District heating system is an important part of the urban energy system, but its large scale and numerous coupling variables bring many difficulties to its regulation and control. At present, rough regulation carried out by manual experience is the common control method, which is hard to guarantee the quality of regulation, thus caused problems about heating comfort and efficiency. The control of the heating substation is mainly to adjust the temperature of secondary supply water and return water by changing the valve opening on the primary side of the heating substation, but it’s hard to make the temperature of secondary supply water reach the set point under constantly changing weather, so it’s essential to establish a model to get the relationships between the temperature of secondary supply water and other variables. Given that numerous variables need to be included in this model and the relationship may be very complex, establishing a white-box model is expensive and time-consuming.This paper proposes a data-driven temperature response prediction model to predict the temperature of secondary supply water considering the historical operating status of heating substation, valve opening degree, and weather forecast data, etc. Non-time sequence models (MLP, XGBoost) and time sequence model (LSTM) prediction models are proposed and tested in this article, which are established under different input steps (6, 12, 24, 48, 72) and prediction steps (6, 12, 24) respectively. Results show that the prediction performance of the XGBoost model of 24 input steps and 24 prediction steps is best, with a mean square error of 0.24°C. Then it is applied to an urban central heating system in northern China for an application example. The prediction results of secondary supply temperature show that the maximum prediction error of the model is 0.37°C, and the average error is 0.27°C, which verifies the validity of this model.Based on this model, a predictive control strategy for guiding the primary side valve opening degree is formed, which determines the valve opening degree under different required secondary supply temperature. By use of this model, the temperature of secondary supply water can reach the set point through adjusting the primary side valve opening degree. In conclusion, the data-driven temperature response prediction model proposed in this article can reduce the dependence on manual experience and improve the control accuracy and heating quality, but for a small number of extreme conditions outside the training sample space, the ability of the data-driven model may be limited.

Investigation of Data-Driven Modeling in District Heating Substations

Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2021/volume/85543

ASME 2021 International Mechanical Engineering Congress and Exposition (IMECE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.