进化策略在1100kV气体绝缘开关设备出线套管内绝缘结构多目标优化中的应用

Application of Evolution Strategy in Multi-objective Optimization of Internal Insulation Structure of 1100 k V GIS Outlet Bushing

针对1 100 k V气体绝缘开关设备(GIS)出线套管内绝缘结构的多目标优化问题,通过(16,110)进化策略(ES)和有限元法(FEM)相结合的方法,对套管内绝缘结构进行了优化;并对比分析了优化前、后套管的电位、电场分布。结果表明,采用(16,110)进化策略后,套管绝缘结构多目标优化问题的评价函数经过20代进化后已达到收敛;进化过程中,亲代群体决策变量的空间演化情况与评价函数的收敛情况相符。(16,110)进化策略亲代群体决策变量的空间演化分析为1 100 k V GIS出线套管的设计提供了一个可选最优空间,且优化后的设计方案更符合气体绝缘套管的设计要求。

The internal insulation structure of 1 100 kV GIS bushing was optimized by jointly using （16,110） evolution strategy （ES） and finite element.method （FEM）.Moreover, the potential and electric field distributions of 1 100 kV GIS bushing were analyzed by comparing the distributions before and after optimization. The results show that the evaluation function of multi-objective optimization problem has converged after evolution of 20 generations by using （16,110） E S. In the process of the evolution, the space evolution of the decision variables of the parent group is consistent with the con- vergence of evaluation function. The optimal space for the design of 1 100 kV GIS bushing has been provided by analyzing the space evolution of the decision variables Of （16,110） ES, and the optimal design is more in line with re- quirements of gas bushing design.