四川轻化工大学学报（自然科学版）2026，Vol.39Issue(3)：49-58,10.DOI:10.11863/j.suse.2026.03.06
基于IWSO和组合模型短期电力负荷预测
Short-term Electricity Load Forecasting Based on IWSO and Combined Model
摘要
Abstract
Short-term load forecasting is of great significance to the operation planning of power systems.In order to further improve the accuracy of load forecasting,a short-term load forecasting model based on improved White Shark Optimizer(IWSO)and the CNN-BiLSTM-SelfAT combined model is proposed.Firstly,a convolutional neural network(CNN)and a self-attention mechanism(SelfAT)are used.The Bidirectional Long-Short-Term Memory(BiLSTM)of SelfAT is combined to fully extract spatial and temporal features of data.At the same time,the periodic mutation strategy,the wavelet fluctuation function,the specular reflection learning of mixed Gaussian levy and the dynamic fitness-distance balance strategy are introduced into the White Shark optimization algorithm to improve the performance of the algorithm,and then the IWSO is used to optimize the hyperparameters of CNN-BiLSTM-SelfAT to obtain the final prediction model.Finally,a simulation experiment is carried out by using the detection data set of one year in a certain area of China.The results show that the RMSE and MAE of the proposed model are respectively 171.5169 and 31.9156 lower than those of WSO-CNN-BiLSTM-SelfAT,and R² is increased by 0.057 43.Compared with other comparison models,the prediction accuracy of the proposed model has been significantly improved,fully verifying the good applicability of the IWSO-CNN-BiLSTM-SelfAT model in short-term power load forecasting.关键词
短期电力负荷预测/改进白鲨优化算法/CNN-BiLSTM-SelfAT/周期突变策略/超参数寻优Key words
short-term power load forecasting/improved White Shark Optimizer/CNN-BiLSTM-SelfAT/periodic mutation strategy/hyper-parameter optimization seeking分类
信息技术与安全科学引用本文复制引用
胡腾,杨毅强,吴凡,邱函,秦海洋..基于IWSO和组合模型短期电力负荷预测[J].四川轻化工大学学报（自然科学版）,2026,39(3):49-58,10.基金项目
四川省科技厅项目(2022YFS0518 （）
2022ZHCG0035 （）
2024ZHCG0006 （）
2024ZYD0265) （）
企业信息化与物联网测控技术四川省高校重点实验室(2022WYY04) （2022WYY04）
