Air conditioner refrigerating capacity regulation and control method based on self-optimization boundary model and related equipment thereof

Abstract

The invention provides an air conditioner refrigerating capacity regulation and control method based on a self-optimization boundary model and related equipment thereof.The air conditioner refrigerating capacity regulation and control method comprises the steps that current working condition data and current environment data of target refrigeration equipment are obtained, and the working condition data comprise exhaust temperature data, electronic expansion valve opening degree data and corresponding operation frequency data; the current environment data comprises humidity and / or temperature; based on a preset judgment network model, the current working condition data are processed according to the current environment data, the optimal working state data of the target refrigeration equipment under the current environment data are determined, and the optimal working state data comprise at least one of the target operation frequency and the target exhaust temperature; and based on the optimal working state data, the current working state of the target refrigeration equipment is adjusted and controlled. Through the steps of the method, dynamic regulation and control over the refrigerating capacity of the air conditioner can be achieved under different environment working conditions, and the operation adaptability and regulation and control precision of the refrigerating system are improved.

Description

Technical Field

[0001] This invention relates to the field of air conditioning system control, and in particular to an air conditioning cooling capacity regulation method, device, electronic device and its storage medium based on a self-optimizing boundary model. Background Technology

[0002] With the widespread application of air conditioning and refrigeration equipment in residential buildings, commercial buildings, and industrial environments, the stable operation and cooling capacity regulation under complex environmental conditions have attracted increasing attention. Especially when external conditions such as ambient temperature and humidity are constantly changing, the operating status of the refrigeration system is prone to fluctuations. How to dynamically regulate the air conditioning cooling capacity based on real-time operating conditions has become an important research direction in the field of air conditioning control.

[0003] Existing methods for regulating air conditioning cooling capacity typically rely on fixed control strategies or empirically set parameters, such as single-variable control based on exhaust temperature or compressor operating frequency. These methods lack system modeling and comprehensive judgment of the inherent relationships between environmental data and multiple operating parameters. Consequently, these approaches struggle to accurately reflect the true optimal operating state of the refrigeration system under complex environmental conditions, easily leading to lag in regulation and insufficient control precision, thus affecting cooling efficiency and system stability.

[0004] Therefore, existing air conditioning cooling capacity control methods based on self-optimizing boundary models have the problem of being unable to determine the optimal operating state data of the target cooling equipment under the corresponding environmental conditions based on the current environmental data and the current operating condition data, and to adaptively control the operating state of the target cooling equipment. Summary of the Invention

[0005] This invention provides an air conditioning cooling capacity control method based on a self-optimizing boundary model, which solves the problem that existing air conditioning cooling capacity control methods based on self-optimizing boundary models cannot determine the optimal operating state data of the target refrigeration equipment under the corresponding environmental conditions based on current environmental data and current operating condition data, and cannot adaptively control the operating state of the target refrigeration equipment.

[0006] In a first aspect, the present invention provides an air conditioning cooling capacity control method based on a self-optimizing boundary model, the method comprising the following steps: Acquire the current operating condition data and current environmental data of the target refrigeration equipment. The operating condition data includes exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data. The current environmental data includes humidity and / or temperature. Based on a preset judgment network model, the current operating condition data is processed according to the current environmental data to determine the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency and the target exhaust temperature. Based on the optimal operating state data, the current operating state of the target refrigeration equipment is adjusted and controlled.

[0007] Optionally, acquiring the current operating status data and current environmental data of the target refrigeration equipment includes: Exhaust temperature data is obtained by collecting data from a temperature sensor pre-set on the exhaust side of the target refrigeration equipment. The opening angle of the electronic expansion valve of the target refrigeration equipment is collected by a preset electronic valve to obtain the opening degree data of the electronic expansion valve; The operating frequency of the target refrigeration equipment is collected in real time to obtain operating frequency data; By using a preset environmental sensor array, the current environmental data of the target refrigeration device is collected in real time to obtain the current environmental data, which includes the current ambient temperature data and the current ambient humidity data.

[0008] Optionally, before processing the current operating condition data based on the current environmental data according to the preset judgment network model to determine the optimal operating state data of the target refrigeration equipment under the current environmental data, the method further includes: Under multiple different environmental data conditions, collect the exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data of the target refrigeration equipment to obtain the corresponding refrigeration performance data. In the refrigeration performance data, the environmental data corresponding to the optimal refrigeration performance data, the exhaust temperature data of the target refrigeration equipment, the electronic expansion valve opening data, and the corresponding operating frequency data are determined, and a label training set is formed. Based on the labeled training set, the untrained preset judgment network model is trained, and the training is completed when the preset loss function converges, thus obtaining the trained preset judgment network model.

[0009] Optionally, the step of training an untrained preset judgment network model based on the label training set, completing the training when the preset loss function converges, and obtaining a trained preset judgment network model includes: The environmental data in the labeled training set is used as the model input sample, and the corresponding exhaust temperature data, electronic expansion valve opening data and operating frequency data are used as the model output sample. The deviation between the model output sample and the corresponding output value in the labeled training set is calculated based on a preset loss function to obtain deviation data; The model parameters in the preset judgment network model are iteratively updated based on the deviation data until the deviation meets the preset convergence condition, thus completing the training of the preset judgment network model and obtaining the trained preset judgment network model.

[0010] Optionally, the step of processing the current operating condition data based on the current environmental data according to the preset judgment network model to determine the optimal operating state data of the target refrigeration equipment under the current environmental data includes: The current environmental data is input into the preset judgment network model for calculation to obtain the corresponding target operating frequency data and / or target exhaust temperature data that match the current environmental data; The target operating frequency and / or target exhaust temperature are compared with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data.

[0011] Optionally, comparing the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data includes: The target operating frequency and / or target exhaust temperature are compared with the operating frequency and exhaust temperature data under the current operating conditions to obtain the difference data; The difference data is subjected to safety limiting and control quantity fusion processing to obtain the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency data and the target exhaust temperature data.

[0012] Optionally, adjusting and controlling the current operating state of the target refrigeration equipment based on the optimal operating state data includes: Based on the optimal operating state data, the compressor operating frequency and the opening degree of the electronic expansion valve of the target refrigeration equipment are coordinated and controlled so that, under the premise of meeting the preset safe operating boundary conditions, the actual operating state of the target refrigeration equipment approaches the operating state corresponding to the optimal operating state data.

[0013] Secondly, the present invention also provides an air conditioning cooling capacity control device based on a self-optimizing boundary model, the air conditioning cooling capacity control device based on the self-optimizing boundary model comprising: The first acquisition module is used to acquire the current operating condition data and current environmental data of the target refrigeration equipment. The operating condition data includes exhaust temperature data, electronic expansion valve opening data and corresponding operating frequency data. The first determining module is used to process the current operating condition data based on the current environmental data according to the preset judgment network model, and determine the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency and the target exhaust temperature. The control module is used to adjust and control the current operating state of the target refrigeration equipment based on the optimal operating state data.

[0014] Thirdly, the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the air conditioning cooling capacity control method based on a self-optimizing boundary model provided by the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps in the air conditioning cooling capacity control method based on a self-optimizing boundary model provided by the invention.

[0016] This invention acquires current operating condition data and current environmental data of a target refrigeration device. The operating condition data includes exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data. The current environmental data includes humidity and / or temperature. Based on a preset judgment network model, the current operating condition data is processed according to the current environmental data to determine the optimal operating state data of the target refrigeration device under the current environmental data. The optimal operating state data includes at least one of a target operating frequency and a target exhaust temperature. Based on the optimal operating state data, the current operating state of the target refrigeration device is adjusted and controlled. Through the above method steps, dynamic regulation of air conditioning cooling capacity can be achieved under different environmental conditions, improving the adaptability and control accuracy of the refrigeration system. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or 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 only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1This is a flowchart of an air conditioning cooling capacity control method based on a self-optimizing boundary model provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an air conditioning cooling capacity control device based on a self-optimizing boundary model provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. 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.

[0020] like Figure 1 As shown, Figure 1 This is a flowchart of an air conditioning cooling capacity control method based on a self-optimizing boundary model provided by an embodiment of the present invention. The air conditioning cooling capacity control method based on a self-optimizing boundary model includes the following steps: 101. Obtain the current operating condition data and current environmental data of the target refrigeration equipment.

[0021] In this embodiment of the invention, the air conditioning cooling capacity control method based on the self-optimizing boundary model can be applied to an air conditioning cooling capacity control platform based on the self-optimizing boundary model. The air conditioning cooling capacity control platform based on the self-optimizing boundary model has functions such as cooling capacity control data processing, cooling capacity control data transmission and reception, and cooling capacity control data memory storage. It can be built based on a server or server cluster. The server or server cluster can be an electronic device with cooling capacity control data processing capabilities.

[0022] The aforementioned target refrigeration equipment can be an air conditioning refrigeration system including a compressor, condenser, evaporator, electronic expansion valve and its supporting control unit, or any other refrigeration device including the aforementioned equipment. The operating state of the aforementioned target refrigeration equipment, as the controlled object, directly determines the cooling capacity output of the air conditioning system. The aforementioned air conditioning cooling capacity control platform based on the self-optimizing boundary model can also use the aforementioned target refrigeration equipment as the control object to jointly control its operating frequency, exhaust temperature and electronic expansion valve opening.

[0023] In this embodiment, the air conditioning cooling capacity control platform based on the self-optimizing boundary model can establish a real-time communication connection with various sensors and control modules installed in the target cooling equipment to synchronously collect the operating status data of the target cooling equipment and the environmental status data. It should be noted that during the data collection process, the air conditioning cooling capacity control platform based on the self-optimizing boundary model can also automatically collect data through a preset sampling period. Generally speaking, when the operating frequency of the target cooling equipment is higher and the exhaust temperature is higher, the sampling period will be automatically shortened, thereby collecting working data in real time in high-power mode and ensuring safe monitoring in high-power mode.

[0024] The aforementioned current operating condition data may refer to a set of key operating parameter data used to characterize the current operating status of the target refrigeration equipment, including but not limited to exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data.

[0025] The aforementioned current environmental data may refer to data characterizing the external environmental state of the target refrigeration equipment, including but not limited to humidity and / or temperature. For example, by setting temperature and humidity sensors around the outdoor unit, the ambient temperature and air humidity values ​​of the target refrigeration equipment can be collected in real time and transmitted to the air conditioning cooling capacity control platform based on the self-optimizing boundary model as the input basis for model judgment.

[0026] 102. Based on the preset judgment network model, the current operating condition data is processed according to the current environmental data to determine the optimal operating state data of the target refrigeration equipment under the current environmental data.

[0027] In this embodiment of the invention, the aforementioned preset judgment network model can refer to an intelligent judgment model trained by the air conditioning cooling capacity control platform based on the self-optimizing boundary model, using historical operating data and cooling performance data. This model is used to learn the mapping relationship between exhaust temperature, electronic expansion valve opening, operating frequency, and cooling performance under different ambient temperature and humidity conditions, and is deployed within the air conditioning cooling capacity control platform based on the self-optimizing boundary model for real-time inference calculations during operation. Specifically, it can be a multi-layer feedforward neural network, convolutional neural network, recurrent neural network, radial basis function neural network, or any combination thereof, used to construct the mapping relationship between current environmental data and optimal operating state data.

[0028] In one possible embodiment, the air conditioning cooling capacity control platform based on the self-optimizing boundary model inputs the acquired current environmental data into the preset judgment network model, and performs a comprehensive analysis of the operating status of the target cooling equipment in combination with the current operating condition data. After parsing and comparing the output results of the preset judgment network model, the target control state parameters of the target cooling equipment under the corresponding conditions of the current environmental data are clearly obtained, that is, the optimal working state data.

[0029] The aforementioned optimal operating state data can refer to the target control data that enables the target refrigeration equipment to reach its optimal state of cooling capacity under the current ambient temperature and / or humidity conditions, provided that the safe operation boundary conditions are met. This optimal operating state data is determined by the air conditioning cooling capacity control platform based on the self-optimizing boundary model, and includes at least one of the target operating frequency data and the target exhaust temperature data.

[0030] The aforementioned target operating frequency can refer to the target speed frequency value calculated and output by the air conditioning cooling capacity control platform based on the self-optimizing boundary model and the current environmental data, which is used to guide the subsequent operation of the compressor.

[0031] The aforementioned target exhaust temperature can refer to the desired temperature value output by the air conditioning cooling capacity control platform based on the current environmental data, which is used as the target for exhaust temperature control.

[0032] 103. Based on the optimal working state data, adjust and control the current working state of the target refrigeration equipment.

[0033] In this embodiment of the invention, the air conditioning cooling capacity control platform based on the self-optimizing boundary model can use the optimal working state data as the control target to perform coordinated control on the compressor operating frequency and electronic expansion valve opening of the target refrigeration equipment, so that the actual operating condition of the target refrigeration equipment gradually approaches the operating condition corresponding to the above-mentioned optimal working state data, under the premise of meeting the safe operating boundary conditions.

[0034] In this embodiment of the invention, current operating condition data and current environmental data of the target refrigeration equipment are acquired. The operating condition data includes exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data. The current environmental data includes humidity and / or temperature. Based on a preset judgment network model, the current operating condition data is processed according to the current environmental data to determine the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency and the target exhaust temperature. Based on the optimal operating state data, the current operating state of the target refrigeration equipment is adjusted and controlled. Through the above method steps, dynamic regulation of the air conditioning cooling capacity can be achieved under different environmental conditions, improving the adaptability and control accuracy of the refrigeration system.

[0035] Through the above methods and steps, based on real-time acquired operating condition data and environmental data, the optimal operating state data of the target refrigeration equipment under different temperature and / or humidity conditions can be intelligently determined. The compressor operating frequency and electronic expansion valve opening of the target refrigeration equipment are then coordinated and adjusted. Under the premise of meeting the safety operation boundary conditions, the actual operating conditions of the target refrigeration equipment gradually approach the optimal operating state, thereby realizing dynamic optimization and control of the cooling capacity and improving the operational stability and control accuracy of the refrigeration system under complex environmental changes.

[0036] Optionally, in the steps of acquiring the current operating condition data and current environmental data of the target refrigeration equipment, the following can also be used: exhaust temperature data can be obtained by pre-setting a temperature sensor on the exhaust side of the target refrigeration equipment; the opening angle of the electronic expansion valve of the target refrigeration equipment can be acquired by pre-setting an electronic valve; the operating frequency of the target refrigeration equipment can be acquired in real time; and the current environmental data, including the current ambient temperature and humidity data, can be acquired in real time using a pre-set environmental sensor array.

[0037] In this embodiment of the invention, the corresponding sensors, acquisition modules, and their operating parameters can be pre-set and calibrated according to the structural form, installation location, and operational requirements of the target refrigeration equipment. For example, the sampling period of the temperature sensor, the installation location of the environmental sensor, and the communication parameters of the electronic expansion valve feedback channel can be pre-set, so that the air conditioning cooling capacity control platform based on the self-optimizing boundary model can directly perform data acquisition operations in a predetermined manner during subsequent operation.

[0038] The aforementioned preset electronic valve can be a drive control unit serving as the actuator of an electronic expansion valve. It drives the electronic expansion valve to perform opening, closing, and opening degree adjustment actions according to control commands, while simultaneously feeding back the corresponding valve drive status information to the air conditioning cooling capacity control platform based on the self-optimizing boundary model. By reading the operating status of the preset electronic valve, the air conditioning cooling capacity control platform based on the self-optimizing boundary model can indirectly obtain the actual opening and closing state of the electronic expansion valve.

[0039] The aforementioned opening angle of the electronic expansion valve can refer to the digital data used to characterize the current opening degree of the electronic expansion valve, calculated by the air conditioning cooling capacity control platform based on the self-optimizing boundary model. For example, if the fully closed state of the electronic expansion valve is defined as 0% and the fully open state as 100%, the corresponding electronic expansion valve opening degree data can be directly used in subsequent operating condition calculations and model judgments.

[0040] In one possible embodiment, the air conditioning cooling capacity control platform based on the self-optimizing boundary model establishes a data communication channel with the temperature sensor, electronic expansion valve drive module, frequency converter drive module and environmental sensor array, and reads and acquires data in real time the exhaust temperature, electronic expansion valve opening, operating frequency and current ambient temperature and humidity of the target refrigeration equipment according to a preset data sampling period.

[0041] The aforementioned operating frequency data can refer to the frequency values ​​collected in real time from the variable frequency drive module of the target refrigeration equipment by the air conditioning cooling capacity control platform based on the self-optimizing boundary model, which are used to characterize the current speed state of the compressor. It can be understood that the aforementioned operating frequency data can be used to reflect the current actual load of the compressor and participate in subsequent model judgment and control decisions as an important part of the current operating condition data.

[0042] The aforementioned preset environmental sensor array can refer to a combination of multiple environmental sensors deployed at a predetermined location in the installation area of ​​the target refrigeration equipment. These sensors are used to synchronously collect temperature and / or humidity parameters in the environment where the target refrigeration equipment is located. This includes, but is not limited to, multiple temperature and humidity sensors, in order to improve the stability and accuracy of environmental data collection. The collected current environmental data is uniformly received by the aforementioned air conditioning cooling capacity control platform based on the self-optimizing boundary model and used to construct model input data.

[0043] By using the above methods and steps, the exhaust temperature, electronic expansion valve opening, operating frequency, and ambient temperature and humidity of the target refrigeration equipment can be acquired synchronously and accurately, providing a reliable data foundation for determining the optimal operating state data, thereby ensuring the real-time performance, accuracy, and stability of the air conditioning cooling capacity control process.

[0044] Optionally, before processing the current operating condition data based on the current environmental data according to the preset judgment network model to determine the optimal operating state data of the target refrigeration equipment under the current environmental data, the process further includes collecting the exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data of the current target refrigeration equipment under multiple different environmental data conditions; determining the environmental data, exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data corresponding to the optimal refrigeration performance data from the refrigeration performance data, and forming a labeled training set; training the untrained preset judgment network model based on the labeled training set, completing the training when the preset loss function converges, and obtaining the trained preset judgment network model.

[0045] In this embodiment of the invention, the operating data of the target refrigeration equipment can be acquired under multiple different external environmental conditions, including at least different ambient temperature conditions and / or different ambient humidity conditions. For example, the air conditioning cooling capacity control platform based on the self-optimizing boundary model described above can collect data on the operating status of the target refrigeration equipment under different temperature conditions such as 25℃, 30℃, 35℃, and 40℃, or under different humidity combinations, to construct training samples covering multiple environmental ranges.

[0046] In one possible embodiment, the air conditioning cooling capacity control platform based on the self-optimizing boundary model simultaneously collects the exhaust temperature data, electronic expansion valve opening data, and operating frequency data of the target refrigeration equipment, and simultaneously acquires the corresponding cooling performance data of the target refrigeration equipment under the combined operating parameters. This cooling performance data is used to characterize the cooling capacity level of the target refrigeration equipment under the current parameter combination. For example, it can be characterized by cooling capacity output value, heat exchange efficiency, or equivalent cooling capacity parameters, and is used as an evaluation basis for judging "good or bad operating conditions".

[0047] In another possible embodiment, the air conditioning cooling capacity control platform based on the self-optimizing boundary model selects cooling performance data samples in the optimal cooling state from multiple sets of collected cooling performance data based on preset cooling performance evaluation criteria, and extracts the environmental data, exhaust temperature data, electronic expansion valve opening data and operating frequency data corresponding to the optimal cooling performance data. The above correspondence is used as a complete set of training label samples, thereby gradually constructing a labeled training set for model training.

[0048] The aforementioned preset refrigeration performance evaluation criteria are used to characterize the evaluation rules for the quality of refrigeration performance of the target refrigeration equipment under the current operating condition parameter combination. These criteria correspond to evaluation index thresholds or extreme value judgment conditions and are used to select sample data with optimal refrigeration effect from multiple sets of refrigeration performance data. For example, under the conditions of an ambient temperature of 38℃ and an ambient humidity of 75%, the aforementioned air conditioning refrigeration capacity control platform based on the self-optimizing boundary model controls the target refrigeration equipment to operate under multiple different combinations of operating frequencies and electronic expansion valve opening parameters, and simultaneously collects the corresponding refrigeration performance data. The collected multiple sets of refrigeration performance data are compared, and the refrigeration performance data with the largest refrigeration capacity output value per unit time is taken as the optimal refrigeration performance data under the environmental conditions. The environmental data, exhaust temperature data, electronic expansion valve opening data, and operating frequency data corresponding to this optimal refrigeration performance data are determined as a set of optimal parameter samples, used to construct the corresponding label training set.

[0049] The aforementioned label training set refers to a set of training data constructed by the air conditioning cooling capacity control platform based on the self-optimizing boundary model under multiple different environmental data conditions, based on the optimal cooling performance selection rules. Each set of data in this label training set includes: environmental data on the input side, and target exhaust temperature data, electronic expansion valve opening data, and operating frequency data corresponding to the optimal cooling performance on the output side. These data are used as standard learning samples for the preset judgment network model, enabling the model to learn the optimal operating boundary rules under different environmental conditions.

[0050] The aforementioned untrained pre-defined decision network model refers to a decision network model that has only completed network structure initialization but has not yet performed parameter learning based on the aforementioned label training set. This untrained pre-defined decision network model only possesses a basic input-output mapping structure.

[0051] The aforementioned preset loss function can refer to an evaluation function pre-set by the air conditioning cooling capacity control platform based on the self-optimizing boundary model during the model training process to measure the degree of deviation between the model output result and the true labeled value in the label training set. Through this preset loss function, the error between the target operating frequency data and / or target exhaust temperature data output by the model and its true optimal value in the label training set can be quantitatively calculated to obtain the basis for iterative updates. Generally, the training can be judged as if the loss function converges infinitely to a certain value.

[0052] Specifically, the aforementioned preset loss function can be obtained through the following formula:

[0053] in, The loss value represents the preset loss function, used to characterize the overall deviation between the output of the current preset judgment network model and the true labeled values ​​in the label training set. N represents the number of samples currently participating in the training calculation, i.e., the number of labeled training sample groups participating in the loss calculation in one training iteration, and i represents the index number of the i-th training sample. This represents the target operating frequency data predicted by the preset judgment network model for the i-th environmental data sample. This represents the labeled value of the true target running frequency corresponding to the i-th environmental data sample in the labeled training set. This represents the target exhaust temperature data predicted and output by the preset judgment network model for the i-th environmental data sample. α represents the true target exhaust temperature label value corresponding to the i-th environmental data sample in the label training set, α represents the weight coefficient of the target operating frequency prediction error term, which is used to adjust the influence ratio of the operating frequency error in the overall loss function, and β represents the weight coefficient of the target exhaust temperature prediction error term, which is used to adjust the influence ratio of the exhaust temperature error in the overall loss function.

[0054] During multiple rounds of training iterations, the loss value calculated by the air conditioning cooling capacity control platform based on the self-optimizing boundary model gradually decreases and stabilizes within the preset threshold range, indicating that the output of the preset judgment network model can stably approximate the optimal labeled value in the label training set. At this point, the model training process is considered to have reached a convergence state.

[0055] The aforementioned pre-trained judgment network model can refer to a judgment network model with stable prediction capabilities that is finally obtained by the air conditioning cooling capacity control platform based on the self-optimizing boundary model after completing multiple rounds of learning and training on the labeled training set and satisfying the convergence condition of the pre-set loss function. It can output the target operating frequency data and / or target exhaust temperature data of the target cooling equipment under the current environmental conditions based on the real-time input current environmental data, which can be used as the basis for judgment in the subsequent optimal working state data determination step.

[0056] Through the above methods and steps, the trained preset judgment network model is able to realistically reflect the optimal cooling boundary law under different environmental conditions, thereby providing a reliable model basis for the accurate determination of subsequent optimal working state data and improving the accuracy and stability of the overall air conditioning cooling capacity control process.

[0057] Optionally, the step of training an untrained preset judgment network model based on a labeled training set, and completing the training when the preset loss function converges to obtain a trained preset judgment network model, further includes using environmental data from the labeled training set as model input samples, and corresponding exhaust temperature data, electronic expansion valve opening data, and operating frequency data as model output samples; calculating the deviation between the model output samples and the corresponding output values ​​in the labeled training set based on the preset loss function to obtain deviation data; iteratively updating the model parameters in the preset judgment network model based on the deviation data until the deviation meets the preset convergence condition, thus completing the training of the preset judgment network model and obtaining a trained preset judgment network model.

[0058] In this embodiment of the invention, the air conditioning cooling capacity control platform based on the self-optimizing boundary model, during the model training process, uses the environmental data in the label training set as model input samples to the untrained preset judgment network model based on the label training samples participating in the training in the current training iteration, to obtain the predicted values ​​of exhaust temperature, electronic expansion valve opening, and operating frequency output by the model. Based on the preset loss function, the difference between the predicted output values ​​of the model and the corresponding real labeled values ​​in the label training set is numerically processed to obtain the deviation data.

[0059] The aforementioned deviation data refers to the error information obtained by the air conditioning cooling capacity control platform based on the self-optimizing boundary model, which calculates the difference between the model output samples and the corresponding output values ​​in the labeled training set based on the aforementioned preset loss function. This deviation data is used to reflect the magnitude of the prediction error of the currently untrained preset judgment network model in prediction dimensions such as target operating frequency, target exhaust temperature, and electronic expansion valve opening.

[0060] In one possible embodiment, after obtaining the aforementioned deviation data, the air conditioning cooling capacity control platform based on the self-optimizing boundary model corrects and updates the model parameters inside the untrained preset judgment network model round by round based on the aforementioned deviation data. It should be noted that the aforementioned air conditioning cooling capacity control platform based on the self-optimizing boundary model updates the model parameters once based on the new deviation data, and re-executes the next round of input, output and deviation calculation after the parameter update is completed, so that the model output result gradually approaches the corresponding optimal labeled value in the label training set.

[0061] The aforementioned preset convergence condition can refer to the stopping condition set before training begins to determine whether the preset judgment network model has completed training. When the aforementioned deviation data is less than or equal to the error threshold corresponding to the aforementioned preset convergence condition in multiple consecutive training iterations, it is determined that the currently untrained preset judgment network model has completed parameter convergence. At this point, the iterative update of the model parameters is stopped, and the trained preset judgment network model is obtained.

[0062] By following the above steps, the trained preset judgment network model can be made to have stable and reliable prediction capabilities, providing a reliable model foundation for the accurate determination of subsequent optimal working state data.

[0063] Optionally, in the step of processing the current operating condition data based on the current environmental data according to the preset judgment network model and determining the optimal operating state data of the target refrigeration equipment under the current environmental data, the method further includes inputting the current environmental data into the preset judgment network model for calculation to obtain the corresponding target operating frequency data and / or target exhaust temperature data that match the current environmental data; comparing the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating condition data to determine the optimal operating state data of the target refrigeration equipment under the current environmental data.

[0064] In this embodiment of the invention, the aforementioned target operating frequency data can refer to the compressor target operating frequency value that is adapted to the environmental conditions, after the air conditioning cooling capacity control platform based on the self-optimizing boundary model inputs the current environmental data (including temperature and / or humidity) into the trained preset judgment network model for calculation. This target operating frequency data is not a fixed threshold, but rather an optimal operating frequency reference value dynamically calculated based on the current environmental conditions. It is used to characterize the compressor operating frequency corresponding to the optimal cooling performance of the target refrigeration equipment under the current environmental conditions, while meeting the safe operating boundary conditions.

[0065] The aforementioned target exhaust temperature data refers to the exhaust temperature control target value corresponding to the current environmental data, output by the preset judgment network model after the air conditioning cooling capacity control platform based on the self-optimizing boundary model inputs the current environmental data for calculation. This target exhaust temperature data serves as a reference benchmark for exhaust temperature control. When the actual exhaust temperature of the target refrigeration equipment is higher than the aforementioned target exhaust temperature data, the air conditioning cooling capacity control platform based on the self-optimizing boundary model will gradually bring the actual exhaust temperature closer to the aforementioned target exhaust temperature data by coordinating the adjustment of the compressor operating frequency and the opening of the electronic expansion valve.

[0066] In one possible embodiment, the air conditioning cooling capacity control platform based on the self-optimizing boundary model described above can compare and analyze the target operating frequency data and / or target exhaust temperature data output by the preset judgment network model with the operating frequency data and exhaust temperature data collected in real time under the current operating conditions, to obtain the difference between the target parameters and the current actual operating parameters. Through this comparison process, the air conditioning cooling capacity control platform based on the self-optimizing boundary model can determine the degree of deviation of the target refrigeration equipment from its optimal operating state under the current environmental conditions.

[0067] Specifically, the aforementioned air conditioning cooling capacity control platform based on the self-optimizing boundary model, after inputting current environmental data into a preset judgment network model and obtaining target operating frequency data and / or target exhaust temperature data, extracts the target operating frequency data and / or target exhaust temperature data from the output of the preset judgment network model. Simultaneously, it extracts the corresponding current operating frequency data and current exhaust temperature data from the current operating condition data, forming a one-to-one correspondence between the target parameter set and the current parameter set. It then calculates the numerical differences between the target operating frequency data and the current operating frequency data, and between the target exhaust temperature data and the current exhaust temperature data, obtaining corresponding operating frequency difference values ​​and exhaust temperature difference values. These values ​​characterize the degree of deviation between the current operating condition and the target operating condition. Based on the positive or negative relationship between the operating frequency difference value and the exhaust temperature difference value, it determines whether the current operating frequency is "too high" or "too low" compared to the target operating frequency, and whether the current exhaust temperature is "exceeding the target" or "below the target" relative to the target exhaust temperature, providing a basis for determining the subsequent control direction.

[0068] More specifically, after completing the single-parameter difference calculation, the air conditioning cooling capacity control platform based on the self-optimizing boundary model further combines the correspondence between the operating frequency difference value and the exhaust temperature difference value to analyze the trend relationship of the influence of the operating frequency change on the exhaust temperature change, thereby establishing the linkage deviation characteristics between the two to reflect the mutual constraint relationship between the compressor load change and the heat load change under the current operating condition. After completing the above numerical comparison, direction determination and linkage relationship analysis, the air conditioning cooling capacity control platform based on the self-optimizing boundary model finally forms a comparison analysis result to characterize the difference between the current operating condition and the target optimal operating condition. The above comparison analysis result includes at least: the operating frequency difference value, the exhaust temperature difference value and the linkage change relationship between the two, which are used as the input basic data for subsequent difference constraint processing and fusion processing.

[0069] Through the above methods and steps, the target operating frequency data and / or target exhaust temperature data output by the preset judgment network model can be aligned and quantified with the actual operating parameters under the current operating conditions, thereby clearly transforming the difference between the "target state" and the "current state" into quantifiable control deviation data, providing reliable data basis for subsequent constraint processing, control quantity fusion processing and final control adjustment under safe boundary conditions.

[0070] Optionally, the step of comparing the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data further includes comparing the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating conditions to obtain difference data; performing safety limiting and control quantity fusion processing on the difference data to obtain the optimal operating state data of the target refrigeration equipment under the current environmental data, wherein the optimal operating state data includes at least one of the target operating frequency data and the target exhaust temperature data.

[0071] In this embodiment of the invention, the aforementioned difference data may refer to the numerical data obtained by the air conditioning cooling capacity control platform based on the self-optimizing boundary model after comparing the target operating frequency data and / or target exhaust temperature data with the actual operating frequency data and actual exhaust temperature data under the current operating conditions, which is used to characterize the degree of deviation between the "target state" and the "current state".

[0072] Specifically, the difference data includes at least the frequency difference between the target operating frequency and the current operating frequency, and the temperature difference between the target exhaust temperature and the current exhaust temperature. This difference data can be used to reflect the deviation of the target refrigeration equipment from its optimal operating state under the current environmental conditions, and is the basic data for subsequent fusion processing of safety constraints and control quantities.

[0073] In one possible embodiment, when processing the aforementioned differential data, the air conditioning cooling capacity control platform based on the self-optimizing boundary model constrains the adjustment range corresponding to the aforementioned differential data by applying upper and lower limits based on pre-set safe operation boundary conditions.

[0074] For example, when the operating frequency adjustment corresponding to the difference data exceeds the safe adjustment range allowed by the compressor, the air conditioning cooling capacity control platform based on the self-optimizing boundary model will limit the adjustment to within the safe upper limit. When the exhaust temperature adjustment corresponding to the difference data may cause the system to overheat, it will also be constrained to the safe range through safety limiting, thereby avoiding overload or abnormal operation of the target refrigeration equipment due to excessive adjustment.

[0075] In another possible embodiment, after completing the safety limiting processing of the difference data, the air conditioning cooling capacity control platform based on the self-optimizing boundary model further performs joint analysis and comprehensive calculation of the difference adjustment amount related to the target operating frequency and the difference adjustment amount related to the target exhaust temperature.

[0076] Through the above-mentioned control quantity fusion processing, the air conditioning cooling capacity control platform based on the self-optimizing boundary model can establish a cooperative relationship between the operating frequency adjustment direction and the exhaust temperature adjustment direction, avoiding the mutual interference problems that may be caused by the independent adjustment of a single parameter, thereby forming a final control adjustment quantity that simultaneously takes into account both cooling performance requirements and safe operation boundary constraints.

[0077] Through the above methods and steps, the difference between the target state and the current operating state can be transformed into quantifiable difference data. With the combined effect of safety limit and control quantity fusion processing, system instability caused by over-adjustment or single parameter adjustment can be avoided. Thus, under the premise of meeting the safety operation boundary conditions, the optimal operating state data of the target refrigeration equipment under the current environmental conditions can be stably output.

[0078] Optionally, in the step of adjusting and controlling the current operating state of the target refrigeration equipment based on the optimal operating state data, the step may further include coordinating the operation frequency of the compressor and the opening degree of the electronic expansion valve of the target refrigeration equipment based on the optimal operating state data, so that the actual operating condition of the target refrigeration equipment approaches the operating condition corresponding to the optimal operating state data, provided that the operating frequency data and the exhaust temperature data meet the preset safe operating boundary conditions.

[0079] In this embodiment of the invention, the aforementioned preset safe operating boundary conditions can refer to a set of operating constraints pre-set based on the structural parameters, rated operating parameters, and safe operating requirements of the target refrigeration equipment before the air conditioning cooling capacity control platform based on the self-optimizing boundary model is put into operation. These constraints are used to limit the safe range of allowable changes in each operating parameter of the target refrigeration equipment during the collaborative control process. They include, but are not limited to, the minimum and maximum operating frequency range allowed by the compressor, the safe upper limit threshold of the exhaust temperature, and the minimum and maximum opening thresholds of the electronic expansion valve. It should be noted that when the air conditioning cooling capacity control platform based on the self-optimizing boundary model detects that any operating parameter may exceed the corresponding safe operating boundary conditions during the collaborative control process, it constrains or limits the corresponding control amount to prevent the target refrigeration equipment from entering an unsafe operating state due to overload, overheating, or abnormal throttling.

[0080] After obtaining the optimal operating state data of the target refrigeration equipment under the current environmental data, the air conditioning cooling capacity control platform based on the self-optimizing boundary model uses the optimal operating state data as the control target to perform coordinated control on the compressor operating frequency and the opening degree of the electronic expansion valve of the target refrigeration equipment, so that the actual operating condition of the target refrigeration equipment gradually approaches the operating condition corresponding to the optimal operating state data.

[0081] Specifically, the target operating frequency data in the aforementioned optimal operating state data is used as the control target for the compressor, while the target exhaust temperature data in the aforementioned optimal operating state data is used as the control reference benchmark for the exhaust temperature. During the coordinated control process, the air conditioning cooling capacity control platform based on the self-optimizing boundary model adjusts the compressor's drive control parameters according to the deviation between the actual operating frequency and the target operating frequency in the current operating condition data, thereby changing the compressor's actual operating frequency. At the same time, according to the deviation between the current exhaust temperature and the target exhaust temperature, the opening of the electronic expansion valve is adjusted to maintain a matching relationship between the refrigerant flow and the compressor load changes.

[0082] Through the above methods and steps, the air conditioning cooling capacity control platform based on the self-optimizing boundary model establishes a linkage regulation relationship between the change in compressor operating frequency and the change in electronic expansion valve opening. This avoids the problems of drastic fluctuations in exhaust temperature, decreased cooling efficiency, or unstable system operation that may be caused by adjusting only the compressor operating frequency or only the electronic expansion valve opening. Thus, under the premise of meeting safe operation constraints, it achieves a smooth transition from the target refrigeration equipment's operating state to the optimal working state.

[0083] like Figure 2 As shown, this embodiment of the invention also provides an air conditioning cooling capacity control device 200 based on a self-optimizing boundary model, the air conditioning cooling capacity control device 200 based on the self-optimizing boundary model includes: The first acquisition module 201 is used to acquire the current operating condition data and current environmental data of the target refrigeration equipment. The operating condition data includes exhaust temperature data, electronic expansion valve opening data and corresponding operating frequency data. The first determining module 202 is used to process the current operating condition data based on the current environmental data according to the preset judgment network model, and determine the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency and the target exhaust temperature. The control module 203 is used to adjust and control the current working state of the target refrigeration equipment based on the optimal working state data.

[0084] Optionally, the first acquisition module 201 mentioned above includes: The first acquisition submodule is used to acquire exhaust temperature data by means of a temperature sensor preset on the exhaust side of the target refrigeration equipment; The second acquisition submodule is used to acquire the opening angle of the electronic expansion valve of the target refrigeration equipment through a preset electronic valve to obtain the opening degree data of the electronic expansion valve. The third acquisition submodule is used to collect the operating frequency of the target refrigeration equipment in real time and obtain the operating frequency data; The fourth acquisition submodule is used to collect the current environmental data of the target refrigeration device in real time through a preset environmental sensor array to obtain the current environmental data, which includes the current ambient temperature data and the current ambient humidity data.

[0085] Optionally, the above-mentioned device further includes: The first training module is used to collect the exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data of the target refrigeration equipment under multiple different environmental data conditions to obtain the refrigeration performance data corresponding to the target refrigeration equipment. The second training module is used to determine the environmental data corresponding to the optimal cooling performance data, the exhaust temperature data of the target refrigeration equipment, the electronic expansion valve opening data, and the corresponding operating frequency data from the cooling performance data, and form a labeled training set. The third training module is used to train the untrained preset judgment network model based on the label training set. The training is completed when the preset loss function converges, and the trained preset judgment network model is obtained.

[0086] Optionally, the third training module mentioned above also includes: The first training submodule is used to take the environmental data in the label training set as the model input sample and the corresponding exhaust temperature data, electronic expansion valve opening data and operating frequency data as the model output sample. The second training submodule is used to calculate the deviation between the model output sample and the corresponding output value in the labeled training set based on a preset loss function, and obtain the deviation data. The third training submodule is used to iteratively update the model parameters in the preset judgment network model based on the deviation data until the deviation meets the preset convergence condition, thereby completing the training of the preset judgment network model and obtaining the trained preset judgment network model.

[0087] Optionally, the first determining module 202 mentioned above includes: The first determining submodule is used to input the current environmental data into the preset judgment network model for calculation to obtain the corresponding target operating frequency data and / or target exhaust temperature data that match the current environmental data; The second determining submodule is used to compare the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data.

[0088] Optionally, the second determining submodule mentioned above includes: The first determining unit is used to compare the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating conditions to obtain difference data; The second determining unit is used to perform safety limiting and control quantity fusion processing on the difference data to obtain the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency data and the target exhaust temperature data.

[0089] Optionally, the control module 203 mentioned above includes: The control submodule is used to coordinately regulate the compressor operating frequency and the opening degree of the electronic expansion valve of the target refrigeration equipment based on the optimal operating state data, so that the actual operating condition of the target refrigeration equipment approaches the operating condition corresponding to the optimal operating state data, provided that the operating frequency data and exhaust temperature data meet the preset safe operating boundary conditions.

[0090] like Figure 3 As shown, this embodiment of the invention also provides an electronic device 300, including a processor, which can execute any of the above-mentioned air conditioning cooling capacity control methods based on a self-optimizing boundary model.

[0091] Specifically, it includes a processor 301 and a memory 302, as well as a computer program stored in the memory 302 and capable of running on the processor 301, which executes an air conditioning cooling capacity control method based on a self-optimizing boundary model, wherein: The processor 301 executes the calculator program for the air conditioning cooling capacity control method based on the self-optimizing boundary model, stored in the memory 302, and performs the following steps: Acquire the current operating condition data and current environmental data of the target refrigeration equipment. The operating condition data includes exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data. The current environmental data includes humidity and / or temperature. Based on a preset judgment network model, the current operating condition data is processed according to the current environmental data to determine the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency and the target exhaust temperature. Based on the optimal operating state data, the current operating state of the target refrigeration equipment is adjusted and controlled.

[0092] Optionally, the processor 301 performs the process of acquiring the current operating condition data and current environmental data of the target refrigeration device, including: Exhaust temperature data is obtained by collecting data from a temperature sensor pre-set on the exhaust side of the target refrigeration equipment. The opening angle of the electronic expansion valve of the target refrigeration equipment is collected by a preset electronic valve to obtain the opening degree data of the electronic expansion valve; The operating frequency of the target refrigeration equipment is collected in real time to obtain operating frequency data; By using a preset environmental sensor array, the current environmental data of the target refrigeration device is collected in real time to obtain the current environmental data, which includes the current ambient temperature data and the current ambient humidity data.

[0093] Optionally, before the processor 301 executes the method based on the preset judgment network model to process the current operating condition data according to the current environmental data and determine the optimal operating state data of the target refrigeration equipment under the current environmental data, the method further includes: Under multiple different environmental data conditions, collect the exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data of the target refrigeration equipment to obtain the corresponding refrigeration performance data. In the refrigeration performance data, the environmental data corresponding to the optimal refrigeration performance data, the exhaust temperature data of the target refrigeration equipment, the electronic expansion valve opening data, and the corresponding operating frequency data are determined, and a label training set is formed. Based on the labeled training set, the untrained preset judgment network model is trained, and the training is completed when the preset loss function converges, thus obtaining the trained preset judgment network model.

[0094] Optionally, the processor 301 executes the training of the untrained preset judgment network model based on the label training set, and completes the training when the preset loss function converges, to obtain a trained preset judgment network model, including: The environmental data in the labeled training set is used as the model input sample, and the corresponding exhaust temperature data, electronic expansion valve opening data and operating frequency data are used as the model output sample. The deviation between the model output sample and the corresponding output value in the labeled training set is calculated based on a preset loss function to obtain deviation data; The model parameters in the preset judgment network model are iteratively updated based on the deviation data until the deviation meets the preset convergence condition, thus completing the training of the preset judgment network model and obtaining the trained preset judgment network model.

[0095] Optionally, the processor 301 executes the preset judgment network model to process the current operating condition data according to the current environmental data, and determines the optimal operating state data of the target refrigeration equipment under the current environmental data, including: The current environmental data is input into the preset judgment network model for calculation to obtain the corresponding target operating frequency data and / or target exhaust temperature data that match the current environmental data; The target operating frequency and / or target exhaust temperature are compared with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data.

[0096] Optionally, the processor 301 performs the step of comparing the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data, including: The target operating frequency and / or target exhaust temperature are compared with the operating frequency and exhaust temperature data under the current operating conditions to obtain the difference data; The difference data is subjected to safety limiting and control quantity fusion processing to obtain the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency data and the target exhaust temperature data.

[0097] Optionally, the processor 301 performs the adjustment and control of the current operating state of the target refrigeration device based on the optimal operating state data, including: Based on the optimal operating state data, the compressor operating frequency and the opening degree of the electronic expansion valve of the target refrigeration equipment are coordinated and controlled so that, under the premise of meeting the preset safe operating boundary conditions, the actual operating state of the target refrigeration equipment approaches the operating state corresponding to the optimal operating state data.

[0098] This invention also provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the various processes of the air conditioning cooling capacity control method based on the self-optimizing boundary model provided in this invention, or the application-side air conditioning cooling capacity control method based on the self-optimizing boundary model, and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0099] Those skilled in the art will understand that implementing all or part of the processes in the above embodiments can be done by a computer program instructing related hardware, and can be stored in a computer-readable storage medium. When executed, the program can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0100] The above description discloses only preferred embodiments of the present invention and should not be construed as limiting the scope of the present invention. Therefore, equivalent variations made in accordance with the claims of the present invention are still within the scope of the present invention.

Claims

1. A method for regulating air conditioning cooling capacity based on a self-optimizing boundary model, characterized in that, include: Acquire the current operating condition data and current environmental data of the target refrigeration equipment. The operating condition data includes exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data. The current environmental data includes humidity and / or temperature. Based on a preset judgment network model, the current operating condition data is processed according to the current environmental data to determine the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency and the target exhaust temperature. Based on the optimal operating state data, the current operating state of the target refrigeration equipment is adjusted and controlled.

2. The air conditioning cooling capacity control method based on a self-optimizing boundary model as described in claim 1, characterized in that, The acquisition of the current operating status data and current environmental data of the target refrigeration equipment includes: Exhaust temperature data is obtained by collecting data from a temperature sensor pre-set on the exhaust side of the target refrigeration equipment. The opening angle of the electronic expansion valve of the target refrigeration equipment is collected by a preset electronic valve to obtain the opening degree data of the electronic expansion valve; The operating frequency of the target refrigeration equipment is collected in real time to obtain operating frequency data; By using a preset environmental sensor array, the current environmental data of the target refrigeration device is collected in real time to obtain the current environmental data, which includes the current ambient temperature data and the current ambient humidity data.

3. The air conditioning cooling capacity control method based on a self-optimizing boundary model as described in claim 1, characterized in that, Before determining the optimal operating state data of the target refrigeration equipment under the current environmental data by processing the current operating condition data based on the current environmental data using a preset judgment network model, the method further includes: Under multiple different environmental data conditions, collect the exhaust temperature data, electronic expansion valve opening data, and corresponding operating frequency data of the target refrigeration equipment to obtain the corresponding refrigeration performance data. In the refrigeration performance data, the environmental data corresponding to the optimal refrigeration performance data, the exhaust temperature data of the target refrigeration equipment, the electronic expansion valve opening data, and the corresponding operating frequency data are determined, and a label training set is formed. Based on the labeled training set, the untrained preset judgment network model is trained, and the training is completed when the preset loss function converges, thus obtaining the trained preset judgment network model.

4. The air conditioning cooling capacity control method based on a self-optimizing boundary model as described in claim 3, characterized in that, The process of training an untrained preset judgment network model based on the labeled training set, completing the training when the preset loss function converges, and obtaining a trained preset judgment network model includes: The environmental data in the labeled training set is used as the model input sample, and the corresponding exhaust temperature data, electronic expansion valve opening data and operating frequency data are used as the model output sample. The deviation between the model output sample and the corresponding output value in the labeled training set is calculated based on a preset loss function to obtain deviation data; The model parameters in the preset judgment network model are iteratively updated based on the deviation data until the deviation meets the preset convergence condition, thus completing the training of the preset judgment network model and obtaining the trained preset judgment network model.

5. The air conditioning cooling capacity control method based on a self-optimizing boundary model as described in claim 1, characterized in that, The process of processing the current operating condition data based on the current environmental data according to the preset judgment network model to determine the optimal operating state data of the target refrigeration equipment under the current environmental data includes: The current environmental data is input into the preset judgment network model for calculation to obtain the corresponding target operating frequency data and / or target exhaust temperature data that match the current environmental data; The target operating frequency and / or target exhaust temperature are compared with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data.

6. The air conditioning cooling capacity control method based on a self-optimizing boundary model as described in claim 5, characterized in that, The step of comparing the target operating frequency and / or target exhaust temperature with the operating frequency and exhaust temperature data under the current operating conditions to determine the optimal operating state data of the target refrigeration equipment under the current environmental data includes: The target operating frequency and / or target exhaust temperature are compared with the operating frequency and exhaust temperature data under the current operating conditions to obtain the difference data; The difference data is subjected to safety limiting and control quantity fusion processing to obtain the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency data and the target exhaust temperature data.

7. The air conditioning cooling capacity control method based on a self-optimizing boundary model as described in claim 1, characterized in that, The adjustment and control of the current operating state of the target refrigeration equipment based on the optimal operating state data includes: Based on the optimal operating state data, the compressor operating frequency and the opening degree of the electronic expansion valve of the target refrigeration equipment are coordinated and controlled so that, under the premise of meeting the preset safe operating boundary conditions, the actual operating state of the target refrigeration equipment approaches the operating state corresponding to the optimal operating state data.

8. An air conditioning cooling capacity control device based on a self-optimizing boundary model, characterized in that, include: The first acquisition module is used to acquire the current operating condition data and current environmental data of the target refrigeration equipment. The operating condition data includes exhaust temperature data, electronic expansion valve opening data and corresponding operating frequency data. The first determining module is used to process the current operating condition data based on the current environmental data according to the preset judgment network model, and determine the optimal operating state data of the target refrigeration equipment under the current environmental data. The optimal operating state data includes at least one of the target operating frequency and the target exhaust temperature. The control module is used to adjust and control the current operating state of the target refrigeration equipment based on the optimal operating state data.

9. An electronic device, characterized in that, include: The memory, the processor, and the computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps in the air conditioning cooling capacity control method based on the self-optimizing boundary model as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps in the air conditioning cooling capacity control method based on a self-optimizing boundary model as described in any one of claims 1 to 7.

Images