基于模拟退火粒子群优化的光伏多峰最大功率跟踪算法  

Multi-peak maximum power point tracking algorithm based on simulated annealing particle swarm optimization for PV systems

作　　者：杨洁[1] 蒋林[1] 蹇清平[2] 张勇军[3] 李永德[1] 

机构地区：[1]西南石油大学电气信息学院,成都610500 [2]西南石油大学机电工程学院,成都610500 [3]昆明理工大学信息工程与自动化学院,昆明650051

出　　处：《计算机应用》2014年第A01期330-333,共4页journal of Computer Applications

基　　金：国家自然科学基金资助项目(51204139);四川省教育厅科技项目(13ZB0199);西南石油大学科研基金资助项目(2012XJ2023)

摘　　要：在光伏发电系统多峰最大功率跟踪中,针对采用干扰观察法跟踪时间长和粒子群优化算法扰动较大等问题,设计了基于模拟退火粒子群优化(SA-PSO)的最大功率跟踪算法。该方法引入粒子速度收缩因子和最优粒子的轮盘赌策略的扰动,有效避免陷入局部最大功率并快速跟踪到全局最大功率,保证了光伏发电系统在光照强度和温度变化时的动态响应和稳态精度。在相同条件下,对3种光伏发电系统最大功率跟踪方法进行仿真对比分析,其结果表明所提方法具有更好的实时性和鲁棒性。To solve the problem of the Perturbation and Observation （ P and O） method for tracking a long time and Particle Swarm Optimization （PSO） algorithm having big perturbation in the Maximum Power Point Tracking （MPPT） of the P＇q （Photo ＂qoltaies） array multi-peak, a Simulated Annealing Particle Swarm Optimization （SA-PSO） algorithm was designed and applied to maximum power point tracking of photovohaic generation system. The method was introduced into the contraction factor of particle velocity and the global best particle roulette strategy disturbance, effectively avoided falling into local maximum power and fast tracked to the global maximum power. The dynamic response and steady state accuracy can be guaranteed for the photovohaic power generation system when light intensity and temperature changes. Under the same conditions, the three kinds of maximum power point tracking algorithms were simulated and comparatively analyzed for PV systems in the no-load and load of system. The results show that the proposed method has better real-time performance and robustness.

关 键 词：光伏 多峰 最大功率跟踪 模拟退火粒子群优化 

分 类 号：TM761[电气工程—电力系统及自动化]