基于IWOA-iTransformer-QR的短期城市电力负荷区间预测模型OA

With the connection of renewable energy and new-type loads,the uncertainty of power load forecasting has significantly increased.In view of this,an Improved whale optimization algorithm(IWOA)-based iTransformer quantile regression model(IWOA-iTransformer-QR)is proposed.This model uses IWOA to automatically optimize hyperparameters,and combines iTransformer and quantile regression to implement feature modeling and synchronous quantification of uncertainty.The results of the calculation examples show that this model significantly outperforms the Long Short-Term Memory(LSTM)network,Gated Recurrent Unit(GRU)and Transformer in short-term urban power load forecasting,with the root mean square difference(RMSE)decreased by more than 70%,and can provide a reliable prediction range.The study has proven that the proposed method has potential in improving prediction accuracy and credibility,thus providing support for the scheduling and operation of power systems.