Optimum Electromagnetism Design Based on Co-Simulation of MATLAB and COMSOL

遗传算法(Genetic Algorithm简称GA)是借鉴生物遗传学和自然选择机理的一种智能的优化算法。MATLAB中的遗传算法工具箱集成了当前比较成熟的各种遗传操作算子，借助它可以方便地完成各种问题的优化。COMSOL Multiphysics是一款大型的高级数值仿真软件，以高效的计算性能和杰出的多场直接耦合能力实现了任意多物理场的高度精确的数值仿真。为使遗传算法更高效的应用于电磁优化设计，我们研究了在MATLAB中调用COMSOL联合仿真的方法，实现了MATLAB与COMSOL的数据传递。我们将该电磁优化设计方法成功的用于一亥姆霍兹线圈，由此表明了此方法的可行性。 The genetic algorithm (GA) is a type of intelligent optimization algorithm that learns from biological genet- ics and natural selection mechanism. There are varieties of mature genetic operators integrated in the genetic algorithm toolbox of the MATLAB, and we can complete variety of optimization problems easily through the toolbox. COMSOL Multiphysics is a large-scale numerical simulation software. Efficient computing performance and outstanding multi- field direct coupling capabilities ensure highly accurate numerical simulation of arbitrary multi-physics fields. In order to make the genetic algorithm apply to optimum electromagnetism design, we studied the method of calling COMSOL in MATLAB, and completed the data transfer between MATLAB and COMSOL. Then we successfully used this opti- mum design method on a Helmholtz coil, and thus verified the feasibility of this method.