Simulation analysis-based high-voltage variable-frequency power supply miniaturization and application correction method

Abstract

The invention relates to the field of power supply design, and discloses a high-voltage variable-frequency power supply miniaturization and application correction method based on simulation analysis, and the method comprises the following steps: carrying out the thermal dynamic simulation and electric field simulation of a transformer, and determining the working conditions of the transformer and the distribution state of a target electric field; based on the working conditions of transformer use and the target electric field distribution state, the purpose of design optimization of the high-frequency transformation power supply is achieved by combining a no-load partial discharge test on the transformer. According to the invention, the miniaturization design of the high-voltage variable-frequency power supply can be realized by carrying out the no-load partial discharge test on the transformer, and the high reliability of the equipment can still be kept in a compact space.

Description

Technical Field

[0001] The invention relates to the field of power supply design, in particular to a method for miniaturization and application correction of a high-voltage variable-frequency power supply based on simulation analysis. Background Art

[0002] High-voltage variable-frequency power supply refers to a variable-frequency power supply with an operating voltage of 6KV or above. It is usually used to drive high-power load equipment, such as large ventilation equipment and water pumps in industrial and mining enterprises. The environment in high-altitude areas is harsh and the air is thin, which puts higher requirements on the insulation performance of electrical equipment. The partial discharge test can evaluate the insulation condition of the transformer under high voltage to ensure that it can operate stably in a high-altitude environment. By optimizing the insulation design, the volume and weight of the transformer can be reduced, thereby realizing the miniaturization of the high-voltage variable-frequency power supply. In high-altitude areas, electrical equipment needs to withstand extreme conditions such as low temperature and strong radiation. Therefore, in the miniaturization design of the transformer, it is necessary to select materials suitable for high-altitude environments, such as low-temperature resistant and radiation-resistant materials. No-load partial discharge tests can help evaluate the performance of different materials under high voltage and provide a basis for material selection. Therefore, a method for miniaturization and application correction of high-voltage variable-frequency power supply based on simulation analysis is proposed to realize the design of high-voltage variable-frequency power supply. Summary of the invention

[0003] The present invention overcomes the shortcomings of the prior art and provides a method for miniaturizing a high-voltage variable-frequency power supply and correcting its application based on simulation analysis.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention provides a method for miniaturization and application correction of a high-voltage variable-frequency power supply based on simulation analysis, comprising the following steps:

[0006] Collect the environmental data of the transformer no-load partial discharge site and the transformer's thermodynamic data, and use the convolutional neural network model to model the transformer's thermodynamic data;

[0007] The standard thermodynamic mathematical model of the target transformer is simulated by Simulink software to simulate the thermal characteristics of the target transformer under different working conditions, and based on the thermal characteristics information under different working conditions, the applicable working conditions of the target transformer at the transformer working position are determined;

[0008] Performing electric field simulation analysis on the target transformer, and determining the target electric field distribution state based on the electric field simulation analysis results;

[0009] A high-voltage variable-frequency power supply is designed to make the electric field distribution state in the qualified electric field distribution transformer equal to the target electric field distribution state, and a no-load partial discharge test is performed on the target transformer. The design of the high-voltage variable-frequency power supply is optimized based on the no-load partial discharge test results.

[0010] Furthermore, in a preferred embodiment of the present invention, the environmental data of the transformer no-load partial discharge site and the thermodynamic data of the transformer are collected, and the thermodynamic data modeling of the transformer is performed in combination with a convolutional neural network model, specifically:

[0011] Determine the target operating altitude of the variable frequency power supply, and locate the on-site location for the transformer no-load partial discharge test within the target operating altitude of the variable frequency power supply, and calibrate it as the transformer operating position;

[0012] A meteorological platform is introduced, and predicted average environmental data of the transformer working position is retrieved from the meteorological platform and calibrated as target predicted environmental data;

[0013] An ultra-high voltage power transformer for no-load partial discharge experiment is introduced and calibrated as a target transformer, and the thermodynamic data range of the target transformer is calculated, wherein the thermodynamic data range of the target transformer includes the real-time temperature range, real-time current range and real-time voltage range when the target transformer is in operation, and a multi-layer feedforward neural network algorithm is introduced;

[0014] Performing time series conversion on the thermodynamic data of the target transformer to obtain the time series thermodynamic data of the target transformer, and combining the thermodynamic data of the target transformer and the target prediction environment data for training based on the multi-layer feedforward neural network algorithm to construct a thermodynamic mathematical model of the target transformer;

[0015] The learning rate and neural layer of the thermodynamic mathematical model of the target transformer are adjusted, and a regularization algorithm is introduced to introduce a penalty term into the thermodynamic mathematical model of the target transformer through the regularization algorithm;

[0016] Fourier series is introduced to decompose the model function of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are adjusted, and different sine functions and cosine functions are obtained. Polynomial regression is performed on the different sine functions and cosine functions to ensure that the model function of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are adjusted is close to the preset value, and the fitting state of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are output;

[0017] A standard fitting state is preset. If the fitting state is equal to the standard fitting state, a penalty term is output, the combined training of the thermodynamic data of the target transformer and the target predicted environmental data is stopped, and a standard thermodynamic mathematical model of the target transformer is output.

[0018] Further, in a preferred embodiment of the present invention, the standard thermodynamic mathematical model of the target transformer is simulated by Simulink software to obtain the corresponding thermal characteristics under different working conditions, and based on the thermal characteristic information under different working conditions, the working conditions applicable to the target transformer at the transformer working position are determined, specifically:

[0019] Obtaining the construction and use instructions of the target transformer, and determining the internal structure of the target transformer based on the construction and use instructions of the target transformer, wherein the internal structure of the target transformer includes the material types of internal components of the target transformer and the arrangement structure of the internal components;

[0020] The Simulink software is introduced, the internal structure of the target transformer is mapped in the Simulink software, a simulation model of the target transformer is constructed, and it is calibrated as a transformer simulation model;

[0021] Importing a standard thermodynamic mathematical model of a target transformer into the transformer simulation model, and converting the standard thermodynamic mathematical model of the target transformer into a Matlab function format in Simulink software to obtain a transformer thermodynamic simulation model;

[0022] Preset a target simulation time and a target simulation step in the transformer thermal power simulation model, run the transformer thermal power simulation model under different operating conditions based on the target simulation time and the target simulation step, and output thermal characteristic information of the transformer thermal power simulation model under different operating conditions;

[0023] The operating conditions are the combinations of different temperatures, currents and voltages within the thermodynamic data range of the target transformer, and the thermal characteristic information is the thermal state of the internal windings and core during the operation of the transformer thermodynamic simulation model;

[0024] Based on the big data network, the standard thermal state of the winding and the core in the transformer thermodynamic simulation model at the transformer working position is determined, and the standard thermal characteristic information is determined by reverse deduction. Based on the standard thermal characteristic information, the operating condition of the target transformer when the thermal characteristic information is equal to the standard thermal characteristic information is determined and calibrated as the target operating condition.

[0025] Furthermore, in a preferred embodiment of the present invention, the target transformer is subjected to electric field simulation analysis, and the target electric field distribution state is determined based on the electric field simulation analysis result, specifically:

[0026] Obtaining electric field simulation software, and constructing an electric field simulation model of the target transformer based on the internal structure of the target transformer in the electric field simulation software;

[0027] Performing parameter setting in the electric field simulation model of the target transformer, wherein the parameter setting includes inputting target operating conditions, transformer working position, and target predicted environmental data to obtain a target electric field simulation model;

[0028] Running the target electric field simulation model, performing grid division in the target electric field simulation model, determining the distribution boundary of the electric field, and outputting an electric field distribution cloud map after the target electric field simulation model is run;

[0029] Calculate the electric field strength values ​​at different positions in the electric field distribution cloud map, and determine whether there is an electric field position with an electric field strength value greater than a preset value. If so, determine that the electric field distribution in the target transformer is uneven, and introduce the Bayesian network algorithm to perform defect traceability analysis on the target transformer's component operating parameters, determine the components that cause the uneven electric field distribution in the target transformer, and mark them as faulty transformer components;

[0030] The material of the faulty transformer element is optimized. If there is still an electric field position where the electric field strength value is greater than the preset value after the material optimization of the faulty transformer element, the insulation performance of the target transformer is judged to be unqualified, and a solution for insulation optimization of the target transformer is retrieved based on the big data network and output to obtain a transformer with qualified electric field distribution;

[0031] In the electric field simulation software, an electric field simulation model of the transformer with qualified electric field distribution is constructed, and the electric field distribution cloud map of the transformer with qualified electric field distribution is obtained to determine the electric field distribution state in the transformer with qualified electric field distribution and calibrate it as the target electric field distribution state.

[0032] Furthermore, in a preferred embodiment of the present invention, the design makes the electric field distribution state in the transformer with qualified electric field distribution equal to the high-voltage variable frequency power supply of the target electric field distribution state, and performs a no-load partial discharge test on the target transformer, and optimizes the design of the high-voltage variable frequency power supply based on the no-load partial discharge test results, specifically:

[0033] Constructing a blank simulation model of a high-voltage variable-frequency power supply and a simulation model of a transformer with qualified electric field distribution in an electric field simulation software, and connecting the blank simulation model of a high-voltage variable-frequency power supply with the simulation model of a transformer with qualified electric field distribution to obtain a power system simulation model;

[0034] Among them, the operating condition of the simulation model of the transformer with qualified electric field distribution is equal to the target operating condition;

[0035] Based on the target electric field distribution state, the model output parameters of the blank simulation model of the high-voltage variable-frequency power supply that enables the electric field distribution qualified transformer simulation model in the power system simulation model to output the target electric field distribution state are retrieved in the big data network, and are calibrated as the target model output parameters, and the control main circuit with the highest utilization rate is retrieved and calibrated as the target control main circuit;

[0036] Mapping the target control main circuit in a blank simulation model of a high-voltage variable-frequency power supply, and adjusting the model output parameters of the blank simulation model of the high-voltage variable-frequency power supply to be equal to the target model output parameters, to obtain a high-voltage variable-frequency target power supply simulation model;

[0037] Based on the high-voltage variable-frequency target power supply simulation model, the power system simulation model is updated to obtain a power system update simulation model, a no-load partial discharge test is performed within the power system update simulation model, and the high-voltage variable-frequency power supply is designed and optimized based on the no-load partial discharge test results.

[0038] Furthermore, in a preferred embodiment of the present invention, the no-load partial discharge test is performed in the power system update simulation model, and the high-voltage variable frequency power supply is designed and optimized based on the no-load partial discharge test results, specifically:

[0039] In the power system update simulation model, the secondary side of the transformer simulation model with qualified electric field distribution is opened, and after the secondary side is opened, a no-load test is performed on the transformer simulation model with qualified electric field distribution in the power system update simulation model, and after the no-load test, the no-load current and magnetic loss of the high-voltage variable frequency target power supply simulation model are calculated;

[0040] Conducting a partial discharge test on a simulation model of a transformer with qualified electric field distribution, and detecting whether the partial discharge amount of a simulation model of a high-voltage variable frequency target power supply after the partial discharge test on the simulation model of a transformer with qualified electric field distribution is maintained within a preset threshold;

[0041] If the no-load current and magnetic loss of the high-voltage variable frequency target power supply simulation model are maintained within the standard threshold, and the partial discharge amount of the high-voltage variable frequency target power supply simulation model after the partial discharge test of the electric field distribution qualified transformer simulation model is maintained within the preset threshold, the corresponding model parameters in the high-voltage variable frequency target power supply simulation model are output;

[0042] If the no-load current and magnetic loss of the high-voltage variable-frequency target power supply simulation model are not maintained within the standard threshold, or the partial discharge of the high-voltage variable-frequency target power supply simulation model after the partial discharge test on the electric field distribution qualified transformer simulation model is not maintained within the preset threshold, then through big data network retrieval, a solution is output that can maintain the no-load current and magnetic loss of the high-voltage variable-frequency target power supply simulation model within the standard threshold, and the partial discharge of the high-voltage variable-frequency target power supply simulation model after the partial discharge test on the electric field distribution qualified transformer simulation model is maintained within the preset threshold.

[0043] The second aspect of the present invention further provides a high-voltage variable frequency power supply miniaturization and application correction system based on simulation analysis, the high-voltage variable frequency power supply miniaturization and application correction system comprising a memory and a processor, the memory stores a high-voltage variable frequency power supply miniaturization and application correction method, and when the high-voltage variable frequency power supply miniaturization and application correction method is executed by the processor, the following steps are implemented:

[0044] Collect the environmental data of the transformer no-load partial discharge site and the transformer's thermodynamic data, and use the convolutional neural network model to model the transformer's thermodynamic data;

[0045] The standard thermodynamic mathematical model of the target transformer is simulated by Simulink software to simulate the thermal characteristics of the target transformer under different working conditions, and based on the thermal characteristics information under different working conditions, the applicable working conditions of the target transformer at the transformer working position are determined;

[0046] Performing electric field simulation analysis on the target transformer, and determining the target electric field distribution state based on the electric field simulation analysis results;

[0047] A high-voltage variable-frequency power supply is designed to make the electric field distribution state in the qualified electric field distribution transformer equal to the target electric field distribution state, and a no-load partial discharge test is performed on the target transformer. The design of the high-voltage variable-frequency power supply is optimized based on the no-load partial discharge test results.

[0048] The present invention solves the technical defects existing in the background technology, and has the following beneficial effects: by performing thermodynamic simulation and electric field simulation on the transformer, the operating conditions of the transformer and the target electric field distribution state are determined, and based on the operating conditions of the transformer and the target electric field distribution state, the design optimization purpose of the high-frequency variable-voltage power supply is achieved in combination with the no-load partial discharge test of the transformer. The present invention can realize the miniaturization design of the high-voltage variable-frequency power supply by performing no-load partial discharge test on the transformer, ensuring that the equipment can still maintain high reliability in a compact space. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0050] Figure 1 A flow chart showing a method for miniaturization and application correction of a high-voltage variable-frequency power supply based on simulation analysis is shown;

[0051] Figure 2A flow chart of a method for designing a high-voltage variable-frequency power supply that makes the electric field distribution state in a qualified electric field distribution transformer equal to the target electric field distribution state is shown;

[0052] Figure 3 The program view of the high-voltage variable-frequency power supply miniaturization and application correction system based on simulation analysis is shown. DETAILED DESCRIPTION

[0053] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0054] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0055] Figure 1 A flowchart of a method for miniaturizing and applying a high-voltage variable-frequency power supply based on simulation analysis is shown, comprising the following steps:

[0056] S102: Collect environmental data of the transformer no-load partial discharge site and the transformer thermodynamic data, and perform transformer thermodynamic data modeling in combination with a convolutional neural network model;

[0057] S104: simulating the thermal characteristics of a standard thermodynamic mathematical model of a target transformer under different operating conditions through Simulink software, and determining the operating conditions applicable to the target transformer at the transformer working position based on the thermal characteristic information under different operating conditions;

[0058] S106: performing electric field simulation analysis on the target transformer, and determining a target electric field distribution state based on the electric field simulation analysis result;

[0059] S108: Design a high-voltage variable-frequency power supply that makes the electric field distribution state in the qualified electric field distribution transformer equal to the target electric field distribution state, and perform a no-load partial discharge test on the target transformer, and optimize the design of the high-voltage variable-frequency power supply based on the no-load partial discharge test results.

[0060] Furthermore, in a preferred embodiment of the present invention, the step S102 is specifically:

[0061] Determine the target operating altitude of the variable frequency power supply, and locate the on-site location for the transformer no-load partial discharge test within the target operating altitude of the variable frequency power supply, and calibrate it as the transformer operating position;

[0062] A meteorological platform is introduced, and predicted average environmental data of the transformer working position is retrieved from the meteorological platform and calibrated as target predicted environmental data;

[0063] An ultra-high voltage power transformer for no-load partial discharge experiment is introduced and calibrated as a target transformer, and the thermodynamic data range of the target transformer is calculated, wherein the thermodynamic data range of the target transformer includes the real-time temperature range, real-time current range and real-time voltage range when the target transformer is in operation, and a multi-layer feedforward neural network algorithm is introduced;

[0064] Performing time series conversion on the thermodynamic data of the target transformer to obtain the time series thermodynamic data of the target transformer, and combining the thermodynamic data of the target transformer and the target prediction environment data for training based on the multi-layer feedforward neural network algorithm to construct a thermodynamic mathematical model of the target transformer;

[0065] The learning rate and neural layer of the thermodynamic mathematical model of the target transformer are adjusted, and a regularization algorithm is introduced to introduce a penalty term into the thermodynamic mathematical model of the target transformer through the regularization algorithm;

[0066] Fourier series is introduced to decompose the model function of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are adjusted, and different sine functions and cosine functions are obtained. Polynomial regression is performed on the different sine functions and cosine functions to ensure that the model function of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are adjusted is close to the preset value, and the fitting state of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are output;

[0067] A standard fitting state is preset. If the fitting state is equal to the standard fitting state, a penalty term is output, the combined training of the thermodynamic data of the target transformer and the target predicted environmental data is stopped, and a standard thermodynamic mathematical model of the target transformer is output.

[0068] It should be noted that since it is necessary to conduct no-load partial discharge tests on transformers in high-altitude areas in order to realize the miniaturization design of high-voltage variable-frequency power supplies, the on-site location of the no-load partial discharge test of the transformer is first determined, and the environmental parameters of the location are determined, because the environmental parameters will affect the accuracy of the experiment. First, a thermodynamic simulation is performed on the transformer. The thermodynamic simulation requires the analysis and modeling of the thermodynamic data of the transformer first, and a multi-layer feedforward neural network is selected to realize the analysis and modeling of the thermodynamic data. The purpose is to predict the thermodynamic data of the transformer after working for a period of time. The purpose of time series conversion of the thermodynamic data of the target transformer is that since the data information corresponding to the thermodynamic data at different time points is different, replacing it with time series data can better improve the data accuracy. If the model is in a state of insufficient fitting or overfitting, it will affect the model prediction ability, so it is necessary to monitor the model training process and adjust the model parameters to prevent overfitting. The regularization method adjusts the model parameters by adding penalty terms to the original model to prevent the model from overfitting. The Fourier series fitting method decomposes the complex function into the sum of multiple simple sine and cosine functions, and then combines polynomial regression to approximate the preset data points. The fitting accuracy is improved by continuously increasing the order of the polynomial, and finally the standard thermodynamic mathematical model of the target transformer is obtained.

[0069] Furthermore, in a preferred embodiment of the present invention, the step S104 is specifically:

[0070] Obtaining the construction and use instructions of the target transformer, and determining the internal structure of the target transformer based on the construction and use instructions of the target transformer, wherein the internal structure of the target transformer includes the material types of internal components of the target transformer and the arrangement structure of the internal components;

[0071] The Simulink software is introduced, the internal structure of the target transformer is mapped in the Simulink software, a simulation model of the target transformer is constructed, and it is calibrated as a transformer simulation model;

[0072] Importing a standard thermodynamic mathematical model of a target transformer into the transformer simulation model, and converting the standard thermodynamic mathematical model of the target transformer into a Matlab function format in Simulink software to obtain a transformer thermodynamic simulation model;

[0073] Preset a target simulation time and a target simulation step in the transformer thermal power simulation model, run the transformer thermal power simulation model under different operating conditions based on the target simulation time and the target simulation step, and output thermal characteristic information of the transformer thermal power simulation model under different operating conditions;

[0074] The operating conditions are the combinations of different temperatures, currents and voltages within the thermodynamic data range of the target transformer, and the thermal characteristic information is the thermal state of the internal windings and core during the operation of the transformer thermodynamic simulation model;

[0075] Based on the big data network, the standard thermal state of the winding and the core in the transformer thermodynamic simulation model at the transformer working position is determined, and the standard thermal characteristic information is determined by reverse deduction. Based on the standard thermal characteristic information, the operating condition of the target transformer when the thermal characteristic information is equal to the standard thermal characteristic information is determined and calibrated as the target operating condition.

[0076] It should be noted that the purpose of building a transformer simulation model through Simulink software is to simulate the thermodynamic data of the transformer through the model, that is, to simulate the operation of the transformer and determine the thermal characteristic response of the transformer under different working conditions. Before building the model, it is necessary to determine the internal structure of the target transformer to ensure the accuracy of the model construction. The purpose of converting the standard thermodynamic mathematical model of the target transformer into the Matlab function format is to ensure that Simulink can accurately identify the trained neural network model. The thermal characteristic information output by the transformer under different working conditions is different. The thermal characteristic information is the thermal state of the internal winding and core during the operation of the transformer thermal dynamic simulation model. Because the transformer generates heat due to the current passing through the winding during operation, and the core loss is also converted into heat, causing the internal temperature of the transformer to rise. This temperature change will affect the performance and life of the transformer, and the winding and core are the main heat sources. The unevenness of heat distribution will affect the thermal stress distribution of the transformer, and then affect its structural strength and insulation performance. Therefore, it is necessary to determine the appropriate working conditions to ensure the safety of the transformer during operation. Finally, according to the standard information of thermal characteristics, the target working conditions of heat. The operating conditions act on the transformer, and the transformer is connected to the high-voltage variable-frequency power supply. Therefore, the operating conditions of the transformer output need to be considered when designing the high-voltage variable-frequency power supply. Therefore, the target operating conditions can be used to help the design of the high-voltage variable-frequency power supply.

[0077] Furthermore, in a preferred embodiment of the present invention, the step S106 is specifically as follows:

[0078] Obtaining electric field simulation software, and constructing an electric field simulation model of the target transformer based on the internal structure of the target transformer in the electric field simulation software;

[0079] Performing parameter setting in the electric field simulation model of the target transformer, wherein the parameter setting includes inputting target operating conditions, transformer working position, and target predicted environmental data to obtain a target electric field simulation model;

[0080] Running the target electric field simulation model, performing grid division in the target electric field simulation model, determining the distribution boundary of the electric field, and outputting an electric field distribution cloud map after the target electric field simulation model is run;

[0081] Calculate the electric field strength values ​​at different positions in the electric field distribution cloud map, and determine whether there is an electric field position with an electric field strength value greater than a preset value. If so, determine that the electric field distribution in the target transformer is uneven, and introduce the Bayesian network algorithm to perform defect traceability analysis on the target transformer's component operating parameters, determine the components that cause the uneven electric field distribution in the target transformer, and mark them as faulty transformer components;

[0082] The material of the faulty transformer element is optimized. If there is still an electric field position where the electric field strength value is greater than the preset value after the material optimization of the faulty transformer element, the insulation performance of the target transformer is judged to be unqualified, and a solution for insulation optimization of the target transformer is retrieved based on the big data network and output to obtain a transformer with qualified electric field distribution;

[0083] In the electric field simulation software, an electric field simulation model of the transformer with qualified electric field distribution is constructed, and the electric field distribution cloud map of the transformer with qualified electric field distribution is obtained to determine the electric field distribution state in the transformer with qualified electric field distribution and calibrate it as the target electric field distribution state.

[0084] It should be noted that after the thermodynamic simulation, the electric field simulation needs to be performed in the transformer. The purpose is to optimize the design of the high-voltage variable frequency power supply according to the results of the electric field simulation, avoid partial discharge, and meet the requirements of partial discharge. After obtaining the electric field distribution state of the transformer, such as the electric field strength and electric field direction, it is crucial to evaluate the insulation performance and partial discharge characteristics of the transformer. First, a target electric field simulation model is constructed for electric field simulation, and grid division is performed in the target electric field simulation model to determine the distribution boundary of the electric field to ensure the accuracy and efficiency of the electric field simulation. After the target electric field simulation model is used, an electric field distribution cloud map is obtained. The electric field direction and strength of the transformer exist on the electric field distribution cloud map. It is judged whether there is an electric field position with an electric field strength value greater than a preset value. If so, the Bayesian network algorithm is introduced to locate the component fault of the transformer, and the material of the faulty transformer component is optimized to ensure that there is no electric field position with an electric field strength value greater than the preset value, and the electric field distribution state of the transformer is normal. When the electric field distribution state of the transformer is normal, it proves that the transformer is fault-free. At this time, the high-voltage variable frequency power supply can be designed and optimized according to the electric field distribution state of the transformer to achieve the purpose of miniaturization.

[0085] Figure 2 A flow chart of a method for designing a high-voltage variable-frequency power supply that makes the electric field distribution state in a qualified electric field distribution transformer equal to the target electric field distribution state is shown, comprising the following steps:

[0086] S202: Design a high-voltage variable-frequency power supply that makes the electric field distribution state in the qualified electric field distribution transformer equal to the target electric field distribution state, and perform a no-load partial discharge test on the target transformer, and optimize the design of the high-voltage variable-frequency power supply based on the no-load partial discharge test results;

[0087] S204: Conduct a no-load partial discharge test in the power system update simulation model, and optimize the design of the high-voltage variable-frequency power supply based on the no-load partial discharge test results.

[0088] Furthermore, in a preferred embodiment of the present invention, the step S202 is specifically:

[0089] Constructing a blank simulation model of a high-voltage variable-frequency power supply and a simulation model of a transformer with qualified electric field distribution in an electric field simulation software, and connecting the blank simulation model of a high-voltage variable-frequency power supply with the simulation model of a transformer with qualified electric field distribution to obtain a power system simulation model;

[0090] Among them, the operating condition of the simulation model of the transformer with qualified electric field distribution is equal to the target operating condition;

[0091] Based on the target electric field distribution state, the model output parameters of the blank simulation model of the high-voltage variable-frequency power supply that enables the electric field distribution qualified transformer simulation model in the power system simulation model to output the target electric field distribution state are retrieved in the big data network, and are calibrated as the target model output parameters, and the control main circuit with the highest utilization rate is retrieved and calibrated as the target control main circuit;

[0092] Mapping the target control main circuit in a blank simulation model of a high-voltage variable-frequency power supply, and adjusting the model output parameters of the blank simulation model of the high-voltage variable-frequency power supply to be equal to the target model output parameters, to obtain a high-voltage variable-frequency target power supply simulation model;

[0093] Based on the high-voltage variable-frequency target power supply simulation model, the power system simulation model is updated to obtain a power system update simulation model, a no-load partial discharge test is performed within the power system update simulation model, and the high-voltage variable-frequency power supply is designed and optimized based on the no-load partial discharge test results.

[0094] It should be noted that a power system simulation model is constructed, and the power system simulation model is a model obtained by connecting a power supply and a transformer, which is used to design a high-voltage variable-frequency power supply. In the power system simulation model, the transformer operating conditions are set to the target operating conditions, and the electric field distribution state is set equal to the target electric field distribution state, because the operating conditions and the electric field distribution state are related to the power supply. The target model output parameters are obtained. Since the power supply affects the transformer output parameters, when the power system simulation model outputs the target model output parameters and the control main circuit is equal to the target control circuit, the power system simulation model can meet the transformer output target operating conditions and the target electric field distribution state, and obtain a high-voltage variable-frequency target power supply simulation model.

[0095] Furthermore, in a preferred embodiment of the present invention, the step S204 is specifically:

[0096] In the power system update simulation model, the secondary side of the transformer simulation model with qualified electric field distribution is opened, and after the secondary side is opened, a no-load test is performed on the transformer simulation model with qualified electric field distribution in the power system update simulation model, and after the no-load test, the no-load current and magnetic loss of the high-voltage variable frequency target power supply simulation model are calculated;

[0097] Conducting a partial discharge test on a simulation model of a transformer with qualified electric field distribution, and detecting whether the partial discharge amount of a simulation model of a high-voltage variable frequency target power supply after the partial discharge test on the simulation model of a transformer with qualified electric field distribution is maintained within a preset threshold;

[0098] If the no-load current and magnetic loss of the high-voltage variable frequency target power supply simulation model are maintained within the standard threshold, and the partial discharge amount of the high-voltage variable frequency target power supply simulation model after the partial discharge test of the electric field distribution qualified transformer simulation model is maintained within the preset threshold, the corresponding model parameters in the high-voltage variable frequency target power supply simulation model are output;

[0099] If the no-load current and magnetic loss of the high-voltage variable-frequency target power supply simulation model are not maintained within the standard threshold, or the partial discharge of the high-voltage variable-frequency target power supply simulation model after the partial discharge test on the electric field distribution qualified transformer simulation model is not maintained within the preset threshold, then through big data network retrieval, a solution is output that can maintain the no-load current and magnetic loss of the high-voltage variable-frequency target power supply simulation model within the standard threshold, and the partial discharge of the high-voltage variable-frequency target power supply simulation model after the partial discharge test on the electric field distribution qualified transformer simulation model is maintained within the preset threshold.

[0100] It should be noted that the high-voltage variable-frequency target power supply simulation model needs to undergo a no-load partial discharge experiment to determine whether its no-load current, magnetic loss, and partial discharge meet the requirements. If the no-load current, magnetic loss, and partial discharge meet the requirements, it proves that it is feasible to design a high-voltage variable-frequency power supply with the model parameters in the current high-voltage variable-frequency target power supply simulation model. If the requirements are not met, it is necessary to adjust the model of the high-voltage variable-frequency target power supply simulation model, including adjusting the discharge gap, optimizing the material selection, using low-temperature resistant and radiation-resistant materials, etc., to improve the life of the power supply. Achieve no-load current, magnetic loss, and partial discharge that meet the requirements.

[0101] like Figure 3As shown, the second aspect of the present invention further provides a high-voltage variable frequency power supply miniaturization and application correction system based on simulation analysis, the high-voltage variable frequency power supply miniaturization and application correction system includes a memory 31 and a processor 32, the memory 31 stores a high-voltage variable frequency power supply miniaturization and application correction method, and when the high-voltage variable frequency power supply miniaturization and application correction method is executed by the processor 32, the following steps are implemented:

[0102] Collect the environmental data of the transformer no-load partial discharge site and the transformer's thermodynamic data, and use the convolutional neural network model to model the transformer's thermodynamic data;

[0103] The standard thermodynamic mathematical model of the target transformer is simulated by Simulink software to simulate the thermal characteristics of the target transformer under different working conditions, and based on the thermal characteristics information under different working conditions, the applicable working conditions of the target transformer at the transformer working position are determined;

[0104] Performing electric field simulation analysis on the target transformer, and determining the target electric field distribution state based on the electric field simulation analysis results;

[0105] A high-voltage variable-frequency power supply is designed to make the electric field distribution state in the qualified electric field distribution transformer equal to the target electric field distribution state, and a no-load partial discharge test is performed on the target transformer. The design of the high-voltage variable-frequency power supply is optimized based on the no-load partial discharge test results.

[0106] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for miniaturization and application correction of high-voltage variable-frequency power supply based on simulation analysis, characterized in that: The following steps are involved: Collect the environmental data of the transformer no-load partial discharge site and the transformer's thermodynamic data, and use the convolutional neural network model to model the transformer's thermodynamic data; The standard thermodynamic mathematical model of the target transformer is simulated by Simulink software to simulate the thermal characteristics of the target transformer under different working conditions, and based on the thermal characteristics information under different working conditions, the applicable working conditions of the target transformer at the transformer working position are determined; Performing electric field simulation analysis on the target transformer, and determining the target electric field distribution state based on the electric field simulation analysis results; A high-voltage variable-frequency power supply is designed to make the electric field distribution state in the qualified electric field distribution transformer equal to the target electric field distribution state, and a no-load partial discharge test is performed on the target transformer. The design of the high-voltage variable-frequency power supply is optimized based on the no-load partial discharge test results.

2. The method for miniaturization and application correction of high-voltage variable-frequency power supply based on simulation analysis according to claim 1 is characterized in that: The environmental data of the transformer no-load partial discharge site and the thermodynamic data of the transformer are collected, and the thermodynamic data modeling of the transformer is performed in combination with the convolutional neural network model, specifically: Determine the target operating altitude of the variable frequency power supply, and locate the on-site location for the transformer no-load partial discharge test within the target operating altitude of the variable frequency power supply, and calibrate it as the transformer operating position; A meteorological platform is introduced, and predicted average environmental data of the transformer working position is retrieved from the meteorological platform and calibrated as target predicted environmental data; An ultra-high voltage power transformer for no-load partial discharge experiment is introduced and calibrated as a target transformer, and the thermodynamic data range of the target transformer is calculated, wherein the thermodynamic data range of the target transformer includes the real-time temperature range, real-time current range and real-time voltage range when the target transformer is in operation, and a multi-layer feedforward neural network algorithm is introduced; Performing time series conversion on the thermodynamic data of the target transformer to obtain the time series thermodynamic data of the target transformer, and combining the thermodynamic data of the target transformer and the target prediction environment data for training based on the multi-layer feedforward neural network algorithm to construct a thermodynamic mathematical model of the target transformer; The learning rate and neural layer of the thermodynamic mathematical model of the target transformer are adjusted, and a regularization algorithm is introduced to introduce a penalty term into the thermodynamic mathematical model of the target transformer through the regularization algorithm; Fourier series is introduced to decompose the model function of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are adjusted, and different sine functions and cosine functions are obtained. Polynomial regression is performed on the different sine functions and cosine functions to ensure that the model function of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are adjusted is close to the preset value, and the fitting state of the thermodynamic mathematical model of the target transformer after the learning rate and the neural layer are output; A standard fitting state is preset. If the fitting state is equal to the standard fitting state, a penalty term is output, the combined training of the thermodynamic data of the target transformer and the target predicted environmental data is stopped, and a standard thermodynamic mathematical model of the target transformer is output.

3. The method for miniaturization and application correction of high-voltage variable-frequency power supply based on simulation analysis according to claim 1 is characterized in that: The standard thermodynamic mathematical model of the target transformer is simulated by Simulink software to respond to the thermal characteristics under different working conditions, and based on the thermal characteristic information under different working conditions, the working conditions applicable to the target transformer at the transformer working position are determined, specifically: Obtaining the construction and use instructions of the target transformer, and determining the internal structure of the target transformer based on the construction and use instructions of the target transformer, wherein the internal structure of the target transformer includes the material types of internal components of the target transformer and the arrangement structure of the internal components; The Simulink software is introduced, the internal structure of the target transformer is mapped in the Simulink software, a simulation model of the target transformer is constructed, and it is calibrated as a transformer simulation model; Importing a standard thermodynamic mathematical model of a target transformer into the transformer simulation model, and converting the standard thermodynamic mathematical model of the target transformer into a Matlab function format in Simulink software to obtain a transformer thermodynamic simulation model; Preset a target simulation time and a target simulation step in the transformer thermal power simulation model, run the transformer thermal power simulation model under different operating conditions based on the target simulation time and the target simulation step, and output thermal characteristic information of the transformer thermal power simulation model under different operating conditions; The operating conditions are the combinations of different temperatures, currents and voltages within the thermodynamic data range of the target transformer, and the thermal characteristic information is the thermal state of the internal windings and core during the operation of the transformer thermodynamic simulation model; Based on the big data network, the standard thermal state of the winding and the core in the transformer thermodynamic simulation model at the transformer working position is determined, and the standard thermal characteristic information is determined by reverse deduction. Based on the standard thermal characteristic information, the operating condition of the target transformer when the thermal characteristic information is equal to the standard thermal characteristic information is determined and calibrated as the target operating condition.

4. The method for miniaturization and application correction of high-voltage variable-frequency power supply based on simulation analysis according to claim 1 is characterized in that: The electric field simulation analysis is performed on the target transformer, and the target electric field distribution state is determined based on the electric field simulation analysis result, specifically: Obtaining electric field simulation software, and constructing an electric field simulation model of the target transformer based on the internal structure of the target transformer in the electric field simulation software; Performing parameter setting in the electric field simulation model of the target transformer, wherein the parameter setting includes inputting target operating conditions, transformer working position, and target predicted environmental data to obtain a target electric field simulation model; Running the target electric field simulation model, performing grid division in the target electric field simulation model, determining the distribution boundary of the electric field, and outputting an electric field distribution cloud map after the target electric field simulation model is run; Calculate the electric field strength values ​​at different positions in the electric field distribution cloud map, and determine whether there is an electric field position with an electric field strength value greater than a preset value. If so, determine that the electric field distribution in the target transformer is uneven, and introduce the Bayesian network algorithm to perform defect traceability analysis on the target transformer's component operating parameters, determine the components that cause the uneven electric field distribution in the target transformer, and mark them as faulty transformer components; The material of the faulty transformer element is optimized. If there is still an electric field position where the electric field strength value is greater than the preset value after the material optimization of the faulty transformer element, the insulation performance of the target transformer is judged to be unqualified, and a solution for insulation optimization of the target transformer is retrieved based on the big data network and output to obtain a transformer with qualified electric field distribution; In the electric field simulation software, an electric field simulation model of the transformer with qualified electric field distribution is constructed, and the electric field distribution cloud map of the transformer with qualified electric field distribution is obtained to determine the electric field distribution state in the transformer with qualified electric field distribution and calibrate it as the target electric field distribution state.

5. The method for miniaturization and application correction of high-voltage variable-frequency power supply based on simulation analysis according to claim 1 is characterized in that: The design makes the electric field distribution state of the qualified transformer equal to the high-voltage variable-frequency power supply of the target electric field distribution state, and performs a no-load partial discharge test on the target transformer. Based on the no-load partial discharge test results, the design of the high-voltage variable-frequency power supply is optimized, specifically: Constructing a blank simulation model of a high-voltage variable-frequency power supply and a simulation model of a transformer with qualified electric field distribution in an electric field simulation software, and connecting the blank simulation model of a high-voltage variable-frequency power supply with the simulation model of a transformer with qualified electric field distribution to obtain a power system simulation model; Among them, the operating condition of the simulation model of the transformer with qualified electric field distribution is equal to the target operating condition; Based on the target electric field distribution state, the model output parameters of the blank simulation model of the high-voltage variable-frequency power supply that enables the electric field distribution qualified transformer simulation model in the power system simulation model to output the target electric field distribution state are retrieved in the big data network, and are calibrated as the target model output parameters, and the control main circuit with the highest utilization rate is retrieved and calibrated as the target control main circuit; Mapping the target control main circuit in a blank simulation model of a high-voltage variable-frequency power supply, and adjusting the model output parameters of the blank simulation model of the high-voltage variable-frequency power supply to be equal to the target model output parameters, to obtain a high-voltage variable-frequency target power supply simulation model; Based on the high-voltage variable-frequency target power supply simulation model, the power system simulation model is updated to obtain a power system update simulation model, a no-load partial discharge test is performed within the power system update simulation model, and the high-voltage variable-frequency power supply is designed and optimized based on the no-load partial discharge test results.

6. The method for miniaturization and application correction of high-voltage variable-frequency power supply based on simulation analysis according to claim 5 is characterized in that: The no-load partial discharge test is carried out in the power system update simulation model, and the design optimization of the high-voltage variable frequency power supply is carried out based on the no-load partial discharge test result, specifically: In the power system update simulation model, the secondary side of the transformer simulation model with qualified electric field distribution is opened, and after the secondary side is opened, a no-load test is performed on the transformer simulation model with qualified electric field distribution in the power system update simulation model, and after the no-load test, the no-load current and magnetic loss of the high-voltage variable frequency target power supply simulation model are calculated; Conducting a partial discharge test on a simulation model of a transformer with qualified electric field distribution, and detecting whether the partial discharge amount of a simulation model of a high-voltage variable frequency target power supply after the partial discharge test on the simulation model of a transformer with qualified electric field distribution is maintained within a preset threshold; If the no-load current and magnetic loss of the high-voltage variable frequency target power supply simulation model are maintained within the standard threshold, and the partial discharge amount of the high-voltage variable frequency target power supply simulation model after the partial discharge test of the electric field distribution qualified transformer simulation model is maintained within the preset threshold, the corresponding model parameters in the high-voltage variable frequency target power supply simulation model are output; If the no-load current and magnetic loss of the high-voltage variable-frequency target power supply simulation model are not maintained within the standard threshold, or the partial discharge of the high-voltage variable-frequency target power supply simulation model after the partial discharge test on the electric field distribution qualified transformer simulation model is not maintained within the preset threshold, then through big data network retrieval, a solution is output that can maintain the no-load current and magnetic loss of the high-voltage variable-frequency target power supply simulation model within the standard threshold, and the partial discharge of the high-voltage variable-frequency target power supply simulation model after the partial discharge test on the electric field distribution qualified transformer simulation model is maintained within the preset threshold.

7. A high-voltage variable-frequency power supply miniaturization and application correction system based on simulation analysis, characterized in that: The high-voltage variable-frequency power supply miniaturization and application correction system includes a memory and a processor, wherein the memory stores a high-voltage variable-frequency power supply miniaturization and application correction method program. When the high-voltage variable-frequency power supply miniaturization and application correction method program is executed by the processor, the high-voltage variable-frequency power supply miniaturization and application correction method steps as described in any one of claims 1 to 6 are implemented.

Images