摘要
针对传统粒子群优化算法存在早熟导致陷入局部最优解,以及后期收敛速度过慢问题,提出了一种基于混沌理论的自适应粒子群优化算法.首先利用混沌思想对粒子群进行初始化,保证粒子随机分布的均匀性,同时提高粒子的质量;其次,通过计算目标函数值,粒子根据自身状态调整惯性权重以增强寻优能力;在此基础上,对每一代全局最优值进行混沌映射,以增加种群的多样性.最后,以电力系统无功补偿控制为例,应用所提算法对控制器进行优化.仿真结果验证了该控制方法的有效性.
An adaptive particle swarm optimization (PSO)algorithm based on chaos theory is presented to make up for the defects of traditional particle swarm optimization including prematurity,the nature of falling into local optimal solution easily and slow convergence rate in the later part.Firstly,the theory of chaos is introduced in the initialization of particle swarm,which not only guarantees the randomness of particles but also improves the quality of particles.Secondly,by calculating the objective function value,the particles adaptively adjust the inertia weights to improve their global and local search capability. On this basis,in each iterative optimization of particle swarm,chaotic optimization is carried out for the global optimum to move the inert particles out of the range of local optimal solution.Finally,taking the reactive power compensation control of power system as an example,the proposed algorithm is applied to optimize the reactive power compensation controller,and the simulation results verify the effectiveness of the proposed method.
作者
赵熙临
汤倩
吴胧胧
徐光辉
何晶晶
ZHAO Xi-lin;TANG Qian;WU Long-long;XU Guang-hui;HE Jing-jing(Dept.of Electrical and Electronic Engineering HuBei University of Technology,Wuhan 430068,China)
出处
《中北大学学报(自然科学版)》
CAS
2018年第6期702-707,716,共7页
Journal of North University of China(Natural Science Edition)
基金
国家自然科学基金资助项目(61473116)
关键词
粒子群优化
混沌优化
无功补偿
自适应惯性权重
particle swarm optimization
chaos optimization
reactive power compensation
adaptive inertia weight