Multi-Sensor Single-Actuator Control of HVAC Systems

Abstract

It is common to control several rooms in a building with a single sensor in one of the rooms and a single actuator driving just one control element such as an air damper. New, low-cost, wireless sensor technology now offers the opportunity to replace the single sensor in one room with a network of sensors having at least one sensor per room. This paper addresses this multi-sensor, single-actuator control problem. We used computer simulations and optimization to study the problem. We designed a computer simulation of the heat transfer behavior of a section of a building that accounted for the effects of weather, building materials, ventilation, and loads from occupants and equipment. We considered ad hoc methods (such as averaging) of using information from multiple sensors. We also developed a new, model-free method of using information from multiple sensors that is based on a simple optimization procedure. The optimization procedure can be configured to optimize comfort or to optimize energy under comfort constraints. We compared the performance of the single-sensor strategy with the ad hoc strategies and optimized strategies using annual simulations of a four-room, perimeter section of a building and weather data from Sacramento, California. We report heating and cooling energy performance along with two comfort metrics, the average number of rooms within the ASHRAE comfort zone and the Predicted Percentage Dissatisfied (PDD). The results show that most of the multi-sensor control strategies do better than the single-sensor strategy on the basis of both energy performance and comfort. The energy-optimal strategy reduces energy consumption by 17% while reducing PDD from 30% to 24%. The comfort-optimal strategy reduces energy consumption by 4% while reducing PPD from 30% to 20%. The performance improvements occur primarily when the average load among all rooms is nearly zero, with some rooms requiring heating while others require cooling. Under these conditions, the single-sensor strategy either overcools or overheats, whereas the multi-sensor strategies use almost no energy.