Experimental evaluation of radiator control based on primary supply temperature for district heating substations

Abstract

In this paper, we evaluate whether the primary supply temperature in district heating networks can be used to control radiator systems in buildings connected to district heating; with the purpose of increasing the Δ T. The primary supply temperature in district heating systems can mostly be described as a function of outdoor temperature; similarly, the radiator supply temperature in houses, offices and industries can also be described as a function of outdoor temperature. To calibrate the radiator control system to produce an ideally optimal radiator supply temperature that produces a maximized Δ T across the substation, the relationship between the primary supply temperature and outdoor temperature must be known. However, even if the relation is known there is always a deviation between the expected primary supply temperature and the actual temperature of the received distribution media. This deviation makes the radiator control system incapable of controlling the radiator supply temperature to a point that would generate a maximized Δ T. Published simulation results show that it is possible and advantageous to utilize the primary supply temperature for radiator system control. In this paper, the simulation results are experimentally verified through implementation of the control method in a real district heating substation. The primary supply temperature is measured by the heat-meter and is shared with the radiator control system; thus no additional temperature sensors were needed to perform the experiments. However additional meters were installed for surveillance purposes. To maintain a stable indoor temperature at times when the primary supply and outdoor temperatures deviates from their assumed relation, the radiator system flow must be controlled by an additional control-loop. The results confirms that it is possible to control the radiator system based on the primary supply temperature while maintaining comfort; however, conclusions regarding improvements in Δ T were hard to distinguish.