Data—Driven Design of a Fault Tolerant Fuzzy Controller for a Simulated Hydroelectric System

Abstract

This paper proposes a data-driven approach oriented to the design of a fault tolerant fuzzy controller for regulating the speed of a Francis turbine included in a hydroelectric system simulated in the Matlab and Simulink environments. The nonlinear characteristics of hydraulic turbine and the inelastic water hammer effects are considered to simulate the dynamic process. This strategy is suggested for enhancing the regulator derivation that could represent an alternative to the standard controllers in typical hydroelectric systems. The controller development requires the knowledge of the dynamic model of the monitored system, which is achieved by means of a fuzzy modelling and identification scheme. This feature of the work, followed by the proposed solution relying on a data-driven approach, represents the key point when on-line implementations are considered for a viable application of the proposed scheme. Moreover, by means of this design scheme, the proposed strategy is also able to provide a fault tolerant controller. In particular, the fault tolerance properties are achieved by using a so-called passive approach. It is assumed that the fault considered in this work affects the electric servomotor used as governor. The performances of the fault tolerant fuzzy controller is compared to that of a PID regulator and an adaptive PID controller scheme already implemented for the simulated hydroelectric system.