改进粒子群算法及其在风电系统中的应用  

An Improved PSO Algorithm and its Utilization in Wind Power Generation System

作　　者：包广清[1] 毛开富[1] 

机构地区：[1]兰州理工大学电气工程与自动化学院,甘肃兰州730050

出　　处：《控制工程》2013年第2期262-266,271,共6页Control Engineering of China

基　　金：国家自然科学基金项目(50877034);国家人力资源与社会保障部"留学人员科技活动优秀项目"(2009)

摘　　要：提出一种基于正交试验的改进粒子群优化算法,以改善传统算法的寻优效率和参数设置。基于随机过程理论,把粒子群算法原有的时变系统转换成概率意义下的线性定常系统进行分析,明确了粒子群算法的全局收敛性及其相关参数取值范围,在此基础上提出一种惯性权重和学习因子非线性、非对称变化的改进粒子群算法,使算法在全局收敛的同时具有较高的优化精度,并引入正交设计的多因素多水平试验,确定该算法参数的最优取值。通过在甘肃酒泉某变速恒频风电机组的最大风能跟踪控制系统应用,有效确定了最优叶尖速比和风机转轴速度,实现了最大年平均风含能量的优化目标,使风电机组的运行效率得到提高,验证了改进粒子群算法的正确性及其工程实用价值。The parameter settings of particle swarm optimization（PSO） often rely on experience or testing, which results in many times iteration and low optimzation efficiency. This paper presents the PSO convergence analysis based on the theory of stochastic processes. By regarding each particle＇ s position on each evolutionary step as a stochastic vector, the convergence-related parameter adjustment range is derived. In addition, an improved PSO（IPSO） with non-linear and asymmetric time-varying parameters is proposed. An orthog onal experiment design method is used to determine the inertia weight and learning factors of the IPSO, in which a minimal number of experiments would have an effect that approximates the full factorial experiments. The optimization reshs of two benchmark functions show the feasibility of IPSO and its superiority over several previous PSO algorithms. When applied in the maximum power point tracking （MPPT） of variable speed constant frequency wind energy conversion sytem, the optimal tip speed ratio and the generator speed are tai lored such that the maximum annual average wind energy and power coefficient are available, which verify the correctness of the IPSO and its practical value.

关 键 词：粒子群算法 收敛 正交试验 变速恒频风电系统 最大风能跟踪控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]