（1.College of Electrical and Information Engineering, Hunan Univ, Changsha, Hunan410082, China;2. Jiangxi Electric Power Research Institute, Nanchang, Jiangxi330096,China) 在知网中查找 在百度中查找 在本站中查找
Affiliation:
Fund Project:
摘要
|
图/表
|
访问统计
|
参考文献
|
相似文献
|
引证文献
|
资源附件
摘要:
边界处理和全局最优引导者选择操作对多目标粒子群算法的性能有重要影响，在考虑不同操作方法特征的基础上，提出了改进的自适应多目标粒子群(multiobjective particle swarm optimization, MOPSO)算法.当算法陷入局部最优时，启用交叉变异操作；当算法收敛性停滞时，轮换修剪边界处理和指数分布边界处理操作；当算法多样性停滞时，轮换反比于拥挤距离和反比于控制粒子数目的全局最优引导者概率选择操作.标准测试函数以及柔性交流输电系统(flexible AC transmission system, FACTS)装置优化配置问题的仿真结果验证了所提算法的有效性.
Boundary handling and global best guider selection operators play an important role in the multiobjective particle swarm optimization (MOPSO) algorithm. Considering the characteristics of different operators, an improved adaptive MOPSO was proposed. When the algorithm falls into a local optimum, enable the crossover and mutation operators; when the convergence of algorithm hasn’t improved in a given duration, switch the boundary handling operator between the truncation and the exponential distribution truncation methods; when the diversity of algorithm hasn’t improved in a given duration, switch the global best guider selection operator between the probability inverse proportion to the crowding distance and the probability inverse proportion to the number of dominating solutions methods. The results of the benchmark functions and the optimal allocation problem of flexible AC transmission system (FACTS) devices confirm the effectiveness of proposed algorithm.
