Convolutional network-based ship fuel system process parameter control method and system
Through the method based on convolutional neural network, the control valve opening control model of the ship fuel system is constructed, which solves the problem of insufficient control capabilities of the traditional PID adjustment method, realizes the overall adjustment of multiple media process parameters, and improves the fuel system Stability and energy efficiency reduce pollution emissions.
Patent Information
- Application Number
- CN202410979510.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-12
AI Technical Summary
The traditional medium parameter control method of marine fuel system is based on PID feedback adjustment, which requires engineers to set parameters and cannot adjust two or more medium process parameters at the same time, resulting in limited control capabilities.
Using a method based on convolutional neural network, the initial control valve opening control model is constructed through data acquisition, numerical processing and model training, and the feature information is extracted through the convolutional layer and L2 regularization is performed to obtain the final control valve opening control model. , realize the overall adjustment of multiple media process parameters.
It realizes stable control of the fuel system, reduces fuel consumption, improves energy efficiency, and reduces air pollution and greenhouse gas emissions.
Smart Images

Figure CN118938659A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of intelligent valve control technology, and in particular, to a method and system for controlling process parameters of a ship fuel system based on a convolutional network and applicable to the ship fuel system. Background Art
[0002] Climate change is one of the most pressing environmental issues in the world today. A large amount of scientific evidence shows that greenhouse gas emissions caused by human activities have caused global temperatures to rise, leading to frequent extreme weather events and rising sea levels. Overexploitation and resource consumption have a negative impact on the ecological balance of the earth.
[0003] By precisely controlling the process parameters of the medium in the ship's fuel system, fuel consumption can be reduced, energy efficiency can be improved, and air pollution and greenhouse gas emissions from ships can be reduced.
[0004] However, the traditional control method of medium parameters of ship fuel system is usually based on PID feedback regulation, and PID feedback regulation has two disadvantages. First, it requires the experience of engineers to set the PID parameters. Second, PID feedback regulation is single, and a PID controller cannot adjust two or more medium process parameters at the same time. Therefore, the control ability of medium process parameters of ship fuel system in traditional control methods is limited, and it is impossible to control from the perspective of the system as a whole.
[0005] Therefore, in order to solve the above-mentioned defects, it is urgent to develop a method and system for controlling process parameters of a ship fuel system based on a convolutional network that overcomes the above-mentioned defects. Summary of the invention
[0006] In view of the above problems, the present invention provides a method for controlling process parameters of a ship fuel system based on a convolutional network, which includes:
[0007] Target process parameter acquisition step: collecting and obtaining target process parameters;
[0008] Model acquisition step: constructing an initial regulating valve opening control model based on a convolutional neural network, and training the initial regulating valve opening control model to obtain a final regulating valve opening control model;
[0009] The regulating valve opening control step: after obtaining the opening value through the final regulating valve opening control model according to the target process parameters, the opening of the regulating valve is adjusted according to the opening value so that the fuel system reaches the target process parameters.
[0010] In the above-mentioned ship fuel system process parameter control method, the model acquisition step includes:
[0011] Data collection step: collecting the real-time medium process parameters and the real-time opening value of each control point of the regulating valve at different openings;
[0012] Data processing step: digitizing the real-time medium process parameters and the real-time opening value and storing them in vector form;
[0013] Model building step: building the initial regulating valve opening control model based on convolutional neural network;
[0014] Model training step: based on the real-time medium process parameters and the real-time opening value after digitization, the initial regulating valve opening control model is trained to obtain the final regulating valve opening control model.
[0015] In the above-mentioned method for controlling process parameters of a ship fuel system, the data processing step comprises:
[0016] The real-time medium process parameters and the real-time opening value are stored in an Excel table in a vector form;
[0017] The data in the Excel table is read and stored in a tensor of PyTorch, and the tensor is used as a data set to be input into the initial regulating valve opening control model.
[0018] In the above-mentioned ship fuel system process parameter control method, the model training step includes:
[0019] Extracting and fusing feature information between different real-time medium process parameters from the data set through a convolutional layer;
[0020] Training the initial regulating valve opening control model through the characteristic information;
[0021] Attenuating the weight of the initial regulating valve opening control model by L2 regularization method;
[0022] The output result of the initial regulating valve opening control model is compared with the real-time opening value, and the weight parameter of the initial regulating valve opening control model is updated by back propagation to obtain the final regulating valve opening control model.
[0023] In the above-mentioned ship fuel system process parameter control method, in the data collection step, the real-time medium process parameters and the real-time opening value are collected by multiple sensors placed in the fuel system.
[0024] The present invention also provides a ship fuel system process parameter control system based on a convolutional network, which includes:
[0025] A target process parameter acquisition unit collects and obtains target process parameters;
[0026] A model acquisition unit, constructing an initial regulating valve opening control model based on a convolutional neural network, and training the initial regulating valve opening control model to obtain a final regulating valve opening control model;
[0027] The regulating valve opening control unit obtains an opening value through the final regulating valve opening control model according to the target process parameter, and then adjusts the opening of the regulating valve according to the opening value so that the fuel system reaches the target process parameter.
[0028] In the above-mentioned ship fuel system process parameter control system, the model acquisition unit comprises:
[0029] A data acquisition module, collecting the real-time medium process parameters and the real-time opening value of each control point of the regulating valve at different openings;
[0030] A data processing module, which performs numerical processing on the real-time medium process parameters and the real-time opening value and stores them in a vector form;
[0031] A model building module, constructing the initial regulating valve opening control model based on a convolutional neural network;
[0032] The model training module trains the initial regulating valve opening control model based on the real-time medium process parameters and the real-time opening value after digitization to obtain the final regulating valve opening control model.
[0033] The above-mentioned ship fuel system process parameter control system, wherein the data processing module stores the real-time medium process parameters and the real-time opening value in an Excel table in the form of a vector, reads the data in the Excel table, and stores it in a pytorch tensor, and uses the tensor as a data set to input into the initial regulating valve opening control model.
[0034] The above-mentioned ship fuel system process parameter control system, wherein the model training module extracts and fuses the characteristic information between different real-time medium process parameters from the data set through a convolution layer; trains the initial control valve opening control model through the characteristic information; attenuates the weight of the initial control valve opening control model through the L2 regularization method; compares the output result of the initial control valve opening control model with the real-time opening value, and updates the weight parameters of the initial control valve opening control model through back propagation to obtain the final control valve opening control model.
[0035] In the above-mentioned ship fuel system process parameter control system, the data acquisition module includes a plurality of sensors placed in the fuel system, and the real-time medium process parameters and the real-time opening value are collected by the sensors.
[0036] Compared with the prior art, the present invention has the following benefits: with respect to ship fuel supply, the present invention proposes a ship fuel system process parameter control method and system based on a convolutional network, obtains the overall process parameters of the ship fuel system in accordance with scientific principles, controls a single actuator through an algorithm to adjust the medium process parameters of two or more fuel systems, thereby controlling the regulating valves in the fuel system over a wide range, taking into account the mutual influence between the medium process parameters and multiple regulating valves, making the overall system control more stable.
[0037] Other features and advantages of the present invention will be described in the following description, and partly become obvious from the description, or be understood by implementing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description and the drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0039] Figure 1 It is a flow chart of the method for controlling process parameters of a ship fuel system of the present invention;
[0040] Figure 2 for Figure 1 The step-by-step flow chart of step S2;
[0041] Figure 3 It is a structural schematic diagram of a process parameter control system of a ship fuel system of the present invention;
[0042] Figure 4 Schematic diagram of model training. DETAILED DESCRIPTION
[0043] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0044] The exemplary embodiments and descriptions of the present invention are used to explain the present invention, but are not intended to limit the present invention. In addition, elements / components with the same or similar reference numerals used in the drawings and embodiments are used to represent the same or similar parts.
[0045] The terms “first”, “second”, “S1”, “S2”, etc. used in this document do not specifically refer to an order or sequence, nor are they used to limit the present invention. They are only used to distinguish elements or operations described with the same technical terms.
[0046] The directional terms used in this document, such as up, down, left, right, front or back, etc., are only used to refer to the directions of the drawings. Therefore, the directional terms used are used to illustrate and not to limit the present invention.
[0047] The words “include,” “including,” “have,” “contain,” etc. used in this document are open-ended terms, meaning including but not limited to.
[0048] As used herein, "and / or" includes any or all combinations of the items described.
[0049] As used herein, “plurality” includes “two” and “more than two”; as used herein, “plurality groups” includes “two groups” and “more than two groups”.
[0050] Certain terms used to describe the present application are discussed below or elsewhere in this specification to provide additional guidance to those skilled in the art regarding the description of the present application.
[0051] See also Figure 1 , Figure 1 FIG. 1 is a flow chart of the method for controlling process parameters of a ship fuel system according to the present invention. Figure 1 As shown, a method for controlling process parameters of a ship fuel system based on a convolutional network of the present invention comprises:
[0052] Target process parameter acquisition step S1: collecting and obtaining target process parameters;
[0053] Model acquisition step S2: constructing an initial regulating valve opening control model based on a convolutional neural network, and training the initial regulating valve opening control model to obtain a final regulating valve opening control model;
[0054] Regulating valve opening control step S3: after obtaining the opening value through the final regulating valve opening control model according to the target process parameter, the opening of the regulating valve is adjusted according to the opening value so that the fuel system reaches the target process parameter.
[0055] The present invention is based on a convolutional neural network control model for regulating valve opening. By inputting target process parameters into the model, wherein the target process parameters include at least one of temperature, pressure and flow, an opening value for controlling the opening of the regulating valve is obtained, and the opening of the regulating valve is controlled by the opening value to achieve the expected temperature, pressure and flow, thereby reducing fuel consumption, improving the stability of medium process parameters of the ship fuel system, improving energy efficiency, and reducing air pollution and greenhouse gas emissions from ships.
[0056] Please refer to Figure 2 , Figure 2 for Figure 1 The step-by-step flow chart of step S2 in FIG. Figure 2 As shown, the model acquisition step S2 includes:
[0057] Data collection step S21: collecting the real-time medium process parameters and the real-time opening value of each control point of the regulating valve at different openings, wherein in the data collection step S21, the real-time medium process parameters and the real-time opening value are collected by multiple sensors placed in the fuel system;
[0058] Data processing step S22: digitizing the real-time medium process parameters and the real-time opening value and storing them in a vector form;
[0059] Model building step S23: building the initial regulating valve opening control model based on a convolutional neural network;
[0060] Model training step S24: Based on the real-time medium process parameters and the real-time opening value after digitization, the initial regulating valve opening control model is trained to obtain the final regulating valve opening control model.
[0061] Furthermore, the data processing step S22 includes:
[0062] The real-time medium process parameters and the real-time opening value are stored in an Excel table in a vector form;
[0063] The data in the Excel table is read and stored in a tensor of PyTorch, and the tensor is used as a data set to be input into the initial regulating valve opening control model.
[0064] Wherein, in the data processing step S22, it includes preprocessing the collected data and integrating the preprocessed data;
[0065] Preprocessing the collected data includes digitizing the real-time medium process parameters and valve opening values in the ship fuel system, and saving them in Excel format, wherein the digitizing of the collected real-time medium process parameters and valve opening includes: 1.0 represents 1% opening of the regulating valve, and 100.0 represents 100% opening of the regulating valve. The collected real-time medium process parameter values are saved in Excel files in floating point form;
[0066] The preprocessed data is integrated by using the pandas toolkit to read the data in the Excel table, and the data exists in the tensor of PyTorch, and the tensor is input as a data set into the initial regulating valve opening control model. For example, the process parameters are stored in the form of vectors through the data acquisition and storage module, [a1, b1, c1, ...; y1, y2, y3, ...]. Wherein, a1, b1, c1, ... are the values of medium process parameter 1, medium process parameter 2, and medium process parameter 3 when the regulating valve opening is y1. [a1, b1, c1, ...,] in the vector is used as training data and input into the multi-layer perceptron network, and y1 is used as the label of the vector. The vector is used as a training data. At the same time, the medium process parameters are collected at different valve openings through the sensor group to form a data set. A convolutional network is used as the control model of the valve opening, and the obtained training set is used as input to the initial regulating valve opening control model.
[0067] The data set includes multiple sets of training data, each set of training data includes: pipeline pressure, pipeline temperature, pipeline flow, and opening annotation information of each regulating valve under real-time pressure, temperature and flow.
[0068] Furthermore, the model training step S24 includes:
[0069] Extracting and fusing feature information between different real-time medium process parameters from the data set through a convolutional layer;
[0070] Training the initial regulating valve opening control model through the characteristic information;
[0071] Attenuating the weight of the initial regulating valve opening control model by L2 regularization method;
[0072] The output result of the initial regulating valve opening control model is compared with the real-time opening value, and the weight parameter of the initial regulating valve opening control model is updated by back propagation to obtain the final regulating valve opening control model.
[0073] Specifically, the initial regulating valve opening control model extracts the features of the data [a1, b1, c1, ...] through forward propagation, and dynamically iterates and updates the weight parameters in the initial regulating valve opening control model through the back propagation algorithm, so that the actual output continuously approaches the label value y; L2 regularization is added to the initial regulating valve opening control model to minimize the error while regularizing the parameters, wherein the error minimization is to make the model better fit the training data, and the regularization is to prevent overfitting; the weight parameters of the trained initial regulating valve opening control model can meet the error requirements of the test set medium process parameter values, and the weight parameters of the initial regulating valve opening control model at this time are saved, and it is called a trained final regulating valve opening control model.
[0074] Furthermore, in the regulating valve opening control step S3, the output information of the final regulating valve opening control model is obtained through the final regulating valve opening control model according to the target process parameters, and the output information y of the control model is used as the opening value of the regulating valve in the ship fuel system, so that the opening of the regulating valve is adjusted according to the opening value so that the fuel system reaches the target process parameters.
[0075] Based on this, the present invention provides a method for controlling ship fuel process parameters based on machine learning, trains the regulating valve opening control model through a neural network method, and prevents the model from overfitting through L2 regularization. It adds a penalty term to the loss function of the model to prevent the model from overfitting the training data, thereby improving the generalization ability of the model, improving the model output accuracy, and automatically adjusting the global regulating valve of the ship fuel system, thereby realizing global regulation of the medium process parameters of the fuel system, achieving constant fuel delivery, and ensuring the safe movement of the ship.
[0076] Please refer to Figure 3 , Figure 3 FIG. 1 is a schematic diagram of the structure of the process parameter control system of the ship fuel system of the present invention. Figure 3As shown, the ship fuel system process parameter control system of the present invention includes: a target process parameter acquisition unit 1, a model acquisition unit 2 and a regulating valve opening control unit 3, wherein the target process parameter is collected and obtained by the target process parameter acquisition unit 1; an initial regulating valve opening control model based on a convolutional neural network is constructed by the model acquisition unit 2, and the initial regulating valve opening control model is trained to obtain a final regulating valve opening control model; after the regulating valve opening control unit 3 obtains an opening value according to the target process parameter through the final regulating valve opening control model, the opening of the regulating valve is adjusted according to the opening value, so that the fuel system reaches the target process parameter.
[0077] Furthermore, the model acquisition unit 2 includes:
[0078] A data acquisition module 21 collects the real-time medium process parameters and the real-time opening value of each control point of the regulating valve at different openings;
[0079] A data processing module 22, which performs numerical processing on the real-time medium process parameters and the real-time opening value and stores them in a vector form;
[0080] A model building module 23 is used to build the initial regulating valve opening control model based on a convolutional neural network;
[0081] The model training module 24 trains the initial regulating valve opening control model based on the real-time medium process parameters and the real-time opening value after digitization to obtain the final regulating valve opening control model.
[0082] Specifically, the data acquisition module 21 includes a plurality of sensors placed in the fuel system, and the real-time medium process parameters and the real-time opening value are collected by the sensors, wherein the plurality of sensors at least include a temperature sensor, a pressure sensor and a flow meter, and the process parameters of each point in the fuel system are collected by the data acquisition module 21, and the opening information of each regulating valve is recorded at the same time;
[0083] After the data processing module 22 stores the real-time medium process parameters and the real-time opening value in an Excel table in the form of a vector, it reads the data in the Excel table and stores it in a tensor of PyTorch, and uses the tensor as a data set to input into the initial regulating valve opening control model. In this embodiment, the pandas toolkit is used to read the data in the Excel table and it exists in the tensor of PyTorch.
[0084] Furthermore, the model training module 24 includes: a convolutional network module 241, an L2 regularization module 242 and an output module 243; the convolutional network module 241 extracts and fuses the characteristic information between different real-time medium process parameters from the data set through a convolution layer; the initial control valve opening control model is trained through the characteristic information; the L2 regularization module 242 attenuates the weight of the initial control valve opening control model through the L2 regularization method; the output module 243 compares the output result of the initial control valve opening control model with the real-time opening value, and updates the weight parameters of the initial control valve opening control model through back propagation to obtain the final control valve opening control model.
[0085] Specifically, combined with Figure 4 , Figure 4 Schematic diagram of model training, the convolutional network module 241 uses the convolutional network to mine and fuse the feature information between different process parameters; the L2 regularization module 242 uses the L2 regularization method to attenuate the model weight and reduce the model generalization error; the output module 243 compares the output with the regulating valve opening label, and updates the weight parameters of the convolutional network through back propagation until the expected values of the medium process parameters in the ship fuel system are met; wherein, the data set includes multiple groups of training data, each group of training data includes: pipeline pressure, pipeline temperature, pipeline flow, and opening labeling information of each regulating valve under real-time pressure, temperature, and flow, the convolutional network module 241 extracts features of pressure, temperature, and flow in the pipeline through convolution, the output module 243 makes a difference between the actual output and the label value of the regulating valve opening, and updates the convolution weight parameters through the back propagation algorithm without iteration until the output value of the training model meets the error requirement, and the training model parameters are saved.
[0086] The regulating valve opening control unit 3 is used to control the opening of the regulating valve in the ship fuel system according to the valve opening information output by the model. The medium process parameters in the ship fuel system are made to reach the expected values. Specifically, the target process parameters collected by the target process parameter acquisition unit 11 are input into the model to obtain the output data of the control model for controlling the regulating valve opening. The output data is assigned to the regulating valve opening control unit 3 to obtain the opening value. The regulating valve opening control unit 3 controls the regulating valve opening according to the opening value to achieve the expected temperature, pressure, flow, etc. Among them, the target process parameters are the expected temperature, pressure, and flow parameters.
[0087] In summary, the present invention obtains the characteristics of the overall process parameters of the ship fuel system through a convolutional network, and obtains the control signal of the overall regulating valve of the system through convolution operation and L2 regularization, which can reduce fuel consumption, improve the stability of the medium process parameters of the ship fuel system, improve energy efficiency, and reduce the air pollution and greenhouse gas emissions of ships. The present invention is particularly suitable for controlling the medium process parameters in the process piping system, obtaining the overall process parameters of the ship fuel system, and controlling the regulating valves in the fuel system in a wide range, taking into account the mutual influence between the medium process parameters and multiple regulating valves, so that the overall control of the system is more stable.
[0088] Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent substitutions for some of the technical features therein; and these modifications or substitutions do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for controlling process parameters of a ship fuel system based on a convolutional network, characterized in that: include: Target process parameter acquisition step: collecting and obtaining target process parameters; Model acquisition step: constructing an initial regulating valve opening control model based on a convolutional neural network, and training the initial regulating valve opening control model to obtain a final regulating valve opening control model; The regulating valve opening control step: after obtaining the opening value through the final regulating valve opening control model according to the target process parameters, the opening of the regulating valve is adjusted according to the opening value so that the fuel system reaches the target process parameters.
2. The method for controlling process parameters of a ship fuel system according to claim 1, characterized in that: The model acquisition step comprises: Data collection step: collecting the real-time medium process parameters and the real-time opening value of each control point of the regulating valve at different openings; Data processing step: digitizing the real-time medium process parameters and the real-time opening value and storing them in vector form; Model building step: building the initial regulating valve opening control model based on convolutional neural network; Model training step: based on the real-time medium process parameters and the real-time opening value after digitization, the initial regulating valve opening control model is trained to obtain the final regulating valve opening control model.
3. The method for controlling process parameters of a ship fuel system according to claim 2, characterized in that: The data processing steps include: The real-time medium process parameters and the real-time opening value are stored in an Excel table in a vector form; The data in the Excel table is read and stored in a tensor of PyTorch, and the tensor is used as a data set to be input into the initial regulating valve opening control model.
4. The method for controlling process parameters of a ship fuel system according to claim 3, characterized in that: The model training step includes: Extracting and fusing feature information between different real-time medium process parameters from the data set through a convolutional layer; Training the initial regulating valve opening control model through the characteristic information; Attenuating the weight of the initial regulating valve opening control model by L2 regularization method; The output result of the initial regulating valve opening control model is compared with the real-time opening value, and the weight parameter of the initial regulating valve opening control model is updated by back propagation to obtain the final regulating valve opening control model.
5. The method for controlling process parameters of a ship fuel system according to claim 2, characterized in that: In the data collection step, the real-time medium process parameters and the real-time opening value are collected by multiple sensors placed in the fuel system.
6. A ship fuel system process parameter control system based on convolutional network, characterized in that: include: A target process parameter acquisition unit collects and obtains target process parameters; A model acquisition unit, constructing an initial regulating valve opening control model based on a convolutional neural network, and training the initial regulating valve opening control model to obtain a final regulating valve opening control model; The regulating valve opening control unit obtains an opening value through the final regulating valve opening control model according to the target process parameter, and then adjusts the opening of the regulating valve according to the opening value so that the fuel system reaches the target process parameter.
7. The ship fuel system process parameter control system according to claim 6, characterized in that: The model acquisition unit comprises: A data acquisition module, collecting the real-time medium process parameters and the real-time opening value of each control point of the regulating valve at different openings; A data processing module, which performs numerical processing on the real-time medium process parameters and the real-time opening value and stores them in a vector form; A model building module, constructing the initial regulating valve opening control model based on a convolutional neural network; The model training module trains the initial regulating valve opening control model based on the real-time medium process parameters and the real-time opening value after digitization to obtain the final regulating valve opening control model.
8. The ship fuel system process parameter control system according to claim 7, characterized in that: After the data processing module stores the real-time medium process parameters and the real-time opening value in an Excel table in the form of a vector, it reads the data in the Excel table and stores it in a tensor of pytorch, and uses the tensor as a data set to input into the initial regulating valve opening control model.
9. The ship fuel system process parameter control system according to claim 8, characterized in that: The model training module extracts and fuses the characteristic information between different real-time medium process parameters from the data set through a convolution layer; trains the initial control valve opening control model through the characteristic information; attenuates the weight of the initial control valve opening control model through an L2 regularization method; compares the output result of the initial control valve opening control model with the real-time opening value, and updates the weight parameters of the initial control valve opening control model through back propagation to obtain the final control valve opening control model.
10. The ship fuel system process parameter control system according to claim 7, characterized in that: The data acquisition module includes a plurality of sensors placed in the fuel system, and the real-time medium process parameters and the real-time opening value are collected through the sensors.
Citation Information
Patent Citations
Control device and method of internal combustion engine, in-vehicle electronic control unit and manufacturing method of electronic control unit, machine learning system, and output parameter calculation device
CN111520241A
LPG gas-liquid separation tank and fuel control system
CN113090417A
Control method for rapidly adjusting flow controller valve based on pressure fluctuation
CN113835449A
Marine liquid ammonia fuel supply system and method with cabin pressure control function
CN116624758A
Estimating method for amount indicating engine state
JP1999351045A
Cited By
Salt pan valve control method, system, equipment and medium
CN119179272A
