摘要
为准确预测光伏系统的发电量,构建了用遗传算法优化的BP神经网络发电量短期预测模型(简称GA-BP模型)。通过遗传算法的迭代,对BP神经网络的权值和阈值进行优化,以实现对算法的改进。对所选择国能日新光伏系统预测大赛的数据进行预处理、归一化,并将数据输入GA-BP模型,进行了实验。对比实验说明,GA-BP模型不管是在预测结果上还是在模型稳定性上都明显优于BP神经网络模型。
This study proposes a Genetic Algorithm-optimized BP(GA-BP)neural network model for accurate short-term forecasting of photovoltaic(PV)power generation.Initially,the genetic algorithm iteratively optimizes the weights and thresholds of the BP neural network to identify optimal values,thereby enhancing the performance of the algorithm.Subsequently,preprocessing and normalization are applied to selected PV prediction data before inputting them into the model.Experimental results indicate that the GA-BP model outperforms the conventional BP model in terms of predictive accuracy and stability.
作者
李燕斌
魏婷婷
贾恒
李淑新
LI Yanbin;WEI Tingting;JIA Heng;LI Shuxin(School of Automation and Electrical Engineering,Zhongyuan University of Technology,Zhengzhou 450007,China)
出处
《中原工学院学报》
CAS
2024年第2期1-5,共5页
Journal of Zhongyuan University of Technology
关键词
光伏发电
发电量预测
遗传算法
BP神经网络
photovoltaic power generation
power generation prediction
genetic algorithm
BP neural network
