Analysis on the Technical Situation and Applied Difficulties of District Heating Load Forecasting

• Published in: Thermal engineering Vol. 69; no. 6; pp. 464 - 472
• Main Authors: Zhao, Bingwen, Jin, Yu, Li, Wan, Zheng, Hanyu
• Format: Journal Article
• Language: English
• Published: Moscow Pleiades Publishing 01.06.2022; Springer Nature B.V
• Subjects: District heating; District Heating Cogeneration and Heat Networks; Eigenvalues; Energy consumption; Engineering; Engineering Thermodynamics; Forecasting; Heat and Mass Transfer; Heating load; Mathematical models; Prediction models; district heating; pipe network; heat load forecast; database; heating demand; smart thermal grids; heat station
• ISSN: 0040-6015; 1555-6301
• DOI: 10.1134/S0040601522060088

As an effective means of energy use, district heating is widely used all over the world. The 4th generation of district heating further puts forward the concept of smart thermal grids and emphasizes the application of heating load prediction technology. This paper mainly reviews the three core elements of data preprocessing, modeling method and input eigenvalue, which affects the accuracy of heating load prediction. At the same time, based on the actual engineering investigation, the difficulties in the application of heating load forecasting in actual engineering are analyzed. On this basis, in order to promote the adoption rate of the 4th generation district heating, the application of heating load prediction in practical engineering is optimized from the following two aspects: enrich and perfect the heating load forecasting database to provide sufficient data support for the heating load forecasting; A partitioning scheme for data set construction of heating load prediction model is proposed in order to improve the accuracy of the prediction model. In order to prove the applicability of the proposed methods, the methods are applied to a district heating system in Kaifeng, and the results show that the prediction effect is obviously improved.