A dynamic control strategy of district heating substations based on online prediction and indoor temperature feedback

Abstract Refined control is significant to ensure on-demand heating and efficient operation in district heating system (DHS). This paper proposes a dynamic control strategy for substations based on online prediction and indoor temperature measurement. Firstly, cross-correlation function method coupled with variable time window, which is a dynamic time lag analysis method, is introduced to analyze the delay time between indoor and comprehensive outdoor temperature. This time lag is used to decide control period. Then, an online multiple linear regression (MLR) model is introduced to predict the supply temperature. The prediction value is adjusted according to the deviation of the consumers’ set point indoor temperature and the actual indoor temperature before sent to the substation controller. The proposed strategy was applied in a practical DHS engineering, and the results showed that the indoor temperature non-uniformity coefficient was reduced from 0.05 to 0.04, the overall heating season heat consumption index was reduced, and the energy saving rate was about 6%.