Decentralized Model Predictive Control for a Cascade of River Power Plants

Abstract

River power plants interrupt the natural flow of a river and induce undesired fluctuations in the water level and water discharge. To prevent the adverse impacts of these fluctuations on the nature as well as on the navigation, the operation of the power plants needs to be regulated to obey certain restrictions imposed by the authorities, i.e., the water levels at specific points in the river have to be kept within certain bounds and large variations of the turbine discharges need to be avoided. In this chapter we present a Model Predictive Control (MPC) scheme to manipulate the turbine discharges of the power plants located in a cascade that will satisfy the restrictions imposed by the authorities. Since a centralized MPC scheme might become computationally infeasible for large cascades, we develop a decentralized MPC scheme, in which the cascade is decomposed into smaller subsystems and each subsystem is controlled by a local MPC scheme. We show through simulations that providing a downstream communication is sufficient to prevent significant performance deterioration in decentralized MPC, which would be expected due to the lack of coordination.