Joint Optimization Design Of Vacuum Interrupter Electromagnetic Field Based On Improved Neural Network

Vacuum circuit breakers play an important protective role in the medium voltage power system.The vacuum interrupter is the core component of the circuit breaker,its internal electrical,magnetic,thermal characteristics and material properties directly affect the insulation performance and breaking capacity of the circuit breaker.In this thesis,the structural parameter design of the vacuum interrupter is taken as the research object,and the numerical simulation of electric and magnetic fields is performed.In order to meet the requirements of insulation performance,parameters optimization and design of structural components inside the vacuum interrupter are performed;an optimization method combining BP neural network and genetic algorithm is proposed,and it is used in the parameterized optimization design of the electric and magnetic fields of the interrupter.For improving the prediction accuracy of the traditional BP neural network model,genetic algorithms are used to optimize the initial weights and thresholds in advance;further orthogonal design and normalized processing methods are used to collect and preprocess the samples;on the basis of the above,using genetic algorithm to complete the optimization.By optimizing typical test functions,the feasibility of the proposed combinatorial optimization strategy is verified.In order to determine the structural parameters of the components of the vacuum interrupter,the material and structure of the insulating shell,shield,contact,corrugated tube,and conductive rod are analyzed by parameterization.The vacuum interrupters with 8 different technical parameters is the object,the influence of structural parameters on the insulation characteristics is compared and analyzed;and the overall structural parameter group of 12 k V/1250A/31.5k A vacuum interrupter is given.For optimizing the electric field inside the vacuum interrupter,a 12 k V/1250A/31.5k A vacuum interrupter is used as the object,a finite element analysis model of the electric field is established,and the electric field prediction model of the vacuum interrupter is obtained by improving the BP neural network method.The genetic algorithm is used to realize the parameterized optimization design of the electric field structure,and the electric field distribution before and after the optimization is compared and analyzed.The inter-phase electric field distribution of the three-phase vacuum interrupter under the rated voltage,power frequency voltage,and lightning impulse voltage is obtained to realize the inter-phase insulation analysis.Based on the insulation structure parameters optimized by the above electric field,the magnetic field optimization analysis of the contact gap of the vacuum interrupter is performed.Build a finite element model of the magnetic field.A single factor analysis method is used to determine the optimization variables.Establish magnetic field to improve BP neural network model.The genetic algorithm is used for parameter optimization of the magnetic field;the magnetic fields on the contact surface and the center plane of the contact gap before and after the optimization are compared and analyzed to verify the feasibility of the optimization scheme,and then improve the breaking performance.