Embedded Control Of Scaled Cooling System

Abstract

Buildings operations account for 41.2 percent of the total energy consumption and greenhouse gas emissions in the U.S [39]. A Large percentage of the energy consumed is wasted due to the high inefficiency of buildings. There are two conventional methods for testing energy efficient technologies in a new building: full scale prototype and computer simulation. Full scale prototypes yield actual results of a building's energy usage. However, this method is extremely expensive and resource consuming. Moreover, it is nearly impossible to compare two different designs due to the difficulties related to reproducing the building and the weather testing conditions. Pure computer simulation is also limited in its accuracy due to constraints on processing power and simplifying assumptions [21]. The overall objective of this research is to design, create and build a scaled test bed that allows engineers to better understand the behavior of buildings and contrast and calibrate computer models with real data collected from the test bed. The test bed consists of a modular and modifiable building envelope, a weather simulation enclosure and a wirelessly controllable Heating, Ventilation and Air Conditioning (HVAC) system. The thesis contributions to the overall objectives are the design and implementation of the hardware and software algorithm for the controller of the scaled HVAC system. This includes implementing Single-Input-Single-Output (SISO) PID loops to control the individual thermal states of the HVAC. A Wireless Sensor Actuator Network iii (WSAN) based on Zigbee protocol is designed and implemented to facilitate two way communications between the PC and the scaled HVAC system. Finally, an equation-based model of a centrifugal pump is created and calibrated using the pump's datasheet. A model based optimization is implemented on a centrifugal pump to maximize its efficiency. iv