Model predictive control of a laboratory gas turbine

This paper deals with the control of a laboratory gas turbine using model predictive control (MPC). The objective is to control the speed of the gas turbine. Firstly, an empirical transfer function model is obtained experimentally, between the fuel flow and shaft speed of the gas turbine. Then, based on the empirical models simple PI controllers are designed. The output responses under PI control are found to be unsatisfactory, with large overshoots and oscillatory behavior. Next, MPC controllers are designed based on the empirical models. Relevant operating constraints are also added as design specifications. It is found that the designed MPC controller delivers good performance.