Compressor automatic control method and device based on AI

Abstract

The invention is suitable for the technical field of air conditioner compressor control, and provides an AI-based compressor automatic control method and device, and the method is applied to an AI-based compressor automatic control system. The system comprises an information acquisition module, a model construction module, a model training module, a parameter determination module, a control application module, a wireless communication module, a memory, an alarm, a processing center and an intelligent mobile terminal. The information acquisition module, the model construction module, the model training module, the parameter determination module, the control application module, the wireless communication module, the memory and the alarm are respectively connected with the processing center; the intelligent mobile terminal is in wireless network connection with the wireless communication module in the range of a wireless network or the Internet. The invention further provides an AI-based compressor automatic control device. The temperature of the air conditioner can be automatically adjusted, operation of the compressor is automatically detected, and operation of the compressor is remotely controlled.

Description

Technical Field

[0001] The present invention relates to the technical field of compressor control, and in particular to an AI-based compressor automatic control method and device. Background Art

[0002] With the rapid development of technologies such as the Internet of Things and artificial intelligence, the application of intelligent technology in the air-conditioning compressor industry will become more extensive; by integrating sensors, remote monitoring and automated control systems, compressors can achieve real-time monitoring, fault prediction and remote maintenance, thereby improving equipment reliability and operating efficiency.

[0003] Existing air-conditioning compressors often malfunction due to problems such as insufficient system pressure, insufficient refrigerant, compressor overheating, air-conditioning filter blockage, component aging, etc., which require investigation one by one, thus wasting a lot of time, and cannot be prevented and alarmed in advance to ensure that the compressor can work normally. Summary of the invention

[0004] In order to address the deficiencies in the prior art, the purpose of the present invention is to provide an AI-based compressor automatic control method and device, which can automatically establish control parameters, automatically obtain compressor parameters, automatically adjust the air-conditioning temperature, and automatically detect the operation of the compressor by setting an information acquisition module, a model construction module, a model training module, a parameter determination module, and a control application module, so as to realize remote control of the operation of the compressor, thereby improving the management efficiency of the compressor and reducing the management cost of the compressor.

[0005] In order to achieve the above object, the technical solution adopted by the present invention is:

[0006] An AI-based compressor automatic control method is applied to an AI-based compressor automatic control system, the system comprising an information acquisition module, a model construction module, a model training module, a parameter determination module, a control application module, a wireless communication module, a memory, an alarm, a processing center, and an intelligent mobile terminal; the information acquisition module, the model construction module, the model training module, the parameter determination module, the control application module, the wireless communication module, the memory, and the alarm are respectively connected to the processing center; the intelligent mobile terminal comprises a smart phone, a tablet computer, and an intelligent remote control, which are respectively connected to the wireless network of the wireless communication module within the range of a wireless network or the Internet;

[0007] The wireless communication module is provided with a wireless network unit, which is responsible for sending and receiving wireless signals and automatically networking with the intelligent mobile terminal within the effective network range;

[0008] The alarm device compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory, and automatically sounds an alarm and notifies to continue training if the standard is not met; and compares the compressor operation test data with the compressor operation quality standard stored in the memory, and automatically sounds an alarm and notifies to readjust the parameters if the standard is not met; and compares the compressor operation quality information with the compressor operation quality standard stored in the memory, and automatically sounds an alarm and notifies to re-adjust the temperature or repair if the standard is not met;

[0009] The memory is responsible for storing information of the information acquisition module, the model construction module, the model training module, the parameter determination module, the control application module, the wireless communication module, and the alarm, as well as the storage of the air-conditioning compressor control category model evaluation sub-standard and the compressor operation quality standard;

[0010] The processing center is responsible for the information transmission of each module, alarm, and memory, and is the hub center of the system. It compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory: if it meets the standard, it is the preset compressor control parameter; if it does not meet the standard, it is passed to the alarm and notified to continue training; and compares the compressor operation test data with the compressor operation quality standard stored in the memory: if it meets the standard, it is set as the official compressor control parameter; if it does not meet the standard, it is passed to the alarm and notified to readjust the parameter; and compares the compressor operation quality information with the compressor operation quality standard stored in the memory: if it meets the standard, it continues to operate; if it does not meet the standard, it is passed to the alarm and notified to re-adjust the temperature or repair;

[0011] The information acquisition module includes a data acquisition unit, a data cleaning unit, a data calibration unit, a normalization standard unit, a filtering and denoising unit, and a data annotation unit, which are responsible for acquiring and preprocessing the normal historical parameter data of the air-conditioning compressor operation and the historical parameter data of various abnormal conditions, and passing them to the model building module;

[0012] The model building module includes a feature extraction unit, a data division unit, a forward propagation unit, a loss function unit, a back propagation unit, and an activation function unit, which are responsible for extracting features and dividing data after processing, inputting the data into the model for data training, and passing it to the model training module;

[0013] The model training module includes a normalized input unit, a convolution output unit, a pooling processing unit, a gradient optimization unit, a learning training unit, and a model evaluation unit, which is responsible for training the model according to the collected data and adjusting the model parameters to optimize the model performance to predetermine the air-conditioning compressor control parameters and pass them to the parameter determination module;

[0014] The parameter determination module includes an exhaust flow unit, a compression efficiency unit, a compression power unit, a refrigeration coefficient unit, a trial debugging unit, and an effect confirmation unit, which is responsible for performing control tests under different environments based on the control parameters predicted by the trained model to obtain the optimal control parameters and pass them to the control application module;

[0015] The control application module includes a temperature control unit, a quality detection unit, and an adjustment and maintenance unit, which are responsible for reasonably controlling the temperature of the air conditioner according to the set control parameters and the program instructions of the air conditioner compressor to ensure that the air conditioner can reach the target temperature under different environmental conditions.

[0016] The present invention provides an AI-based compressor automatic control method, comprising the following steps:

[0017] S10, before operation, the information acquisition module acquires and pre-processes the normal historical parameter data of the air-conditioning compressor operation and the historical parameter data of various abnormal conditions for subsequent model construction, and transmits them to the model construction module;

[0018] S20, the model building module extracts features and divides the processed data and inputs the data into the model for data training to ensure the accuracy of model recognition and pass it to the model training module;

[0019] S30, the model training module trains the model according to the collected data and adjusts the model parameters to optimize the model performance to predetermine the compressor control parameters, ensure the stability of the compressor temperature control, and pass them to the parameter determination module;

[0020] S40, the parameter determination module performs control tests under different environments according to the control parameters predicted by the trained model to obtain the best control parameters to ensure the quality of subsequent formal operation, and passes them to the control application module;

[0021] S50, the control application module controls the air-conditioning compressor to reasonably control the temperature of the air-conditioning according to the set control parameters and the program instructions, so as to ensure that the air-conditioning can reach the target temperature under different environmental conditions.

[0022] Further, the step S10 comprises the following steps:

[0023] S11. The data acquisition unit obtains historical data of pressure, temperature, refrigerant quantity, current and voltage of similar air-conditioning compressors by connecting to the industry database, and transmits the data to the data cleaning unit;

[0024] S12. The data cleaning unit cleans the collected historical data to remove abnormal values, duplicate values ​​or missing values ​​in the data to improve the quality and accuracy of the data, and transmits the data to the data calibration unit;

[0025] S13, the data calibration unit obtains standardized data according to the data standardization processing formula "z = (x-μ) / σ, z is standardized data, x is original data, μ is the mean of the data, σ is the standard deviation of the data" to improve the stability of the model, and passes it to the normalization standard unit;

[0026] S14, the normalization standard unit obtains the normalized data according to the normalization calculation formula "x'=(x-min(x)) / (max(x)-min(x)), x' is the normalized data, x is the original data, min(x) and max(x) are the minimum and maximum values ​​of the data respectively" to improve the model performance, and passes it to the filtering and denoising unit;

[0027] S15, filtering and denoising unit calculates the formula " G(x,y) is the two-dimensional Gaussian function pixel, (x,y) is the pixel coordinate, and σ is the standard deviation. The denoised image is obtained for color space conversion and passed to the data annotation unit;

[0028] S16. The data annotation unit automatically labels the preprocessed data for classification, target detection, and semantic segmentation to facilitate the extraction of key features in the subsequent data, which is more conducive to model training, testing, and verification.

[0029] Further, the step S20 comprises the following steps:

[0030] S21, the feature extraction unit converts the labeled data into feature vectors useful for model training and extracts key features of temperature, pressure, vibration frequency, and current fluctuation for subsequent data division, and transmits them to the data division unit;

[0031] S22, the data division unit divides the data set with key feature vectors after extraction into a training set, a validation set and a test set and establishes a convolutional neural network structure for subsequent model training, and passes it to the forward propagation unit;

[0032] S23, the forward propagation unit calculates the forward propagation formula of the convolutional layer according to the formula " a' is the first

[0033] Layer output, σ is the activation function, Wlm is the l-th layer convolution kernel, Xm is the input feature map, bl is the bias term" to obtain the convolution layer output and pass it to the loss function unit;

[0034] S24, the loss function unit obtains the loss function value according to the loss function calculation formula "L = max(0, m + y * (f (x) - b)", L is the loss function value, m is the hyperparameter, y is the sample label (0 or 1), f (x) is the predicted value of the model, and b is the classification boundary" for subsequent back propagation and passes it to the back propagation unit;

[0035] S25, the back propagation unit obtains the output of the convolution layer and the fully connected layer according to the back propagation calculation formula "dz[l] = da[l] * g[l]′(z[l]), da[l] is the input data, da[1] is the output data, and g[l]′ is the sigmoid derivative of the activation function", and passes it to the activation function unit;

[0036] S26. The activation function unit obtains the output of the Tanh function according to the ReLU function calculation formula "f(x)=max(0,x), f(x) is the output of the ReLU function, and x is the input of the ReLU function".

[0037] Further, the step S30 comprises the following steps:

[0038] S31, the normalized input unit calculates the formula "h = φ (B N (Wx+b)), h is the output of the fully connected layer, φ is the activation function, B N is the batch normalization operator, W is the weight parameter, x is the input of the fully connected layer, and b is the bias parameter. The output of Batch Norm is obtained and passed to the convolution output unit;

[0039] S32, the convolution output unit calculates the convolution output according to the formula " O(n,k,ox,oy) convolution output data, C is the convolution channel, R is the convolution kernel height, S is the convolution kernel width, I(n,c,ix,iy), W(k,c,r,s) are the convolution input data" to obtain the convolution output data and pass it to the pooling processing unit;

[0040] S33, the pooling processing unit calculates the maximum pooling formula

[0041] “ M(I,j) is the (I,j)th element of the output feature map, u and v are indexes that vary in the range [0, f-1], and s is the step size. The spatial dimension of the feature map is obtained and passed to the gradient optimization unit;

[0042] S34, the gradient optimization unit calculates the formula " Xn is the parameter value of the nth iteration, α is the learning rate or step size, Get the gradient descent data for the gradient of function f at Xn and pass it to the learning training unit;

[0043] S35, the learning training unit determines the training model according to the gradient descent data, and inputs the data in the training set into the model for simulation training under different environmental conditions to ensure the accuracy of model recognition, and transmits it to the model evaluation unit;

[0044] S36, the model evaluation unit obtains the comprehensive evaluation score F according to the model evaluation score calculation formula "F = 2*(Ac*Re) / (Ac+Re), F is the training model evaluation score, Ac is the precision, and Re is the recall rate", and transmits it to the processing center;

[0045] S37. The processing center compares the actual evaluation score of the model test with the compressor control category model evaluation score standard stored in the memory: if it meets the standard, it is the predetermined compressor control parameter; if it does not meet the standard, it is transmitted to the alarm and notified to continue training until the standard is met.

[0046] Further, the step S40 comprises the following steps:

[0047] S41, exhaust flow unit according to the compressed exhaust volume flow calculation formula "L = A * υ0 * sqrt (P0 / P), L is the compressed exhaust volume flow (m 3 / s), A is the area of ​​the compressor inlet pipe (m 2 ), υ0 is the flow rate of the medium under atmospheric pressure (m / s), P0 is the atmospheric pressure (N), and P is the pressure of the medium in the intake pipe (N)” to obtain the exhaust flow of the air conditioning compressor for subsequent debugging and transfer to the compression efficiency unit;

[0048] S42, the compression efficiency unit is calculated according to the compressor efficiency formula "η=P / (4π*D 2 *L*n 2 *(p1 / p2))*100%, η is the compressor efficiency (%), P is the exhaust power, D is the cylinder diameter, L is the stroke, n is the number of compressor cylinders, p1 and p2 are the compressor exhaust pressure and intake pressure respectively" to obtain the air conditioning compressor efficiency for subsequent debugging and use, and pass it to the compression power unit;

[0049] S43, the compression power unit is calculated according to the compression power calculation formula "P = (L*k) / (k-1)*p2*[(p1 / p2) (k-1) / k -1]*(1 / η), P is the compression power (W), L is the compressed exhaust volume flow, k is the gas constant, p1 is the exhaust pressure, p2 is the intake pressure, η is the compressor efficiency" to obtain the compression power for subsequent debugging and transfer to the refrigeration coefficient unit;

[0050] S44, the refrigeration coefficient unit obtains the compressor refrigeration coefficient according to the refrigeration coefficient calculation formula "C0 = m*(h2-h1) / (m*(h2-h1) / C1), C0 is the refrigeration coefficient, m is the refrigerant mass flow rate (kg / s), h2 is the refrigerant enthalpy value at the evaporator outlet (kJ / kg), h1 is the refrigerant enthalpy value at the evaporator inlet (kJ / kg), C1 is the power coefficient", and transmits it to the trial debugging unit;

[0051] S45, the trial debugging unit debugs the current, voltage, pressure, and refrigerant quantity of the air-conditioning compressor under different environmental conditions according to the predetermined control parameters to ensure the normal operation of the air-conditioning compressor to ensure the required temperature requirements, and transmits them to the effect confirmation unit;

[0052] S46, the effect confirmation unit obtains the compressor exhaust pressure, surface temperature and air conditioner temperature, oil pressure, operating current, abnormal sound through the set intelligent sensors, and transmits them to the processing center;

[0053] S47. The processing center compares the compressor operation test data with the compressor operation quality standard stored in the memory: if the standard is met, it is set as the official compressor control parameter; if the standard is not met, it is transmitted to the alarm and notified to adjust the parameter operation.

[0054] Further, the step S50 comprises the following steps:

[0055] S51, the temperature control unit inputs the actual temperature in the air-conditioning compressor into the model to obtain the temperature control mode and implements the corresponding temperature and its temperature control scheme to control the air-conditioning to ensure the temperature required by the user, and transmits it to the quality inspection unit;

[0056] S52, the quality inspection unit obtains various parameter indicators, various performance indicators and whether the filter is blocked of the compressor through intelligent sensors, life testers and gas flow testers, and transmits them to the processing center;

[0057] S53, the processing center compares the compressor operation quality information with the compressor operation quality standard stored in the memory: if the standard is met, the compressor continues to operate; if the standard is not met, the information is transmitted to the alarm and the temperature is re-adjusted or repaired;

[0058] S54, adjusting the maintenance unit to adjust the corresponding temperature within the specified range according to the difference between the actual temperature of the air conditioner and the target temperature, or repairing and replacing the compressor parts to ensure that the compressor can be normally controlled.

[0059] The system described in the present invention also includes a computer-readable storage medium containing a memory; the memory stores a computer program, and when the above functional modules execute the computer program, the steps of an AI-based compressor automatic control method described in any one of the above are implemented; the computer-readable storage medium stores a computer program, and when the computer program is executed by each functional module, the steps of an AI-based compressor automatic control method described in any one of the above are implemented.

[0060] The present invention also provides an AI-based compressor automatic control device, which is implemented using the above-mentioned AI-based compressor automatic control method.

[0061] The beneficial effects of the present invention compared with the prior art are as follows:

[0062] By setting up information acquisition modules, model construction modules, model training modules, parameter determination modules, control application modules and other modules, control parameters can be automatically established, compressor parameters can be automatically obtained, air-conditioning temperature can be automatically adjusted, compressor operation can be automatically detected, and remote control of compressor operation can be achieved, thereby improving compressor management efficiency and reducing compressor management costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or exemplary technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0064] Figure 1 It is a schematic diagram of the system module of the present invention;

[0065] Figure 2 This is a schematic diagram of an information acquisition module of the present invention;

[0066] Figure 3 It is a schematic diagram of the model construction module of the present invention;

[0067] Figure 4 This is a schematic diagram of a model training module of the present invention;

[0068] Figure 5 It is a schematic diagram of a parameter determination module of the present invention;

[0069] Figure 6 It is a schematic diagram of the control application module of the present invention;

[0070] Figure 7 It is a schematic diagram of the method flow control program of the present invention;

[0071] Figure 8This is a schematic diagram of step S10 in the method flow of the present invention;

[0072] Fig. 9 This is a schematic diagram of step S20 in the method flow of the present invention;

[0073] Fig.10 This is a schematic diagram of step S30 in the method flow of the present invention;

[0074] Fig.11 This is a schematic diagram of step S40 in the method flow of the present invention;

[0075] Fig.12 It is a program schematic diagram of step S50 in the method flow of the present invention. DETAILED DESCRIPTION

[0076] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0077] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments:

[0078] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0079] It should be noted that when a module is referred to as being "disposed on" another module, it can be directly on the other module or indirectly on the other module. When a module is referred to as being "connected to" another module, it can be directly connected to the other module or indirectly connected to the other module.

[0080] In the description of the present application, "multiple" means two or more, unless otherwise clearly defined. "Several" means one or more, unless otherwise clearly defined.

[0081] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "connected" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances. The terms "include", "comprise", "have" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0082] See also Figure 1 As shown, the present invention provides an AI-based compressor automatic control method, which is applied to an AI-based compressor automatic control system, wherein the system includes an information acquisition module, a model construction module, a model training module, a parameter determination module, a control application module, a wireless communication module, a memory, an alarm, a processing center, and an intelligent mobile terminal; the information acquisition module, the model construction module, the model training module, the parameter determination module, the control application module, the wireless communication module, the memory, and the alarm are respectively connected to the processing center; the intelligent mobile terminal includes a smart phone, a tablet computer, and a smart remote control, which are respectively connected to the wireless communication module wireless network within the range of a wireless network or the Internet.

[0083] The wireless communication module is provided with a wireless network unit, which is responsible for sending and receiving wireless signals, and automatically connects with the smart mobile terminal within the effective network range. The wireless signals include 8 types of Internet of Things signals such as MQTT, CoAP, HTTP, REST API, Zigbee, LoRaWAN, NB-IoT, and Bluetooth.

[0084] The alarm compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory, and automatically sounds an alarm and notifies to continue training if the standard is not met; and compares the compressor operation test data with the compressor operation quality standard stored in the memory, and automatically sounds an alarm and notifies to adjust the operation parameters if the standard is not met; and compares the compressor operation quality information with the compressor operation quality standard stored in the memory, and automatically sounds an alarm and notifies to re-adjust the temperature or perform maintenance if the standard is not met.

[0085] The memory is responsible for information storage of the information acquisition module, model construction module, model training module, parameter determination module, control application module, wireless communication module, alarm, and storage of air-conditioning compressor control category model evaluation sub-standards and compressor operation quality standards.

[0086] The processing center is responsible for information acquisition module, model construction module, model training module, parameter determination module, control application module, wireless communication module, alarm, and information transmission of memory. It is the hub center of the system and compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory: if it meets the standard, it is the predetermined compressor control parameter; if it does not meet the standard, it is passed to the alarm and notified to continue training; and compares the compressor operation test data with the compressor operation quality standard stored in the memory: if it meets the standard, it is set as the official compressor control parameter; if it does not meet the standard, it is passed to the alarm and notified to adjust the parameter operation; and compares the compressor operation quality information with the compressor operation quality standard stored in the memory: if it meets the standard, it continues to operate; if it does not meet the standard, it is passed to the alarm and notified to re-adjust the temperature or repair.

[0087] See also Figure 2 As shown, the information acquisition module includes a data acquisition unit, a data cleaning unit, a data calibration unit, a normalization standard unit, a filtering and denoising unit, and a data labeling unit, which are responsible for acquiring and preprocessing the normal historical parameter data of the air-conditioning compressor operation and the historical parameter data of various abnormal conditions, and passing them to the model building module.

[0088] Furthermore, the data acquisition unit obtains historical data of pressure, temperature, refrigerant quantity, filter cleaning frequency, component life and current and voltage of similar air-conditioning compressors by networking with the industry database, and transmits it to the data cleaning unit; the data cleaning unit cleans the collected historical data to remove abnormal values, duplicate values ​​or missing values ​​in the data, and transmits it to the data calibration unit; the data calibration unit obtains standardized data according to the data standardization processing formula "z=(x-μ) / σ", and transmits it to the normalization standard unit; the normalization standard unit obtains normalized data according to the normalization calculation formula "x'=(x-min(x)) / (max(x)-min(x))", and transmits it to the filtering and denoising unit; the filtering and denoising unit calculates the data according to the Gaussian filtering method formula The denoised image is obtained and passed to a data annotation unit; the data annotation unit automatically labels the preprocessed data for classification, target detection, and semantic segmentation to facilitate the extraction of key features in subsequent data, which is more conducive to model training, testing, and verification.

[0089] See also Figure 3 As shown, the model building module includes a feature extraction unit, a data partitioning unit, a forward propagation unit, a loss function unit, a back propagation unit, and an activation function unit, which are responsible for extracting features and partitioning the processed data, inputting the data into the model for data training, and passing it to the model training module.

[0090] Furthermore, the feature extraction unit converts the labeled data into feature vectors useful for model training and extracts key features of temperature, pressure, vibration frequency, and current fluctuation, and transmits them to the data division unit; the data division unit divides the data set with key feature vectors after extraction into training set, verification set, and test set and establishes a convolutional neural network structure, and transmits it to the forward propagation unit; the forward propagation unit calculates the forward propagation of the convolution layer according to the forward propagation formula The output of the convolution layer is obtained and passed to the loss function unit; the loss function unit obtains the loss function value according to the loss function calculation formula "L = max(0, m + y * (f (x) - b))" and passes it to the back propagation unit; the back propagation unit obtains the output of the convolution layer and the fully connected layer according to the back propagation calculation formula "dz [l] = da [l] * g [l] ′ (z [l])", and passes it to the activation function unit; the activation function unit obtains the output of the Tanh function according to the ReLU function calculation formula "f (x) = max (0, x)".

[0091] See also Figure 4 As shown, the model training module includes a normalized input unit, a convolution output unit, a pooling processing unit, a gradient optimization unit, a learning training unit, and a model evaluation unit, which is responsible for training the model according to the collected data and adjusting the model parameters to optimize the model performance to predetermine the air-conditioning compressor control parameters and pass them to the parameter determination module.

[0092] Furthermore, the normalized input unit is calculated according to the Batch Norm layer formula “h=φ(B N (Wx+b))” obtains the output of Batch Norm and passes it to the convolution output unit; the convolution output unit calculates the convolution output according to the convolution output formula Obtain the convolution output data and pass it to the pooling processing unit; the pooling processing unit calculates the maximum pooling formula The spatial dimension of the feature map is obtained and passed to the gradient optimization unit; the gradient optimization unit calculates the formula according to the gradient descent method Gradient descent data is obtained and passed to a learning and training unit; the learning and training unit determines a training model according to the gradient descent data, inputs the data in the training set into the model for simulation training under different environmental conditions, and passes the model evaluation unit; the model evaluation unit calculates the model evaluation score according to the model evaluation formula "F = 2*(Ac*Re) / (Ac+

[0093] Re)” to obtain a comprehensive evaluation score of F and pass it to the processing center.

[0094] See also Figure 5As shown, the parameter determination module includes an exhaust flow unit, a compression efficiency unit, a compression power unit, a refrigeration coefficient unit, a trial debugging unit, and an effect confirmation unit. It is responsible for conducting control tests under different environments based on the control parameters predicted by the trained model to obtain the optimal control parameters and pass them to the control application module.

[0095] Further, the exhaust flow unit obtains the exhaust flow of the air conditioner compressor according to the compressed exhaust volume flow calculation formula "L = A*υ0*sqrt(P0 / P)", and transmits it to the compression efficiency unit; the compression efficiency unit calculates the compressor efficiency according to the compressor efficiency calculation formula "η = P / (4π*D 2 *L*n 2 *(p1 / p2))*100%” to obtain the air conditioner compressor efficiency and transmit it to the compression power unit; the compression power unit calculates the compression power according to the compression power calculation formula “P=(L*k) / (k-1)*p2*[(p1 / p2) (k-1) / k -1]*(1 / η)” to obtain compression power and transmit it to the refrigeration coefficient unit; the refrigeration coefficient unit calculates the refrigeration coefficient according to the refrigeration coefficient calculation formula “C0=m*(h2-h1) / (m*(h2

[0096] -h1) / C1)" to obtain the refrigeration coefficient of the compressor, and transmit it to the trial debugging unit; the trial debugging unit debugs the current, voltage, pressure and refrigerant quantity of the air-conditioning compressor under different environmental conditions according to the predetermined control parameters to ensure the normal operation of the air-conditioning compressor, and transmits it to the effect confirmation unit; the effect confirmation unit obtains the compressor exhaust pressure, surface temperature and air-conditioning temperature as well as the oil pressure, operating current and abnormal sound through the set intelligent sensors, and transmits them to the processing center.

[0097] See also Figure 6 As shown, the control application module includes a temperature control unit, a quality detection unit, and an adjustment and maintenance unit, which are responsible for reasonably controlling the temperature of the air conditioner according to the set control parameters and the program instructions of the air conditioner compressor to ensure that the air conditioner can reach the target temperature under different environmental conditions.

[0098] Furthermore, the temperature control unit inputs the actual temperature in the air-conditioning compressor into the model to obtain the temperature control mode and implements the corresponding temperature and its temperature control scheme to control the air-conditioning, and transmits it to the quality detection unit; the quality detection unit obtains various parameter indicators, various performance indicators and whether the filter is blocked of the compressor through intelligent sensors, life testers, and gas flow testers, and transmits them to the processing center; the adjustment and maintenance unit adjusts the corresponding temperature within the specified range according to the difference between the actual temperature of the air-conditioning and the target temperature, or repairs and replaces the compressor accessories to ensure that the compressor can be controlled normally.

[0099] System operation working principle:

[0100] Before operation, the information acquisition module obtains the normal historical parameter data of the air-conditioning compressor operation and the historical parameter data of various abnormal conditions and pre-processes them for subsequent model construction, and passes them to the model construction module; the model construction module extracts features and divides the processed data and inputs them into the model for data training to ensure the accuracy of model recognition, and passes them to the model training module; then the model training module trains the model according to the collected data and adjusts the model parameters to optimize the model performance to predetermine the air-conditioning compressor control parameters, ensure the stability of the air-conditioning compressor temperature control, and pass it to the parameter determination module; the parameter determination module conducts control tests under different environments according to the control parameters predicted by the trained model to obtain the optimal control parameters to ensure the quality of subsequent formal operation, and passes them to the control application module; then the control application module performs reasonable temperature control of the air-conditioning compressor according to the set control parameters according to the program instructions to ensure that the air-conditioning can reach the target temperature under different environmental conditions.

[0101] When the operator or manager is far away or out of town, he or she can use a smart phone to automatically connect to the wireless network or the Internet through the wireless communication module, thereby integrating with smart mobile terminals, the Internet and Internet of Things technologies. The operation of the compressor can be remotely monitored or adjusted in real time through the mobile phone APP, which improves equipment management efficiency and reduces equipment management costs.

[0102] See also Figure 7 As shown, the present invention provides an AI-based compressor automatic control method, comprising the following steps:

[0103] S10, before operation, the information acquisition module acquires and pre-processes the normal historical parameter data of the air-conditioning compressor operation and the historical parameter data of various abnormal conditions for subsequent model construction, and transmits them to the model construction module;

[0104] See also Figure 8 As shown, the step S10 comprises the following steps:

[0105] S11. The data acquisition unit obtains historical data of pressure, temperature, refrigerant quantity, filter cleaning frequency, component life, current and voltage of similar air-conditioning compressors by networking with the industry database to meet the model training requirements, and transmits it to the data cleaning unit;

[0106] It is further explained that various historical parameter data sets of air-conditioning compressors of similar specifications are collected from the industry database through data collection software; the industry database includes historical data of similar air-conditioning compressor application companies and production companies, and also includes historical data of air-conditioning compressors of similar specifications obtained from third-party sharing platforms such as the Internet, social media, or public data sets, so as to obtain high-quality, diverse and rich normal and abnormal historical data of air-conditioning compressors of similar specifications, including pressure, temperature, refrigerant amount, filter cleaning frequency, component life and its corresponding current, voltage, etc., as well as historical data of ambient temperature, humidity, etc.

[0107] S12. The data cleaning unit cleans the collected historical data to remove abnormal values, duplicate values ​​or missing values ​​in the data to improve the quality and accuracy of the data, and transmits the data to the data calibration unit;

[0108] It is further explained that the data is formatted and converted before cleaning to ensure the consistency and integrity of the data and the quality of the data, so that the data type can be identified and converted automatically later; due to sensor failure, recording errors or system abnormalities in the operation of the air-conditioning compressor, abnormal values ​​may occur, which may affect the accuracy of subsequent analysis; missing values ​​are processed for data with missing parts; each feature in the data set is detected by writing code to determine the missing values, and the missing value pattern is identified. According to the number and impact of the missing values, the code is written to select to fill or delete samples or variables with missing values, and a new category is created for categorical feature data to represent the "unknown" state. For time series data, forward filling or backward filling is used to fill the missing values; the processing effect is verified by comparing indicators such as data quality before and after processing, accuracy and reliability of the model, and if the processing effect is not good, the processing method is reselected or the data is re-cleaned until the best effect is achieved, so as to remove duplicate, erroneous, and invalid data to effectively process missing values ​​for subsequent model construction and analysis.

[0109] S13, the data calibration unit obtains standardized data according to the data standardization processing formula "z = (x-μ) / σ, z is standardized data, x is original data, μ is the mean of the data, σ is the standard deviation of the data" to improve the stability of the model, and passes it to the normalization standard unit;

[0110] It is further explained that the collected raw data is calibrated, including time synchronization, unit conversion, range adjustment, etc., such as ensuring that all timestamps are in a unified format and synchronized with the system clock; converting data collected by different sensors to the same unit to ensure data accuracy; using the written code to use the above data standardization processing formula "z = (x-μ) / σ" to convert dates, numbers, texts, etc. into a standard normal distribution with a mean of 0 and a standard deviation of 1, subtracting the mean and dividing by the standard deviation to make it a unified standard format to ensure data consistency and comparability for subsequent processing; encoding unstructured data and converting it into structured data, or converting text data into numerical representation for model processing and analysis, thereby processing the raw data into a form that meets specific standards for subsequent model training and analysis, improving the quality and consistency of the data, and making the data easier to understand and use.

[0111] S14, the normalization standard unit obtains the normalized data according to the normalization calculation formula "x'=(x-min(x)) / (max(x)-min(x)), x' is the normalized data, x is the original data, min(x) and max(x) are the minimum and maximum values ​​of the data respectively" to improve the model performance, and passes it to the filtering and denoising unit;

[0112] It is further explained that the above normalization calculation formula "x'=(x-min(x)) / (max(x)-min(x))" is used by the written code to proportionally map the cleaned data to the specified range of [0,1] to eliminate the total amount difference between different samples or features and the dimensional influence between the data, make the distribution of the data consistent, and confirm whether the data has been normalized to the specified range or distribution as expected, and verify the normalization effect through statistical descriptions such as maximum value and minimum value; the data includes multiple types such as text, pictures, audio, video, etc., to eliminate the dimensional differences between different data, thereby improving the accuracy and efficiency of data analysis.

[0113] S15, filtering and denoising unit calculates the formula " G(x,y) is the two-dimensional Gaussian function pixel, (x,y) is the pixel coordinate, and σ is the standard deviation. The denoised image is obtained for color space conversion and passed to the data annotation unit;

[0114] It is further explained that the Gaussian filter output pixel value obtained according to the Gaussian filter calculation formula is the weighted average of the input pixel values ​​in its neighborhood, and the weight is given by the Gaussian function, where the standard deviation σ determines the width of the Gaussian function, and the Gaussian function is discretized, that is, the weight is calculated within a certain window size, and then these weights are applied to the corresponding neighborhood pixel values ​​of the input image to obtain the output pixel value, and the pixels are arranged from small to large according to the grayscale value and the median is taken as the new grayscale value; the sampling value in the input signal is checked to determine whether it represents the signal itself, and the values ​​in the window are sorted by using an observation window composed of an odd number of samples, the middle value is taken as the output, the earliest value is discarded, a new sample is obtained, and the above calculation process is repeated to reduce the random noise in the image and make the image smoother.

[0115] S16. The data annotation unit automatically labels the preprocessed data for classification, target detection, and semantic segmentation to facilitate the extraction of key features in the subsequent data, which is more conducive to model training, testing, and verification.

[0116] To further explain, the standardized data is classified through the written code, and then the classified data is automatically labeled, and accurate labels or annotations are added to the data, which are used as target variables or features in subsequent model training, and as training sets and validation sets. This ensures that the model has accurate data references during the training process, so that it learns how to generate corresponding labels based on data features to improve labeling efficiency and accuracy; for some data, users are required to enter the system remotely through their mobile phones to manually add new labels or modify existing labels to ensure that they accurately reflect the essential characteristics of the data, and deploy the labeled data to the corresponding system platform for subsequent applications or analysis to improve the accuracy and reliability of the labels.

[0117] S20, the model building module extracts features and divides the processed data and inputs the data into the model for data training to ensure the accuracy of model recognition and pass it to the model training module;

[0118] See also Fig. 9 As shown, the step S20 comprises the following steps:

[0119] S21, the feature extraction unit converts the labeled data into feature vectors useful for model training and extracts key features of temperature, pressure, vibration frequency, and current fluctuation for subsequent data division, and transmits them to the data division unit;

[0120] It is further explained that the most representative features that can reflect the essential content of the data and have distinguishing and descriptive powers are selected from the preprocessed data to reduce the dimension of the features, reduce the computational complexity, and improve the performance and generalization ability of the model; after feature extraction, principal component analysis is used to perform feature dimensionality reduction to obtain a large amount of feature information in order to reduce the dimension of the features and reduce the computational complexity; after feature dimensionality reduction, the features are binary encoded to improve the interpretability of the features and the performance of the model; since some original data information may be lost after feature extraction and dimensionality reduction, feature reconstruction such as principal component reconstruction or least squares reconstruction is performed to restore the data information as much as possible; evaluation is performed through cross-validation, and the feature extraction method and parameters are adjusted according to the evaluation results to evaluate the quality and effect of the extracted features; through feature fusion, the features extracted by various methods are fully utilized.

[0121] S22, the data division unit divides the data set with key feature vectors after extraction into a training set, a validation set and a test set and establishes a convolutional neural network structure for subsequent model training, and passes it to the forward propagation unit;

[0122] It is further explained that the ratio of training set, validation set and test set is determined according to the pre-selected segmentation strategy, and the data distribution between each subset is ensured to be as consistent as possible to avoid introducing bias; in some application scenarios, due to the time series characteristics of the data, the data set is divided into a training set for training the model, a validation set for adjusting model parameters and selecting the best model, and a test set for evaluating the final performance of the model according to the written code to ensure the temporal consistency of the training and test sets; the data set is passed through the convolutional neural network deep learning method in a 9:1 relationship to establish the training set and test set required for the training of the temperature pattern prediction model under different environments, thereby establishing a deep convolutional neural network structure with 3 convolutional layers, 3 maximum pooling layers, 2 fully connected layers, 1 softmax layer, and 1 output layer.

[0123] S23, the forward propagation unit calculates the forward propagation formula of the convolutional layer according to the formula " a' is the output of the lth layer, σ is the activation function, Wlm is the convolution kernel of the lth layer, Xm is the input feature map, and bl is the bias term. The convolution layer output is obtained and passed to the loss function unit;

[0124] Further explanation: the forward pass calculation formula of the convolutional layer is Through "a 2 =σ(z 2 )=σ(a 2 *W 2 +b 2)”, where the superscript 2 is the number of layers 2, the asterisk * is convolution, b is bias, and σ is activation function. In the convolutional neural network, forward propagation includes: the data of the input layer is convolved through the convolution kernel to generate a feature map; the output of the convolution layer is nonlinearly transformed through the activation function; the output of the activation layer is downsampled through the pooling operation to reduce the number of parameters and the amount of calculation; the output of the pooling layer is flattened and input to the fully connected layer to finally obtain the output result.

[0125] S24, the loss function unit obtains the loss function value according to the loss function calculation formula "L = max(0, m + y * (f (x) - b)", L is the loss function value, m is the hyperparameter, y is the sample label (0 or 1), f (x) is the predicted value of the model, and b is the classification boundary" for subsequent back propagation and passes it to the back propagation unit;

[0126] To further explain, in the loss function calculation formula "L=max(0,m+y*(f(x)-b))", when f(x)≤b, then L=m+y(f(x)-b), when f(x)>b, then L=0; in machine learning or optimization problems, f(x) is an optimization target, b is a threshold, m and y are parameters; when the target value f(x) is less than or equal to the threshold b, the value of L is equal to m+y(f(x)-b); when the target value f(x) is greater than the threshold b, the value of L is 0; use the training set data to train the model, adjust the model parameters to minimize the loss function, so that the predicted value of the positive sample is greater than a certain threshold, and the predicted value of the negative sample is less than a certain threshold.

[0127] S25, the back propagation unit obtains the output of the convolution layer and the fully connected layer according to the back propagation calculation formula "dz[l] = da[l] * g[l]′(z[l]), da[l] is the input data, da[1] is the output data, and g[l]′ is the sigmoid derivative of the activation function", and passes it to the activation function unit;

[0128] It is further explained that the output is dw[l]=dz[l]*a[l-1], db[l]=dz[l], da[l-1]=W[l]T*dz[l]; the back propagation of the convolution layer transfers the error term da[l]da[l] to the previous layer through a convolution operation, flips the convolution kernel and applies it to the error map, calculates the local gradient through the derivative of the activation function, and implements it through matrix operations. The input and the convolution kernel are deformed into matrices and matrix multiplication is performed, that is, the numbers in the convolution window are pulled into a row to form a column vector and matrix multiplication is performed; the back propagation of the fully connected layer calculates the gradient of the weight and the gradient of the bias through the chain rule. For each node, the error term is transferred to the node of the previous layer through the weight matrix, and the local gradient is calculated through the derivative of the activation function. It is implemented through matrix operations. The error term is multiplied by the output of the current layer and multiplied by the input of the previous layer to update it in the direction of minimizing the loss function.

[0129] S26. The activation function unit obtains the output of the Tanh function according to the ReLU function calculation formula "f(x)=max(0,x), f(x) is the output of the ReLU function, and x is the input of the ReLU function".

[0130] It is further explained that before data activation, linear transformation output is performed through the linear transformation calculation formula "y=W*X+b", where X is input data, y is output data, W is weight matrix, and b is bias vector; then function activation is performed through the ReLU function calculation formula, and the input x of the ReLU function is the output of the convolution layer or other layers, which is a nonlinear transformation activation function; when the ReLU function is in forward propagation, the gradient is 1 when the input is greater than 0, otherwise it is 0. Since the calculation is simple, the problem of gradient disappearance will not occur, so it performs better in deep networks and maintains the stability of the gradient during propagation. For each element in the input x, if it is greater than 0, the element itself is output; after function activation, the neural network TDropout output is obtained through the Dropout layer calculation formula "y=f(W*X+b)", where y is the neural network output, f is the activation function, W is the weight matrix, X is the input data, and b is the bias vector, which can effectively prevent overfitting.

[0131] S30, the model training module trains the model according to the collected data and adjusts the model parameters to optimize the model performance to predetermine the air-conditioning compressor control parameters, ensure the stability of the air-conditioning compressor temperature control, and pass them to the parameter determination module;

[0132] See also Fig.10 As shown, the step S30 comprises the following steps:

[0133] S31, the normalized input unit calculates the formula "h = φ (BN (Wx+b)), h is the output of the fully connected layer, φ is the activation function, B N is the batch normalization operator, W is the weight parameter, x is the input of the fully connected layer, and b is the bias parameter. The output of Batch Norm is obtained and passed to the convolution output unit;

[0134] Further explanation: the “h=φ(B N (Wx+b))” through “B N (x) = γ⊙[(x-μ B ) / σ B ]+β”, where μ B is the mean, σ B is the standard deviation, γ is the stretch parameter, β is the offset parameter, (x-μ B ) / σ B To standardize the normal distribution; the Batch Norm layer (referred to as "BN layer") is placed between the affine transformation and the activation function in the fully connected layer, and the output value of each layer in the network is standardized to make it more obedient to the normal distribution, which can accelerate the training speed of the neural network and help prevent gradient disappearance and gradient explosion; the BN layer reduces the correlation between input data by standardizing the data of each mini-batch, so that the mean of each sample is 0 and the variance is 1, thereby reducing the problem of internal covariate shift, helping to accelerate the training process, improve the stability and generalization ability of the model, and avoid reducing the problem of reducing the representation ability of the neural network.

[0135] S32, the convolution output unit calculates the convolution output according to the formula " O(n,k,ox,oy) convolution output data, C is the convolution channel, R is the convolution kernel height, S is the convolution kernel width, I(n,c,ix,iy), W(k,c,r,s) are the convolution input data" to obtain the convolution output data and pass it to the pooling processing unit;

[0136] It is further explained that the number of convolution channels C is obtained according to the design; the convolution kernel height R and the convolution kernel width S are calculated by the formulas "R = (R1-F + 2P) / L + 1" and "S = (S1-F + 2P) / L + 1", respectively, where R1 and R are the heights of the kernel before and after the convolution, F is the convolution kernel size, P is the image plus zero thickness, L is the distance of each movement, S1 and S are the widths of the kernel before and after the convolution, respectively; each input channel corresponds to a convolution template, which is used to generate a feature map, and the elements of the feature maps corresponding to all input channels at the same position are added to form an output channel, and the number of output channels is equal to the number of convolution templates (or convolution kernels).

[0137] S33, the pooling processing unit calculates the maximum pooling formula

[0138] “ M(I,j) is the (I,j)th element of the output feature map, u and v are indexes that vary in the range [0, f-1], and s is the step size. The spatial dimension of the feature map is obtained and passed to the gradient optimization unit;

[0139] To further explain, select an appropriate window size according to needs, such as 2*2, 3*3, etc., determine the step size of each movement, usually the step size is consistent with the window size, and for the data in each window, calculate the maximum value or average value, and use the calculation result as the output, and each output position corresponds to an area in the input; the maximum pooling algorithm uses an overlapping 2*2 window for the pooling layer to reduce the size of the feature map, thereby reducing the amount of calculation, while retaining the most significant features, and reducing the spatial dimension of the feature map by taking the maximum value in the local area, but does not change the number of channels, and selects the largest value in a local window of the input feature map as the output of the window. Its parameters include: pooling kernel size, step size, padding, expansion, and whether to round up. It can reduce the spatial dimension of the feature map and reduce the size of the feature map, thereby reducing the amount of calculation and the risk of overfitting.

[0140] S34, the gradient optimization unit calculates the formula " Xn is the parameter value of the nth iteration, α is the learning rate or step size, Get the gradient descent data for the gradient of function f at Xn and pass it to the learning training unit;

[0141] Further explanation: the gradient descent method can minimize the differentiable function f(x) to obtain the minimum value x by iteratively updating the parameters and gradually approaching the optimal solution. In the formula, the learning rate or step size α is used to control the step size of each iteration, and the gradient of the function f at Xn is That is, the derivative, starting from any initial point, determines the next position according to the gradient direction of the current position (that is, the direction in which the function decreases fastest) and the learning rate, and iterates until the stopping condition is met, including: starting from any initial point, calculating the gradient of the current point Update Parameters Check the stopping condition and end the iteration if it is satisfied, otherwise recalculate the gradient.

[0142] S35, the learning training unit determines the training model according to the gradient descent data, and inputs the data in the training set into the model for simulation training under different environmental conditions to ensure the accuracy of model recognition, and transmits it to the model evaluation unit;

[0143] It is further explained that the loss function is minimized and the prediction accuracy of the model is improved by continuously adjusting the model parameters to ensure that the model can accurately identify the adaptability of the temperature control of the air-conditioning compressor in different environments; the model is trained through the data set in the training set to fit the data analysis rules, that is, to determine the various learning parameters such as the model weight and bias; and the model parameters and hyperparameters are adjusted during the model training process through the data set in the validation set to optimize the model performance and avoid overfitting. The model is selected, but it does not participate in the determination of the learning parameters, and the model parameters and hyperparameters with smaller model errors are selected; then the data set in the test set is used to evaluate the generalization ability of the model on unknown data after the model training is completed, and it is used once after training to evaluate the effect of the final model, without participating in the learning parameter process or the hyperparameter selection process.

[0144] S36, the model evaluation unit obtains the comprehensive evaluation score F according to the model evaluation score calculation formula "F = 2*(Ac*Re) / (Ac+Re), F is the training model evaluation score, Ac is the precision, and Re is the recall rate", and transmits it to the processing center;

[0145] Further explanation, the Ac and Re are respectively expressed by "Ac=(T P +T N ) / (T P +F P +F N +T N ), Re=T P / (T P +F N )”, where T P is the number of true positive samples, F P is the actual number of false positive samples, F N is the number of false negative samples, T N is the number of true negative samples; after the model training is completed, a test is performed to verify the accuracy and reliability of the model. During the test, problems with the model may be found and optimized and adjusted; the training model evaluation score, namely the F value, is the harmonic mean of precision and recall, which can evaluate the performance of the classification model. The value range is 0 to 1, where 1 is the best performance and 0 is the worst performance. The higher the F value, the better the prediction effect of the model, and vice versa. Therefore, if the F value is close to 1, it indicates that the model performs well while maintaining a balance between precision and recall; adjust the hyperparameters according to the model's learning rate, regularization coefficient and other performance to optimize the model performance.

[0146] S37. The processing center compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory: if it meets the standard, it is the predetermined compressor control parameter; if it does not meet the standard, it is transmitted to the alarm and notified to continue training until it meets the standard.

[0147] It is further explained that the air-conditioning compressor control category identification training model test evaluation score standard is best when it is greater than 0.8, which is specifically determined according to the environmental location of the air-conditioning compressor and its air-conditioning and the climate temperature difference. For example, the control category training model evaluation score standard for air-conditioning compressors in areas with uniform temperature differences throughout the year and less dust in the air is 1 or close to 1; while the control model evaluation score standard for air-conditioning compressors in the Northeast, Northwest desert areas or tropical areas of South China where the temperature difference throughout the year is very different is relatively low. Due to the long and cold winters in the Northeast and the long summers in South China, the air-conditioning compressor must always maintain a stable temperature, with sufficiently low temperatures in the summer and sufficiently high temperatures in the winter. The amount of dust in the air must also be small to avoid clogging of the air-conditioning filter, thereby ensuring the normal operation of the air-conditioning compressor.

[0148] S40, the parameter determination module performs control tests under different environments according to the control parameters predicted by the trained model to obtain the best control parameters to ensure the quality of subsequent formal operation, and passes them to the control application module;

[0149] See also Fig.11 As shown, the step S40 comprises the following steps:

[0150] S41, exhaust flow unit according to the compressed exhaust volume flow calculation formula "L = A * υ0 * sqrt (P0 / P), L is the compressed exhaust volume flow (m 3 / s), A is the area of ​​the compressor inlet pipe (m 2 ), υ0 is the flow rate of the medium under atmospheric pressure (m / s), P0 is the atmospheric pressure (N), and P is the pressure of the medium in the intake pipe (N)” to obtain the exhaust flow of the air conditioning compressor for subsequent debugging and transfer to the compression efficiency unit;

[0151] To further explain, the compressed exhaust volume flow calculation formula "L=A*υ0*sqrt(P0 / P)" is obtained by substituting "υ=υ0*sqrt(P0 / P)" into "L=A*υ", where υ is the average flow rate of compressed air in the pipeline (m / s); the compressor intake pipeline area A is a known design parameter of the air-conditioning compressor manufacturer, and the medium flow rate υ0 at atmospheric pressure is verifiable data; the atmospheric pressure P0 and the medium pressure P in the intake pipeline are both obtained by detecting through a pressure sensor; the medium includes air, nitrogen, oxygen, hydrogen, argon and other gases that need to be compressed in industrial production, and the flow rates υ0 of various medium gases at atmospheric pressure are different, and the pressures P of various media in the intake pipeline are also different.

[0152] S42, the compression efficiency unit is calculated according to the compressor efficiency formula "η=P / (4π*D 2*L*n 2 *(p1 / p2))*100%, η is the compressor efficiency (%), P is the exhaust power, D is the cylinder diameter, L is the stroke, n is the number of compressor cylinders, p1 and p2 are the compressor exhaust pressure and intake pressure respectively" to obtain the air conditioning compressor efficiency for subsequent debugging and use, and pass it to the compression power unit;

[0153] Further explanation: the compressor efficiency calculation formula is “η=P / (4π*D 2 *L*n 2 *(p1 / p2))” is derived from “P=4π*D 2 *L*n 2 *(p1 / p2)*η”; the “P=4π*D 2 *L*n 2 *(p1 / p2)*η” is derived from “S=4π*D 2 *L*n" into "V = S*n" and then into "P = V*(p1 / p2)*η", where P is the exhaust power (W), V is the effective displacement of the cylinder in the compressor (m 3 ), S is the cross-sectional area of ​​the compressor exhaust port (m 2 ); the compressor exhaust power P, compressor cylinder displacement V, compressor exhaust port cross-sectional area S, cylinder diameter D, stroke L, compressor cylinder number n are all design parameters of the air-conditioning compressor manufacturer; the average pressure p0 in the exhaust pipe is calculated through historical experience data; the compressor exhaust pressure p1 and intake pressure p2 are both detected by the set pressure sensor, and the unit is N / m 2 ; The units of the cylinder diameter D and stroke L are both m; the compressor efficiency calculation formula is calculated using the numerical values ​​of each parameter.

[0154] S43, the compression power unit is calculated according to the compression power calculation formula "P = (L*k) / (k-1)*p2*[(p1 / p2) (k-1) / k -1]*(1 / η), P is the compression power (W), L is the compressed exhaust volume flow, k is the gas constant, p1 is the exhaust pressure, p2 is the intake pressure, η is the compressor efficiency" to obtain the compression power for subsequent debugging and transfer to the refrigeration coefficient unit;

[0155] Further explanation: the “p2*[(p1 / p2) (k-1) / k -1]” is related to the suction pressure, exhaust pressure and gas constant, indicating the energy required for pressure change; the compressed exhaust volume flow L (m 3 / s) is calculated in step S41 above; the gas constant k is obtained by querying the industry database, where the k value of air is 1.4; the exhaust pressure p1 (N / m2 )、Intake pressure p2(N / m 2 ) are respectively obtained by detecting through the provided pressure sensors; the compressor efficiency η is obtained by calculation in the above step S42; the compressor efficiency calculation formula is calculated using the numerical values ​​of each parameter.

[0156] S44, the refrigeration coefficient unit obtains the compressor refrigeration coefficient according to the refrigeration coefficient calculation formula "C0 = m*(h2-h1) / (m*(h2-h1) / C1), C0 is the refrigeration coefficient, m is the refrigerant mass flow rate (kg / s), h2 is the refrigerant enthalpy value at the evaporator outlet (kJ / kg), h1 is the refrigerant enthalpy value at the evaporator inlet (kJ / kg), C1 is the power coefficient", and transmits it to the trial debugging unit;

[0157] Further explanation: the refrigeration coefficient calculation formula is obtained by substituting "Qc = m*(h2-h1)", "P = m*(h2-h1) / C1" into "C0 = Qc / P", where Qc is the compressor cooling capacity and P is the compressor exhaust power; the refrigeration coefficient C0 is the ratio between the compressor cooling effect and energy consumption; the C1 is the power coefficient, which is the ratio of the unit cooling capacity to the compressor cooling power, calculated from historical data; the refrigerant enthalpy value h2 / h1 at the evaporator inlet / outlet is designed by the air-conditioning compressor manufacturer Parameters; the refrigerant mass flow rate m is calculated by the formula "m=(V*p*N*3600) / (1000*η)", where m is the refrigerant mass flow rate, V is the compressor displacement, p is the refrigerant density, N is the compressor speed, and η is the compression efficiency; the refrigerant mass flow rate m and the compressor displacement V are design parameters of the air-conditioning compressor manufacturer; the refrigerant density p is obtained by network query; the compressor speed N is obtained by detection by a set sensor; the compression efficiency η is calculated in the above step S42.

[0158] S45, the trial debugging unit debugs the current, voltage, pressure, and refrigerant quantity of the air-conditioning compressor under different environmental conditions according to the predetermined control parameters to ensure the normal operation of the air-conditioning compressor to ensure the required temperature requirements, and transmits them to the effect confirmation unit;

[0159] Further explanation, referring to the parameters calculated in the above steps S41-S44, the air-conditioning compressor is controlled under simulated environmental conditions such as high temperature, low temperature, high humidity, and high pressure. The air-conditioning compressor is controlled closely or remotely through the APP software on the smart mobile terminal or the remote control software on the computer. Low-temperature control and high-temperature control are performed in this simulated environment to ensure that the air-conditioning compressor can heat and cool normally to meet the temperature requirements of the user. Through the system, managers can remotely control and monitor all air-conditioning compressors, view daily operating conditions, data changes, equipment activity logs, etc., and remotely connect and control the air-conditioning compressor through pre-made deployment codes or manually adding deployment codes. The operation of the remote air-conditioning compressor can be viewed through the smart mobile terminal or the computer, and can be controlled using the local keyboard and mouse.

[0160] S46, the effect confirmation unit obtains the compressor exhaust pressure, surface temperature and air conditioner temperature, oil pressure, operating current, abnormal sound through the set intelligent sensors, and transmits them to the processing center;

[0161] It is further explained that the intelligent sensor is provided with multiple sensor probes. The pressure sensor probe can be used to detect the exhaust pressure of the air-conditioning compressor, so that the exhaust pressure of the compressor is stable and can meet the working pressure requirements of the equipment; the temperature sensor probe can be used to detect the surface temperature of the air-conditioning compressor and the temperature of the air-conditioning exhaust respectively to ensure that the cooling capacity of the compressor should meet the cooling requirements of the equipment, and the refrigeration performance coefficient (COP) is high, indicating a high energy efficiency ratio, which can save energy and reduce operating costs; the current and voltage sensor probes are used to detect the current and voltage of the air-conditioning compressor; whether there are metal friction sounds, buzzing sounds, knocking sounds and other noises when the compressor is running.

[0162] S47. The processing center compares the compressor operation test data with the compressor operation quality standard stored in the memory: if it meets the standard, it is set as the official compressor control parameter; if it does not meet the standard, it is transmitted to the alarm and notified to adjust the parameter operation until the temperature control effect meets the standard.

[0163] It is further explained that the compressor operation quality standards include: performance parameters such as cooling capacity, energy consumption, exhaust pressure, input power, refrigeration performance coefficient and so on meet the standards, there is no abnormal sound and temperature, pressure and other temperature control indicators and vibration amplitude meet the standards, durability, vibration resistance, corrosion resistance, temperature rise test and other comprehensive performance meet the standards, among which, it has a high cooling capacity to meet the cooling needs of the equipment, has a high energy efficiency ratio and saves energy to reduce operating costs, the exhaust pressure is stable and can meet the working pressure requirements of the equipment, the refrigeration performance coefficient meets the standards, there is no abnormal sound during operation of the main bearing, crankcase, slide, cylinder and other parts, temperature, pressure and other temperature control indicators meet the specified requirements and the excess rate is less than 2‰, the body amplitude does not exceed the specified value and there is no obvious vibration phenomenon in the auxiliary machine and pipeline, the performance is stable after long-term operation, it has stability and reliability under vibration conditions, it has durability and corrosion resistance in a specific corrosive environment, the temperature rise during operation meets the standards, the filter is not blocked, and the components are not aged.

[0164] S50, the control application module controls the air-conditioning compressor to reasonably control the temperature of the air-conditioning according to the set control parameters and the program instructions, so as to ensure that the air-conditioning can reach the target temperature under different environmental conditions.

[0165] See also Fig.12 As shown, the step S50 comprises the following steps:

[0166] S51, the temperature control unit inputs the actual temperature in the air-conditioning compressor into the model to obtain the temperature control mode and implements the corresponding temperature and its temperature control scheme to control the air-conditioning to ensure the temperature required by the user, and transmits it to the quality inspection unit;

[0167] Further explanation: According to the national air-conditioning temperature standard, the indoor air-conditioning temperature standard in summer is 26-28℃, and the indoor air-conditioning temperature standard in winter is 18-20℃. However, in order to meet the special needs of some customers, indoor temperature control solutions are established for different environments and user needs: 26-28℃ in summer, 22-24℃ in spring and autumn, and 18-20℃ in winter at home; 24-26℃ in summer, 22-24℃ in spring and autumn, and 20-22℃ in winter in offices or commercial places. When the indoor temperature is lower than these standard ranges detected by the temperature sensor, the compressor is automatically controlled to control the air-conditioning to heat and increase the temperature. When the temperature is higher than these standard ranges, the air-conditioning is automatically controlled to cool and reduce the temperature. According to the needs of some special users, temporary adjustments can be made outside these ranges. A temporary temperature control range can be established through wireless control of smart mobile terminals or remote control of the target air-conditioning. After the set time, it will be restored to the original temperature control range to save energy and reduce emissions.

[0168] S52, the quality inspection unit obtains various parameter indicators, various performance indicators and whether the filter is blocked of the compressor through intelligent sensors, life testers and gas flow testers, and transmits them to the processing center;

[0169] Further explanation: Through the intelligent sensor installed in the compressor, various parameters such as the compressor's exhaust temperature, shell temperature, condensing temperature, evaporating temperature, oil temperature and oil pressure, operating current, abnormal sound, etc. can be obtained. The compressor's power, pressure, flow rate, compression ratio, refrigeration capacity, vibration, energy consumption and other performance indicators can be detected through a life tester. The gas flow tester can be used to detect whether the filter is blocked. If the compressor emits a uniform and low operating sound, it is normal, and the life and performance of the compressor can be scientifically evaluated to avoid production interruptions caused by faults. If the compressor exhaust temperature is abnormally high, it may be caused by the compressor's suction temperature or condensation temperature being too high. If the exhaust temperature is too low, it may be caused by excessive humidity during the operation of the compressor or too little refrigerant in the system. If the upper casing of the compressor is affected by the inhaled steam, the temperature is low and there is condensation, while the temperature in the lower casing is high due to the heat generated by the motor and the friction heat brought out by the refrigeration oil. If the casing surface temperature rises abnormally, it is caused by the excessively high suction temperature of the refrigeration system.

[0170] S53, the processing center compares the compressor operation quality information with the compressor operation quality standard stored in the memory: if the standard is met, the compressor continues to operate; if the standard is not met, the information is transmitted to the alarm and the temperature is re-adjusted or repaired;

[0171] Further explanation, in addition to the content described in the above step S47, the compressor operation quality standard also includes that the exhaust temperature should generally not exceed 105°C in summer, and if it is R22 refrigerant, it should not exceed 125°C; the normal operating temperature range of the compressor shell is generally between 50-90°C, and the detailed temperature range is set according to the compressor type, working conditions and external environment; the condensing temperature of the compressor is 40°C, and the evaporation temperature is 3-5°C lower than the outlet temperature of the chilled water; the oil temperature of the compressor is maintained between 40 and 55°C; the oil pressure of the compressor is between 0.2 and 0.3MPa; the operating current of the compressor is between 70% and 100% of the rated current, and the detailed range is set according to the compressor model, use environment, and load conditions to avoid burning of the motor; the operating sound of the compressor should be smooth and uniform, without knocking and abnormal sounds; the filter is not blocked.

[0172] S54, adjusting the maintenance unit to adjust the corresponding temperature within the specified range according to the difference between the actual temperature of the air conditioner and the target temperature, or repairing and replacing the compressor parts to ensure that the compressor can be normally controlled.

[0173] To further explain, the indoor temperature standard is: 26-28℃ in summer, 22-24℃ in spring and autumn, and 18-20℃ in winter at home; 24-26℃ in summer, 22-24℃ in spring and autumn, and 20-22℃ in winter in offices or commercial places; when the indoor temperature is lower than the minimum temperature standard in summer, the compressor is controlled to automatically adjust the air conditioner to increase the high temperature to achieve the minimum temperature standard of indoor air conditioning; when the indoor temperature is higher than the maximum temperature standard in winter, the compressor is controlled to automatically adjust the air conditioner to lower the temperature to achieve the maximum temperature standard of indoor air conditioning; if the indoor temperature in summer cannot reach the minimum standard or the indoor temperature in winter cannot reach the maximum standard, it proves that the air-conditioning compressor is faulty or aging, or other quality standards are not met, and maintenance is required. It may be one or more of the faults such as the air-conditioning compressor cannot refrigerate, abnormal noise, poor refrigeration effect, increased power consumption, condenser or evaporator freezing, aging of accessories, etc. If there are no faults, it means that the compressor is aging and needs to be replaced.

[0174] It is further explained that the above-mentioned steps are displayed in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise clearly stated in this document, there is no strict order restriction for the execution of these steps. These steps can also be executed according to other orders, and some steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed and completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0175] To further illustrate, the present invention is described according to the content implemented by the software program of the AI-based compressor automatic control system. Each AI-based compressor automatic control method is divided into several modules or units to implement the software program instructions generated by each step. The software program instructions include the above-mentioned AI-based compressor automatic control method.

[0176] The system described in the present invention also includes a computer-readable storage medium containing a memory; the memory stores a computer program, and when the above functional modules execute the computer program, the steps of an AI-based compressor automatic control method described in any one of the above are implemented; the computer-readable storage medium stores a computer program, and when the computer program is executed by each functional module, the steps of an AI-based compressor automatic control method described in any one of the above are implemented.

[0177] It is further explained that the computer program is divided into multiple modules / units, and the multiple modules / units are stored in the memory and executed by each module / unit to complete the present invention; the multiple modules / units can be a series of computer program instructions that can complete specific functions, and the instructions are used to describe the execution process of the computer program in each module / unit.

[0178] Further explanation: the content described above in the present invention is based on the "Zhongneng Refrigeration Compressor Refrigeration and Fresh Air Linkage Energy Saving Intelligent Switching System (Registration No.: 2020SR0486)" which was applied for by the company's compressor refrigeration system project in May 2020.

[0179] 686)", "Zhongneng Refrigeration Compressor Refrigeration and Heat Exchange Linkage Energy Saving Intelligent Switching System (Registration No.: 2020SR0487002)", "Zhongneng Refrigeration Single-phase AC Compressor Drive (2000W) System (Registration No.: 2020SR0488581)", "Zhongneng Refrigeration Single-phase AC Compressor Drive Servo Drive Parameter Configuration System (Registration No.: 2020SR0495905)", "Zhongneng Refrigeration DC48V Single-phase DC Compressor Drive (300W) System (Registration No.: 2020SR0494619)", "Compressor Reciprocating State Control System (Registration No.: 2022SR1182406)", "Compressor Flow Regulation Control System (Registration No.: 2022SR1182348)", "Compressor Variable Frequency Starter Application Software (Registration No.: 2022SR

[0180] 1321515)" and other software copyrights.

[0181] The present invention also provides an AI-based compressor automatic control device, which is implemented by the above-mentioned AI-based compressor automatic control method.

[0182] It is further explained that the control device is composed of the above-mentioned related equipment or devices, and can be made into a complete set of compressor production and testing equipment together with the compressor production and testing equipment according to the needs of the compressor manufacturer or the user; it can also be made into a unit device of each functional module belonging to each step or the entire control device, and during installation, each control device can be connected to each compressor production and testing equipment through a wireless connection; the existing old air-conditioning compressor production and testing equipment can also be upgraded and modified by adding the control device, thereby saving equipment costs. The above equipment structures are not described in detail here; the control device can also be used in professional compressor testing institutions to improve detection efficiency.

[0183] It is obvious to those skilled in the art that the present application is not limited to the details of the above exemplary embodiments, and that the present application can be implemented in other specific forms without departing from the spirit or basic features of the present application. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive, and the scope of the present application is defined by the attached claims rather than the above description, and therefore it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present application.

[0184] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present application and are not intended to limit it. Although the present application has been described in detail with reference to the preferred embodiments, a person skilled in the art should understand that the technical solution of the present application can be modified or replaced by equivalents, and all should be included in the scope of protection of the present application.

Claims

1. A compressor automatic control method based on AI, characterized in that The method is applied to an AI-based compressor automatic control system, the system comprising an information acquisition module, a model construction module, a model training module, a parameter determination module, a control application module, a wireless communication module, a memory, an alarm, a processing center, and an intelligent mobile terminal; the information acquisition module, the model construction module, the model training module, the parameter determination module, the control application module, the wireless communication module, the memory, and the alarm are respectively connected to the processing center; the intelligent mobile terminal comprises a smart phone, a tablet computer, and a smart remote control, which are wirelessly connected to the wireless communication module within the range of a wireless network or the Internet; the method comprises the following steps: S10, before operation, the information acquisition module acquires and pre-processes the normal historical parameter data of the air-conditioning compressor operation and the historical parameter data of various abnormal conditions for subsequent model construction, and transmits them to the model construction module; S20, the model building module extracts features and divides the processed data and inputs the data into the model for data training to ensure the accuracy of model recognition and pass it to the model training module; S30, the model training module trains the model according to the collected data and adjusts the model parameters to optimize the model performance to predetermine the air-conditioning compressor control parameters, ensure the stability of the air-conditioning compressor temperature control, and pass them to the parameter determination module; S40, the parameter determination module performs control tests under different environments according to the control parameters predicted by the trained model to obtain the best control parameters to ensure the quality of subsequent formal operation, and passes them to the control application module; S50, the control application module controls the air-conditioning compressor to reasonably control the temperature of the air-conditioning according to the set control parameters and the program instructions, so as to ensure that the air-conditioning can reach the target temperature under different environmental conditions.

2. The AI-based compressor automatic control method according to claim 1, characterized in that: The system further comprises: the wireless communication module is provided with a wireless network unit, which is responsible for sending and receiving wireless signals and automatically networking with the intelligent mobile terminal within an effective network range; The alarm device compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory, and automatically sounds an alarm and notifies to continue training if the standard is not met; and compares the compressor operation test data with the compressor operation quality standard stored in the memory, and automatically sounds an alarm and notifies to adjust the operation parameters if the standard is not met; and compares the compressor operation quality information with the compressor operation quality standard stored in the memory, and automatically sounds an alarm and notifies to re-adjust the temperature or repair if the standard is not met; The memory is responsible for storing information of the information acquisition module, the model construction module, the model training module, the parameter determination module, the control application module, the wireless communication module, and the alarm, as well as the storage of the air-conditioning compressor control category model evaluation sub-standard and the compressor operation quality standard; The processing center is responsible for the information transmission of each module, alarm, and memory, and is the hub center of the system. It compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory: if it meets the standard, it is the preset compressor control parameter; if it does not meet the standard, it is passed to the alarm and notified to continue training; and compares the compressor operation test data with the compressor operation quality standard stored in the memory: if it meets the standard, it is set as the official compressor control parameter; if it does not meet the standard, it is passed to the alarm and notified to readjust the parameter; and compares the compressor operation quality information with the compressor operation quality standard stored in the memory: if it meets the standard, it continues to operate; if it does not meet the standard, it is passed to the alarm and notified to re-adjust the temperature or repair; The information acquisition module includes a data acquisition unit, a data cleaning unit, a data calibration unit, a normalization standard unit, a filtering and denoising unit, and a data labeling unit, which are responsible for acquiring and preprocessing the normal historical parameter data of the air-conditioning compressor operation and the historical parameter data of various abnormal conditions, and passing them to the model building module; The model building module includes a feature extraction unit, a data partitioning unit, a forward propagation unit, a loss function unit, a back propagation unit, and an activation function unit, which are responsible for extracting features and partitioning the processed data, inputting the data into the model for data training, and passing it to the model training module; The model training module includes a normalized input unit, a convolution output unit, a pooling processing unit, a gradient optimization unit, a learning training unit, and a model evaluation unit, which is responsible for training the model according to the collected data and adjusting the model parameters to optimize the model performance to predetermine the air-conditioning compressor control parameters and pass them to the parameter determination module; The parameter determination module includes an exhaust flow unit, a compression efficiency unit, a compression power unit, a refrigeration coefficient unit, a trial debugging unit, and an effect confirmation unit, which is responsible for performing control tests under different environments based on the control parameters predicted by the trained model to obtain the optimal control parameters and pass them to the control application module; The control application module includes a temperature control unit, a quality detection unit, and an adjustment and maintenance unit, which are responsible for reasonably controlling the temperature of the air conditioner according to the set control parameters and the program instructions of the air conditioner compressor to ensure that the air conditioner can reach the target temperature under different environmental conditions.

3. The AI-based compressor automatic control method according to claim 1, characterized in that: Described step S20, comprises the following steps: S21, the feature extraction unit converts the labeled data into feature vectors useful for model training and extracts key features of temperature, pressure, vibration frequency, and current fluctuation for subsequent data division, and transmits them to the data division unit; S22, the data division unit divides the data set with key feature vectors after extraction into a training set, a validation set and a test set and establishes a convolutional neural network structure for subsequent model training, and passes it to the forward propagation unit; S23, forward propagation unit according to the convolutional layer forward propagation calculation formula a' is the first Layer output, σ is the activation function, Wlm is the l-th layer convolution kernel, Xm is the input feature map, bl is the bias term" to obtain the convolution layer output and pass it to the loss function unit; S24, the loss function unit obtains the loss function value according to the loss function calculation formula "L = max(0, m + y * (f (x) - b)), L is the loss function value, m is the hyperparameter, y is the sample label (0 or 1), f (x) is the predicted value of the model, and b is the classification boundary" for subsequent back propagation and passes it to the back propagation unit; S25, the back propagation unit obtains the output of the convolution layer and the fully connected layer according to the back propagation calculation formula "dz[l] = da[l] * g[l]′(z[l]), da[l] is the input data, da[1] is the output data, and g[l]′ is the sigmoid derivative of the activation function", and passes it to the activation function unit; S26, the activation function unit obtains the output of the Tanh function according to the ReLU function calculation formula "f(x)=max(0,x), f(x) is the output of the ReLU function, and x is the input of the ReLU function"; The step S40 comprises the following steps: S41, exhaust flow unit according to the compressed exhaust volume flow calculation formula "L = A * υ0 * sqrt (P0 / P), L is the compressed exhaust volume flow (m 3 / s), A is the area of ​​the compressor inlet pipe (m 2 ), υ0 is the flow rate of the medium under atmospheric pressure (m / s), P0 is the atmospheric pressure (N), and P is the pressure of the medium in the intake pipe (N)” to obtain the exhaust flow of the air conditioning compressor for subsequent debugging and transfer to the compression efficiency unit; S42, compression efficiency unit according to the compressor efficiency calculation formula "η=P / (4π*D 2 *L*n 2 *(p1 / p2))*100%, η is the compressor efficiency (%), P is the exhaust power, D is the cylinder diameter, L is the stroke, n is the number of compressor cylinders, p1 and p2 are the compressor exhaust pressure and intake pressure respectively" to obtain the air conditioning compressor efficiency for subsequent debugging and use, and pass it to the compression power unit; S43, compression power unit according to the compression power calculation formula "P=(L*k) / (k-1)*p2*[(p1 / p2) (k-1) / k -1]*(1 / η), P is the compression power (W), L is the compressed exhaust volume flow, k is the gas constant, p1 is the exhaust pressure, p2 is the intake pressure, η is the compressor efficiency" to obtain the compression power for subsequent debugging and transfer to the refrigeration coefficient unit; S44, the refrigeration coefficient unit obtains the compressor refrigeration coefficient according to the refrigeration coefficient calculation formula "C0 = m*(h2-h1) / (m*(h2-h1) / C1), C0 is the refrigeration coefficient, m is the refrigerant mass flow rate (kg / s), h2 is the refrigerant enthalpy value at the evaporator outlet (kJ / kg), h1 is the refrigerant enthalpy value at the evaporator inlet (kJ / kg), C1 is the power coefficient", and transmits it to the trial debugging unit; S45, the trial debugging unit debugs the current, voltage, pressure, and refrigerant quantity of the air-conditioning compressor under different environmental conditions according to the predetermined control parameters to ensure the normal operation of the air-conditioning compressor to ensure the required temperature requirements, and transmits them to the effect confirmation unit; S46, the effect confirmation unit obtains the compressor exhaust pressure, surface temperature and air conditioner temperature, oil pressure, operating current, abnormal sound through the set intelligent sensors, and transmits them to the processing center; S47. The processing center compares the compressor operation test data with the compressor operation quality standard stored in the memory: if the standard is met, it is set as the official compressor control parameter; if the standard is not met, it is transmitted to the alarm and notified to adjust the parameter operation.

4. The AI-based compressor automatic control method according to claim 1, characterized in that: Described step S30, comprises the following steps: S31, the normalized input unit is calculated according to the Batch Norm layer formula "h=φ(B N (Wx+b)), h is the output of the fully connected layer, φ is the activation function, B N is the batch normalization operator, W is the weight parameter, x is the input of the fully connected layer, and b is the bias parameter. The output of Batch Norm is obtained and passed to the convolution output unit; S32, the convolution output unit is calculated according to the convolution output formula O(n,k,ox,oy) convolution output data, C is the convolution channel, R is the convolution kernel height, S is the convolution kernel width, I(n,c,ix,iy), W(k,c,r,s) are the convolution input data" to obtain the convolution output data and pass it to the pooling processing unit; S33, the pooling processing unit calculates the maximum pooling formula M(I,j) is the (I,j)th element of the output feature map, u and v are indexes that vary in the range [0, f-1], and s is the step size. The spatial dimension of the feature map is obtained and passed to the gradient optimization unit; S34, the gradient optimization unit calculates the formula according to the gradient descent method Xn is the parameter value of the nth iteration, α is the learning rate or step size, Get the gradient descent data for the gradient of function f at Xn and pass it to the learning training unit; S35, the learning training unit determines the training model according to the gradient descent data, and inputs the data in the training set into the model for simulation training under different environmental conditions to ensure the accuracy of model recognition, and transmits it to the model evaluation unit; S36, the model evaluation unit obtains the comprehensive evaluation score F according to the model evaluation score calculation formula "F = 2*(Ac*Re) / (Ac+Re), F is the training model evaluation score, Ac is the precision, and Re is the recall rate", and transmits it to the processing center; S37, the processing center compares the actual evaluation score of the model test with the air-conditioning compressor control category model evaluation score standard stored in the memory: if it meets the standard, the compressor control parameters are preset; if it does not meet the standard, it is transmitted to the alarm and notified to continue training until it meets the standard; Described step S50 comprises the following steps: S51, the temperature control unit inputs the actual temperature in the air-conditioning compressor into the model to obtain the temperature control mode and implements the corresponding temperature and its temperature control scheme to control the air-conditioning to ensure the temperature required by the user, and transmits it to the quality inspection unit; S52, the quality inspection unit obtains various parameter indicators, various performance indicators and whether the filter is blocked of the compressor through intelligent sensors, life testers and gas flow testers, and transmits them to the processing center; S53, the processing center compares the compressor operation quality information with the compressor operation quality standard stored in the memory: if the standard is met, the compressor continues to operate; if the standard is not met, the information is transmitted to the alarm and the temperature is re-adjusted or repaired; S54, adjusting the maintenance unit to adjust the corresponding temperature according to the difference between the actual temperature of the air conditioner and the target temperature to ensure that it is within the specified range, or repairing and replacing the compressor parts to ensure that the compressor can be normally controlled.

5. According to the AI-based compressor automatic control method described in claims 1-4, it is characterized in that : The system also includes a computer-readable storage medium including a memory; the memory stores a computer program, and when the above functional modules execute the computer program, the steps of the AI-based compressor automatic control method described in any one of claims 1 to 4 are implemented; the computer-readable storage medium stores a computer program, and when the computer program is executed by the functional modules, the steps of the AI-based compressor automatic control method described in any one of claims 1 to 4 are implemented.

6. An AI-based compressor automatic control device, characterized in that: This is achieved by using an AI-based compressor automatic control method as described in any one of claims 1 to 4 above.

Images