基于自适应混沌粒子群算法的光伏电池模型参数辨识

发布时间：2018-03-14 22:18

本文选题：太阳能电池模型　切入点：参数估计　出处：《电工技术学报》2014年09期 　论文类型：期刊论文

【摘要】：光伏电池模型参数的快速准确辨识在光伏阵列的输出功率预测、最大功率点跟踪以及电池故障模型的特性研究方面具有非常重要的工程意义。针对大部分传统智能算法用于系统参数辨识时的辨识精确受参数初值影响较大,而且算法易陷入早熟的问题,利用自适应混沌粒子群算法(SA-CPSO)对光伏电池模型参数进行辨识。将混沌算法与粒子群算法融合,对粒子群进行混沌初始化并促使陷入局部最优的粒子进行混沌搜索,引导其跳出局部极值从而搜索到更好的解;同时引入自适应调整策略来有效控制全局与局部搜索,提高了进化后期算法的收敛精度。经过仿真和实验测试,证明SA-CPSO算法在光伏电池模型参数辨识方面具有较高的精确度和快速性。还通过实验探讨了辐照度变化对太阳能电池参数的影响。
[Abstract]:The fast and accurate identification of photovoltaic cell model parameters is used to predict the output power of photovoltaic array. The research of maximum power point tracking and the characteristics of battery fault model is of great engineering significance. The accuracy of most traditional intelligent algorithms used in system parameter identification is greatly affected by the initial parameters. Moreover, the algorithm is easy to fall into premature problem. The parameters of photovoltaic cell model are identified by using adaptive chaotic particle swarm optimization algorithm (SA-CPSO), and the chaos algorithm and particle swarm optimization algorithm are fused. The chaos of particle swarm is initialized and chaotic search is made to lead the particle to jump out of the local extremum to find a better solution, and the adaptive adjustment strategy is introduced to effectively control the global and local search. The convergence accuracy of the late evolutionary algorithm is improved. It is proved that the SA-CPSO algorithm has high accuracy and rapidity in the parameter identification of photovoltaic cell model, and the influence of irradiance on the parameters of solar cell is also discussed through experiments.
【作者单位】：天津大学电气与自动化工程学院;
【基金】：国家基础研究项目(973计划)(2009CB219700) 天津市重点科技支撑项目(09ZCGYGX01100) 天津市太阳能光电建筑应用示范项目(2011E1-002)资助
【分类号】：TM914.4

【相似文献】