Hybrid predictive supervisory control based on genetic algorithms for a gas turbine of combined cycle power plants

This work considers the optimization of the gas turbine for a combined cycle power plant by using a supervisory level. The design of a hybrid predictive supervisory controller is based on state space discrete model including the switching behavior of PI control system. The control design is based on an objective function that represents the economic and regulatory performance of a gas turbine by using a dynamic optimal set-point for the regulatory level. The hybrid predictive supervisory control problem considers the hybrid behaviour using a mixed logical dynamical systems model within the optimization problem and solved by Genetic Algorithms. The proposed algorithms are applied to the gas turbine of a thermal power plant and successfully compared with the regulatory control strategy with constant optimal set-points.