基于主成分分析和遗传优化BP神经网络的光伏输出功率短期.PDFVIP

第44 卷 第22 期 电力系统保护与控制 Vol.44 No.22 2016 年 11 月 16 日 Power System Protection and Control Nov. 16, 2016 DOI: 10.7667/PSPC160074 基于主成分分析和遗传优化 BP 神经网络的 光伏输出功率短期预测 许童羽，马艺铭, 曹英丽, 唐 瑞，陈俊杰 (沈阳农业大学信息与电气工程学院，辽宁 沈阳 110161) 摘要：针对光伏系统输出功率的波动性和间歇性特点，提出一种基于主成分分析(PCA)和遗传算法(GA)优化的 BP 神经网络功率短期预测方法。通过历史功率数据和实时气象因素对输出功率进行直接预测，利用主成分分析法将 多个原始变量降维成少数彼此独立的变量，作为神经网络的输入。同时利用遗传算法的全局搜索特性在解空间中 定位一个较好的空间，优化 BP 的初始权值阈值，克服了传统 BP 神经网络易陷入局部极小点、学习收敛速度慢的 问题。通过建立不同预测模型进行对比，验证了所提算法和模型的有效性。 关键词：主成分分析；遗传算法；功率预测；BP 神经网络；光伏系统 Short term forecasting of photovoltaic output power based on principal component analysis and genetic optimization of BP neural network XU Tongyu, MA Yiming, CAO Yingli, TANG Rui, CHEN Junjie (College of Information and Electrical Engineering, Shenyang Agricultural University, Shenyang 110161, China) Abstract: In view of the fluctuation and the intermittence of the output power of the photovoltaic system, based on principal component analysis (PCA) and genetic algorithm (GA) optimization, a short term forecasting method of BP neural network power is proposed. Direct forecasting of output power is done by historical power data and real time meteorological factors. It uses principal component analysis to reduce the dimension of multiple original variables into a few independent variables, so that it can optimize the initial weights of back-propagations threshold and overcome the traditional BP neural network easy to fall into local minimum point, and the