Design of fusion physics system based on resampling particle swarm optimisation algorithm

Abstract This paper presents a fusion physics system design method based on the resampling particle swarm optimisation algorithm. The method optimises the design of the fusion physics system by establishing the functionality function for a specific optimised fusion physics problem, initialising the particle population, performing resampling operations on the particle population, renovating the particle locations and velocities, renewing the historical optimum location and the population optimal location of each particle, and iterative operations. It includes the following steps: i: establish the functionality function for the specific optimised fusion physics problem; ii: initialise the particle population; iii: perform the resampling operation on the particle population; iv: update the particle locations and velocities; v: update the historical optimal location and the group optimal location of each particle; vi: if the accuracy requirement is not met and maximum iterations has not yet been achieved, the number of iterations is increased by one and return to step iii, otherwise record and output the results; the system design can effectively deal with the practical problems of fusion physics system optimal design, and the introduction of resampling technology, on the one hand, accelerates the convergence speed and improves the efficiency, on the other hand, improves the ability of global search, improves the optimisation accuracy, and ultimately achieves the purpose of efficiently and accurately solving the problems of optimal design of fusion physics system. The results show that the system design method can solve the fusion physics system optimal design problem efficiently and accurately.