RULE-BASED THERMOSTAT DESIGN FOR NIGHT SETBACK OF HEAT-PUMPS

We introduce a computer model for testing heat-pump thermostat performance with regard to energy and cost savings due to night setback. Also introduced are two new control strategies for heat-pump night setback: 1. (i) an inexpensive variation of existing control strategies and 2. (ii) a rule-based control strategy using techniques of artificial intelligence. The computer model is based on a transient system model of the residence, heat-pump, and controller, in response to actual weather data. Seasonal energy usage and costs are computed for each control strategy, subject to the effects of variable heat-pump sizing, building thermal inertia, building heat loss, climatic region, and electric rate structure. Results show that night setback can reduce heating season costs in residential heat-pump systems if the control strategy takes into account the characteristics of both the heat-pump system and the building. If these factors are not considered, night setback can increase heat-pump operating costs. The rule-based control strategy consistently provides positive cost savings above other setback methods, and has the added potential of providing diagnostic information about both the heat-pump system and the residence. © 1990.