摘要
日照的不稳定性导致光伏系统的输出受到影响,为了尽可能准确地预测光伏系统的功率输出,就需要一种估测日照量的方法。文章中提出了一种使用神经网络预测日照量的方法,并通过计算机仿真证实了这种方法的合理性。同时介绍了3种神经网络法:FFNN、RBFNN和RNN,最后还对这3种方法的仿真结果进行了适当的比较。
Output of photo-voltaic （PV） system is influenced by unconstant solar radiation. In order to forecast the power output of PV system as accurately as possible, a method of solar quantity estimation is required. Therefore, it is proposed a method of neural network to estimate solar quantity, and the rationality is confirmed by computer simulations. Then three kinds of neural networks： FFNN, RBFNN and RNN are introduced, and finally the simulation results are properly compared.
出处
《大功率变流技术》
2010年第3期28-32,共5页
HIGH POWER CONVERTER TECHNOLOGY
关键词
神经网络
光伏系统的功率输出
日照预测
neural network（NN）
power output for PV system
solar quantity estimation
