A high-voltage switch mechanical characteristic parameter optimization platform

Abstract

The present invention discloses a high-voltage switch mechanical characteristic parameter optimization platform. The present invention relates to the technical field of high-voltage switch mechanical characteristic parameter optimization, and includes a parameter acquisition module for real-time acquisition of the mechanical characteristic parameters of the high-voltage switch, and a data analysis module for analyzing the correlation between the performance of the high-voltage switch and the mechanical characteristic parameters through an adaptive optimization algorithm based on the acquired mechanical characteristic parameters, and generating optimization reference data. This high-voltage switch mechanical characteristic parameter optimization platform can accurately determine the influence degree and mutual relationship of each parameter on the performance; based on the generated optimization reference data and adjustment reference data, parameters such as the opening time, closing time, and tripping time of the high-voltage switch can reach better values, thereby improving the comprehensive performance index, being able to comprehensively evaluate the parameter combination under various working conditions, being able to discover potential performance problems and adjust in a timely manner, so that the high-voltage switch always maintains good performance in a complex actual operating environment.

Description

Technical Field

[0001] The present invention relates to the technical field of optimizing mechanical characteristic parameters of high-voltage switches, and particularly to an optimization platform for mechanical characteristic parameters of high-voltage switches. Background Art

[0002] In the power system, high-voltage switches are one of the key devices to ensure the safe transmission and distribution of electric power; the mechanical characteristic parameters of high-voltage switches, such as opening time, closing time, tripping time, stroke, etc., play a crucial role in their performance and reliability; accurately adjusting and optimizing these mechanical characteristic parameters can improve the operation accuracy of high-voltage switches, extend their service life, and enhance the stability of the power system. However, at present, in the determination and adjustment of mechanical characteristic parameters of high-voltage switches, it mainly relies on traditional empirical methods and on-site commissioning; technicians usually initially set parameters according to past experience and some basic calculation formulas, but this method has great limitations.

[0003] On the one hand, high-voltage switches of different models and specifications have differences in structure, materials, and working environments, etc. It is difficult to accurately determine the optimal parameter combination suitable for a specific switch simply relying on experience; moreover, empirical methods are often relatively conservative and may not be able to fully exert the performance potential of high-voltage switches. On the other hand, although on-site commissioning can optimize parameters to a certain extent, this method is time-consuming and laborious, requires a large number of tests and adjustments, not only increases the cost and time investment, but also may cause unnecessary losses to the equipment; it is difficult to determine the best parameters at one time and often requires multiple repeated attempts, with low efficiency; that is to say, with the continuous development of the power system and the continuous improvement of the performance requirements for high-voltage switches, the existing parameter determination methods are already difficult to meet the actual needs; there is an urgent need for a technical means that can accurately predict the relationship between the mechanical characteristic parameters and performance of high-voltage switches, so as to quickly determine the optimal parameter combination during the design, commissioning, and operation processes, and improve the performance and reliability of high-voltage switches. Summary of the Invention

[0004] To solve the above technical problems, the present invention is realized through the following technical solutions: An optimization platform for mechanical characteristic parameters of high-voltage switches, including a parameter acquisition module for real-time acquisition of mechanical characteristic parameters of high-voltage switches, and the mechanical characteristic parameters include opening time, closing time, tripping time, stroke, contact stroke, opening and closing current, speed, and pressure, and these parameters comprehensively reflect the working state and performance of high-voltage switches;

[0005] A data analysis module, which is used to analyze the correlation between the performance of the high-voltage switch and the mechanical characteristic parameters based on the collected mechanical characteristic parameters through an adaptive optimization algorithm, and generate optimization reference data. The adaptive optimization algorithm includes a genetic algorithm and a particle swarm optimization algorithm to improve the prediction accuracy and optimization efficiency between the mechanical characteristic parameters and the performance of the high-voltage switch;

[0006] A parameter optimization module, which adopts a feedback mechanism. By monitoring the operating state of the high-voltage switch in real time and combining the optimization reference data, it adjusts the mechanical characteristic parameters to optimize the performance of the high-voltage switch and obtains adjustment reference data. The parameter optimization process realizes the automation of parameter adjustment through an iterative algorithm; among them, the feedback mechanism combines the actual operating data of the high-voltage switch under different environmental conditions to adjust the parameter weights of the optimization model to ensure that the optimization results adapt to different working environments;

[0007] A simulation test module, which is used to simulate the operation of the high-voltage switch based on the adjustment reference data, obtain simulation results, and evaluate the performance under this parameter combination;

[0008] A feedback adjustment module, which has a closed-loop control mechanism, and is used to further optimize and adjust the parameters according to the real-time simulation results and in combination with the preset performance targets to form optimal parameters.

[0009] Preferably, the simulation test module can simulate the working states of the high-voltage switch under various working conditions, including voltage fluctuations, mechanical wear, and temperature changes, evaluate the stability and performance of the mechanical characteristic parameters under different environmental conditions, and provide comprehensive parameter optimization feedback. The simulation test module integrates finite element analysis and multibody dynamics simulation technologies to accurately simulate the dynamic behavior of the high-voltage switch under different working conditions and provide simulation results for quantitative performance evaluation.

[0010] Preferably, the process for the data analysis module to obtain the optimization reference data is as follows:

[0011] Perform algorithm selection and model construction, and adopt an adaptive optimization algorithm including a genetic algorithm and a particle swarm optimization algorithm; these algorithms can generate adaptive optimization schemes for high-voltage switches of different models and specifications to improve the prediction accuracy and optimization efficiency between the mechanical characteristic parameters and the performance of the high-voltage switch;

[0012] Construct an optimization model: construct a multi-objective optimization model based on the collected mechanical characteristic parameters and the selected optimization algorithm; this model takes the performance of the high-voltage switch as the objective and the mechanical characteristic parameters as variables, and establishes a mathematical relationship model between the two;

[0013] First, train the model using historical data: Train the constructed model with continuously updated historical data; By enabling the model to learn a large amount of historical data, it can understand the internal relationship between different combinations of mechanical characteristic parameters and the performance of high-voltage switches, thereby improving the prediction ability of the model;

[0014] During the training process, through the method of cross-validation, evaluate the performance of the model under different parameter settings, continuously adjust the parameters of the model, select the optimal parameter combination, and obtain a trained multi-objective optimization model to improve the accuracy and stability of the model; Enable the model to better adapt to high-voltage switches of different models and specifications;

[0015] Based on the trained multi-objective optimization model, analyze the correlation between the performance of high-voltage switches and mechanical characteristic parameters; Determine the influence degree and mutual relationship of each mechanical characteristic parameter on the performance of high-voltage switches, and obtain the analysis results, such as which parameters have a greater impact on the breaking performance and which parameters have a strong correlation, etc.; According to the analysis results, generate optimization reference data; The optimization reference data includes information on the direction and approximate range of adjustment of each mechanical characteristic parameter to achieve better performance under the current state of the high-voltage switch, which can provide a reference basis for the parameter optimization module to achieve the optimization of the performance of high-voltage switches.

[0016] Preferably, the process of generating optimization reference data based on the machine learning-based multi-objective optimization model is as follows:

[0017] Let the performance index of the high-voltage switch be y, and the performance index y includes comprehensive performance indexes such as breaking reliability and closing stability. The mechanical characteristic parameter vector is x = [x 1 , x 2 ,..., x n , where x 1 is the breaking time, x 2 is the closing time, and so on. n is the number of mechanical characteristic parameters;

[0018] Then the multi-objective optimization model is expressed as: y = f(x, ω), where f is a prediction function constructed based on a machine learning algorithm, and ω is the parameter vector of the multi-objective optimization model. The parameter vector is continuously optimized and adjusted during the training process;

[0019] Obtain the optimal individual based on the genetic algorithm formula in the adaptive optimization algorithm, and obtain the optimal particle position based on the formula of the particle swarm optimization algorithm;

[0020] After the iteration terminates, based on the finally obtained optimal particle position and the optimal individual in the genetic algorithm, the optimal solution is selected therefrom, and the combination of mechanical characteristic parameters corresponding to the optimal solution is combined with the analysis of the multi-objective optimization model, such as the sensitivity analysis of the influence of parameters on performance, etc., to generate optimization reference data; for example, the adjustment direction and adjustment amplitude suggestions for each mechanical characteristic parameter are obtained, so that the parameter optimization module can perform subsequent parameter adjustment operations.

[0021] Preferably, the genetic algorithm formula in the adaptive optimization algorithm:

[0022] Coding and initializing the population: Through binary coding, each mechanical characteristic parameter x i , is encoded; it is set that each parameter is encoded as a binary string of length l, then the encoding of the individual is represented as p = [p 1 , p 2 ,..., p n×l , where p 1 to p l is the encoding of x 1 , p l+1 to p 2l is the encoding of x 2 , and so on; among them, the individual represents a set of mechanical characteristic parameter combinations;

[0023] Initializing the population P(0), randomly generating m individuals, that is, P(0) = {p 1 (0), p 2 (0),..., p m (0)};

[0024] Fitness evaluation: Calculate the fitness value F(p i ) of each individual. The fitness function is defined according to the performance index y of the high-voltage switch. F(p i ) = -|y i - y target | k , where y i is the predicted performance index value corresponding to the individual p i , y target is the expected performance target value, and k is a constant. Through such a definition, the larger the fitness value, the closer the individual is to the optimal solution;

[0025] Selection operation: Use the roulette wheel selection method to select parent individuals; the probability that the individual p i is selected

[0026] Crossover and mutation operations: Use single-point crossover to perform crossover operations on the selected parent individuals. Let the two parent individuals be p a and p b, randomly select a crossover point c (1 ≤ c ≤ n×l), then the offspring individuals p' a and p' b are:

[0027] p' a = [p a (1),..., p a (c), p b (c + 1),..., p b (n×l)], p' b = [p b (1),..., p b (c), p a (c + 1),..., p a (n×l)];

[0028] Perform a mutation operation on the offspring individuals after crossover, and change the binary values of some gene positions in the individuals with a certain mutation probability p m ; for a certain gene position p i in the individual p i,j , if the random number r < p m , where r is a random number between 0 and 1, then p i,j = 1 - p i,j ;

[0029] Update the population and the iteration termination condition: After selection, crossover, and mutation operations, a new generation of population P(t + 1) is obtained. Repeat the fitness evaluation, selection, crossover, and mutation operations until the global optimal solution no longer changes significantly, which means reaching the iteration termination condition and obtaining the optimal individual.

[0030] Preferably, the formula of the particle swarm optimization algorithm:

[0031] Each particle i is represented as a vector of mechanical characteristic parameters x i = [x i1 , x i2 ,..., x in , and at the same time each particle has a velocity vector v i = [v i1 , v i2 ,..., v in ; Initialize the position and velocity of the particles. The position is randomly initialized within the reasonable value range of the mechanical characteristic parameters, and the velocity is randomly initialized within a certain range;

[0032] Update the individual optimal and global optimal: Calculate the fitness value F(x i ) of each particle, where the fitness function is defined in the same way as in the genetic algorithm and is defined according to the performance indicators of the high-voltage switch; for each particle i, if Among them is the optimal position in the history of particle i, then Meanwhile, find the particle with the smallest fitness value among all particles, that is, the particle with the best performance, and its position is denoted as g best ;

[0033] Velocity and position update: The velocity and position update formulas of the particle are as follows:

[0034]

[0035] x ij (t + 1) = x ij (t) + v ij (t + 1);

[0036] where w is the inertia weight, c 1 and c 2 are learning factors, r 1 and r 2 are random numbers between 0 and 1, t is the number of iterations, j = 1, 2,..., n; v ij : the j - th dimension velocity of particle i; x ij : the j - th dimension position of particle i;

[0037] Repeat the operations of individual - optimal and global - optimal updates and velocity and position updates until the iteration terminates, that is, the global optimal solution no longer changes significantly, and the optimal particle position is obtained.

[0038] Preferably, the process by which the parameter optimization module obtains the adjusted reference data is as follows:

[0039] First, obtain the optimization reference data from the data analysis module. By real - time monitoring the operating state of the high - voltage switch, obtain the actual operating data of the current high - voltage switch. The actual operating data includes the real - time values of opening time, closing time, tripping time, stroke, contact stroke, switching current, speed, and pressure; The feedback mechanism adjusts the parameter weights of the optimization model in combination with the actual operating data of the high - voltage switch under different environmental conditions. Among them, different environmental conditions include voltage fluctuation, mechanical wear, and temperature change;

[0040] When it is found that mechanical characteristic parameters such as tripping time have a more significant impact on the switch performance in a high - temperature environment among environmental conditions, then increase the weight of this parameter in the current optimization process;

[0041] Suppose the original weight vector is ω = [ω 1 , ω 2 ,..., ω n , n is the number of mechanical characteristic parameters, and the weight vector adjusted according to environmental factors is ω′ = [ω′ 1 , ω′ 2,..., ω' n ;

[0042] Calculate the parameter adjustment amount: Based on the optimized reference data and the adjusted weights, calculate the adjustment amount Δx = [Δx 1 , Δx 2 ,..., Δx n for each mechanical characteristic parameter;

[0043] During the calculation, combine the gap between the current parameter value and the target value, the parameter weight, and the operating state of the high-voltage switch. The target value is obtained based on the optimized reference data;

[0044] Among them, the target value is the ideal value or value range preset for each mechanical characteristic parameter during the optimization process of the high-voltage switch mechanical characteristic parameters. These values are designed to make the performance of the high-voltage switch reach the best state. For example, for the mechanical characteristic parameter of breaking time, its target value may be set to 5 ms, which means that during the optimization process, it is expected to adjust the breaking time to about 5 ms to achieve the best breaking performance, such as minimizing the arc energy and ensuring fast and reliable circuit interruption;

[0045] For a certain mechanical characteristic parameter x i , its adjustment amount Δx i 's calculation formula Δx i = k × ω' i × (x i,target - x i,current ), where k is the adjustment coefficient, x i,target is the target value of this parameter obtained according to the optimized reference data, and x i,current is the current actual value;

[0046] Apply the calculated adjustment amount to the current mechanical characteristic parameter to obtain the adjusted parameter value x' = [x' 1 , x' 2 ,..., x' n , where x' i = x i,current + Δx i ;

[0047] Combine the adjusted parameter values and related adjustment information to generate adjusted reference data, such as adjustment amount, adjustment basis, etc. The adjusted reference data will be used as the input of the subsequent simulation test module to evaluate the performance of the high-voltage switch under this parameter combination, and at the same time provide basic data for the feedback adjustment module to further optimize the parameters.

[0048] Preferably, the process of the simulation test module obtaining the simulation result is as follows:

[0049] Based on finite element analysis and multibody dynamics simulation techniques, finite element analysis and multibody dynamics simulation models are constructed: Using the finite element analysis method to model and analyze the physical fields of electric field, magnetic field, stress and strain of high-voltage switches, and determine the changes in the physical state inside the high-voltage switches under different working conditions; At the same time, using multibody dynamics simulation techniques to construct the mechanical motion model of high-voltage switches, and simulate their dynamic behaviors during operation, including the motion trajectories of contacts, speed changes, and interactions between mechanical components; For example, for the contact system of high-voltage switches, calculate the electric field distribution and the forces between contacts under different voltage and current conditions through finite element analysis, and use multibody dynamics simulation to simulate the opening and closing action processes of contacts, including collisions, bounces during closing, and the effects of arc generation and extinction processes during opening on the motion of contacts, etc.; According to actual requirements, set the working state parameters of high-voltage switches under various working conditions, and the working state parameters include voltage fluctuations: such as setting different voltage amplitudes, frequency change ranges, mechanical wear: simulating the switch performance in different wear stages by changing the friction coefficient and wear degree parameters of mechanical components, temperature changes: setting the environmental temperature range and the temperature rise caused by the self-heating of the switch during operation; Use the adjustment reference data generated by the parameter optimization module as the input parameters of the simulation, and set the initial state and operating conditions of the high-voltage switch in the simulation model according to these parameters;

[0050] Start the simulation program, and based on the constructed finite element analysis and multibody dynamics simulation models, combined with the input adjustment reference data and the set working condition parameters, perform simulation calculations on the operation process of the high-voltage switch; During the simulation process, the model will calculate various state variables of the high-voltage switch at different times according to physical laws and mathematical algorithms, such as electric field strength, current distribution, displacements, speeds, accelerations of mechanical components, and interaction forces, etc.;

[0051] During the simulation calculation process, collect the data related to the performance of the high-voltage switch in real time to obtain simulation data, and the simulation data includes electrical performance data: such as breaking capacity, insulation performance, conduction resistance, mechanical performance data: such as contact wear amount, mechanical life, operation force change, and dynamic characteristic data: such as opening and closing speed curves, contact bounce time, vibration frequency; The collected simulation data is used for subsequent evaluation of the performance of the high-voltage switch; According to the collected simulation data, calculate the indicators used to evaluate the performance of the high-voltage switch; Integrate and analyze the calculated performance indicators to generate simulation results; The simulation results include the specific numerical values and performance curves of various performance indicators of the high-voltage switch under the given adjustment reference data and working condition parameters, such as speed-time curves, stress-strain curves; And performance evaluation conclusions, such as whether it meets the design requirements, and where there are performance advantages or deficiencies, etc.

[0052] Preferably, the process of the feedback adjustment module obtaining the optimal parameters mainly includes the following steps:

[0053] Receive the simulation results provided by the simulation test module. At the same time, obtain the performance goals preset by the user through the customization mechanism. Among them, the user sets the priority of specific performance indicators according to their own needs: including setting the opening reliability as the highest priority, followed by closing stability, and the parameter adjustment range: including, for example, specifying the upper and lower limits of the allowable adjustment of a certain mechanical characteristic parameter. These preset information reflect the expected performance of the high-voltage switch in different application scenarios. Compare and analyze each performance indicator in the simulation results with the preset performance goals, and calculate the deviation value of each performance indicator. For example, for the performance indicator of opening time, if the preset goal is that the opening time should be less than or equal to 10 ms under specific conditions, and the simulation results show that the actual opening time is 12 ms, then the deviation value is 2 ms. In this way, determine the gap between the performance of the high-voltage switch under the current parameter settings and the expected performance. Determine the parameter adjustment strategy according to the deviation value and the performance indicator priority. If the deviation of a high-priority performance indicator is large, for example, the opening reliability does not meet the requirements and the deviation is significant, then focus on adjusting the mechanical characteristic parameters that affect the opening reliability, such as opening speed, contact pressure, etc. At the same time, consider the limitations of the parameter adjustment range to avoid problems caused by adjusting beyond the reasonable range. For example, if the adjustment range of the opening speed is set to [5 m / s, 10 m / s], when the current speed is 7 m / s and the opening reliability needs to be improved, only adjust within this range, such as increasing to 8 m / s. Adjust the mechanical characteristic parameters according to the determined adjustment strategy. The adjustment amplitude and direction are based on the analysis of the simulation results and the guidance of the optimization algorithm.

[0054] Input the adjusted parameters into the simulation test module again for simulation testing to obtain new simulation results. Then repeat the above process of deviation analysis, adjustment strategy determination, and parameter adjustment, and continuously iterate and optimize until the termination conditions are met. The termination conditions can include that the deviations of all performance indicators are within the acceptable range, the number of iterations reaches the preset upper limit, or the performance improvement is not obvious after several consecutive iterations. After meeting the termination conditions, verify the finally obtained parameter combination. Ensure that this parameter combination can stably achieve good performance in various situations by conducting additional simulation tests under different working conditions or referring to actual operation experience. At the same time, check whether the parameters are within the preset adjustment range and whether they meet the actual constraint conditions of the physical characteristics and operating principles of the high-voltage switch. After verification, determine that the mechanical characteristic parameters at this time are the optimal parameters.

[0055] Preferably, the parameter acquisition module is also used for data adjustment, including archiving and version management of historical mechanical characteristic parameters, which ensures that historical data can be traced during multiple iterative optimizations of the platform and further analysis can be performed based on these data.

[0056] The present invention provides a high-voltage switch mechanical characteristic parameter optimization platform, which has the following beneficial effects:

[0057] First, this high-voltage switch mechanical characteristic parameter optimization platform can accurately determine the influence degree and mutual relationship of each parameter on performance by collecting various mechanical characteristic parameters of the high-voltage switch in real time and analyzing their correlation with performance using an adaptive optimization algorithm; based on the generated optimization reference data and adjustment reference data, parameters such as the opening time, closing time, and tripping time of the high-voltage switch can reach better values, thereby improving comprehensive performance indicators such as opening reliability and closing stability.

[0058] Second, this high-voltage switch mechanical characteristic parameter optimization platform can comprehensively evaluate parameter combinations under various working conditions through the simulation test module, discover potential performance problems and adjust them in a timely manner, so that the high-voltage switch always maintains good performance in a complex actual operating environment.

[0059] Third, this high-voltage switch mechanical characteristic parameter optimization platform uses a multi-objective optimization algorithm based on machine learning in the data analysis module, trains the model using continuously updated historical data, improves the prediction accuracy, and reduces the blindness in the optimization process; at the same time, the parameter optimization module adjusts the weights through a feedback mechanism in combination with real-time monitoring data and environmental factors, realizes the automation and intelligence of parameter adjustment, accelerates the optimization process, quickly finds a better parameter combination, and improves the efficiency of high-voltage switch performance optimization.

[0060] Fourth, this high-voltage switch mechanical characteristic parameter optimization platform can adapt to high-voltage switches of different models and specifications, generate adaptive optimization solutions according to their respective characteristics, and there is no need to develop optimization methods separately for each model, saving R & D costs and time, and enhancing the versatility and practicability of the platform; for example, whether it is a small-scale distribution switch or a large-scale transmission switch, the platform can effectively optimize according to its specific parameters and performance requirements. Brief Description of the Drawings

[0061] Figure 1 It is a flow block diagram of a high-voltage switch mechanical characteristic parameter optimization platform of the present invention. Detailed Embodiment

[0062] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are given for purposes of illustration and description, and are not exhaustive or limit the present invention to the disclosed form. Many modifications and variations are obvious to those of ordinary skill in the art. The embodiments are selected and described to better illustrate the principles and practical applications of the present invention, and enable those of ordinary skill in the art to understand the present invention and design various embodiments with various modifications suitable for specific purposes.

[0063] As Figure 1 shown, the present invention provides a technical solution: a high-voltage switch mechanical characteristic parameter optimization platform, including a parameter acquisition module for real-time acquisition of the mechanical characteristic parameters of the high-voltage switch. The mechanical characteristic parameters include opening time, closing time, tripping time, stroke, contact stroke, opening and closing current, speed, and pressure. These parameters comprehensively reflect the working state and performance of the high-voltage switch.

[0064] A data analysis module for analyzing the correlation between the performance of the high-voltage switch and the mechanical characteristic parameters based on the acquired mechanical characteristic parameters through an adaptive optimization algorithm to generate optimization reference data. The adaptive optimization algorithm includes a genetic algorithm and a particle swarm optimization algorithm to improve the prediction accuracy and optimization efficiency between the mechanical characteristic parameters and the performance of the high-voltage switch.

[0065] A parameter optimization module that adopts a feedback mechanism. By real-time monitoring the operating state of the high-voltage switch and combining the optimization reference data, it adjusts the mechanical characteristic parameters to optimize the performance of the high-voltage switch and obtains adjustment reference data. The parameter optimization process realizes the automation of parameter adjustment through an iterative algorithm. Among them, the feedback mechanism combines the actual operating data of the high-voltage switch under different environmental conditions to adjust the parameter weights of the optimization model to ensure that the optimization results adapt to different working environments.

[0066] A simulation test module for simulating the operating conditions of the high-voltage switch based on the adjustment reference data to obtain simulation results and evaluate the performance under this parameter combination.

[0067] A feedback adjustment module with a closed-loop control mechanism for further optimizing and adjusting the parameters according to the real-time simulation results in combination with the preset performance targets to form optimal parameters.

[0068] The simulation test module can simulate the working state of the high-voltage switch under various working conditions, including voltage fluctuations, mechanical wear, and temperature changes, evaluate the stability and performance of the mechanical characteristic parameters under different environmental conditions, and provide comprehensive parameter optimization feedback. The simulation test module integrates finite element analysis and multibody dynamics simulation technologies to accurately simulate the dynamic behavior of the high-voltage switch under different working conditions and provide quantitative performance evaluation simulation results.

[0069] The process by which the data analysis module obtains the optimization reference data is as follows:

[0070] Perform algorithm selection and model construction, and adopt an adaptive optimization algorithm including a genetic algorithm and a particle swarm optimization algorithm; these algorithms can generate adaptive optimization schemes for high-voltage switches of different models and specifications to improve the prediction accuracy and optimization efficiency between mechanical characteristic parameters and the performance of high-voltage switches;

[0071] Construct an optimization model: construct a multi-objective optimization model based on the collected mechanical characteristic parameters and the selected optimization algorithm; this model takes the performance of the high-voltage switch as the objective and the mechanical characteristic parameters as variables, and establishes a mathematical relationship model between the two;

[0072] Model training and optimization:

[0073] First, use historical data to train the model: use continuously updated historical data to train the constructed model; by enabling the model to learn a large amount of historical data, it can understand the internal relationship between different combinations of mechanical characteristic parameters and the performance of high-voltage switches, thereby improving the prediction ability of the model;

[0074] Then optimize the model parameters: during the training process, through the method of cross-validation, evaluate the performance of the model under different parameter settings, continuously adjust the parameters of the model, select the optimal parameter combination, and obtain the trained multi-objective optimization model to improve the accuracy and stability of the model; enable the model to better adapt to high-voltage switches of different models and specifications;

[0075] Analyze the relevance: based on the trained multi-objective optimization model, analyze the relevance between the performance of the high-voltage switch and the mechanical characteristic parameters; determine the influence degree and mutual relationship of each mechanical characteristic parameter on the performance of the high-voltage switch, and obtain the analysis results, such as which parameters have a greater impact on the breaking performance and which parameters have a strong correlation; according to the analysis results, generate optimization reference data; the optimization reference data includes information on the direction and approximate range of adjustment of each mechanical characteristic parameter to achieve better performance under the current state of the high-voltage switch, which can provide a reference basis for the parameter optimization module to optimize the performance of the high-voltage switch.

[0076] The process of generating optimization reference data based on a machine learning-based multi-objective optimization model is as follows:

[0077] Let the performance index of the high-voltage switch be y, and the performance index y includes comprehensive performance indexes such as breaking reliability and closing stability. The mechanical characteristic parameter vector is x = [x 1 , x 2 ,..., x n , where x 1 is the breaking time, x 2is the closing time, and so on. n is the number of mechanical characteristic parameters;

[0078] Then the multi-objective optimization model is expressed as: y = f(x, ω), where f is a prediction function constructed based on a machine learning algorithm, and ω is the parameter vector of the multi-objective optimization model. The parameter vector is continuously optimized and adjusted during the training process;

[0079] The optimal individual is obtained based on the genetic algorithm formula in the adaptive optimization algorithm, and the optimal particle position is obtained based on the formula of the particle swarm optimization algorithm;

[0080] After the iteration termination, according to the finally obtained optimal particle position and the optimal individual in the genetic algorithm, the optimal solution is selected therefrom, and the mechanical characteristic parameter combination corresponding to the optimal solution is combined with the analysis of the multi-objective optimization model, such as the sensitivity analysis of the parameter's influence on the performance, etc., to generate optimization reference data; for example, the adjustment direction and adjustment amplitude suggestions for each mechanical characteristic parameter are obtained, so that the parameter optimization module can perform subsequent parameter adjustment operations.

[0081] Genetic algorithm formula in the adaptive optimization algorithm:

[0082] Coding and initializing the population: Through binary coding, each mechanical characteristic parameter x i , is encoded; it is set that each parameter is encoded as a binary string of length l, then the encoding of the individual is expressed as p = [p 1 , p 2 ,..., p n×l , where p 1 to p l is the encoding of x 1 , p l+1 to p 2l is the encoding of x 2 , and so on; among them, the individual represents a set of mechanical characteristic parameter combinations;

[0083] Initializing the population P(0), randomly generating m individuals, that is, P(0) = {p 1 (0), p 2 (0),..., p m (0)};

[0084] Fitness evaluation: Calculate the fitness value F(p i ) of each individual. The fitness function is defined according to the performance index y of the high-voltage switch. F(p i ) = -|y i - y target | k , where y i is the predicted performance index value corresponding to the individual p i , y targetis the expected performance target value, and k is a constant. With such a definition, the larger the fitness value, the closer the individual is to the optimal solution;

[0085] Selection operation: The roulette wheel selection method is used to select parental individuals; individual p i The probability of being selected

[0086] Crossover and mutation operations: The single-point crossover method is used to perform crossover operations on the selected parental individuals. Let the two parental individuals be p a and p b , randomly select the crossover point c (1 ≤ c ≤ n × l), then the offspring individuals p' a and p' b are:

[0087] p' a = [p a (1),..., p a (c), p b (c + 1),..., p b (n × l)], p' b = [p b (1),..., p b (c), p a (c + 1),..., p a (n × l)];

[0088] Perform mutation operations on the offspring individuals after crossover, and change the values of some gene bit binary bits in the individuals with a certain mutation probability p m ; For a certain gene bit p i in the individual p i,j , if the random number r < p m , where r is a random number between 0 and 1, then p i,j = 1 - p i,j ;

[0089] Update the population and iteration termination condition: After selection, crossover, and mutation operations, a new generation of population P(t + 1) is obtained. Repeat the fitness evaluation, selection, crossover, and mutation operations until the global optimal solution no longer changes significantly, which is considered to reach the iteration termination condition, and the optimal individual is obtained.

[0090] The formula of the particle swarm optimization algorithm:

[0091] Particle representation and initialization: Each particle i is represented as a vector of mechanical characteristic parameters x i = [x i1 , x i2 ,..., x in , and at the same time each particle has a velocity vector v i= [v i1 , v i2 ,..., v in ; Initialize the positions and velocities of the particles. The positions are randomly initialized within the reasonable value range of the mechanical characteristic parameters, and the velocities are randomly initialized within a certain range;

[0092] Individual best and global best update: Calculate the fitness value F(x i ), where the definition of the fitness function is the same as in the genetic algorithm and is defined according to the performance indicators of the high-voltage switch; For each particle i, if where is the best position in the history of particle i, then Meanwhile, find the particle with the minimum fitness value among all particles, that is, the particle with the best performance, and its position is denoted as g best ;

[0093] Velocity and position update: The velocity and position update formulas of the particles are:

[0094]

[0095] x ij (t + 1) = x ij (t) + v ij (t + 1);

[0096] where w is the inertia weight, c 1 and c 2 are the learning factors, r 1 and r 2 are random numbers between 0 and 1, t is the number of iterations, j = 1, 2,..., n; v ij : The j-th dimension velocity of particle i; x ij : The j-th dimension position of particle i;

[0097] Iteration termination condition: Repeat the operations of individual best and global best update and velocity and position update until the iteration terminates, that is, the global optimal solution no longer changes significantly, and the optimal particle position is obtained.

[0098] The process by which the parameter optimization module obtains the adjusted reference data is as follows:

[0099] First, obtain the optimization reference data from the data analysis module. By real-time monitoring the operating state of the high-voltage switch, obtain the actual operating data of the current high-voltage switch. The actual operating data includes the real-time values of the opening time, closing time, tripping time, stroke, contact stroke, switching current, velocity, and pressure; The feedback mechanism adjusts the parameter weights of the optimization model in combination with the actual operating data of the high-voltage switch under different environmental conditions, where different environmental conditions include voltage fluctuations, mechanical wear, and temperature changes;

[0100] When it is found that mechanical characteristic parameters such as the opening time have a more significant impact on the switch performance in a high-temperature environment among environmental conditions, then increase the weight of this parameter in the current optimization process;

[0101] Let the original weight vector be ω = [ω 1 , ω 2 ,..., ω n , where n is the number of mechanical characteristic parameters, and the weight vector adjusted according to environmental factors is ω' = [ω' 1 , ω' 2 ,..., ω' n ;

[0102] Calculate the parameter adjustment amount: Based on the optimization reference data and the adjusted weights, calculate the adjustment amount Δx = [Δx 1 , Δx 2 ,..., Δx n for each mechanical characteristic parameter;

[0103] During the calculation process, combine the gap between the current parameter value and the target value, the parameter weight, and the operating state of the high-voltage switch. The target value is obtained based on the optimization reference data;

[0104] It should be further noted that in the specific implementation process, the target value is the ideal value or value range preset for each mechanical characteristic parameter during the optimization process of the mechanical characteristic parameters of the high-voltage switch. These values are intended to make the performance of the high-voltage switch reach the best state; for example, for the mechanical characteristic parameter of the opening time, its target value may be set to 5 ms, which means that during the optimization process, it is expected to adjust the opening time to about 5 ms to achieve the best opening performance, such as minimizing the arc energy and ensuring a fast and reliable circuit interruption;

[0105] And the setting of the target value is based on the design requirements, operating standards, and actual application scenario requirements of the high-voltage switch; different application scenarios may have different focuses on the performance of the high-voltage switch, resulting in differences in the target value; for example, in a scenario with extremely high requirements for power supply continuity, the target value of the closing time may be set shorter to reduce the power outage time; while in some scenarios with high requirements for equipment life, the target value of the parameters related to contact wear may be set at a lower level to extend the service life of the equipment;

[0106] The optimization reference data is generated by the data analysis module through adaptive optimization algorithms, including genetic algorithms and particle swarm optimization algorithms, after analyzing the correlation between the performance of the high-voltage switch and the mechanical characteristic parameters; it contains information on the direction and approximate amplitude of adjustment for each mechanical characteristic parameter to achieve better performance under the current state of the high-voltage switch; this information provides a reference basis for determining the target value;

[0107] For example, the optimized reference data indicates that the current opening time is too long and needs to be shortened. Based on historical data and algorithm analysis, a reasonable range for shortening is given. Based on this, when setting the target value of the opening time, this recommended adjustment direction and range will be referred to, and combined with the actual operating limits and performance requirements of the high-voltage switch, a specific target value will be determined, such as shortening from the current 8 ms to 5 ms.

[0108] That is to say, the target value is to convert the general adjustment suggestions in the optimized reference data into specific and operable numerical targets. The optimized reference data is a relatively abstract adjustment trend and range, while the target value clarifies the precise state that each mechanical characteristic parameter is expected to reach during the optimization process. In the parameter optimization module, when calculating the parameter adjustment amount, it is precisely based on these target values, combined with the current parameter values and other factors, such as the weight coefficient, to determine the specific adjustment range.

[0109] For a certain mechanical characteristic parameter x i , its adjustment amount Δx i has the calculation formula Δx i = k × ω′ i × (x i,target - x i,current ), where k is the adjustment coefficient, x i,target is the target value of this parameter obtained according to the optimized reference data, and x i,current is the current actual value.

[0110] Apply the calculated adjustment amount to the current mechanical characteristic parameter to obtain the adjusted parameter value x′ = [x′ 1 , x′ 2 ,..., x′ n , where x′ i = x i,current + Δx i ;

[0111] Combine the adjusted parameter values and related adjustment information to generate adjusted reference data, such as the adjustment amount, adjustment basis, etc. The adjusted reference data will be used as the input for the subsequent simulation test module to evaluate the performance of the high-voltage switch under this parameter combination, and at the same time provide basic data for the feedback adjustment module to further optimize the parameters. When generating the adjusted reference data, operations such as format conversion and adding time stamps will also be performed on the data to facilitate the subsequent processing and data management of the system.

[0112] For example, the format of the adjusted reference data is {[x′ 1 , x′ 2 ,..., x′ n , [Δx 1 , Δx 2,..., Δx n , t, reason}, where t is the timestamp and reason is the reason for adjustment, such as: calculated after adjusting the weight according to the optimization reference data and environmental factors;

[0113] It should be further noted that in the specific implementation process, through the above process, the parameter optimization module can generate reasonable adjustment reference data based on the information provided by the data analysis module and the actual operation conditions of the high-voltage switch, providing strong support for the optimization of the high-voltage switch performance.

[0114] The process by which the simulation test module obtains the simulation results is as follows:

[0115] Based on finite element analysis and multi-body dynamics simulation technology, construct a finite element analysis and multi-body dynamics simulation model: Use the finite element analysis method to model and analyze the physical fields of the electric field, magnetic field, stress and strain of the high-voltage switch, and determine the physical state changes inside the high-voltage switch under different working conditions; At the same time, use multi-body dynamics simulation technology to construct a mechanical motion model of the high-voltage switch, and simulate its dynamic behavior during operation, including the movement trajectory, speed change, and interaction between mechanical components of the contact; For example, for the contact system of the high-voltage switch, calculate the electric field distribution and the force on the contacts under different voltage and current conditions through finite element analysis, and use multi-body dynamics simulation to simulate the opening and closing action process of the contacts, including the collision, bounce during closing, and the influence of the arc generation and extinction process during opening on the contact movement, etc.; According to actual needs, set the working state parameters of the high-voltage switch under various working conditions. The working state parameters include voltage fluctuation: such as setting different voltage amplitudes, frequency change ranges, mechanical wear: simulating the switch performance in different wear stages by changing the friction coefficient and wear degree parameters of mechanical components, temperature change: setting the environmental temperature range and the temperature rise caused by the high-voltage switch's own heat generation during operation; The setting of these working conditions is aimed at comprehensively evaluating various complex situations that the high-voltage switch may encounter in actual operation to ensure that the optimized parameters can make the high-voltage switch operate stably and reliably under different environmental conditions; Use the adjustment reference data generated by the parameter optimization module as the input parameters of the simulation, and set the initial state and operation conditions of the high-voltage switch in the simulation model according to these parameters;

[0116] Start the simulation program, and based on the constructed finite element analysis and multi-body dynamics simulation model, combined with the input adjustment reference data and the set working conditions, perform simulation calculation on the operation process of the high-voltage switch; During the simulation process, the model will calculate various state variables of the high-voltage switch at different times according to physical laws and mathematical algorithms, such as electric field strength, current distribution, displacement, speed, acceleration of mechanical components, and interaction forces, etc.

[0117] During the simulation calculation process, data related to the performance of the high-voltage switch is collected in real time to obtain simulation data. The simulation data includes electrical performance data such as breaking capacity, insulation performance, and conduction resistance, mechanical performance data such as contact wear amount, mechanical life, and operation force change, and dynamic characteristic data such as opening and closing speed curves, contact bounce time, and vibration frequency. The collected simulation data is used for subsequent evaluation of the performance of the high-voltage switch. According to the collected simulation data, indicators for evaluating the performance of the high-voltage switch are calculated. For example, for the breaking performance, indicators such as the accuracy of the breaking time, the arc energy during the breaking process, and the insulation recovery voltage between contacts after breaking can be calculated. For the closing performance, indicators such as the stability of the closing speed, the collision energy of the contacts during closing, and the stability of the contact resistance after closing can be calculated. For the mechanical performance, indicators such as the stress level of mechanical components, the predicted fatigue life, and the motion smoothness of the operating mechanism are calculated. The calculated performance indicators are integrated and analyzed to generate simulation results.

[0118] It should be further noted that in the specific implementation process, the simulation results include the specific numerical values and performance curves of various performance indicators of the high-voltage switch under given adjustment reference data and working conditions, such as speed-time curves and stress-strain curves; and performance evaluation conclusions, such as whether the design requirements are met, and where there are performance advantages or disadvantages. These simulation results will provide a basis for the feedback adjustment module to further optimize the mechanical characteristic parameters of the high-voltage switch, and at the same time provide important references for the design, manufacturing, operation, and maintenance of the high-voltage switch.

[0119] The process for the feedback adjustment module to obtain the optimal parameters mainly includes the following steps:

[0120] Receive the simulation results provided by the simulation test module. At the same time, obtain the performance goals preset by the user through the customization mechanism. Among them, the user sets specific performance indicator priorities according to their own needs, including setting the breaking reliability as the highest priority, followed by the closing stability, and parameter adjustment ranges, including, for example, specifying the upper and lower limits of the allowable adjustment of a certain mechanical characteristic parameter. These preset information reflect the user's expected performance of the high-voltage switch in different application scenarios.

[0121] Compare and analyze the various performance indicators in the simulation results with the preset performance goals, and calculate the deviation value of each performance indicator. For example, for the performance indicator of the breaking time, if the preset goal is that the breaking time should be less than or equal to 10 ms under specific conditions, and the simulation results show that the actual breaking time is 12 ms, then the deviation value is 2 ms. In this way, determine the gap between the performance of the high-voltage switch and the expected performance under the current parameter settings. According to the deviation value and the performance indicator priorities, determine the parameter adjustment strategy.

[0122] If the deviation of a certain high-priority performance indicator is large, for example, the opening reliability fails to meet the requirements and the deviation is significant, then the mechanical characteristic parameters that affect the opening reliability, such as the opening speed, contact pressure, etc., will be focused on for adjustment; at the same time, considering the limitations of the parameter adjustment range, avoid adjusting beyond the reasonable range and causing other problems; for example, if the adjustment range of the opening speed is set to [5m / s, 10m / s], when the current speed is 7m / s and the opening reliability needs to be improved, it can only be adjusted within this range, such as increasing to 8m / s; according to the determined adjustment strategy, adjust the mechanical characteristic parameters; the adjustment amplitude and direction are based on the analysis of the simulation results and the guidance of the optimization algorithm; for example, if it is found that the closing time is too long, the value of the mechanical characteristic parameter of the closing speed may be appropriately increased, but the adjustment amplitude will comprehensively consider other relevant factors, such as mechanical shock, wear, etc.; during the adjustment process, the method in a similar parameter optimization module will be adopted, and the specific adjustment amount of each parameter will be calculated in combination with the weight coefficients (these weight coefficients may be dynamically adjusted according to the performance indicator deviation situation and priority).

[0123] Input the adjusted parameters into the simulation test module again for simulation testing to obtain new simulation results; then repeat the above process of deviation analysis, determination of adjustment strategy, and parameter adjustment, and continuously iterate and optimize until the termination condition is met; the termination condition may include that the deviations of all performance indicators are within the acceptable range, the number of iterations reaches the preset upper limit, or the performance improvement is not obvious after several consecutive iterations; after meeting the termination condition, verify the finally obtained parameter combination; ensure that this parameter combination can stably achieve good performance in various situations through additional simulation testing under different working conditions or referring to actual operation experience; at the same time, check whether the parameters are within the preset adjustment range and whether they meet the actual constraint conditions of the physical characteristics and operating principles of the high-voltage switch; after verification, determine that the mechanical characteristic parameters at this time are the optimal parameters; these optimal parameters will be provided as the final result to the high-voltage switch manufacturing system or operation and maintenance personnel to guide the design improvement of the high-voltage switch, parameter setting during the manufacturing process, or parameter adjustment during the operation process, so as to optimize the performance of the high-voltage switch, meet the user's needs, and ensure reliable operation in different application scenarios.

[0124] It should be further noted that in the specific implementation process, through the above series of steps, the feedback adjustment module can continuously adjust and optimize the mechanical characteristic parameters according to the simulation results and the user's preset performance goals, and finally obtain the optimal parameters that meet the requirements, realizing the continuous improvement of the performance of the high-voltage switch.

[0125] The parameter acquisition module is also used for data adjustment, including archiving and version management of historical mechanical characteristic parameters, which ensures that historical data can be traced during multiple iterations and optimizations of the platform, and further analysis can be performed based on these data. At the same time, it has the ability to deeply analyze multi-dimensional data. By methods such as clustering analysis and correlation analysis, hidden parameter correlations are discovered, providing more accurate inputs for subsequent optimization algorithms.

[0126] It should be further noted that in the specific implementation process, by real-time collecting various mechanical characteristic parameters of the high-voltage switch and using the adaptive optimization algorithm to analyze their correlation with performance, the influence degree and mutual relationship of each parameter on performance can be accurately determined. Based on the generated optimization reference data and adjustment reference data, parameters such as the opening time, closing time, and tripping time of the high-voltage switch can reach better values, thereby improving comprehensive performance indicators such as opening reliability and closing stability. For example, the optimized opening time is shorter and more stable, which can effectively reduce the arc energy, reduce the ablation of the switch contacts, extend the service life of the equipment, and at the same time ensure that the circuit is quickly and reliably cut off in case of a fault, guaranteeing the safe and stable operation of the power system.

[0127] It should be further noted that in the specific implementation process, by comprehensively evaluating the parameter combinations under various working conditions through the simulation test module, potential performance problems can be discovered and adjusted in a timely manner, enabling the high-voltage switch to always maintain good performance in a complex actual operating environment. For example, in the case of large voltage fluctuations, by optimizing the parameters, it is ensured that the insulation performance of the switch is not affected, avoiding faults such as insulation breakdown caused by voltage fluctuations. And this platform can adapt to high-voltage switches of different models and specifications, generating adaptive optimization solutions according to their respective characteristics, without the need to develop optimization methods separately for each model, saving R & D costs and time, and enhancing the versatility and practicality of the platform. For example, whether it is a small distribution switch or a large transmission switch, the platform can effectively optimize according to its specific parameters and performance requirements.

[0128] Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art and related fields without creative efforts shall fall within the protection scope of the present invention. Structures, devices, and operation methods not specifically described and explained in the present invention shall be implemented according to the conventional means in the art without special instructions and limitations.

Claims

1. A high-voltage switch mechanical characteristic parameter optimization platform, characterized in that: include: A parameter acquisition module is used to acquire the mechanical characteristic parameters of the high-voltage switch in real time, wherein the mechanical characteristic parameters include breaking time, closing time, opening time, stroke, contact stroke, opening and closing current, speed and pressure; The data analysis module is used to analyze the correlation between the high-voltage switch performance and the mechanical characteristic parameters based on the collected mechanical characteristic parameters through an adaptive optimization algorithm to generate optimized reference data, wherein the adaptive optimization algorithm includes a genetic algorithm and a particle swarm optimization algorithm; the process of obtaining the optimized reference data by the data analysis module is as follows: Perform algorithm selection and model building, using adaptive optimization algorithms including genetic algorithm and particle swarm optimization algorithm; Constructing an optimization model: Constructing a multi-objective optimization model based on the collected mechanical characteristic parameters and the selected optimization algorithm; the model takes the high-voltage switch performance as the target and the mechanical characteristic parameters as the variables, and establishes a mathematical relationship model between the two; Use the continuously updated historical data to train the constructed model; During the training process, the performance of the model under different parameter settings is evaluated through the cross-validation method, the parameters of the model are continuously adjusted, and the optimal parameter combination is selected to obtain a trained multi-objective optimization model; Based on the trained multi-objective optimization model, analyze the correlation between the high-voltage switch performance and mechanical characteristic parameters; determine the influence of each mechanical characteristic parameter on the high-voltage switch performance and the relationship between them, obtain the analysis results, and generate optimization reference data based on the analysis results; The parameter optimization module adopts a feedback mechanism to adjust the mechanical characteristic parameters by real-time monitoring the operating status of the high-voltage switch in combination with the optimization reference data to obtain the adjustment reference data; wherein the feedback mechanism adjusts the parameter weights of the optimization model in combination with the actual operating data of the high-voltage switch under different environmental conditions; A simulation test module, used to simulate the operation of the high-voltage switch based on the adjusted reference data to obtain a simulation result; The feedback adjustment module has a closed-loop control mechanism, which is used to further optimize and adjust parameters to form optimal parameters based on real-time simulation results combined with preset performance goals.

2. A high-voltage switch mechanical characteristic parameter optimization platform according to claim 1, characterized in that: The simulation test module can simulate the working state of the high-voltage switch under various working conditions, including voltage fluctuations, mechanical wear, and temperature changes. The simulation test module integrates finite element analysis and multi-body dynamics simulation technology to accurately simulate the dynamic behavior of the high-voltage switch under different working conditions and provide simulation results for quantitative performance evaluation.

3. A high-voltage switch mechanical characteristic parameter optimization platform according to claim 2, characterized in that: The process of generating optimization reference data based on the multi-objective optimization model of machine learning is as follows: Assume that the performance index of the high-voltage switch is y, which includes the breaking reliability and closing stability. The mechanical characteristic parameter vector is x = [x1, x2, ..., x n ], where x1 is the breaking time, x2 is the closing time, and so on, and n is the number of mechanical characteristic parameters; Then the multi-objective optimization model is expressed as: y = f(x, ω) where f is the prediction function built based on the machine learning algorithm, and ω is the parameter vector of the multi-objective optimization model; The optimal individual is obtained based on the genetic algorithm formula in the adaptive optimization algorithm, and the optimal particle position is obtained based on the particle swarm optimization algorithm formula; After the iteration is terminated, the optimal solution is selected based on the final optimal particle position and the optimal individual in the genetic algorithm. The mechanical characteristic parameters corresponding to the optimal solution are combined and combined with the analysis of the multi-objective optimization model to generate optimization reference data.

4. A high-voltage switch mechanical characteristic parameter optimization platform according to claim 3, characterized in that: Genetic algorithm formula in adaptive optimization algorithm: Encoding and initialization population: Through binary encoding, for each mechanical characteristic parameter x i , perform encoding operation; assume that each parameter is encoded as a binary string of length l, then the encoding of the individual is represented as p = [p1, p2, ..., p n×l ], where p1 to p l is the encoding of x1, p l+1 to p 2l is the code of x2, and so on; where each individual represents a set of mechanical characteristic parameter combinations; Initialize the population P(0) and randomly generate m individuals, that is, P(0) = {p1(0), p2(0), ..., p m (0)}; Fitness evaluation: Calculate the fitness value F(p i ), the fitness function is defined according to the high-voltage switch performance index y, F(p i )=-|y i -y target | k , where y i is an individual p i The corresponding prediction performance index value, y target is the expected performance target value, k is a constant; Selection operation: Use roulette wheel selection method to select parent individuals; individual p i Probability of being selected Crossover and mutation operations: Use single-point crossover to perform crossover operations on the selected parent individuals. Suppose the two parent individuals are p a and p b , randomly select the crossover point c (1≤c≤n×l), then the offspring individual p′ after crossover a and p′ b for: p′ a =[p a (1),...,p a (c),p b (c+1),...,p b (n×l)],p′ b =[p b (1),...,p b (c),p a (c+1),...,p a (n×l)]; The offspring individuals after crossover are mutated with a certain mutation probability p m Change the value of some gene positions in individuals; for individual p i A gene position p in i,j , if the random number r<p m , where r is a random number between 0 and 1, then p i,j =1-p i,j ; Update population and iteration termination conditions: After selection, crossover and mutation operations, a new generation of population P(t+1) is obtained. Repeat fitness evaluation, selection, crossover and mutation operations until the global optimal solution no longer changes significantly and the optimal individual is obtained.

5. A high-voltage switch mechanical characteristic parameter optimization platform according to claim 4, characterized in that: The formula of particle swarm optimization algorithm is: Each particle i is represented by a mechanical characteristic parameter vector x i =[x i1 , x i2 , ..., x in ], and each particle has a velocity vector v i =[v i1 , v i2 , ..., v in ]; Initialize the position and velocity of the particle. The position is randomly initialized within the reasonable value range of the mechanical characteristic parameters, and the velocity is randomly initialized within a certain range; Individual optimal and global optimal update: Calculate the fitness value F(x i ), where the fitness function is defined in the same way as in the genetic algorithm, and is defined based on the high-voltage switch performance index; for each particle i, if in is the optimal position of particle i in history, then At the same time, find the particle with the smallest fitness value among all particles, that is, the particle with the best performance, and its position is recorded as g best ; Speed ​​and position update: The speed and position update formula of the particle is: x ij (t+1)=x ij (t)+v ij (t+1); Where w is the inertia weight, c1 and c2 are learning factors, r1 and r2 are random numbers between 0 and 1, t is the number of iterations, j = 1, 2, ..., n; v ij : the j-th velocity of particle i; x ij : j-th dimension position of particle i; Repeat the operations of updating individual optimal and global optimal solutions and updating speed and position until the iteration is terminated, that is, the global optimal solution no longer changes significantly, and the optimal particle position is obtained.

6. A high-voltage switch mechanical characteristic parameter optimization platform according to claim 5, characterized in that: The process of obtaining the adjustment reference data by the parameter optimization module is as follows: First, the optimization reference data is obtained from the data analysis module. By real-time monitoring of the operating status of the high-voltage switch, the actual operating data of the current high-voltage switch is obtained. The actual operating data includes the real-time values ​​of the breaking time, closing time, opening time, stroke, contact stroke, opening and closing current, speed and pressure. The feedback mechanism combines the actual operating data of the high-voltage switch under different environmental conditions to adjust the parameter weights of the optimization model. When it is found that the mechanical characteristic parameter has a significant impact on the switch performance under environmental conditions, the weight of this parameter is increased in the current optimization process; Assume the original weight vector is ω = [ω1, ω2, ..., ω n ], n is the number of mechanical characteristic parameters, and the weight vector adjusted according to environmental factors is ω′=[ω′1,ω′2,...,ω′ n ]; Calculate the parameter adjustment amount: Based on the optimized reference data and the adjusted weights, calculate the adjustment amount of each mechanical characteristic parameter Δx = [Δx1, Δx2, ..., Δx n ]; The calculation process combines the gap between the current parameter value and the target value, the parameter weight, and the operating status of the high-voltage switch. The target value is obtained based on the optimized reference data; For the mechanical characteristic parameter x i , the adjustment amount Δx i The calculation formula of Δx i = k × ω′ i ×(x i,target -x i,current ), where k is the adjustment coefficient, x i,target is the target value of the parameter obtained based on the optimized reference data, x i,current is the current actual value; Apply the calculated adjustment amount to the current mechanical characteristic parameter to obtain the adjusted parameter value x′=[x′1, x′2, ..., x′ n ], where x′ i =x i,current +Δx i ; Combine the adjusted parameter values ​​and related adjustment information to generate adjustment reference data.

7. A high voltage switch mechanical characteristic parameter optimization platform according to claim 6, characterized in that: The process of obtaining the simulation results by the simulation test module is as follows: Based on finite element analysis and multi-body dynamics simulation technology, finite element analysis and multi-body dynamics simulation models are constructed: the finite element analysis method is used to model and analyze the physical fields of the electric field, magnetic field, and stress and strain of the high-voltage switch to determine the changes in the physical state inside the high-voltage switch under different working conditions; at the same time, the multi-body dynamics simulation technology is used to construct a mechanical motion model of the high-voltage switch to simulate its dynamic behavior during operation, including the motion trajectory of the contacts, speed changes, and the interaction between mechanical components; according to actual needs, the working state parameters of the high-voltage switch under various working conditions are set, and the working state parameters include voltage fluctuations, mechanical wear, and temperature changes; The adjustment reference data generated by the parameter optimization module are used as input parameters for the simulation, and the initial state and operating conditions of the high-voltage switch in the simulation model are set according to these parameters; Start the simulation program, and simulate the operation process of the high-voltage switch based on the constructed finite element analysis and multi-body dynamics simulation model, combined with the input adjustment reference data and the set working conditions; During the simulation, the model calculates various state variables of the high-voltage switch at different times based on physical laws and mathematical algorithms; During the simulation calculation process, data related to the high-voltage switch performance is collected in real time to obtain simulation data, which includes electrical performance data, mechanical performance data and dynamic characteristic data; the collected simulation data is used for subsequent evaluation of the performance of the high-voltage switch; based on the collected simulation data, indicators for evaluating the performance of the high-voltage switch are calculated; the calculated performance indicators are integrated and analyzed to generate simulation results.

8. A high voltage switch mechanical characteristic parameter optimization platform according to claim 7, characterized in that: The process of obtaining the optimal parameters by the feedback adjustment module mainly includes the following steps: Receive the simulation results provided by the simulation test module, and at the same time, obtain the performance targets preset by the user through the customization mechanism; wherein the user sets specific performance indicator priorities according to their own needs: including setting the breaking reliability as the highest priority, followed by the closing stability, and the parameter adjustment range: including, for example, specifying the upper and lower limits of the allowable adjustment of a certain mechanical characteristic parameter; Compare and analyze each performance indicator in the simulation results with the preset performance target, and calculate the deviation value of each performance indicator; Determine the parameter adjustment strategy based on the deviation value and performance indicator priority; if the deviation of a high-priority performance indicator is large, then focus on adjusting the mechanical characteristic parameters that affect the breaking reliability; adjust the mechanical characteristic parameters based on the determined adjustment strategy; The adjusted parameters are input into the simulation test module again for simulation testing to obtain new simulation results; then the above-mentioned deviation analysis, adjustment strategy determination and parameter adjustment process are repeated, and the optimization is continuously iterated until the termination condition is met.

9. A high-voltage switch mechanical characteristic parameter optimization platform according to claim 8, characterized in that: The termination conditions include that the deviations of the performance indicators are within an acceptable range, the number of iterations reaches a preset upper limit, or the performance improvement is not obvious after several consecutive iterations; After the termination condition is met, the final parameter combination is verified; By conducting additional simulation tests under different operating conditions, we can ensure that the parameter combination can stably achieve good performance in various situations. At the same time, we can check whether the parameters are within the preset adjustment range and whether they meet the actual constraints of the physical characteristics and operating principles of the high-voltage switch. After verification, it is determined that the mechanical characteristic parameters at this time are the optimal parameters.

Images