Research On The Combined Compensation Method Of Insulated Core Transformer High-Voltage Power Supply Based On Evolutionary Algorithms

Irradiation processing technology has been widely used in various fields of national economy and has a broad prospect of development.In irradiation processing,the electron irradiation accelerator with insulated core transformer(ICT)high-voltage power supply is an advantage equipment in the application of irradiation processing in low and medium energy region due to its high energy conversion efficiency,stability and reliability.ICT power supply is the most important component of the accelerator and has the characteristic of segmented magnetic core,which causes serious magnetic flux leakage,resulting in nonuniformity of disk output voltages.The voltage compensation method is introduced to solve this problem,and it is one of the key technical studies in the design of ICT high-voltage power supply.In order to compensate the output voltages of secondary coils in ICT high-voltage power supply,a combined compensation method based on the turn compensation and the capacitance compensation was proposed in Huazhong University of Science and Technology(HUST).It can effectively optimize the output voltage uniformity and improve the power supply performance.However,the method involves many compensation parameters.Its design and optimization often depend on the experience of designers,and all the parameters need to be adjusted manually.Therefore,the design and the optimization are inefficient and tedious.On the other hand,in recent years,the evolutionary algorithm(EA)has been widely used in various parameter optimization problems and has achieved good results.Therefore,it is of great significance to introduce EA into the compensation parameter optimization of ICT high-voltage power supply.It can help to realize automatic iterative optimization,improve the design optimization efficiency,get better optimization parameters,and finally improve the performance of ICT high-voltage power supply.In this paper,the combined compensation parameter optimization based on EA is studied deeply,and the parameter optimization of an ICT high-voltage power supply design is completed,the following work is carried out specifically:This paper takes genetic algorithm(GA)and particle swarm optimization(PSO)as examples of evolutionary algorithms,introduces the basic principles and summarizes the key technologies in detail.A 350 kV/50 mA ICT high-voltage power supply(HUST-ICT)was designed in HUST.Base on this prototype,the physical model and simulation model of HUST-ICT are established,which lays a foundation and platform for the application of EA.The simulation model is established based on the finite element analysis(FEA)software Maxwell.Based on the model,the single turn mutual inductance matrix is obtained.Based on MATLAB/Simulink,the simulation of the no-load output voltage and the full-load output voltage of HUST-ICT is carried out.For the combined compensation method,the performance evaluation parameters are designed in this paper,which can directly reflect the quality of the set of combined compensation parameters,and lay a foundation for the establishment of fitness function in the EA.GA and PSO are designed for the compensation parameter optimization of HUST-ICT.The key technical methods of the two algorithms are designed in detail,and the key control parameters are analyzed and selected.The programming based on GA is completed,and the simulation design is carried out.Through automatic iterative optimization,the optimization parameters that fully meet the requirements are obtained.Finally,the feasibility and effectiveness of the EA in the optimization of the combined compensation parameters of ICT high-voltage power supply are successfully verified.