电力信息与通信技术2024，Vol.22Issue(11)：34-42,9.DOI:10.16543/j.2095-641x.electric.power.ict.2024.11.05
基于LEA-LSTM的光伏发电功率短期预测方法
A Short-term Prediction Method of Photovoltaic Power Generation Based on LEA-LSTM
摘要
Abstract
The photovoltaic power generation is easily affected by meteorological and environmental factors,resulting in significant randomness and uncertainty,which has a certain impact on the scheduling of the power sector and the safety of the power grid.Therefore,accurate prediction of photovoltaic power generation is of great significance for the stable operation of the power system.This paper utilizes the strong global search ability of the love evolution algorithm(LEA)to optimize the long short term memory(LSTM)network for more accurate short-term prediction of photovoltaic power generation.Firstly,Pearson correlation coefficient was used to calculate the correlation degree between various meteorological factors and photovoltaic power generation,and determine the input characteristics of the model.Secondly,the LEA algorithm was used to optimize the initial hyperparameters of the LSTM network,obtain the optimal parameter combination,and establish the LEA-LSTM model.Finally,to verify the universality and superiority of the proposed method,experiments were conducted using datasets under different weather conditions,and compared with three models including LSTM,PSO-LSTM,and WOA-LSTM.The average absolute error,mean square error,root mean square error,average absolute percentage error,and coefficient of determination of the four models were calculated as five error evaluation indicators.The experimental results show that compared with LSTM,PSO-LSTM,and WOA-LSTM,the LEA-LSTM model proposed in this paper has higher prediction accuracy and smaller prediction errors.关键词
光伏功率预测/皮尔逊相关系数/爱情进化算法/长短期记忆网络Key words
photovoltaic power prediction/Pearson correlation coefficient/love evolution algorithm/long short-term memory network分类
信息技术与安全科学引用本文复制引用
赵佳蕊,王玲芝,李晨阳..基于LEA-LSTM的光伏发电功率短期预测方法[J].电力信息与通信技术,2024,22(11):34-42,9.基金项目
国家自然科学基金项目(52177194). （52177194）
