ANFIS‐based sensor fault‐tolerant control for hybrid grid

Abstract

This paper describes an intelligent sensor fault detection and compensation (FDC) scheme for a hybrid grid involving renewable energy (RE) sources with power electronic interfaces. To evaluate the sensor FDC scheme, a wound rotor induction generator (WRIG)-based wind energy system connected to the hybrid grid is examined. The system employs a WRIG with excitation from DC RE grid via a single, rotor side converter. To analyse the dynamic performance, a decoupled voltage vector control is employed. The sensors involved in the control algorithm may not be ideal and if fault occurs might lead to system collapse. To overcome electrical sensor failure, a robust, intelligent sensor fault control scheme is proposed with ANFIS. A hardware-free solution with PLL block is proposed for mechanical sensor fault. The ANFIS-based FDC scheme involves a bank of observers that computes residuals, detects and isolates faults. When fault occurs, the observer output is utilised for the execution of control algorithm thereby avoiding system collapse. The focused system is simulated in MATLAB and experimentally tested with a dSPACE controller for various fault scenarios. The results show the effectiveness of the FDC scheme.