Frequency and phase tracking for excitation capacitor fault detection in a self-excited induction generator using the ZCT method

Abstract

This paper presents a methodology for detecting excitation capacitor faults in self-excited induction generators (SEIGs) used in wind energy systems operating in isolated or rural areas. The proposed approach combines frequency and phase tracking with zero-crossing time rate analysis for real-time anomaly detection. The SEIG is modelled using the finite element method (FEM) to simulate its electromechanical behaviour under various fault conditions. Fault scenarios are introduced by progressively disconnecting excitation capacitors connected in parallel and series, thus simulating failures ranging from one to five units. Monitoring zero-crossing intervals and phase variations enables accurate fault detection without relying on computationally intensive spectral techniques. The results confirm the effectiveness of the proposed methodology in identifying fault-induced distortions and improving the reliability of SEIG-based wind energy systems.