![](/ii/tfec2019.jpg)
ISSN Online: 2379-1748
ISBN Flash Drive: 978-1-56700-483-0
ISBN Online: 978-1-56700-482-3
4th Thermal and Fluids Engineering Conference
ON-LINE FUZZY CONTROL OF A MULTI-ROOM BUILDING FACILITY
Abstract
We extend our previous work to develop effective strategies for temperature control of a multi-room subscaled building facility. The main goal, here, is to maintain the temperature in each of the rooms of the building at specific settings, while taking into account energy coupling from adjacent rooms. The system consists of eight rooms, distributed in two floors, with a cooling unit that provides cold air to the system, and 25W light bulbs that act as heat source in each room. Two K-type thermocouples are placed in each room to gather temperature-readings, and eight dampers are used deliver the airflow from the cooling unit. The
controller structure is built upon the fuzzy logic technique, with expert-based rules and linguistic variables
describing the behavior of the system. The fuzzy sets are quantified via triangular membership functions, and the rules built from experimental data. A Mamdani inference method is used to defuzzify the output signals
providing a crisp value to the actuators. The control variables are average temperatures inside the rooms, and
the manipulated ones are flow rates of the airflow. The structure of the controller is developed in a way that
it achieves a specific temperature setpoint and assists the adjacent rooms in decreasing the time needed to attain their respective setpoints. Temperature readings and control actions are executed via LabVIEW, and MATLAB is used to implement the fuzzy controllers, while experiments are conducted to assess their relative performance. Results demonstrate that this fuzzy controller can effectively provide temperature control of the test-bed, while providing faster response than that of the controllers developed in our previous work [3].