沈阳农业大学学报2025，Vol.56Issue(2)：131-143,13.DOI:10.3969/j.issn.1000-1700.2025.02.014
基于AWOA-BI-LSTM的光伏发电功率预测
Photovoltaic Power Generation Prediction Based on AWOA-BI-LSTM
摘要
Abstract
[Objective]Accurate prediction of photovoltaic(PV)power generation is crucial for integrating renewable energy into the grid,energy markets,and building energy management systems.To improve prediction accuracy,this study proposed a hybrid model,denoted as AWOA-Bi-LSTM,which combines an improved Whale Optimization Algorithm(AWOA)with a Bidirectional Long Short-Term Memory network(Bi-LSTM).To solve the problems of low optimization accuracy and slow convergence of the traditional Whale Optimization Algorithm(WOA),two improvement strategies,dynamic weight factors and adaptive parameter adjustment,are introduced to enhance the model's global search capability and convergence efficiency.[Methods]Based on the real PV power generation data and measured meteorological data,comparative experiments were conducted among AWOA-Bi-LSTM,WOA-Bi-LSTM,and GRNN.[Results]The R² values of AWOA-Bi-LSTM model on the test set and training set are 0.997 01 and 0.998 43,respectively;the RMSE values are 1.585 and 0.900 63,respectively;and the RPD are 20.160 4 for the test set and 25.935 7 for the training set.[Conclusion]The results demonstrate that the AWOA-Bi-LSTM outperforms conventional methods in terms of goodness of fit,prediction accuracy,and stability.It more effectively captures complex patterns and trends in time series data,substantially enhancing prediction performance.关键词
光伏发电/功率预测/LSTM/BI-LSTM/WOA算法Key words
photovoltaic power generation/power prediction/LSTM/BI-LSTM/WOA algorithm分类
信息技术与安全科学引用本文复制引用
吴仕宏,张璧臣,吴佳文,武兴宇..基于AWOA-BI-LSTM的光伏发电功率预测[J].沈阳农业大学学报,2025,56(2):131-143,13.基金项目
国家自然科学基金项目(61903264) （61903264）
