Heat load forecasting model considering two dimensional changes of time series

Abstract

The regulation strategy of a district heating system is adjusted based on accurate heat load prediction, which not only effectively reduces energy consumption but also improves system efficiency and user comfort. In order to improve the accuracy of heat load forecasting, a heat load forecasting model considering the two-dimensional change of time series is introduced in this paper. Firstly, the original heat load data is denoised by SVMD decomposition, and several stationary and regular modal components are obtained. Then, three strategies were used to enhance the BWO algorithm, and the IBWO-TimesNet prediction model was established to extract the hidden information of time series data from a two-dimensional perspective. Finally, the prediction performance of the model is evaluated in detail through case analysis. The results show that MAE, RMSE and R2 of SVMD-IBWO-TimesNet model are 0.647, 1.190 and 99.1%, respectively. Compared with other mainstream prediction models, this model has higher prediction accuracy. In addition, even if the training samples are reduced, the SVMD-IBWO-TimesNet model can still effectively predict the heat load and has strong generalization ability. Therefore, the performance of the model is verified, which provides a reference for the accurate control of heat load. Practical application Heat load forecasting is a vital task particularly in relation to its impact on the management of building energy efficiency. The contribution of this paper is to provide a heat load forecasting model based on advanced algorithms and data analysis. This insight derived from advanced modelling will assist professionals in the pursuit of more accurate prediction of the heat needs of buildings, thereby optimizing the design and operation of heating systems. The practical application of this technology could save energy costs, reduce carbon emissions, and improve the comfort and sustainability of buildings.