An electric scroll compressor regulating control system

Abstract

The present application relates to the field of electric scroll compressors, and specifically to an electric scroll compressor regulation and control system. It obtains the temperature target offset between the ambient temperature and the target value, as well as the real-time flow value of the refrigerant and the real-time speed value of the compressor motor through real-time monitoring, and introduces data processing and analysis algorithms based on artificial intelligence and deep learning at the back end to perform time-series forward feature extraction and collaborative correlation analysis on the temperature target offset, the real-time flow value of the refrigerant and the real-time speed value of the compressor motor, so as to capture the implicit connection and interactive influence between the temperature target offset characteristics and the target regulation parameter characteristics of the electric scroll compressor, thereby facilitating the use of the relationship between the two to perform refrigerant flow value regulation and control at the next time point. The present application enables the compressor regulation and control system to better cope with different working conditions and environments, and ensure the stability of indoor temperature.

Description

Technical Field

[0001] The present application relates to the field of electric scroll compressors, and in particular to an electric scroll compressor regulation and control system. Background Art

[0002] With the increasing global attention to energy efficiency and environmental protection, electric scroll compressors are widely used in air conditioners, refrigerators and other refrigeration systems as efficient and low-noise refrigeration equipment. Traditional compressor control systems mostly rely on PID control algorithms to adjust indoor temperature. PID control is a widely used feedback control method that generates a control signal to adjust the behavior of the system by proportionally amplifying, integrating and differentiating the error signal. Although the PID control method is mature, it mainly relies on preset control parameters. When faced with complex dynamic environments and different working conditions, it may not be able to respond quickly and accurately to temperature changes and load fluctuations, making it difficult to maintain a stable indoor temperature.

[0003] Therefore, an optimized electric scroll compressor adjustment control system is desired to solve the above technical problems. Summary of the invention

[0004] This application is made in view of the above problems. One object of this application is to provide an electric scroll compressor adjustment control system.

[0005] The embodiment of the present application provides an electric scroll compressor adjustment control system, which includes:

[0006] The temperature data acquisition module is used to obtain the time queue of the real-time ambient temperature value collected by the temperature sensor and set the target temperature value;

[0007] A flow data acquisition module, used to obtain a time queue of real-time flow values ​​of refrigerant collected by a flow agent;

[0008] The motor speed data acquisition module is used to obtain the time queue of the real-time speed value of the compressor motor collected by the speed monitor;

[0009] A temperature target offset calculation module, used for calculating the difference between each real-time ambient temperature value in the time queue of the real-time ambient temperature value and the set target temperature value to obtain a time queue of the temperature target offset;

[0010] A parameter processing module, for obtaining a global representation vector of a fine-grained decoupling and enhanced interaction of a temperature target offset and a target adjustment parameter based on a time queue of the temperature target offset, a time queue of the real-time flow value of the refrigerant, and a time queue of the real-time speed value of the compressor motor;

[0011] The refrigerant flow control module is used to determine the control instruction based on the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector, and the control instruction is used to indicate whether the refrigerant flow value at the next time point should increase, decrease or remain unchanged.

[0012] For example, the parameter processing module includes:

[0013] A parameter timing correlation feature extraction module is used to perform timing feature analysis on the time queue of the temperature target offset, the time queue of the refrigerant real-time flow value, and the time queue of the compressor motor real-time speed value to obtain a temperature target offset timing forward correlation feature vector, a refrigerant real-time flow timing forward correlation feature vector, and a compressor motor real-time speed forward timing correlation feature vector;

[0014] A refrigerant flow-motor speed timing feature collaborative attention module, used for inputting the refrigerant real-time flow timing forward correlation feature vector and the compressor motor real-time speed forward timing correlation feature vector into a fine-grained feature value correlation distinguishing attention fusion module to obtain a refrigerant flow-motor speed timing collaborative representation vector;

[0015] The temperature target offset-target adjustment parameter timing interaction module is used to input the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward correlation feature vector into a local fine-grained sequence interaction response aggregation network based on feature decoupling to obtain a temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector.

[0016] For example, according to the electric scroll compressor regulation control system of an embodiment of the present application, the parameter timing association feature extraction module is used to: input the time queue of the temperature target offset, the time queue of the refrigerant real-time flow value and the time queue of the compressor motor real-time speed value into a sequence encoder based on a forward Bi-LSTM model respectively to obtain the temperature target offset timing forward association feature vector, the refrigerant real-time flow timing forward association feature vector and the compressor motor real-time speed forward timing association feature vector.

[0017] For example, according to the electric scroll compressor regulation control system of an embodiment of the present application, the refrigerant flow-motor speed timing feature collaborative attention module includes: a global eigenvalue association matrix calculation unit, which is used to calculate the global eigenvalue association matrix between the refrigerant real-time flow timing forward association eigenvector and the compressor motor real-time speed forward timing association eigenvector to obtain the refrigerant flow-motor speed global eigenvalue association matrix;

[0018] A weight matrix calculation unit, used for inputting the refrigerant flow-motor speed global eigenvalue association matrix into a learnable gating function to obtain a refrigerant flow-motor speed correlation distinguishable weight matrix;

[0019] a mask calculation unit, for using the refrigerant flow-motor speed correlation distinguishable weight matrix as a fusion mask matrix, and respectively calculating the matrix products between the refrigerant real-time flow forward correlation feature vector and the compressor motor real-time speed forward correlation feature vector and the refrigerant flow-motor speed correlation distinguishable weight matrix to obtain a distinguished enhanced refrigerant real-time flow forward correlation feature vector and a distinguished enhanced compressor motor real-time speed forward correlation feature vector;

[0020] A vector fusion unit is used to fuse the forward correlation feature vector for distinguishing and enhancing the real-time flow rate of the refrigerant and the forward correlation feature vector for distinguishing and enhancing the real-time speed of the compressor motor to obtain the refrigerant flow-motor speed timing collaborative representation vector.

[0021] For example, according to the electric scroll compressor regulation control system of an embodiment of the present application, the weight matrix calculation unit is used to: use the negative number of each position eigenvalue in the refrigerant flow-motor speed global eigenvalue association matrix as the exponential power, calculate the natural exponential function value with the natural constant e as the base to obtain the refrigerant flow-motor speed global eigenvalue association class support matrix;

[0022] Calculate the reciprocal of the sum of the eigenvalues ​​of each position in the refrigerant flow-motor speed global eigenvalue association class support matrix and the constant 1 to obtain a refrigerant flow-motor speed global eigenvalue association weight matrix;

[0023] The refrigerant flow-motor speed correlation distinguishable weight matrix is ​​obtained by calculating the position point multiplication between the refrigerant flow-motor speed global eigenvalue association weight matrix and the refrigerant flow-motor speed global eigenvalue association matrix.

[0024] For example, according to the electric scroll compressor regulation control system of an embodiment of the present application, the temperature target offset-target adjustment parameter timing interaction module includes: a feature decoupling unit, which is used to perform feature decoupling on the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward correlation feature vector to obtain a sequence of refrigerant flow-motor speed timing collaborative local semantic representation vectors and a sequence of temperature target offset timing forward correlation local semantic representation vectors;

[0025] A local semantic association interaction unit is used to calculate the local semantic association interaction vector between each group of corresponding refrigerant flow-motor speed timing coordination local semantic representation vectors and the temperature target offset timing forward association local semantic representation vector in the sequence of the refrigerant flow-motor speed timing coordination local semantic representation vectors and the sequence of the temperature target offset timing forward association local semantic representation vectors to obtain a sequence of temperature target offset-target adjustment parameter local timing semantic association interaction vectors;

[0026] A fine-grained attention weight calculation unit, used to calculate the fine-grained attention weight of each temperature target offset-target adjustment parameter local temporal semantic association interaction vector in the sequence of the temperature target offset-target adjustment parameter local temporal semantic association interaction vector to obtain a sequence of temperature target offset-target adjustment parameter local temporal fine-grained attention weights;

[0027] A random deactivation unit, used for inputting the sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights into a random deactivation module to obtain a sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights after deactivation processing;

[0028] A weighted sum calculation unit is used to calculate the weighted sum of the sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights after the deactivation process as the weight distribution, so as to obtain the temperature target offset-target adjustment parameter temporal fine-grained decoupling reinforcement interaction global representation vector.

[0029] For example, according to the electric scroll compressor regulation control system of an embodiment of the present application, the local semantic association interaction unit is used to: perform position difference, position dot multiplication and position addition processing on each group of corresponding refrigerant flow-motor speed timing coordination local semantic representation vector and temperature target offset timing forward association local semantic representation vector to obtain a temperature target offset-target adjustment parameter local timing semantic difference representation vector, a temperature target offset-target adjustment parameter local timing semantic dot product representation vector and a temperature target offset-target adjustment parameter local timing semantic summation representation vector;

[0030] Cascading the temperature target offset-target adjustment parameter local temporal semantic difference representation vector, the temperature target offset-target adjustment parameter local temporal semantic dot product representation vector and the temperature target offset-target adjustment parameter local temporal semantic sum representation vector to obtain a temperature target offset-target adjustment parameter local temporal semantic association multidimensional representation fusion vector;

[0031] The temperature target offset-target adjustment parameter local temporal semantic association multidimensional representation fusion vector is one-dimensionally convolutionally encoded and then passed through a maximum pooling layer to obtain a temperature target offset-target adjustment parameter local temporal semantic association interaction vector.

[0032] For example, according to the electric scroll compressor regulation control system of an embodiment of the present application, the fine-grained attention weight calculation unit is used to: perform matrix multiplication on the temperature target offset-target regulation parameter local temporal semantic association interaction vector and the weight matrix, and then add the matrix to the bias vector by position to obtain a linear transformation temperature target offset-target regulation parameter local temporal semantic association interaction vector;

[0033] The linear transformation temperature target offset-target adjustment parameter local temporal semantic association interaction vector is activated using a tanh function and then vector-multiplied with a weight coefficient vector to obtain the temperature target offset-target adjustment parameter local temporal fine-grained attention weight.

[0034] For example, according to the electric scroll compressor regulation control system of the embodiment of the present application, the refrigerant flow control module is used to: input the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector into the classifier-based refrigerant flow control module to obtain a control instruction, and the control instruction is used to indicate that the refrigerant flow value at the next time point should increase, decrease or remain unchanged.

[0035] The present application has at least the following technical effects: compared with the prior art, the electric scroll compressor regulation and control system of the embodiment of the present application obtains the temperature target offset between the ambient temperature and the target value, as well as the real-time flow value of the refrigerant and the real-time speed value of the compressor motor through real-time monitoring, and introduces data processing and analysis algorithms based on artificial intelligence and deep learning at the back end to perform time-series forward feature extraction and collaborative correlation analysis on the temperature target offset, the real-time flow value of the refrigerant and the real-time speed value of the compressor motor, so as to capture the implicit connection and interactive influence between the temperature target offset characteristics and the target regulation parameter characteristics of the electric scroll compressor, thereby helping to utilize the relationship between the two to perform refrigerant flow value regulation and control at the next time point, enabling the compressor regulation and control system to better cope with different working conditions and environments and ensure the stability of indoor temperature. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings of the embodiments of the present application are briefly introduced below. Obviously, the drawings described below only relate to some embodiments of the present application, and are not intended to limit the present application.

[0037] Figure 1A schematic diagram of the application architecture of the electric scroll compressor adjustment control system in an embodiment of the present application is shown;

[0038] Figure 2 A schematic diagram of the structure of the electric scroll compressor adjustment control system in an embodiment of the present application is shown;

[0039] Figure 3 A schematic diagram of the structure of a parameter processing module of an electric scroll compressor adjustment control system in an embodiment of the present application is shown;

[0040] Figure 4 A schematic diagram of the structure of a refrigerant flow-motor speed timing characteristic collaborative attention module of an electric scroll compressor regulation and control system in an embodiment of the present application is shown;

[0041] Figure 5 A schematic diagram of the structure of a temperature target offset-target adjustment parameter timing interaction module of an electric scroll compressor adjustment control system in an embodiment of the present application is shown;

[0042] Figure 6 A flow chart of a method for regulating and controlling an electric scroll compressor in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0044] The terms used in this specification are those common terms currently widely used in the art in consideration of the functions of the present application, but these terms may vary according to the intentions of those skilled in the art, precedents, or new technologies in the art. In addition, specific terms may be selected, and in this case, their detailed meanings will be described in the detailed description of the present application. Therefore, the terms used in the specification should not be understood as simple names, but rather as a general description based on the meaning of the terms and the present application.

[0045] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0046] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0047] Figure 1 A schematic diagram of the application architecture of the electric scroll compressor regulation control system in an embodiment of the present application is shown, including a server 100 and a terminal device 200.

[0048] The terminal device 200 and the server 100 can be connected via the Internet to achieve mutual communication. Optionally, the above-mentioned Internet uses standard communication technology and / or protocol. The Internet is usually the Internet, but it can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network. In some embodiments, the data exchanged through the network can be represented by technologies and / or formats including Hyper Text Markup Language (HTML), Extensible Markup Language (XML), etc. In addition, conventional encryption technologies such as Secure Socket Layer (SSL), Transport Layer Security (TLS), Virtual Private Network (VPN), Internet Protocol Security (IPsec), etc. can be used to encrypt all or some links. In other embodiments, customized and / or dedicated data communication technologies can also be used to replace or supplement the above-mentioned data communication technologies.

[0049] The server 100 can provide various network services for the terminal device 200, wherein the server 100 can be a single server, a server cluster consisting of several servers, or a cloud computing center. Specifically, the server 100 may include a processor 110 (Center Processing Unit, CPU), a memory 120, an input device 130, and an output device 140, etc. The input device 130 may include a keyboard, a mouse, a touch screen, etc., and the output device 140 may include a display device, such as a liquid crystal display (Liquid Crystal Display, LCD), a cathode ray tube (Cathode Ray Tube, CRT), etc.

[0050] The memory 120 may include a read-only memory (ROM) and a random access memory (RAM), and provides the processor 110 with program instructions and data stored in the memory 120. In the embodiment of the present application, the memory 120 may be used to store the corresponding method of the electric scroll compressor adjustment control system in the embodiment of the present application.

[0051] The processor 110 calls the program instructions stored in the memory 120, and the processor 110 is used to execute the steps of any one of the electric scroll compressor adjustment control methods in the embodiments of the present application according to the obtained program instructions.

[0052] In addition, the application architecture diagram in the embodiment of the present application is intended to more clearly illustrate the technical solution in the embodiment of the present application, and does not constitute a limitation on the technical solution provided in the embodiment of the present application. Of course, for other application architectures and business applications, the technical solution provided in the embodiment of the present application is also applicable to similar problems.

[0053] The following is a non-restrictive description of the electric scroll compressor adjustment control system provided according to at least one embodiment of the present application through several examples or embodiments. As described below, different features in these specific examples or embodiments can be combined with each other without conflicting with each other to obtain new examples or embodiments, and these new examples or embodiments also fall within the scope of protection of the present application.

[0054] In practical applications, the stability of indoor temperature depends not only on the operating status of the compressor, but also on multiple factors such as external ambient temperature, refrigerant flow rate, and compressor motor speed. Existing technologies usually ignore the complex relationship between these variables, resulting in response lag or instability during the adjustment process, making it difficult to maintain the stability of indoor temperature.

[0055] Based on this, in the technical solution of the present application, an electric scroll compressor regulation and control system is proposed, which can adopt an advanced algorithm based on deep learning to establish and learn the timing coordination and correlation characteristics between the ambient temperature, the speed of the compressor motor and the refrigerant flow rate, so as to realize a more intelligent temperature control system and improve the performance of the electric scroll compressor regulation and control system.

[0056] Specifically, the technical concept of the present application is to obtain the temperature target offset between the ambient temperature and the target value, as well as the real-time refrigerant flow value and the real-time speed value of the compressor motor through real-time monitoring, and introduce data processing and analysis algorithms based on artificial intelligence and deep learning at the back end to perform time-series forward feature extraction and collaborative correlation analysis on the temperature target offset, the real-time refrigerant flow value and the real-time speed value of the compressor motor, so as to capture the implicit connection and interactive influence between the temperature target offset characteristics and the target adjustment parameter characteristics of the electric scroll compressor, thereby helping to utilize the relationship between the two to adjust and control the refrigerant flow value at the next time point, and can predictively and adaptively control the refrigerant flow according to the current ambient temperature offset and the compressor motor speed, thereby realizing more intelligent electric scroll compressor control and temperature control regulation, so as to improve the control accuracy and response speed of the compressor. In this way, the compressor adjustment and control system can better cope with different working conditions and environments to ensure the stability of indoor temperature.

[0057] Figure 2 The schematic diagram of the structure of the electric scroll compressor control system 800 in the embodiment of the present application is shown. The electric scroll compressor control system 800 includes: a temperature data acquisition module 810, which is used to obtain a time queue of real-time ambient temperature values ​​collected by a temperature sensor and a set target temperature value; a flow data acquisition module 820, which is used to obtain a time queue of real-time refrigerant flow values ​​collected by a flow agent; a motor speed data acquisition module 830, which is used to obtain a time queue of real-time compressor motor speed values ​​collected by a speed monitor; a temperature target offset calculation module 840, which is used to calculate the difference between each real-time ambient temperature value in the time queue of the real-time ambient temperature value and the set target temperature value to obtain a time queue of temperature target offset; a parameter processing module 850, which is used to obtain a global representation vector of fine-grained decoupling and enhanced interaction of temperature target offset-target adjustment parameters based on the time queue of the temperature target offset, the time queue of the real-time refrigerant flow value and the time queue of the real-time compressor motor speed value; specifically, as Figure 3As shown, the parameter processing module 850 includes a parameter timing correlation feature extraction module 851, which is used to perform timing feature analysis on the time queue of the temperature target offset, the time queue of the refrigerant real-time flow value, and the time queue of the compressor motor real-time speed value to obtain a temperature target offset timing forward correlation feature vector, a refrigerant real-time flow timing forward correlation feature vector, and a compressor motor real-time speed forward timing correlation feature vector; a refrigerant flow-motor speed timing feature collaborative attention module 852, which is used to input the refrigerant real-time flow timing forward correlation feature vector and the compressor motor real-time speed forward timing correlation feature vector into a fine-grained feature value correlation distinction attention fusion module Block to obtain the refrigerant flow-motor speed timing collaborative representation vector; temperature target offset-target adjustment parameter timing interaction module 853, used to input the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward correlation feature vector into the local fine-grained sequence interaction response aggregation network based on feature decoupling to obtain the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector; refrigerant flow control module 860, used to determine the control instruction based on the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector, the control instruction is used to indicate that the refrigerant flow value at the next time point should increase, decrease or remain unchanged.

[0058] More specifically, in the technical solution of the present application, first, a time queue of real-time ambient temperature values ​​collected by a temperature sensor and a set target temperature value are obtained, a time queue of real-time flow values ​​of a refrigerant collected by a flow meter is obtained, and a time queue of real-time speed values ​​of a compressor motor collected by a speed monitor is obtained. Then, the difference between each real-time ambient temperature value in the time queue of the real-time ambient temperature value and the set target temperature value is calculated to obtain a time queue of a temperature target offset.

[0059] Among them, the temperature data acquisition module 810 is responsible for real-time monitoring and recording of the ambient temperature, which is one of the basic inputs of the control system, because the working state of the compressor needs to be adjusted according to the ambient temperature. In one example, the temperature data acquisition module 810 can use a digital temperature sensor, such as DS18B20, which can provide high-precision temperature readings and communicate with the control system through a single bus protocol. The sensor periodically measures the ambient temperature and converts the data into digital signals and sends them to the data acquisition system. The data acquisition module 810 receives these signals, converts them into a time series of temperature values, and records the timestamp for subsequent analysis.

[0060] Among them, the flow data acquisition module 820 is responsible for monitoring the flow of the refrigerant. The flow data is crucial to the operating efficiency and cooling effect of the compressor. The amount of flow directly affects the cooling capacity. In one example, the flow data acquisition module 820 can use a flow meter, such as an ultrasonic flow meter or an electromagnetic flow meter, which can measure the flow rate based on the physical properties of the fluid. The flow meter can be installed on the refrigerant pipeline to monitor and measure the refrigerant flow passing through in real time. The flow data acquisition module 820 receives the output signal of the flow meter, converts it into a time series of digital flow values, and also records the timestamp.

[0061] Among them, the motor speed data acquisition module 830 is responsible for measuring and recording the speed of the compressor motor. The speed is a key indicator of the compressor performance. By monitoring the speed, the working state of the compressor can be understood and adjusted accordingly. Specifically, the motor speed data acquisition module 830 can use an encoder or a Hall effect sensor to monitor the speed of the motor. The encoder or sensor is installed on the shaft of the motor and can detect the rotation speed of the shaft. The data acquisition module 830 receives these signals, converts them into a time series of motor speed values, and records the corresponding time information.

[0062] Among them, the temperature target offset calculation module 840 is responsible for calculating the deviation between the real-time ambient temperature and the set target temperature. This deviation is the basis for the control system to make adjustment decisions. By comparing the actual temperature and the target temperature, the system can decide whether it is necessary to adjust the working state of the compressor to approach or reach the target temperature.

[0063] Then, the time queue of the temperature target offset, the time queue of the refrigerant real-time flow value, and the time queue of the compressor motor real-time speed value are respectively input into the sequence encoder based on the forward Bi-LSTM model to obtain the temperature target offset time series forward correlation feature vector, the refrigerant real-time flow time series forward correlation feature vector, and the compressor motor real-time speed forward time series correlation feature vector. It should be understood that the Bi-LSTM model can effectively capture the long-term dependencies in the time series. In addition, considering that each variable (temperature target offset, refrigerant flow, and motor speed) has its own time dynamic characteristics, therefore, by using the sequence encoder based on the forward Bi-LSTM model to process the time queue of the temperature target offset, the time queue of the refrigerant real-time flow value, and the time queue of the compressor motor real-time speed value, the forward time series correlation characteristics of the temperature target offset, the refrigerant real-time flow value, and the compressor motor real-time speed value in the time dimension can be captured respectively.

[0064] Correspondingly, the parameter timing association feature extraction module 851 is used to: input the time queue of the temperature target offset, the time queue of the refrigerant real-time flow value and the time queue of the compressor motor real-time speed value into the sequence encoder based on the forward Bi-LSTM model respectively to obtain the temperature target offset timing forward association feature vector, the refrigerant real-time flow timing forward association feature vector and the compressor motor real-time speed forward timing association feature vector.

[0065] Specifically, refrigerant flow and motor speed are two key variables that affect compressor performance. There is a complex relationship between them. This correlation relationship is the key to subsequent control tasks and achieving room temperature stability. Based on this, in the technical solution of the present application, the refrigerant real-time flow forward correlation feature vector and the compressor motor real-time speed forward correlation feature vector are further input into the fine-grained feature value correlation distinguishing attention fusion module to obtain the refrigerant flow-motor speed timing synergistic representation vector. The fine-grained feature value correlation distinguishing attention fusion module can learn and capture the timing synergistic effect between the refrigerant flow and the motor speed, so as to more comprehensively understand the dynamic behavior and performance of the compressor. At the same time, the fine-grained feature value correlation distinguishing attention fusion module can also use the temporal fine-grained association mask distinguishable attention mechanism to automatically identify and emphasize the most important features for the current refrigerant flow control and room temperature adjustment tasks. The model can adaptively focus on the key features and synergistic changes of the refrigerant flow or motor speed at different time points, ensuring that the most relevant information can be used more comprehensively and effectively when performing subsequent control tasks.

[0066] Accordingly, if Figure 4As shown, the refrigerant flow-motor speed timing feature collaborative attention module 852 includes: a global eigenvalue association matrix calculation unit 8521, which is used to calculate the global eigenvalue association matrix between the refrigerant real-time flow timing forward association eigenvector and the compressor motor real-time speed forward timing association eigenvector to obtain the refrigerant flow-motor speed global eigenvalue association matrix; a weight matrix calculation unit 8522, which is used to input the refrigerant flow-motor speed global eigenvalue association matrix into a learnable gating function to obtain a refrigerant flow-motor speed correlation distinguishable weight matrix; a mask calculation unit 8523, which is used to calculate the refrigerant flow-motor speed correlation with the global eigenvalue association matrix. The distinguishable weight matrix is ​​used as a fusion mask matrix to calculate the matrix product between the refrigerant real-time flow timing forward correlation feature vector and the compressor motor real-time speed forward timing correlation feature vector and the refrigerant flow-motor speed correlation distinguishable weight matrix to obtain the distinguished and enhanced refrigerant real-time flow timing forward correlation feature vector and the distinguished and enhanced compressor motor real-time speed forward timing correlation feature vector; the vector fusion unit 8524 is used to fuse the distinguished and enhanced refrigerant real-time flow timing forward correlation feature vector and the distinguished and enhanced compressor motor real-time speed forward timing correlation feature vector to obtain the refrigerant flow-motor speed timing collaborative representation vector.

[0067] Among them, the weight matrix calculation unit 8522 is used to: use the negative number of the eigenvalues ​​at each position in the refrigerant flow-motor speed global eigenvalue association matrix as the exponential power, calculate the natural exponential function value with the natural constant e as the base to obtain the refrigerant flow-motor speed global eigenvalue association class support matrix; calculate the reciprocal of the sum of the eigenvalues ​​at each position in the refrigerant flow-motor speed global eigenvalue association class support matrix and the constant 1 to obtain the refrigerant flow-motor speed global eigenvalue association weight matrix; calculate the position point multiplication between the refrigerant flow-motor speed global eigenvalue association weight matrix and the refrigerant flow-motor speed global eigenvalue association matrix to obtain the refrigerant flow-motor speed correlation distinguishable weight matrix.

[0068] In one example, the refrigerant flow-motor speed timing feature collaborative attention module 852 is used to: input the refrigerant real-time flow timing forward correlation feature vector and the compressor motor real-time speed forward timing correlation feature vector into the fine-grained feature value correlation distinguishing attention fusion module for processing using the following distinguishable attention fusion formula to obtain the refrigerant flow-motor speed timing collaborative representation vector; wherein the distinguishable attention fusion formula is:

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075] in, and are respectively the refrigerant real-time flow rate time series forward correlation feature vector and the compressor motor real-time speed forward time series correlation feature vector, is the transpose operation, is the global eigenvalue correlation matrix of refrigerant flow-motor speed, is the exponential operation with base e, is the refrigerant flow-motor speed global eigenvalue association weight matrix, is the refrigerant flow-motor speed correlation distinguishable weight matrix, and They are respectively distinguishing and enhancing the forward correlation feature vector of the time series of the real-time flow rate of the refrigerant and distinguishing and enhancing the forward correlation feature vector of the time series of the real-time speed of the compressor motor. is the cascade operation of vectors, is a vector representing the refrigerant flow-motor speed timing coordination.

[0076] Furthermore, since the refrigerant flow-motor speed timing synergistic representation vector and the temperature target offset timing forward correlation feature vector respectively represent the time correlation between the refrigerant flow and the motor speed, and the deviation change trend between the ambient temperature and the target temperature, the timing synergistic representation between the refrigerant flow and the motor speed can reflect the timing dynamic semantics of the target adjustment parameter, while the temperature target offset timing feature reflects the timing change semantics of the temperature offset from the target value in the time dimension, there is an implicit correlation relationship and interactive influence between the two. In addition, the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward association feature vector both contain a lot of redundant and information and semantics that are irrelevant to subsequent control tasks. In order to be able to assign low weights to these redundant information and noises and to highlight key features, thereby improving the comprehensiveness and effectiveness of the temperature target offset-target adjustment parameter timing feature expression, in the technical solution of the present application, the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward association feature vector are further input into the local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interactive global representation vector. Through the processing of the local fine-grained sequence interactive response aggregation network based on feature decoupling, the processing capability of the temperature target offset-target adjustment parameter timing detail features can be enhanced through fine-grained structural operations, and the ability to capture the dynamics of the local features of the two is enhanced, thereby facilitating the subsequent better interactive analysis of the temperature target offset timing features and the target adjustment parameter timing features.

[0077] Next, the interaction features between the fine-grained decoupled refrigerant flow-motor speed timing collaborative local semantic representation vector and the temperature target offset timing forward correlation local semantic representation vector are calculated, and the potential correlation between fine-grained features is captured through a one-dimensional convolutional neural network and maximum pooling, providing a quantitative perspective for understanding how features interact at the local level. Furthermore, the fine-grained attention weights between each local semantic representation vector and other semantic representation vectors are calculated to highlight key features during the global feature fusion process, so that the model can focus on the most important feature interactions, improve the quality of the fine-grained feature representation of the temperature target offset-target adjustment parameter timing and the prediction accuracy of the model.

[0078] Then, the sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights is input into the random dropout module to achieve weight sparsification and enhance the generalization ability of the model. The random dropout process achieves pruning optimization by randomly "closing" a part of the weights, reducing the risk of overfitting of the model while retaining the most important temperature target offset-target adjustment parameter temporal fine-grained feature interaction information.

[0079] Finally, the sequence of the fine-grained attention weights of the local temporal sequence of the temperature target offset-target adjustment parameters after the inactivation process is used as the weight distribution, and the sequence of the local temporal semantic association interaction vectors of the temperature target offset-target adjustment parameters is weighted summed. This step combines the local temporal semantic interaction information and attention weights of the temperature target offset-target adjustment parameters to generate a global representation vector of the fine-grained decoupling and enhanced interaction of the temperature target offset-target adjustment parameters, and enhances the model's understanding of the overall feature structure. In this way, by more accurately understanding the interaction between the temporal features of the temperature target offset-target adjustment parameters, the model can make more accurate control of the refrigerant flow value at the next time point and improve the generalization performance of the model.

[0080] Accordingly, if Figure 5 As shown, the temperature target offset-target adjustment parameter timing interaction module 853 includes: a feature decoupling unit 8531, which is used to perform feature decoupling on the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward association feature vector to obtain a sequence of refrigerant flow-motor speed timing collaborative local semantic representation vectors and a sequence of temperature target offset timing forward association local semantic representation vectors; a local semantic association interaction unit 8532, which is used to calculate the local semantic association interaction vector between each group of corresponding refrigerant flow-motor speed timing collaborative local semantic representation vectors and the temperature target offset timing forward association local semantic representation vector in the sequence of the refrigerant flow-motor speed timing collaborative local semantic representation vectors and the sequence of the temperature target offset timing forward association local semantic representation vectors to obtain a sequence of temperature target offset-target adjustment parameter local timing semantic association interaction vectors; a fine-grained attention weight calculation unit 8533 , used to calculate the fine-grained attention weights of each temperature target offset-target adjustment parameter local temporal semantic association interaction vector in the sequence of the temperature target offset-target adjustment parameter local temporal semantic association interaction vector to obtain a sequence of temperature target offset-target adjustment parameter local temporal fine-grained attention weights; a random deactivation unit 8534, used to input the sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights into the random deactivation module to obtain a sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights after the deactivation process; a weighted sum calculation unit 8535, used to use the sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights after the deactivation process as the weight distribution, and calculate the weighted sum of the sequence of the temperature target offset-target adjustment parameter local temporal semantic association interaction vector to obtain a global representation vector of the temperature target offset-target adjustment parameter temporal fine-grained decoupling reinforcement interaction.

[0081] Among them, the local semantic association interaction unit 8532 is used to: perform positional differentiation, positional dot multiplication and positional addition processing on each group of corresponding refrigerant flow-motor speed timing coordination local semantic representation vector and temperature target offset timing forward association local semantic representation vector to obtain temperature target offset-target adjustment parameter local timing semantic difference representation vector, temperature target offset-target adjustment parameter local timing semantic dot product representation vector and temperature target offset-target adjustment parameter local timing semantic addition representation vector; cascade the temperature target offset-target adjustment parameter local timing semantic difference representation vector, the temperature target offset-target adjustment parameter local timing semantic dot product representation vector and the temperature target offset-target adjustment parameter local timing semantic addition representation vector to obtain the temperature target offset-target adjustment parameter local timing semantic association multidimensional representation fusion vector; perform one-dimensional convolution encoding on the temperature target offset-target adjustment parameter local timing semantic association multidimensional representation fusion vector and pass it through the maximum pooling layer to obtain the temperature target offset-target adjustment parameter local timing semantic association interaction vector.

[0082] Among them, the fine-grained attention weight calculation unit 8533 is used to: matrix multiply the temperature target offset-target adjustment parameter local temporal semantic association interaction vector and the weight matrix, and then add them positionally with the bias vector to obtain the linear transformation temperature target offset-target adjustment parameter local temporal semantic association interaction vector; use the tanh function to activate the linear transformation temperature target offset-target adjustment parameter local temporal semantic association interaction vector, and then vector multiply it with the weight coefficient vector to obtain the temperature target offset-target adjustment parameter local temporal fine-grained attention weight.

[0083] In one example, the temperature target offset-target adjustment parameter timing interaction module 853 is used to: input the refrigerant flow-motor speed timing synergy representation vector and the temperature target offset timing forward correlation feature vector into the local fine-grained sequence interaction response aggregation network based on feature decoupling and process them with the following fine-grained interaction response aggregation formula to obtain the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector; wherein the fine-grained interaction response aggregation formula is:

[0084]

[0085]

[0086]

[0087]

[0088]

[0089]

[0090] in, is the refrigerant flow-motor speed timing synergy representation vector, is the temperature target offset timing forward correlation feature vector, For feature decoupling operations, They are the first and second in the sequence of the refrigerant flow-motor speed timing coordinated local semantic representation vector , No. indivual, , No. A refrigerant flow-motor speed timing coordinated local semantic representation vector, are the first and second in the sequence of the local semantic representation vector of the temperature target offset timing forward association , No. indivual, , No. The temperature target offset timing is forward associated with the local semantic representation vector, For cascade operation, is a one-dimensional convolution operation, For positional subtraction, For point-by-position multiplication, To add by position, For maximum pooling processing, is the temperature target offset-target adjustment parameter local temporal semantic association interaction vector, and are the weight matrix and bias vector respectively, for function, is the weight coefficient vector, The local temporal fine-grained attention weights for the temperature target offset-target adjustment parameters, For random dropout, is the temperature target offset after inactivation processing-target adjustment parameter local temporal fine-grained attention weight, A global representation vector is provided for the fine-grained decoupling of the temperature target offset-target adjustment parameter timing to strengthen the interaction.

[0091] Specifically, the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector is input into the refrigerant flow control module based on the classifier to obtain a control instruction, and the control instruction is used to indicate that the refrigerant flow value at the next time point should increase, decrease or remain unchanged. That is, the temporal fine-grained decoupling enhanced interaction characteristics between the temperature target offset and the target adjustment parameter are used for classification processing, so as to adjust the refrigerant flow value at the next time point. In this way, the refrigerant flow can be predictively and adaptively controlled according to the current ambient temperature offset and the compressor motor speed, thereby realizing more intelligent electric scroll compressor control and temperature control regulation, so as to improve the control accuracy and response speed of the compressor. Specifically, in a specific example of the present application, the refrigerant flow can be adjusted by a solenoid valve. In this way, the compressor adjustment control system can better cope with different working conditions and environments to ensure the stability of the indoor temperature.

[0092] Correspondingly, the refrigerant flow control module 860 is used to: input the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector into the classifier-based refrigerant flow control module to obtain a control instruction, and the control instruction is used to indicate that the refrigerant flow value at the next time point should increase, decrease or remain unchanged.

[0093] Specifically, the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression, SVM, etc. are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is inefficient. Commonly used multi-classification methods include the Softmax classification function.

[0094] In the technical solution of the present application, the refrigerant flow-motor speed timing synergy representation vector and the temperature target offset timing forward correlation feature vector respectively represent the timing fine-grained correlation features between the refrigerant flow and the motor speed and the timing correlation features of the temperature target offset value. Among them, the refrigerant flow and the motor speed are internal causes and the temperature regulation is an external result. Therefore, the refrigerant flow-motor speed timing synergy representation vector and the temperature target offset timing forward correlation feature vector are further input into a local fine-grained sequence interactive response aggregation network based on feature decoupling, so as to capture the timing nonlinear implicit association between the internal cause and the external result through the local fine-grained sequence interactive response aggregation network based on feature decoupling. However, considering that there is a characteristic mode and characteristic fine-grained correspondence offset between the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward correlation feature vector, this will cause the obtained temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global representation vector to have a distribution regression mode offset relative to the classifier, that is, it will cause the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global representation vector input to the control instruction obtained by the refrigerant flow control module based on the classifier to overflow, affecting its accuracy.

[0095] Based on this, the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector is input into the classifier-based refrigerant flow control module to obtain the control instruction, which specifically includes:

[0096] Calculate the sum of the absolute values ​​of each eigenvalue of the global representation vector of the temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction to obtain a first global spatial structure value of the temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction, and calculate the square root of the sum of the squares of each eigenvalue of the global representation vector of the temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction to obtain a second global spatial structure value of the temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction.

[0097] Each eigenvalue of the global representation vector of the temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction is multiplied with the first temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global spatial structure value and the second temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global spatial structure value to obtain the first temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global structure reference value and the second temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global structure reference value corresponding to each eigenvalue.

[0098] Each eigenvalue of the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global representation vector is multiplied by the length of the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global representation vector and the square root of the length to obtain a first temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global scale transformation value and a second temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global scale transformation value corresponding to each eigenvalue.

[0099] The first temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global structure reference value is divided by the difference between the first temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global spatial structure value and the first temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global scale transformation value to obtain the first temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global transformation adjustment value.

[0100] The second temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global structure reference value is divided by the difference between the second temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global spatial structure value and the second temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global scale transformation value to obtain the second temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global transformation adjustment value.

[0101] Calculate the weighted sum of the first temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global transformation adjustment value and the second temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global transformation adjustment value to obtain each eigenvalue of the optimized temperature target offset - target adjustment parameter timing fine-grained decoupling and enhanced interaction global representation vector.

[0102] The optimized temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector is input into a classifier-based refrigerant flow control module to obtain a control instruction.

[0103] Among them, the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector is denoted as The optimization expression is:

[0104]

[0105]

[0106]

[0107]

[0108]

[0109]

[0110]

[0111]

[0112] in, A global representation vector for the fine-grained decoupling and enhanced interaction of the temperature target offset-target adjustment parameter timing, represents the set of real numbers, The first global representation vector of the temperature target offset-target adjustment parameter temporal fine-grained decoupling and enhanced interaction The eigenvalues ​​at the positions, represents the scale of the vector, It represents the first temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global spatial structure value, It represents the second temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global space structure value, It means point multiplication by position. It means adding by position. represents the first temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global transformation adjustment vector, It represents the first temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global transformation adjustment value, It represents the second temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global transformation adjustment vector, It represents the second temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global transformation adjustment value, represents the weighted hyperparameter, Represents the optimized temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction global representation vector.

[0113] Therefore, in the preferred example, with respect to the spatial structure information of the feature set of the global representation vector of the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction in the high-dimensional space, a scale-based box transformation of each eigenvalue of the global representation vector of the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction is performed by taking the class-norm spatial structured representation of the global representation vector of the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction as a reference window, and a spatial structure-based box attention weight adjustment of each eigenvalue of the global representation vector of the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction is implemented to ensure that the global representation vector of the temperature target offset-target adjustment parameter timing fine-grained decoupling and enhanced interaction is transformed in the feature space. The invariance of spatial transformations (translation, scaling and rotation) under each other is improved, thereby improving the convergence and generalization effect of the feature set of the temperature target offset-target adjustment parameter temporal fine-grained decoupling and enhanced interactive global representation vector for classification regression under complex spatial structure representation, thereby avoiding the invalid overflow of the classification result caused by the deviation of the distribution pattern of the temperature target offset-target adjustment parameter temporal fine-grained decoupling and enhanced interactive global representation vector, so as to improve the accuracy of the control instructions obtained by the refrigerant flow control module based on the classifier, and can more accurately predict the refrigerant flow according to the current ambient temperature offset and the compressor motor speed, thereby realizing more intelligent electric scroll compressor control and temperature control adjustment, so as to improve the control accuracy and response speed of the compressor and ensure the stability of the indoor temperature.

[0114] Based on the above embodiments, see Figure 6 As shown in FIG. 1 , a flow chart of a method for adjusting and controlling an electric scroll compressor in an embodiment of the present application is shown. For example, the method for adjusting and controlling an electric scroll compressor can be executed by a server, which can be Figure 1 The server 100 shown in FIG. Figure 6As shown, according to the electric scroll compressor regulation control method of the embodiment of the present application, the steps include: S510, obtaining a time queue of real-time ambient temperature values ​​collected by a temperature sensor and a set target temperature value; S520, obtaining a time queue of real-time refrigerant flow values ​​collected by a flow meter; S530, obtaining a time queue of real-time speed values ​​of the compressor motor collected by a speed monitor; S540, calculating the difference between each real-time ambient temperature value in the time queue of the real-time ambient temperature value and the set target temperature value to obtain a time queue of temperature target offset; S550, performing timing feature analysis on the time queue of the temperature target offset, the time queue of the real-time refrigerant flow value, and the time queue of the real-time speed value of the compressor motor, respectively, to obtain a temperature target offset timing forward correlation feature vector, a refrigerant real-time flow timing forward correlation feature vector, and a pressure The forward timing association feature vector of the real-time speed of the compressor motor; S560, the refrigerant real-time flow timing forward association feature vector and the compressor motor real-time speed forward timing association feature vector are input into the fine-grained feature value correlation distinguishing attention fusion module to obtain the refrigerant flow-motor speed timing collaborative representation vector; S570, the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward association feature vector are input into the local fine-grained sequence interactive response aggregation network based on feature decoupling to obtain the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector; S580, based on the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector, a control instruction is determined, and the control instruction is used to indicate that the refrigerant flow value at the next time point should increase, decrease or remain unchanged.

[0115] Specifically, the specific operations of each step in the above electric scroll compressor adjustment control method have been described in the above reference. Figures 2 to 6 The electric scroll compressor adjustment control system 800 has been described in detail, and therefore, its repeated description will be omitted.

[0116] Based on the above embodiments, another exemplary embodiment of an electronic device is also provided in the embodiments of the present application. In some possible implementations, the electronic device in the embodiments of the present application may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor can implement the steps of the electric scroll compressor adjustment control method in the above embodiments when executing the program.

[0117] For example, in the case of electronic equipment Figure 1 Taking the server 100 in the example as an example, the processor in the electronic device is the processor 110 in the server 100, and the memory in the electronic device is the memory 120 in the server 100.

[0118] The embodiment of the present application also provides a computer-readable storage medium, on which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, the electric scroll compressor adjustment control method according to the embodiment of the present application described with reference to the above figures can be executed. The computer-readable storage medium includes, but is not limited to, for example, volatile memory and / or non-volatile memory. The volatile memory may, for example, include a random access memory (RAM) and / or a cache memory (cache), etc. The non-volatile memory may, for example, include a read-only memory (ROM), a hard disk, a flash memory, etc.

[0119] The embodiment of the present application also provides a computer program product or a computer program, which includes computer executable instructions, and the computer executable instructions are stored in a computer readable storage medium. The processor of the computer device reads the computer executable instructions from the computer readable storage medium, and the processor executes the computer executable instructions, so that the computer device performs the electric scroll compressor adjustment control method according to the embodiment of the present application.

[0120] Those skilled in the art will appreciate that the contents disclosed in this application may be subject to various variations and improvements. For example, the various devices or components described above may be implemented by hardware, or by software, firmware, or a combination of some or all of the three.

[0121] In addition, although the present application makes various references to certain units in the system according to embodiments of the present application, any number of different units can be used and run on the client and / or server. The units are only illustrative, and different aspects of the system and method can use different units.

[0122] Those skilled in the art will appreciate that all or part of the steps in the above method can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module / unit in the above embodiment can be implemented in the form of hardware or in the form of software function modules. The present application is not limited to any particular form of combination of hardware and software.

[0123] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0124] The above is an explanation of the present application and should not be considered as limiting thereof. Although several exemplary embodiments of the present application have been described, those skilled in the art will readily appreciate that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application.

Claims

1. An electric scroll compressor adjustment control system, characterized in that: include: The temperature data acquisition module is used to obtain the time queue of the real-time ambient temperature value collected by the temperature sensor and set the target temperature value; A flow data acquisition module, used to obtain a time queue of real-time flow values ​​of refrigerant collected by a flow agent; The motor speed data acquisition module is used to obtain the time queue of the real-time speed value of the compressor motor collected by the speed monitor; A temperature target offset calculation module, used for calculating the difference between each real-time ambient temperature value in the time queue of the real-time ambient temperature value and the set target temperature value to obtain a time queue of the temperature target offset; A parameter processing module, for obtaining a global representation vector of a fine-grained decoupling and enhanced interaction of a temperature target offset and a target adjustment parameter based on a time queue of the temperature target offset, a time queue of the real-time flow value of the refrigerant, and a time queue of the real-time speed value of the compressor motor; A refrigerant flow control module, used to determine a control instruction based on the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector, wherein the control instruction is used to indicate whether the refrigerant flow value at the next time point should increase, decrease, or remain unchanged; Wherein, the parameter processing module includes: The parameter time series correlation feature extraction module is used to analyze the time series characteristics of the temperature target offset, the time series of the refrigerant real-time flow value and the time series of the compressor motor real-time speed value to obtain the temperature target offset time series forward correlation feature vector , refrigerant real-time flow time series forward correlation feature vector and the forward time series correlation feature vector of the real-time speed of the compressor motor ; The refrigerant flow-motor speed time series feature collaborative attention module is used to forward associate the refrigerant real-time flow time series feature vector and the forward time series correlation feature vector of the compressor motor real-time speed Input fine-grained feature value correlation distinguishing attention fusion module to obtain the refrigerant flow-motor speed timing synergistic representation vector ; Specifically, the refrigerant real-time flow rate time series forward correlation feature vector and the compressor motor real-time speed forward correlation feature vector are input into the fine-grained feature value correlation distinguishing attention fusion module for processing using the following distinguishable attention fusion formula to obtain the refrigerant flow rate-motor speed time series collaborative representation vector; wherein the distinguishable attention fusion formula is: in, and are respectively the refrigerant real-time flow rate time series forward correlation feature vector and the compressor motor real-time speed forward time series correlation feature vector, is the transpose operation, is the global eigenvalue correlation matrix of refrigerant flow-motor speed, is the exponential operation with base e, is the refrigerant flow-motor speed global eigenvalue association weight matrix, is the refrigerant flow-motor speed correlation distinguishable weight matrix, and They are respectively distinguishing and enhancing the forward correlation feature vector of the time series of the real-time flow rate of the refrigerant and distinguishing and enhancing the forward correlation feature vector of the time series of the real-time speed of the compressor motor. is the cascade operation of vectors, A vector representing the refrigerant flow rate-motor speed timing coordination; The temperature target offset-target adjustment parameter timing interaction module is used to represent the refrigerant flow-motor speed timing coordination vector and the temperature target offset timing forward correlation feature vector Input the local fine-grained sequence interaction response aggregation network based on feature decoupling to obtain the global representation vector of temperature target offset-target adjustment parameter temporal fine-grained decoupling reinforcement interaction; Specifically, the refrigerant flow-motor speed timing synergy representation vector and the temperature target offset timing forward correlation feature vector are input into the local fine-grained sequence interaction response aggregation network based on feature decoupling and processed with the following fine-grained interaction response aggregation formula to obtain the temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector; wherein the fine-grained interaction response aggregation formula is: in, is the refrigerant flow-motor speed timing synergy representation vector, is the temperature target offset timing forward correlation feature vector, For feature decoupling operations, They are the first, second, ..., and third vectors in the sequence of the refrigerant flow-motor speed timing coordinated local semantic representation vector. , ..., A refrigerant flow-motor speed timing coordinated local semantic representation vector, are the first, second, ..., and third vectors in the sequence of the temperature target offset timing forward association local semantic representation vectors. , ..., The temperature target offset timing is forward associated with the local semantic representation vector, For cascade operation, is a one-dimensional convolution operation, For positional subtraction, For point-by-position multiplication, To add by position, For maximum pooling processing, is the temperature target offset-target adjustment parameter local temporal semantic association interaction vector, and are the weight matrix and bias vector respectively, for function, is the weight coefficient vector, The local temporal fine-grained attention weights for the temperature target offset-target adjustment parameters, For random dropout, is the temperature target offset after inactivation processing-target adjustment parameter local temporal fine-grained attention weight, A global representation vector for the fine-grained decoupling and enhanced interaction of the temperature target offset-target adjustment parameter timing; Wherein, the parameter time series correlation feature extraction module is used to: The time queue of the temperature target offset, the time queue of the real-time refrigerant flow value, and the time queue of the real-time speed value of the compressor motor are respectively input into the sequence encoder based on the forward Bi-LSTM model to obtain the temperature target offset timing forward correlation feature vector, the refrigerant real-time flow timing forward correlation feature vector, and the compressor motor real-time speed forward timing correlation feature vector.

2. The electric scroll compressor adjustment control system according to claim 1, characterized in that: The refrigerant flow-motor speed timing characteristics collaborative attention module includes: A global eigenvalue association matrix calculation unit, used for calculating the global eigenvalue association matrix between the real-time refrigerant flow rate time series forward association eigenvector and the compressor motor real-time speed forward association eigenvector to obtain a refrigerant flow rate-motor speed global eigenvalue association matrix; A weight matrix calculation unit, used for inputting the refrigerant flow-motor speed global eigenvalue association matrix into a learnable gating function to obtain a refrigerant flow-motor speed correlation distinguishable weight matrix; a mask calculation unit, for using the refrigerant flow-motor speed correlation distinguishable weight matrix as a fusion mask matrix, and respectively calculating the matrix products between the refrigerant real-time flow forward correlation feature vector and the compressor motor real-time speed forward correlation feature vector and the refrigerant flow-motor speed correlation distinguishable weight matrix to obtain a distinguished enhanced refrigerant real-time flow forward correlation feature vector and a distinguished enhanced compressor motor real-time speed forward correlation feature vector; A vector fusion unit, used for fusing the time series forward correlation feature vector for distinguishing and enhancing the real-time flow rate of the refrigerant and the time series forward correlation feature vector for distinguishing and enhancing the real-time speed of the compressor motor to obtain the time series synergistic representation vector of the refrigerant flow rate and the motor speed; Wherein, the weight matrix calculation unit is used to: Taking the negative number of each position eigenvalue in the refrigerant flow-motor speed global eigenvalue association matrix as the exponential power, calculating the natural exponential function value with the natural constant e as the base to obtain the refrigerant flow-motor speed global eigenvalue association class support matrix; Calculate the reciprocal of the sum of the eigenvalues ​​of each position in the refrigerant flow-motor speed global eigenvalue association class support matrix and the constant 1 to obtain a refrigerant flow-motor speed global eigenvalue association weight matrix; The refrigerant flow-motor speed correlation distinguishable weight matrix is ​​obtained by calculating the position point multiplication between the refrigerant flow-motor speed global eigenvalue association weight matrix and the refrigerant flow-motor speed global eigenvalue association matrix.

3. The electric scroll compressor adjustment control system according to claim 2, characterized in that: The temperature target offset-target adjustment parameter timing interaction module includes: A feature decoupling unit, used for feature decoupling the refrigerant flow-motor speed timing collaborative representation vector and the temperature target offset timing forward correlation feature vector to obtain a sequence of refrigerant flow-motor speed timing collaborative local semantic representation vectors and a sequence of temperature target offset timing forward correlation local semantic representation vectors; A local semantic association interaction unit is used to calculate the local semantic association interaction vector between each group of corresponding refrigerant flow-motor speed timing coordination local semantic representation vectors and the temperature target offset timing forward association local semantic representation vector in the sequence of the refrigerant flow-motor speed timing coordination local semantic representation vectors and the sequence of the temperature target offset timing forward association local semantic representation vectors to obtain a sequence of temperature target offset-target adjustment parameter local timing semantic association interaction vectors; A fine-grained attention weight calculation unit, used to calculate the fine-grained attention weight of each temperature target offset-target adjustment parameter local temporal semantic association interaction vector in the sequence of the temperature target offset-target adjustment parameter local temporal semantic association interaction vector to obtain a sequence of temperature target offset-target adjustment parameter local temporal fine-grained attention weights; A random deactivation unit, used for inputting the sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights into a random deactivation module to obtain a sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights after deactivation processing; A weighted sum calculation unit, used to calculate the weighted sum of the sequence of the temperature target offset-target adjustment parameter local temporal fine-grained attention weights after the deactivation process as the weight distribution, to obtain the temperature target offset-target adjustment parameter temporal fine-grained decoupling reinforcement interaction global representation vector; Wherein, the local semantic association interaction unit is used to: Perform positional difference, positional dot multiplication and positional addition processing on the corresponding refrigerant flow-motor speed timing coordination local semantic representation vector and the temperature target offset timing forward association local semantic representation vector of each group to obtain the temperature target offset-target adjustment parameter local timing semantic difference representation vector, the temperature target offset-target adjustment parameter local timing semantic dot product representation vector and the temperature target offset-target adjustment parameter local timing semantic addition representation vector; Cascading the temperature target offset-target adjustment parameter local temporal semantic difference representation vector, the temperature target offset-target adjustment parameter local temporal semantic dot product representation vector and the temperature target offset-target adjustment parameter local temporal semantic sum representation vector to obtain a temperature target offset-target adjustment parameter local temporal semantic association multidimensional representation fusion vector; The temperature target offset-target adjustment parameter local temporal semantic association multidimensional representation fusion vector is subjected to one-dimensional convolution encoding and then passed through a maximum pooling layer to obtain a temperature target offset-target adjustment parameter local temporal semantic association interaction vector; Wherein, the fine-grained attention weight calculation unit is used to: The temperature target offset-target adjustment parameter local temporal semantic association interaction vector and the weight matrix are matrix-multiplied and then added to the bias vector by position to obtain a linearly transformed temperature target offset-target adjustment parameter local temporal semantic association interaction vector; The linear transformation temperature target offset-target adjustment parameter local temporal semantic association interaction vector is activated using a tanh function and then vector-multiplied with a weight coefficient vector to obtain the temperature target offset-target adjustment parameter local temporal fine-grained attention weight.

4. The electric scroll compressor adjustment control system according to claim 3, characterized in that: The refrigerant flow control module is used to: The temperature target offset-target adjustment parameter timing fine-grained decoupling enhanced interaction global representation vector is input into a classifier-based refrigerant flow control module to obtain a control instruction, wherein the control instruction is used to indicate whether the refrigerant flow value at the next time point should increase, decrease or remain unchanged.

Images