基于RBF的光伏系统发电量智能预测研究  

Study of intelligent prediction for power generation in PV system based on RBF nerve network

作　　者：易灵芝[1] 刘珊[1] 王雅慧[2] 刘颉[1] 王书颢[1] 

机构地区：[1]湘潭大学信息工程学院,湖南湘潭411105 [2]湖南大学电气与信息工程学院,湖南长沙410082

出　　处：《电源技术》2013年第4期606-608,611,共4页Chinese Journal of Power Sources

基　　金：国家能源局-国家发展改革委发改2011(1952)号;湖南省自科基金项目(11JJ8004);湖南省教育厅重点项目(10A114)

摘　　要：针对传统太阳电池建模要求频繁设置开路电压、短路电流、最大功率点电压/电流等参数的问题,找出太阳电池材料各常数相互关系,并通过RBF神经网络实现各参数逼近;建立基于禁带宽度的太阳电池新型通用模型,参考电池温度和日照强度,自动调整参数,得出新环境的电池输出特性。在光伏发电短期预测中,智能预测方法能有效降低预测误差。The traditional methods of solar cell must set many parameters frequently, such as open circuit voltage, short circuit current, voltage and current of the maximum power point, and so on. By studying the interrelationship of all parameters in every solar cell material, the parameters could be approached by RBF nerve network, a new solar general simulation model could be built, which could automatically adjust these parameters, consulting the cell temperature and the intensity of sunshine, the output characteristic of solar cell under new environment could be got. In short-term prediction, intelligent method can reduce error effectively.

关 键 词：太阳电池材料 禁带宽度 RBF神经网络 智能预测 

分 类 号：TM911.4[电气工程—电力电子与电力传动]