A Data-driven Model of Pipe Diameter and Insulation Thickness Optimization for District Heating Systems. Issue 1 (1st January 2022)

• Record Type: Journal Article
• Authors: Chen, Kailun; Hu, Jiayi; Yu, Libin; Zheng, Menglian; Sun, Shien; He, Dong; Lin, Junguang
• Is Part Of: Journal of physics. Volume 2166:Issue 1(2022)
• Publication Date: 2022-01-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2166/1/012046
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: One of the major downsides of district heating systems is the large thermal loss through the distribution pipe network, which is expected to be alleviated through the optimization for the thickness of the insulation layer and the pipe diameter. Although a few previous studies have developed a diversity of techno-economic models for optimizing the pipe diameter or the insulation thickness of the district heating pipeline, few of them were able to capture the complicated relationship among the heat loss, pipe diameter, insulation thickness, operation parameters, and complex exterior conditions in the real world (such as the conditions of the soil and degradation condition of the insulation layer). The present study devises a combined model that first bases on the heat transfer model to theoretically estimate heat losses and then bases on the data-driven artificial-neural-network model to fill in the gap between the measured and theoretical heat losses. Results show that the predicted heat losses via the developed combined model reproduce the measured data well. Furthermore, compared with the original pipe diameter and insulation layer thickness, the optimized diameters of the pipe and thickness values of the insulation layers via the proposed combined model are all elevated.