Water cooling control method of gas test transformer, terminal and storage medium

Abstract

The invention belongs to the technical field of power grid equipment control, and particularly relates to a water-cooling control method of a gas test transformer, a terminal and a storage medium, and the method comprises the steps: collecting multi-dimensional parameters of the gas test transformer and a transformer matched water-cooling system in real time through a sensor network; the multi-dimensional parameters comprise transformer temperature, load voltage and current, transformer winding temperature, cooling medium temperature, cooling medium flow velocity, environment temperature and humidity and rotating speed of a variable frequency fan in the water cooling system; the temperature distribution of the transformer can be accurately predicted in combination with the heat conduction model established based on the Fourier heat conduction law, the corresponding PID parameters are extracted from the pre-stored temperature distribution and PID parameter corresponding table, and the PID parameters are optimized through the pre-stored LMS algorithm.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid equipment control, and particularly relates to a water cooling control method, a terminal and a storage medium for a gas test transformer. Background Art

[0002] In the power system, as a key high-voltage test equipment, the stability and safety of the gas test transformer are directly related to the reliability of the entire power system. However, under the existing technical background, there are still many deficiencies in the water cooling control system of the gas test transformer, which restricts the further improvement of its performance.

[0003] The application of the traditional PID control algorithm in the water cooling system of the gas test transformer mainly relies on manual fixed setting of PID parameters. Although this method is simple and direct, its limitations are particularly obvious when facing complex and changeable working conditions. Since the PID parameters are difficult to automatically adjust once set, when external conditions such as load current and ambient temperature change, the traditional PID control system often fails to respond in a timely manner, resulting in poor cooling effect and even potential safety hazards such as transformer overheating.

[0004] In addition, there are also significant deficiencies in the temperature monitoring of the traditional water cooling system. They often rely only on a single temperature monitoring method, such as only monitoring the cooling water temperature or the transformer shell temperature. This monitoring method has a limited coverage area and cannot comprehensively reflect the true temperature conditions inside the transformer, especially the temperature changes in key parts such as windings. Since the windings are the areas where heat is most concentrated in the transformer, their temperature changes have a crucial impact on the operating state of the transformer. Therefore, the limitations of the traditional water cooling system in temperature monitoring make it difficult to accurately grasp the actual operating state of the transformer, thereby affecting the accuracy and effectiveness of control decisions. Summary of the Invention

[0005] Aiming at the defects in the prior art that the application of the traditional PID control algorithm in the water cooling system of the gas test transformer mainly relies on manual fixed setting of PID parameters, and when external conditions such as load current and ambient temperature change, it often fails to respond in a timely manner, resulting in poor cooling effect; and it relies on a single temperature monitoring method and cannot comprehensively reflect the true temperature conditions inside the transformer, the present invention provides a water cooling control method, a terminal and a storage medium for a gas test transformer to solve the above technical problems.

[0006] In the first aspect, the present invention provides a water cooling control method for a gas test transformer, including: Real-time collect multi-dimensional parameters of the gas test transformer and the transformer supporting water cooling system through the sensor network. The multi-dimensional parameters include transformer temperature, load voltage and current, transformer winding temperature, cooling medium temperature, cooling medium flow rate, ambient temperature and humidity, and the rotational speed of the variable-frequency fan in the water cooling system; Use the local PLC to preprocess the collected multi-dimensional parameters and compress and transmit them to the upper computer; the preprocessing includes outlier filtering and missing value filling; The upper computer performs feature selection on the preprocessed multi-dimensional parameters based on L1 regularization to eliminate redundant data; Input the multi-dimensional data after eliminating redundant data into the heat conduction model established based on Fourier's law of heat conduction to predict the temperature distribution of the gas test transformer; Extract the corresponding PID parameters from the pre-stored corresponding table of temperature distribution and PID parameters according to the temperature distribution, and optimize the extracted PID parameters based on the pre-stored LMS algorithm to adjust the operating state of the transformer supporting water cooling system.

[0007] A further improvement of this technical solution is that multi-dimensional parameters of the gas test transformer and the transformer supporting water cooling system are real-time collected through the sensor network. The multi-dimensional parameters include transformer temperature, load voltage and current, transformer winding temperature, cooling medium temperature, cooling medium flow rate, ambient temperature and humidity, and the rotational speed of the variable-frequency fan in the water cooling system. The method includes: Use a PT100 platinum resistance sensor to collect the surface temperature of the transformer ; Use a fiber Bragg grating sensor to collect the winding temperature ; Use a Rogowski coil to collect the load current , use a voltage transformer to collect the load voltage ; Use a thermocouple sensor to collect the cooling medium temperature ; Use an ultrasonic flowmeter to collect the cooling medium flow rate ; Use a digital temperature and humidity sensor to collect the ambient temperature and relative humidity ; Use a Hall effect sensor to collect the rotational speed of the variable-frequency fan in the water cooling system .

[0008] A further improvement of this technical solution is that the upper computer performs feature selection on the preprocessed multi-dimensional parameters based on L1 regularization to eliminate redundant data. The method includes: Form the preprocessed multi-dimensional parameters into a feature vector , each dimension corresponds to a collected parameter and is normalized; Construct an objective function containing the L1 regularization term, and solve for the eigenvector by minimizing the weighted sum of the prediction error and parameter sparsity; and use the coordinate descent method to iteratively update the weights corresponding to the eigenvector; Sort according to the absolute value of the feature weights, eliminate the features with weights less than the preset weight threshold, and retain several features with the strongest correlation with the temperature distribution.

[0009] A further improvement of this technical solution is that the objective function containing the L1 regularization term is: ; where n is the number of samples; is the eigenvector of the i-th sample; is the target value of the i-th sample; Mean squared error loss function; is the feature weight vector, with dimension m; is the regularization parameter; is the L1 norm penalty term.

[0010] A further improvement of this technical solution is that the multi-dimensional data after removing redundant data is input into a heat conduction model established based on Fourier's law of heat conduction to predict the temperature distribution of a gas test transformer. The method includes: Based on the pre-stored physical structure parameters of the gas test transformer, divide the interior of the gas test transformer into three-dimensional grid cells in the sulfur hexafluoride gas region, the metal layer region of the transformer housing, and the cooling pipe region, and assign an initial heat conduction coefficient to each cell; Based on the differential form of Fourier's law of heat conduction Establish a heat conduction model, iteratively calculate the heat flux density vector of each grid cell in the heat conduction model, and solve the temperature gradient equation corresponding to each cell by the gradient descent method , and generate a corresponding real-time temperature distribution map; where is the direction and magnitude of the heat flow through a unit area per unit time; is the heat conduction coefficient corresponding to the grid cell, with the unit of W / (m·K), and the larger the value, the stronger the heat conduction performance; is the temperature field in the grid cell; is the temperature gradient; is the heat source term; Input the transformer winding temperature into the heat conduction model to initialize the temperature field of the heat conduction model; Input the cooling medium flow rate into the heat conduction model to dynamically adjust the heat conduction coefficient corresponding to the cooling pipe region; The load current and voltage are converted into the heating power of the winding and input as the heat source term of the temperature gradient equation to adjust the temperature field in the sulfur hexafluoride gas region; The ambient temperature and humidity are input into the heat conduction model to adjust the heat dissipation boundary conditions in the metal layer region of the transformer housing; The variable-frequency fan speed is input into the heat conduction model to adjust the heat dissipation boundary conditions in the metal layer region of the transformer housing; The heat conduction model outputs a three-dimensional temperature cloud map including the transformer winding, the transformer housing and the cooling medium, and marks the areas where the temperature exceeds the preset temperature threshold.

[0011] A further improvement of this technical solution is that, according to the temperature distribution, the corresponding PID parameters are extracted from the pre-stored temperature distribution and PID parameter correspondence table, and the extracted PID parameters are optimized based on the pre-stored LMS algorithm to adjust the operating state of the water-cooling system supporting the transformer. The method includes: According to the position of the high-temperature area marked in the three-dimensional temperature cloud map and the temperature gradient value, retrieve the matching load differential coefficient, integral coefficient and threshold setting combination from the pre-stored temperature distribution and PID parameter correspondence table; Based on the pre-stored LMS algorithm, adjust the extracted PID parameters, and update the weight coefficient in real time by minimizing the mean square error value between the current temperature distribution and the target temperature distribution; According to the optimized PID parameters, synchronously adjust the flow rate of the cooling medium in the water-cooling system supporting the transformer and the speed of the variable-frequency fan in the water-cooling system .

[0012] A further improvement of this technical solution is that, based on the pre-stored LMS algorithm, adjust the extracted PID parameters, and update the weight coefficient in real time by minimizing the mean square error value between the current temperature distribution and the target temperature distribution. The method includes: Based on the real-time three-dimensional temperature cloud map output by the heat conduction model, extract the actual temperature values of each grid cell, and compare them point by point with the pre-stored target temperature distribution data; By calculating the mean square error of all cells, quantify the difference between the current temperature distribution and the ideal state. The calculation formula is: , where N is the total number of network cells, is the temperature prediction value of the i-th cell, is the target temperature threshold corresponding to the i-th cell; and integrate the input signal vector u(n) under the current working conditions, including load current, cooling medium flow rate and ambient temperature and humidity data; Based on the pre-stored LMS algorithm, according to the mean square error e(n) and the input signal vector u(n), iteratively update the weight vector w(n) of the PID parameters.

[0013] A further improvement of this technical solution is that the formula for iteratively updating the weight vector w(n) of the PID parameters is: , where is the step size parameter in the LMS algorithm, which controls the amplitude of weight update.

[0014] In a second aspect, the present invention provides a terminal, including: a processor and a memory, where the memory is used to store a computer program, the processor is used to call and run the computer program from the memory, so that the terminal executes the method of the above-mentioned terminal.

[0015] In a third aspect, the present invention provides a computer storage medium, and instructions are stored in the computer-readable storage medium. When it runs on a computer, the computer is made to execute the methods described in the above aspects.

[0016] The beneficial effects of the present invention are as follows: The present invention collects multi-dimensional parameters of the gas test transformer and the water cooling system in real time through a sensor network, that is, uses a variety of sensors to comprehensively monitor the surface temperature of the transformer, winding temperature, cooling medium temperature, etc., and realizes the accurate grasp of the internal temperature distribution of the transformer. This multi-dimensional and high-precision temperature monitoring method provides a reliable basis for the precise control of the cooling system.

[0017] The present invention performs feature selection on the pre-processed multi-dimensional parameters by the upper computer based on L1 regularization, eliminates redundant data, and improves the generalization ability and prediction accuracy of the model. Combining with the heat conduction model established based on Fourier's law of heat conduction, it can accurately predict the temperature distribution of the transformer, extract the corresponding PID parameters from the pre-stored correspondence table of temperature distribution and PID parameters, and optimize the PID parameters through the pre-stored LMS algorithm. Compared with the traditional PID control method that relies on manual fixed setting of parameters, the present invention can dynamically adjust the PID parameters according to the real-time working conditions, achieve more precise control, effectively cope with the changes of external conditions such as load current and ambient temperature, thereby improving the cooling effect and reducing the risk of transformer overheating. Description of the Drawings

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0019] Figure 1 It is a schematic flowchart of the method according to an embodiment of the present invention.

[0020] Figure 2 A schematic structural diagram of a terminal provided by an embodiment of the present invention. Detailed implementation manners

[0021] To make the objectives, features, and advantages of the present invention more obvious and understandable, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the specific embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present application.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field of the present invention. The terms used in the description of the present invention in this specification are only for the purpose of describing specific embodiments, and are not intended to limit the present invention.

[0023] PLC, Programmable Logic Controller, a programmable logic controller.

[0024] L1 regularization, L1 Regularization, L1 regularization (also known as Lasso regularization), is a technique commonly used in machine learning and statistics to prevent model overfitting, and realizes feature selection by adding an L1 norm penalty term to the loss function.

[0025] PID, Proportional-Integral-Derivative, a proportional-integral-derivative controller, is a control algorithm widely used in industrial control systems, and realizes precise control of the system by adjusting three parameters: proportional, integral, and derivative.

[0026] RTD, Resistance Temperature Detector, a resistance temperature detector, is a sensor that measures temperature by utilizing the characteristic that resistance changes with temperature, and is commonly used in occasions where high-precision temperature measurement is required.

[0027] Figure 1 It is a schematic flowchart of a water cooling control method for a gas test transformer provided by the present invention. Among them, according to different requirements, the order of the steps in this flowchart can be changed, and some can be omitted.

[0028] Such as Figure 1 shown, the method includes: Step 110, collect multi-dimensional parameters of the gas test transformer and the transformer supporting water-cooling system in real time through the sensor network. The multi-dimensional parameters include transformer temperature, load voltage and current, transformer winding temperature, cooling medium temperature, cooling medium flow rate, ambient temperature and humidity, and the rotational speed of the variable-frequency fan in the water-cooling system; Step 120, preprocess the collected multi-dimensional parameters using the local PLC and compress and transmit them to the host computer; the preprocessing includes outlier filtering and missing value filling; Step 130, the host computer performs feature selection on the preprocessed multi-dimensional parameters based on L1 regularization to eliminate redundant data; Step 140, input the multi-dimensional data after eliminating redundant data into the heat conduction model established based on Fourier's law of heat conduction to predict the temperature distribution of the gas test transformer; Step 150, extract the corresponding PID parameters from the pre-stored corresponding table of temperature distribution and PID parameters according to the temperature distribution, and optimize the extracted PID parameters based on the pre-stored LMS algorithm to adjust the operating state of the transformer supporting water-cooling system.

[0029] For the convenience of understanding the present invention, the following further describes the water-cooling control method of the gas test transformer provided by the present invention in combination with the principle of the water-cooling control method of the gas test transformer of the present invention and the process of controlling the water-cooling system in the gas test transformer in the embodiment.

[0030] Specifically, the method for collecting multi-dimensional parameters of the gas test transformer and the transformer supporting water-cooling system in real time through the sensor network, where the multi-dimensional parameters include transformer temperature, load voltage and current, transformer winding temperature, cooling medium temperature, cooling medium flow rate, ambient temperature and humidity, and the rotational speed of the variable-frequency fan in the water-cooling system, includes: Use a PT100 platinum resistance sensor to collect the surface temperature of the transformer ; Use a fiber Bragg grating sensor to collect the winding temperature ; Use a Rogowski coil to collect the load current , use a voltage transformer to collect the load voltage ; Use a thermocouple sensor to collect the cooling medium temperature ; Use an ultrasonic flowmeter to collect the cooling medium flow rate ; Use a digital temperature and humidity sensor to collect the ambient temperature and relative humidity ; Use a Hall effect sensor to collect the rotational speed of the variable-frequency fan in the water-cooling system 。

[0031] PT100 platinum resistance sensors are evenly distributed on the surface of the transformer housing (top, side, and bottom), with one PT100 platinum resistance sensor installed every 0.5 meters; the PT100 platinum resistance sensor outputs a resistance signal, which is converted into a temperature value (accuracy ±0.5°C) by the RTD module of the local PLC, and the sampling frequency is 10Hz.

[0032] Fiber Bragg grating sensors are embedded inside the transformer winding coils, and temperature measurement points are arranged at intervals of 20cm along the axial direction of the winding; the fiber sensors analyze temperature changes through a wavelength demodulator (integrated in the local PLC), and the data is transmitted to the host computer in real time, with a sampling frequency of 50Hz.

[0033] The Rogowski coil is sleeved on the primary side high-voltage wire of the transformer; the output current signal is converted into an analog quantity through an integrator, and the AI module of the local PLC collects and filters it (cut-off frequency 1kHz), with a sampling frequency of 1kHz.

[0034] The capacitive voltage transformer is connected in parallel between the primary side high-voltage terminal and the grounding terminal of the transformer; the output signal is conditioned by a voltage dividing circuit and then input into the PLC, with an accuracy of ±0.2%.

[0035] K-type thermocouple sensors are embedded at the inlet and outlet of the cooling pipeline, and a group of sensors are arranged every 1 meter; the thermocouple signal is processed by a cold junction compensation circuit and then input into the PLC, with a temperature resolution of 0.1°C and a sampling frequency of 5Hz.

[0036] The ultrasonic flowmeter is installed on the straight pipe section of the cooling pipeline (meeting the straight pipe requirements of 10D in the front and 5D in the back, where D is the pipeline diameter); the flow rate is measured based on the time difference method, and a 4-20mA signal is output to the PLC, with an accuracy of ±1% and a response time ≤100ms.

[0037] Digital temperature and humidity sensors (SHT35) are arranged within 1 meter around the transformer body (avoiding the heat radiation area); the I²C digital signal is directly input into the PLC, with a temperature range of -40 to 125°C, a humidity range of 0 to 100%RH, and a sampling frequency of 1Hz.

[0038] Hall effect sensors (integrated in the fan motor): A magnet is installed at the motor shaft, and the Hall sensor is fixed to the motor housing; a fixed pulse is output per revolution (such as 60 pulses per revolution), and the PLC calculates the real-time speed through the high-speed counter module (resolution ±1rpm), with a sampling frequency of 100Hz.

[0039] All sensor data is preprocessed by the PLC, including: Outlier filtering: The sliding window method (window size 50 sampling points) is used to identify and remove outliers; Missing value filling: Linear interpolation between adjacent timestamps is used for completion; Standardization processing: Normalize each parameter to the interval [0, 1] to eliminate the dimension difference; Data transmission: The PLC compresses and transmits the preprocessed data to the host computer through the Ethernet communication protocol (Modbus TCP). The transmission period is 200 ms to ensure data real-time performance.

[0040] The present invention collects multi-dimensional parameters of the gas test transformer and the water cooling system in real time through a multi-type and high-precision sensor network, ensuring the comprehensiveness and accuracy of the data. Each type of sensor adopts a professional measurement method for different monitoring objects, improving the reliability of the data. At the same time, the local PLC preprocesses the collected data, including outlier filtering, missing value filling, and standardization processing, further improving the data quality. The preprocessed data is compressed and transmitted to the host computer through the Ethernet communication protocol, which not only ensures data real-time performance but also reduces the transmission bandwidth requirement. This technical means provides a reliable data basis for subsequent precise control, fault prediction, etc., and helps to improve the operation stability and security of the system.

[0041] After that, the host computer performs feature selection on the preprocessed multi-dimensional parameters based on L1 regularization to eliminate redundant data. The methods include: Form the preprocessed multi-dimensional parameters into a feature vector , each dimension corresponds to a collected parameter, and standardization processing is performed; Construct an objective function containing the L1 regularization term, and solve the feature vector by minimizing the weighted sum of the prediction error and parameter sparsity; and use the coordinate descent method to iteratively update the weights corresponding to the feature vector; Sort according to the absolute value of the feature weights, eliminate the features with weights less than the preset weight threshold, and retain several features with the strongest correlation with the temperature distribution.

[0042] Furthermore, the objective function containing the L1 regularization term is: ; where n is the number of samples; is the i-th sample feature vector; is the target value of the i-th sample; Mean square error loss function; is the feature weight vector, with dimension m; is the regularization parameter; is the L1 norm penalty term.

[0043] Initialize the weight vector as a zero vector; update the weights in turn according to the feature dimensions: , if , then , otherwise, ; where is the correlation between the j-th feature and the model residual; is the variance of the j-th feature, which is used to standardize the magnitude of weight update.

[0044] In the present invention, the host computer performs feature selection on the preprocessed multi-dimensional parameters based on L1 regularization, effectively eliminating redundant data and improving the data processing efficiency and model performance. Specifically, after forming a feature vector from the multi-dimensional parameters and performing standardization processing, an objective function including an L1 regularization term is constructed, and the feature weights are optimized by minimizing the weighted sum of the prediction error and parameter sparsity. The coordinate descent method is used to iteratively update the weights, ensuring the convergence and stability of the algorithm. Finally, the features are sorted according to the absolute value of the feature weights, and the low-weight features are eliminated, retaining the features with the strongest correlation with the temperature distribution, providing high-quality and low-redundancy input data for subsequent heat conduction model prediction and PID parameter optimization, and enhancing the overall reliability and control accuracy of the system.

[0045] In addition, the multi-dimensional data after eliminating redundant data is input into a heat conduction model established based on Fourier's law of heat conduction to predict the temperature distribution of the gas test transformer. The method includes: Based on the pre-stored physical structure parameters of the gas test transformer, the internal part of the gas test transformer is divided into three-dimensional grid cells of the sulfur hexafluoride gas region, the metal layer region of the transformer shell, and the cooling pipe region, and an initial heat conduction coefficient is assigned to each cell; Based on the differential form of Fourier's law of heat conduction A heat conduction model is established, and the heat flux density vector of each grid cell is iteratively calculated in the heat conduction model, and the temperature gradient equation corresponding to each cell is solved by the gradient descent method , and a corresponding real-time temperature distribution map is generated; where is the direction and magnitude of the heat flow passing through a unit area per unit time; is the heat conduction coefficient corresponding to the grid cell, with the unit of W / (m·K), and the larger the value, the stronger the heat conduction performance; is the temperature field in the grid cell; is the temperature gradient; is the heat source term; The transformer winding temperature is input into the heat conduction model to initialize the temperature field of the heat conduction model; The flow rate of the cooling medium is input into the heat conduction model to dynamically adjust the heat conduction coefficient corresponding to the cooling pipe region; The load current and voltage are converted into the winding heating power and input as the heat source term of the temperature gradient equation to adjust the temperature field of the sulfur hexafluoride gas region; Input the ambient temperature and humidity into the heat conduction model to adjust the heat dissipation boundary conditions in the metal layer area of the transformer housing; input the variable-frequency fan speed into the heat conduction model to adjust the heat dissipation boundary conditions in the metal layer area of the transformer housing. The heat conduction model outputs a three-dimensional temperature nephogram including the transformer windings, the transformer housing, and the cooling medium, and marks the areas where the temperature exceeds the preset temperature threshold.

[0046] Heat source term , although the load voltage parameter U is not directly used in the calculation of the winding heating power, it still has the following key functions: Exceeding the standard of the load voltage (such as overvoltage or undervoltage) may trigger the protection mechanism, adjust the cooling strategy in advance or cut off the power supply to prevent equipment damage. For example, when the voltage is abnormal, even if the current is normal, the cooling intensity may be actively increased due to the increased risk of insulation breakdown; in addition, it can be combined with the load current I and voltage U to calculate the actual power , to assist in evaluating the operating efficiency of the transformer. The system adjusts the cooling strategy through the power factor. For example, when the power factor is high, the cooling energy consumption is reduced. The load voltage U is used as a component of the feature vector and participates in the L1 regularization feature selection. Although it may ultimately be excluded (if the weight is low), its redundancy or correlation needs to be verified through data analysis.

[0047] Variable-frequency fan speed Adjust the forced convective heat transfer coefficient h on the surface of the housing, which affects the heat dissipation boundary conditions in the metal layer area of the transformer housing. The calculation formula for the forced convective heat transfer coefficient h is .

[0048] The ambient temperature and humidity correct the natural convective heat transfer boundary conditions on the surface of the housing. When the humidity is higher than the preset threshold, the heat dissipation efficiency is reduced, which affects the heat dissipation boundary conditions in the metal layer area of the transformer housing.

[0049] Sulfur hexafluoride gas area: filled with insulating gas, the default value of the thermal conductivity (k1) is 0.015 W / (m·K); Transformer housing metal layer area (aluminum shell): k2 = 237 W / (m·K); Cooling pipe area (fluoride liquid flowing in a stainless steel pipe): k3 = 50 W / (m·K).

[0050] Initialize the temperature field ; Based on the current temperature gradient , update the temperature distribution ; Iterate until the convergence condition (the temperature change is less than 0.1°C) or the maximum number of iterations (such as 1000 times).

[0051] Adjust the equivalent thermal conductivity of the cooling pipe area according to the flow rate data .

[0052] By dividing the interior of the transformer into three-dimensional grid cells of different material regions in detail and assigning reasonable initial thermal conductivity coefficients to each, the model can accurately simulate the complex heat conduction process inside the transformer, including the sulfur hexafluoride gas region, the metal layer region of the transformer housing, and the cooling pipe region, ensuring the accuracy and comprehensiveness of the simulation. Using the differential form of Fourier's law of heat conduction to construct a heat conduction model and iteratively calculating the heat flux density vector and temperature gradient equations of each grid cell, the model can dynamically reflect the temperature changes of the transformer under different operating conditions and generate a real-time temperature distribution map. This not only helps to detect potential high-temperature regions in a timely manner but also provides accurate data support for subsequent cooling control. The model output includes three-dimensional temperature cloud maps of the transformer windings, housing, and cooling medium, and particularly marks the regions where the temperature exceeds the preset threshold, enabling operators to intuitively understand the thermal state of the transformer and take measures in a timely manner to prevent overheating. Combining the cooling medium flow rate data collected by an ultrasonic flowmeter, optimizing the cooling path through fluid dynamics simulation, and dynamically adjusting the coefficients in the heat conduction model further improve the prediction accuracy and practicality of the model. This adaptive adjustment mechanism ensures that the cooling system can be intelligently adjusted according to the actual operating conditions, effectively improving the operating efficiency and safety of the transformer.

[0053] In addition, according to the temperature distribution, the corresponding PID parameters are extracted from the pre-stored correspondence table of temperature distribution and PID parameters, and the extracted PID parameters are optimized based on the pre-stored LMS algorithm to adjust the operating state of the water-cooling system supporting the transformer. The method includes: According to the positions of the high-temperature regions marked in the three-dimensional temperature cloud map and the temperature gradient values, retrieve the matching load differential coefficient, integral coefficient, and threshold setting combination from the pre-stored correspondence table of temperature distribution and PID parameters; Based on the pre-stored LMS algorithm, adjust the extracted PID parameters, and update the weight coefficients in real time by minimizing the mean square error value between the current temperature distribution and the target temperature distribution; According to the optimized PID parameters, synchronously adjust the flow rate of the cooling medium of the water-cooling system supporting the transformer and the rotational speed of the variable-frequency fan in the water-cooling system .

[0054] Based on historical experimental data, establish a mapping relationship between temperature distribution characteristics (such as the positions of high-temperature regions and temperature gradient values) and PID parameters, as shown in Table 1.

[0055] Table 1: Mapping relationship table of temperature distribution characteristics and PID parameters

[0056] The retrieval logic from Table 1 is as follows: The host computer matches the corresponding PID parameter combination (Kd = 1.0, Ki = 1.5, threshold 95°C) according to the high-temperature area marked in the real-time three-dimensional temperature cloud map (such as the winding hot spot accounting for 25%) and the temperature gradient (such as 85°C / m).

[0057] Furthermore, based on the pre-stored LMS algorithm, the extracted PID parameters are adjusted. By minimizing the mean square error value between the current temperature distribution and the target temperature distribution, the weight coefficient is updated in real time. The method includes: Based on the real-time three-dimensional temperature cloud map output by the heat conduction model, the actual temperature values of each grid cell are extracted and compared point by point with the pre-stored target temperature distribution data; By calculating the mean square error of all cells, the difference between the current temperature distribution and the ideal state is quantified. The calculation formula is: , where N is the total number of network cells, is the temperature prediction value of the i-th cell, is the target temperature threshold corresponding to the i-th cell; and the input signal vector u(n) under the current working condition is integrated, including load current, cooling medium flow rate, and ambient temperature and humidity data; Based on the pre-stored LMS algorithm, according to the mean square error e(n) and the input signal vector u(n), the weight vector w(n) of the PID parameters is iteratively updated.

[0058] Specifically, the formula for iteratively updating the weight vector w(n) of the PID parameters is: , where, is the step size parameter in the LMS algorithm, which controls the amplitude of weight update.

[0059] Through the establishment of a mapping relation table between the temperature distribution characteristics and the PID parameters, the system can quickly retrieve and match the PID parameter combinations according to the real-time three-dimensional temperature cloud map, achieving an accurate response to the cooling requirements under different working conditions. This mapping relation based on historical experimental data ensures the scientificity and rationality of the PID parameter selection, avoiding the blindness and inefficiency of manual parameter adjustment. Secondly, the LMS algorithm is used to optimize the extracted PID parameters in real time. By minimizing the mean square error value between the current temperature distribution and the target temperature distribution, the weight coefficient is dynamically adjusted, enabling the cooling system to adaptively adjust according to the actual temperature conditions. This optimization mechanism not only improves the cooling effect but also effectively reduces energy waste and enhances the operating efficiency of the system. Finally, the flow rate of the cooling medium and the rotational speed of the variable-frequency fan are synchronously adjusted according to the optimized PID parameters, achieving precise control of the operating state of the water-cooling system. This closed-loop control strategy ensures that the transformer can maintain a stable operating temperature under various working conditions, effectively preventing the occurrence of safety hazards such as overheating, extending the service life of the transformer, and improving the overall reliability of the power system. At the same time, this technical means also has high flexibility and scalability, and can adapt to gas test transformers of different models and specifications, providing strong support for the intelligent management of power equipment.

[0060] Figure 2 FIG. 4 is a schematic structural diagram of a terminal 200 provided by an embodiment of the present invention. The terminal 200 can be used to execute the water-cooling control method for a gas test transformer provided by the embodiment of the present invention.

[0061] Among them, the terminal 200 may include: a processor 210, a memory 220, and a communication module 230. These components communicate through one or more buses. Those skilled in the art can understand that the structure of the server shown in the figure does not constitute a limitation to the present invention. It can be a bus structure, a star structure, and may also include more or fewer components than shown in the figure, or combine some components, or have different component arrangements.

[0062] Among them, the memory 220 can be used to store the execution instructions of the processor 210. The memory 220 can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory 220 are executed by the processor 210, the terminal 200 can execute some or all of the steps in the above method embodiments.

[0063] The processor 210 is the control center of the storage terminal, connecting various parts of the entire electronic terminal through various interfaces and circuits. By running or executing software programs and / or modules stored in the memory 220, and by invoking the data stored in the memory, it executes various functions of the electronic terminal and / or processes data. The processor may be composed of an integrated circuit (IC), for example, it may be composed of a single packaged IC, or it may be composed of multiple packaged ICs with the same or different functions connected together. For example, the processor 210 may only include a central processing unit (CPU). In the embodiments of the present invention, the CPU may be a single arithmetic core or may include multiple arithmetic cores.

[0064] The communication module 230 is used to establish a communication channel so that the storage terminal can communicate with other terminals. It receives user data sent by other terminals or sends user data to other terminals.

[0065] The present invention also provides a computer storage medium. Among them, this computer storage medium can store a program, and when the program is executed, it may include some or all of the steps in the various embodiments provided by the present invention. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0066] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions in the embodiments of the present invention, in essence, or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product is stored in a storage medium such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc., which can store program codes, and includes several instructions to enable a computer terminal (which may be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention.

[0067] For the same or similar parts between the various embodiments in this specification, reference can be made to each other. In particular, for the terminal embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can refer to the description in the method embodiments.

[0068] Although the present invention has been described in detail by reference to the accompanying drawings and in conjunction with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, those of ordinary skill in the art can make various equivalent modifications or substitutions to the embodiments of the present invention, and all such modifications or substitutions should be within the scope of the present invention / Any person skilled in the art within the technical scope disclosed by the present invention can easily conceive of changes or substitutions, which should all be covered within the protection scope of the present invention.

Claims

1. A water-cooling control method for a gas test transformer, characterized in that Including: Collecting multi-dimensional parameters of the gas test transformer and the transformer supporting water cooling system in real time through a sensor network. The multi-dimensional parameters include transformer temperature, load voltage and current, transformer winding temperature, cooling medium temperature, cooling medium flow rate, ambient temperature and humidity, and the speed of the variable-frequency fan in the water cooling system. Using the local PLC to preprocess the collected multi-dimensional parameters and compress and transmit them to the upper computer. The preprocessing includes outlier filtering and missing value filling. The upper computer performs feature selection on the preprocessed multi-dimensional parameters based on L1 regularization to eliminate redundant data. Inputting the multi-dimensional data after eliminating redundant data into the heat conduction model established based on Fourier's law of heat conduction to predict the temperature distribution of the gas test transformer. Extracting the corresponding PID parameters from the pre-stored correspondence table of temperature distribution and PID parameters according to the temperature distribution, and optimizing the extracted PID parameters based on the pre-stored LMS algorithm to adjust the operating state of the transformer supporting water cooling system.

2. The water cooling control method of the gas test transformer according to claim 1, wherein, Collecting multi-dimensional parameters of the gas test transformer and the transformer supporting water cooling system in real time through a sensor network. The multi-dimensional parameters include transformer temperature, load voltage and current, transformer winding temperature, cooling medium temperature, cooling medium flow rate, ambient temperature and humidity, and the speed of the variable-frequency fan in the water cooling system. The method includes: Use a PT100 platinum resistance sensor to collect the surface temperature of the transformer ; Use fiber Bragg grating sensors to collect winding temperature ; Collect the load current through a Rogowski coil , and collect the load voltage through a voltage transformer ; Use a thermocouple sensor to collect the temperature of the cooling medium ; Use an ultrasonic flowmeter to collect the flow rate of the cooling medium ; Use a digital temperature and humidity sensor to collect the ambient temperature and relative humidity ; Use a Hall effect sensor to collect the rotational speed of the variable-frequency fan in the water cooling system .

3. The water-cooling control method of the gas test transformer according to claim 2, characterized in that, The upper computer performs feature selection on the preprocessed multi-dimensional parameters based on L1 regularization to eliminate redundant data. The method includes: Form a feature vector from the preprocessed multi-dimensional parameters , where each dimension corresponds to a collected parameter and is standardized; Constructing an objective function containing an L1 regularization term, solving the eigenvector by minimizing the weighted sum of the prediction error and parameter sparsity, and iteratively updating the weight corresponding to the eigenvector using the coordinate descent method. Sorting according to the absolute value of the feature weight, eliminating the features with weights less than the preset weight threshold, and retaining several features with the strongest correlation with the temperature distribution.

4. The water-cooling control method of the gas test transformer according to claim 3, characterized in that, The objective function containing the L1 regularization term is: ； Among them, n is the number of samples; is the i-th sample feature vector; is the target value of the i-th sample; Mean squared error loss function; is the feature weight vector, with dimension m; is the regularization parameter; is the L1 norm penalty term.

5. The water cooling control method of the gas test transformer according to claim 3, characterized in that Inputting the multi-dimensional data after eliminating redundant data into the heat conduction model established based on Fourier's law of heat conduction to predict the temperature distribution of the gas test transformer. The method includes: Based on the pre-stored physical structure parameters of the gas test transformer, dividing the interior of the gas test transformer into three-dimensional grid cells of the sulfur hexafluoride gas region, the transformer housing metal layer region, and the cooling pipe region, and assigning an initial heat conduction coefficient to each cell. Differential form based on Fourier's law of heat conduction Establish a heat conduction model, iteratively calculate the heat flux density vector of each grid cell in the heat conduction model, and solve the temperature gradient equation corresponding to each cell by the gradient descent method , and generate a corresponding real-time temperature distribution map; where is the direction and magnitude of the heat flow passing through a unit area per unit time; is the heat conduction coefficient corresponding to the grid cell, with the unit of W / (m·K), and the greater the value, the stronger the heat conduction performance; is the temperature field in the grid cell; is the temperature gradient; is the heat source term; Inputting the transformer winding temperature into the heat conduction model to initialize the temperature field of the heat conduction model. Inputting the cooling medium flow rate into the heat conduction model to dynamically adjust the heat conduction coefficient corresponding to the cooling pipe region. Converting the load current and voltage into the heat generation power of the winding and inputting it as the heat source term of the temperature gradient equation to adjust the temperature field of the sulfur hexafluoride gas region. Inputting the ambient temperature and humidity into the heat conduction model to adjust the heat dissipation boundary condition of the transformer housing metal layer region; inputting the speed of the variable-frequency fan into the heat conduction model to adjust the heat dissipation boundary condition of the transformer housing metal layer region. The heat conduction model outputs a three-dimensional temperature cloud map including the transformer winding, the transformer housing, and the cooling medium, and marks the regions where the temperature exceeds the preset temperature threshold.

6. The water-cooling control method of the gas test transformer according to claim 5, characterized in that, Extract the corresponding PID parameters from the pre-stored temperature distribution and PID parameter correspondence table according to the temperature distribution, and optimize the extracted PID parameters based on the pre-stored LMS algorithm to adjust the operating state of the water-cooling system supporting the transformer. The method includes: According to the positions of the high-temperature regions marked in the three-dimensional temperature cloud map and the temperature gradient values, retrieve the matching load differential coefficient, integral coefficient, and threshold setting combination from the pre-stored temperature distribution and PID parameter correspondence table; Adjust the extracted PID parameters based on the pre-stored LMS algorithm, and update the weight coefficients in real time by minimizing the mean square error value between the current temperature distribution and the target temperature distribution; Synchronously adjust the flow rate of the cooling medium of the water-cooling system supporting the transformer and the rotational speed of the variable-frequency fan in the water-cooling system according to the optimized PID parameters and the rotational speed of the variable-frequency fan in the water-cooling system .

7. The water-cooling control method of the gas test transformer according to claim 6, characterized in that, Adjust the extracted PID parameters based on the pre-stored LMS algorithm, and update the weight coefficients in real time by minimizing the mean square error value between the current temperature distribution and the target temperature distribution. The method includes: Based on the real-time three-dimensional temperature cloud map output by the heat conduction model, extract the actual temperature values of each grid cell and compare them point by point with the pre-stored target temperature distribution data; By calculating the mean square error of all units, the difference between the current temperature distribution and the ideal state is quantified, and its calculation formula is: , where N is the total number of network units, is the temperature prediction value of the i-th unit, is the target temperature threshold corresponding to the i-th unit; and the input signal vector u(n) under the current working condition is integrated, including load current, cooling medium flow rate, and ambient temperature and humidity data; Based on the pre-stored LMS algorithm, according to the mean square error e(n) and the input signal vector u(n), iteratively update the weight vector w(n) of the PID parameters.

8. The water-cooling control method of the gas test transformer according to claim 7, characterized in that The formula for iteratively updating the weight vector w(n) of the PID parameters is as follows: , where is the step size parameter in the LMS algorithm, which controls the amplitude of weight update.

9. A terminal, characterized in that, Including: A processor; A memory for storing the execution instructions of the processor; Wherein, the processor is configured to execute the method according to any one of claims 1-8.

10. A computer-readable storage medium storing a computer program, characterized in that, When the program is executed by the processor, it implements the method according to any one of claims 1-8.

Images