基于ARPSO的高温超导分层储能磁体优化设计研究

Research on optimized design of high temperature superconducting layered energy storage magnet based on ARPSO

提出利用双饼线圈轴向阶梯分层的方法,通过改变每层线圈的平均半径来改变径向磁场分布从而优化磁体结构,减少用线量,且能解决阶梯电流法的多电源供电问题。在COMSOL中利用2G高温超导带材建立储能磁体初始模型;在MATLAB中调用已建立完成的模型,与自适应粒子群优化算法(ARPSO)相结合,以用线量为目标,储能量及临界电流为约束,寻找每组的最佳平均半径;在COMSOL中建立优化后模型,进一步做电磁分析。分别对1、3、5阶梯型的分层储能磁体进行仿真。仿真结果表明:采用此方法得出的优化后模型,在通入相同电流的情况下,径向磁场减小约21%;在储能量为10 MJ的前提下,用线量减小约48%,应力及漏磁场亦满足要求,验证了此方法的有效性。

This paper proposed a method of using the axial layering of the double-pie coil to change the radial magnetic field distribution by changing the average radius of each layer of the coil to optimize the structure of the magnet,reduce the amount of wires used,and solve the problem of multi-power supply.Firstly,2 G HTS tape in COMSOL was used to establish the initial model of the energy storage magnet.Secondly,the model in MATLAB was called and combine with the Adaptive Particle Swarm Optimization Algorithm(ARPSO),the wire quantity was as the target,the energy storage and the critical current were the constraints,and the best average radius of each layer was found.Finally,an optimized model was established in COMSOL for further electromagnetic analysis.This article simulated 1,3 and 5 stepped energy storage magnets respectively.The simulation results show that the optimized model obtained by this method can reduce the radial magnetic field by about 21%,when the same current is applied;On the premise that the energy storage is 10 MJ,the line consumption is reduced by about 48%,which verifies the effectiveness of this method.