摘要
光伏发电存在着波动性和不确定性,对光伏发电系统的功率预测是提高光伏发电的利用率和经济效益的重要举措。通过构建蚁群算法(ACO)优化后的BP神经网络预测模型进行短期光伏功率预测研究,引入灰色关联度分析,确定影响光伏发电的主要因素,提高模型的预测准确性。该模型综合了ACO的寻优能力和BP神经网络的自学习、自适应能力。将训练好的模型用于光伏发电短期功率预测研究,对比仿真结果得出ACO-BP神经网络模型在晴天时的预测误差为8.60%,多云时的预测误差为12.53%,雨天时的预测误差为26.27%,其预测精度均优于原BP神经网络模型。
Photovoltaic power generation has volatility and uncertainty,and predicting the power of photovoltaic power generation systems is an important measure to improve the utilization rate and economic benefits of photovoltaic power generation.This article conducts short-term photovoltaic power prediction research by constructing a BP neural network prediction model optimized by Ant Colony Optimization(ACO),introducing grey correlation analysis to identify the main factors affecting photovoltaic power generation and improve the prediction accuracy of the model.This model combines the optimization ability of ACO and the self-learning and adaptive ability of BP neural network.The trained model was used for short-term power prediction research in photovoltaic power generation.Comparing the simulation results,it was found that the prediction error of the ACO-BP neural network model was 8.60%on sunny days,12.53%on cloudy days,and 26.27%on rainy days,all of which had better prediction accuracy than the original BP neural network model.
作者
钟安德
吴自玉
谢宗效
毛玉明
杨留方
ZHONG Ande;WU Ziyu;XIE Zongxiao;MAO Yuming;YANG Liufang(School of Electrical and Information Technology,Yunnan University of Nationalities,Kunming 650031,China)
出处
《电子设计工程》
2024年第18期82-86,共5页
Electronic Design Engineering
关键词
光伏发电
蚁群算法
BP神经网络
参数优化
短期功率预测
photovoltaic power generation
Ant Colony Optimization
BP neural network
parameter optimization
short-term power prediction
