Design of a hydraulic turbine control system by numerical optimization

Abstract

© 2015 Dime universita di genova, dimeg university of calabria. All rights reserved.This work proposes the use of numerical optimization based on Direct Transcription as a method for the design of the control system for hydroelectric turbines. The control design for this application involves constraints but the usual control techniques do not allow explicitly to incorporate constraints in its formulation or are very sensitive to initial estimates of the optimization problem implying in convergence issues. The Direct Transcription is an alternative optimization-based design where the dynamics are discretized and included as constraints of the optimization problem causing the errors due to the quality of the initial estimate to be diluted over the discretization nodes. The constraints considered are related to the actuator and other operating limits during a change maneuver of operating point.