Studies on the heating capacity control of a multi-type heat pump system applying a multi-input multi-output (MIMO) method

This paper describes a method to control capacity of a multi-type heat pump system. Because indoor units are interconnected, alteration of the heating or cooling setting of one indoor unit can influence the performance of the other indoor units. Proportional-integral (PI) controllers were used to assess system performance, and the PI controllers’ gain values were optimized by using a genetic algorithm. A system model was established following a system identification optimization process, and a pseudo random binary sequence was selected as the system identification input signal. A multi-input multi-output (MIMO) controller was more effective in reducing such cross-coupling effects than a single-input single-output (SISO) controller. The temperature at the secondary fluid outlet in the condenser and the saturation pressure at the evaporator were selected as the control variables in the MIMO controller. The experimental results showed that an optimized MIMO controller could reduce overshoot by up to 40% compared with the results using a SISO controller.