Model Predictive Control of an HVAC System Based on Dynamic Tracking and Optimization of Energy Use

U.S. Department of Energy affirms that HVAC systems consume approximately 40% of the total energy used in commercial-building sector. These types of systems are complex because they are composed of a large number of interconnected subsystems. The analysis shown in this paper is established on a building geographically located at the Caribbean coast region of Colombia in a region with tropical savanna climate and it is exposed to constant thermal load changes associated to high wall temperatures and direct sunlight incidence. Under this perspective, an energetic analysis is performed for the HVAC in order to implement a Model Predictive Control (MPC) strategy to enhance the system efficiency under the previously mentioned external conditions. The model predictive strategy is implemented as a system supervisor in order to minimize a cost function that measures the ratio of water consumption to air temperature change in the cooling coil. The strategy manipulates the required temperature of supply water to cooling coil from the chiller, perceiving as input perturbation the outdoor temperature, the desired temperatures for the classrooms and the desired temperature of the air supply to the different zones. The comparison and selection of thermodynamical states for analysis are conducted according to the dynamic characteristics of the entire system and individual components, and the energy assessment is performed including the system transient response. The accomplishment of the supervisory control strategy has demonstrated that dynamic energetic analysis and assessment is an auxiliary tool for HVAC performance management. The analysis performed shows that the supervisory strategy can reduce properly the energy performance index as a consequence, the energy consumption of the fan has a reduction of a 0.65%, while the water required shows a reduction of 66.93%.