them.

[0043] In one implementation, the objective function of the control model is related to the energy consumption parameters, comfort parameters, and equipment loss parameters of the air conditioner.

[0044] In one implementation, the objective function (i.e., the loss function) of the control model is calculated as follows:

[0045] in, To optimize the objective, Energy consumption parameters For comfort parameters, These are equipment loss parameters. , and These are weighting coefficients, energy consumption parameters related to the actual power of the air conditioner in the current cycle, comfort parameters related to indoor ambient temperature, indoor humidity and air conditioner noise, and equipment loss parameters related to vibration data, actual power and continuous operation time of the equipment in the current cycle.

[0046] In one implementation, the formula for calculating the comfort parameter is:

[0047] in, Let be the indoor ambient temperature in the i-th cycle (i.e., the current cycle). Let be the indoor humidity during the i-th cycle. Let i be the indoor noise level during the i-th cycle. , and These are the weighting coefficients.

[0048] In one implementation, the formula for calculating the equipment loss parameters is:

[0049] in, Let be the power in the i-th cycle. The vibration data for the i-th cycle (such as vibration frequency or vibration velocity). Let be the continuous operating time of the device in the i-th cycle (i.e., the total duration from the start time to the current cycle). , and These are the weighting coefficients.

[0050] In one implementation, the formula for calculating energy consumption parameters is:

[0051] in, Let be the power in the i-th cycle. It represents the duration of the i-th cycle (i.e., the duration after the start of the i-th cycle). The actual power can be calculated by detecting the operating voltage and current of the air conditioner in the current cycle i, or it can be calculated using the following formula:

[0052] in, This represents the power during the (i-1)th cycle (i.e., the power of the air conditioner in the previous cycle). This is the set of optimal control parameters output from the control model of the previous cycle and applied to the air conditioner. The result function is the optimal set of control parameters output by the control model in the (i-1)th cycle, applied to the air conditioner, representing the impact on the air conditioner's power.

[0053] In one embodiment, prior to the step of inputting the future operating condition prediction results into the pre-trained control model, the method provided in this embodiment further includes: After the air conditioner operates based on the optimal control parameter set of the previous cycle, the actual power of the air conditioner in the current cycle is obtained, and the actual power of the current cycle is input into the control model to update the objective function of the control model.

[0054] By updating the control model based on the actual operating parameters of the air conditioner, the control model has the ability to continuously learn. As the operating time increases, the control strategy is continuously optimized, achieving "the more it is used, the smarter it becomes".

[0055] The adaptive control method for air conditioners provided in this embodiment dynamically adapts to scene changes through cloud-based deep learning. Compared with traditional fixed parameter control, it reduces energy consumption by 15%-30%, achieving precise energy saving. The control model adopts a globally optimal objective function that balances comfort and equipment wear, avoiding extreme parameter operation. User perceived temperature fluctuations are controlled within ±0.5℃, and the service life of core equipment components is extended by more than 20%. The prediction model based on multi-dimensional data can filter environmental noise and accidental operational interference, resulting in high stability of the control strategy and a misadjustment rate of less than 5%. It has wide adaptability, supporting air conditioning equipment of different brands and models. By adjusting the model input parameters and weight coefficients, it can adapt to different scenario needs such as home and commercial use. The dual models (prediction model and control model) have continuous learning capabilities. As the operating time increases, the control strategy is continuously optimized, achieving "the more you use it, the smarter it becomes."

[0056] Corresponding to the adaptive control method for air conditioners provided in the above embodiments, this embodiment of the invention provides an adaptive control system for air conditioners, which includes: a control cloud and an edge control terminal, wherein the control cloud and the edge control terminal are communicatively connected, and the edge control terminal is disposed in the air conditioner; The control cloud includes a computer-readable storage medium storing a computer program and a processor. The computer program is read and executed by the processor to implement the adaptive control method for the air conditioner provided in the above embodiments.

[0057] The adaptive control system of the air conditioner provided in this embodiment predicts future operating conditions by inputting equipment status time-series data, time-series environmental data, and historical user habit data into a prediction model. It can accurately sense multi-dimensional data such as environment, equipment, and user habits. By inputting the prediction results of future operating conditions into the control model, the optimal parameter combination under different operating conditions can be calculated. It can dynamically adapt to scene changes and optimize control parameters, thereby reducing air conditioning energy consumption, achieving precise energy saving, and improving indoor environmental comfort.

[0058] In one implementation, the edge control terminal provided in this embodiment includes a data acquisition module; The data acquisition module is used to acquire time-series data of the air conditioner's equipment status, time-series environmental data, and historical data of user habits.

[0059] In one embodiment, the adaptive control system provided in this embodiment further includes: a communication link; and a control cloud including a cloud database. The edge control terminal is used to encrypt and send the collected device status time-series data, time-series environmental data, and user habit history data to the cloud database via a communication link.

[0060] The control cloud uses a distributed server cluster, equipped with GPU computing units for model training, and the database storage unit uses a time-series database to support efficient storage and retrieval of massive historical data.

[0061] In one embodiment, the edge control terminal provided in this embodiment further includes: an edge computing unit; The control cloud is also used to send the optimal set of control parameters to the edge computing unit via a communication link; The edge computing unit is also used to parse the optimal set of control parameters sent from the cloud and drive the execution module to adjust the parameters based on the optimal set of control parameters.

[0062] The edge computing unit uses an embedded processor as the core of edge computing and has a built-in lightweight model inference engine. It can quickly parse the optimal set of control parameters sent from the cloud and drive the execution module to complete parameter adjustment. It also has a local data caching function to prevent the system from losing control when the network is interrupted.

[0063] The device provided in this embodiment has the same implementation principle and technical effect as the aforementioned embodiments. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the aforementioned method embodiment.

[0064] This embodiment also provides a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the various processes of the adaptive control method embodiment for the air conditioner described above, and achieves the same technical effect. To avoid repetition, it will not be described again here. The computer-readable storage medium may be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0065] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by computer-controlled devices. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The storage medium can be a memory, a disk, an optical disk, etc.

[0066] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

[0067] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0068] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. An adaptive control method for an air conditioner, characterized in that, An adaptive control system for an air conditioner, comprising a control cloud and an edge control terminal, wherein the control cloud and the edge control terminal are communicatively connected, and the edge control terminal is disposed in the air conditioner, and the adaptive control method of the air conditioner includes: Acquire the device status time-series data of the air conditioner, acquire time-series environmental data and user habit history data; The device status time-series data, the time-series environmental data, and the user habit history data are input into a pre-trained prediction model. Based on the prediction model, future operating conditions are predicted to obtain future operating condition prediction results. The future operating condition prediction results include future indoor environmental conditions, expected heat load of the air conditioner, future equipment performance degradation prediction results, and future equipment operating power prediction results. The predicted results of the future operating conditions are input into the pre-trained control model to determine the optimal set of control parameters corresponding to the maximum value of the objective function of the control model under the future operating conditions. The optimal control parameter set is sent to the edge control terminal so that the edge control terminal drives the air conditioner to operate based on the optimal control parameter set.

2. The method according to claim 1, characterized in that, The control model includes a neural network model, and the training steps of the control model include: Obtain the training set of the control model; wherein the training set of the control model includes multiple operating condition data and the corresponding set of optimal control parameters under each operating condition data; The training set of the control model is input into the control model for model training until the objective function of the control model reaches the optimal value, thus obtaining the trained control model.

3. The method according to claim 2, characterized in that, The objective function of the control model is related to the energy consumption parameters, comfort parameters, and equipment loss parameters of the air conditioner.

4. The method according to claim 2, characterized in that, The objective function of the control model is calculated using the following formula: in, To optimize the objective, Energy consumption parameters For comfort parameters, These are equipment loss parameters. , and These are weighting coefficients, respectively. The energy consumption parameter is related to the actual power of the air conditioner in the current cycle, the comfort parameter is related to the indoor ambient temperature, indoor humidity and air conditioner noise, and the equipment loss parameter is related to the vibration data of the current cycle, the actual power and the continuous operation time of the equipment.

5. The method according to claim 4, characterized in that, Before the step of inputting the future operating condition prediction results into the pre-trained control model, the method further includes: After the air conditioner operates based on the optimal control parameter set of the previous cycle, the actual power of the air conditioner in the current cycle is obtained, and the actual power of the current cycle is input into the control model to update the objective function of the control model.

6. The method according to claim 1, characterized in that, The optimal control parameter set includes compressor frequency, fan speed, and set temperature; The device status time-series data includes the cumulative running time of the air conditioner's compressor and the trend of filter pressure difference; The time-series environmental data includes outdoor ambient temperature, humidity, and light intensity data for a future preset duration. The user habit history data includes historical temperature settings and historical wind speed settings set by the user.

7. An adaptive control system for an air conditioner, characterized in that, include: The system includes a control cloud and an edge control terminal, wherein the control cloud and the edge control terminal are communicatively connected, and the edge control terminal is installed in the air conditioner. The control cloud includes a computer-readable storage medium storing a computer program and a processor, the computer program being read and executed by the processor to implement the method as described in any one of claims 1-6.

8. The system according to claim 7, characterized in that, The edge control terminal includes a data acquisition module; The data acquisition module is used to acquire the time-series data of the air conditioner's device status, time-series environmental data, and historical user habit data.

9. The system according to claim 8, characterized in that, Also includes: Communication link; the control cloud includes a cloud database; The edge control terminal is used to encrypt and send the collected device status timing data, timing environment data, and user habit history data to the cloud database through the communication link.

10. The system according to claim 7, characterized in that, The edge control terminal also includes: an edge computing module; The control cloud is also used to send the optimal control parameter set to the edge computing module via the communication link; The edge computing module is also used to parse the optimal control parameter set sent by the control cloud, and drive the execution module to adjust the parameters based on the optimal control parameter set.

Images