Identifying supply-demand mismatches in district heating system based on association rule mining

Abstract

Supply-demand mismatches in district heating system (DHS) may affect system's safe and stable operation without reasonable regulation actions in time. The accurate identification of supply-demand matching states is the prerequisite for heat balance regulation. However, this job is quite difficult and time-consuming, as DHS is typically a large time delay system and operates under various conditions. This paper proposes an association rule mining-based method to identify supply-demand mismatch from DHS operation data. Considering DHS's thermal inertia and fault-tolerance, the supply-demand mismatch rate is firstly defined along with its quantitative evaluation model. Then the theoretical intervals of the heating parameters are obtained by establishing association rules and rules post-processing. And a reasonable error band is decided by comparing the occurrence times of mismatches to balance identification accuracy and sensitivity. The proposed method is tested and verified in a typical DHS. The rules between primary supply temperature and relevant operation parameters are mined. An error band of ±7.5% is determined by comparison with historical operation data. The supply-demand mismatch rate is calculated to be −10%–15%. The identification results are validated by analyzing the heat substations' operation data including supply and return temperature, flow rate, and valve opening.