粒子群算法应用于避雷器均压环配置优化

Application of Particle Swarm Optimization Algorithm to Configuration Optimization of Grading Rings for Arrester

合适的均压环配置即均压环数量和参数能使避雷器电阻片的电压承担率分布更加均匀,均压环数量、参数与电压承担率之间存在着复杂的多维非线性关系,均压环最优配置难于求解但易于评估。为了研究该问题,应用粒子群算法对均压环配置进行优化计算,以最大电压承担率为目标,建立不同个数均压环的优化模型,根据个体适应度值自适应调节惯性权重以提高种群的全局与局部寻优能力。算例结果表明:对500 kV避雷器配置3个均压环和配置2个均压环进行优化计算,优化后最大电压承担率分别为1.104和1.143,均小于1.15,满足相关要求。其中,前者优化后均压效果更为显著;后者优化后结构更为简单。

Appropriate configuration of grading rings, including number and parameters, can achieve more uniform distribution of voltage assignment ratio on arrester＇s resistor. However, since the relation among the number and parameters of grading rings and the voltage assignment ratio is complicated and multidimensional nonlinear, optimum configuration of grading rings is difficult to be resolved but is easy to be evaluated. Therefore, particle swarm optimization algorithm is employed to optimize configuration of grading rings in this paper. Taking maximum voltage assignment ratio as the objective, optimal models of different number of grading rings are built, and the inertia weight is adjusted adaptively in response to individual fitness value to improve the population＇s ability of global and local search. A 500 kV arrester with respective three and two grading rings is optimized, and the results show that the maximum voltage assignment ratios get 1.104 and 1.146, respectively, satisfying the requirement of less than 1.15. The arrester with three grading rings gains more uniform voltage assignment, and the one with two grading rings has simpler structure after optimization.