【摘要】 针对电磁仿真过程中电磁结构计算量大的问题，将智能优化算法与电磁仿真软件相结合以减少电磁场计算量。以电调滤波器为例，针对滤波器的环形电感结构进行参数化设计，并将相关参数作为约束条件，引入带精英策略的非支配排序遗传算法（Non-dominated Sorting Genetic Algorithm Ⅱ,NSGA-Ⅱ），以电调滤波器的损耗、带宽和传输零点等指标作为优化算法的优化目标，建立了电磁环境下集成电感结构参数与频率响应的优化方法。算法在第6次迭代后，电调滤波器的性能指标达到设计要求。研究结果表明，该算法显著减少了电磁环境中的计算量，且优化结果与电磁仿真结果基本一致，方法合理有效。更多还原

【Abstract】 To address the issue of high computational cost in electromagnetic simulation of electrically tunable filter structures, the intelligent optimization algorithm is integrated with the electromagnetic simulation software to reduce computational load of electromagnetic field. The parametric design of the filter’s ring inductor structure is carried out and used as a constraint condition. The non-dominated sorting genetic algorithm with an elitism strategy is introduced. The loss, bandwidth and transmission zeros of the electrically tunable filter are taken as the debugging targets of the optimization algorithm. The optimization method and system for structural parameters and frequency response of the integrated inductor in the electromagnetic environment are established. After the sixth iteration of the algorithm, the performance index of the electrically tunable filter meets the design requirements. The results show that the algorithm significantly reduces computational cost in the electromagnetic simulation, and the optimization results are basically consistent with electromagnetic simulation results. The method is reasonable and effective.更多还原