基于STL-Informer-BiLSTM-XGB模型的供热负荷预测

供热负荷预测是指导供热系统调控的重要手段。提高供热负荷预测精度十分重要,针对机器学习中输出目标的分解预测,提出了一种基于季节和趋势分解(Seasonal and Trend decomposition using Loess, STL)的供热负荷预测方法,构建了适用于供热负荷预测的输出目标。首先利用STL算法将供热负荷时间序列数据分解为趋势分量、周期分量和残差分量,分别训练Informer、BiLSTM和XGB三个模型,将构建好的三个分量预测模型的输出叠加作为初步预测结果,分析误差序列,以BiLSTM预测误差提高模型精度,构建出STL-Informer-BiLSTM-XGB预测模型。将上述模型与常用预测模型进行对比,结果表明所构建的STL-Informer-BiLSTM-XGB 模型的MAPE、MAE和MSE,分别为0.871%、96.18和13202.2,预测效果最优,验证了所提出的方法具有较高的供热负荷预测精度。

Heating load forecasting based on STL-Informer -BiLSTM-XGB model

Heating load forecasting is a key method to guide the regulation of heating system. Improving the accuracy of heating load forecasting is very essential. For the decomposition prediction of output targets in machine learning, proposes a heating load forecasting method based on Seasonal and Trend decomposition using Loss (STL), and constructs output targets suitable for heating load forecasting. Initially, using the STL algorithm, the heating load time series data is decomposed into trend components, periodic components, and residual components. Using these components to train Informer, BiLSTM and XGB models respectively. The output of the constructed three component forecasting models is superimposed as the preliminary forecasting results, and the error sequence is analyzed to improve the model accuracy with BiLSTM forecasting errors. Finally the STL-Informer-BiLSTM-XGB prediction model is constructed. The above model is compared with commonly used forecasting models, and the results showed that the constructed STL-Informer-BiLSTM-XGB model had MAPE, MAE, and MSE of 0.871%, 96.18%, and 13202.2%, respectively, which showed the best forecasting effect and verified the high accuracy of the heating load forecasting method proposed in this paper.