Open phase fault-tolerant support vector machine predictive power control for six-phase induction generator WECS

Wind energy systems are often located in remote areas or offshore, making maintenance and repair both expensive and logistically challenging. Fault-tolerant systems can assist in minimizing the frequency and urgency of maintenance, ultimately reducing operational costs. This study aims to design a simple and efficient Open-Phase Fault Tolerant (OPFT) control for a wind power system (WPS) based on an asymmetric six-phase induction generator using Finite-Set Predictive Power Control (FS-PPC). The suggested technique involves three key steps. Firstly, the current harmonics in the ( x, y ) plane are analyzed to discover faults. Then, open-phase localization is achieved using the Support Vector Machine (SVM) with hyperparameter Bayesian Optimization (BO). Finally, the phase that forms 90 degrees with the faulty phase is opened to restore the system’s stability. Importantly, this approach does not require reconfiguring the control algorithm while preserving the system’s effective performance. Simulation results demonstrate the effectiveness of the OPFT-SVM-PPC control strategy in preserving control over the machine while ensuring high energy quality for the grid with a THD of 2.71%. By implementing this fault tolerance control, the system can operate reliably and deliver high-quality power, even in the presence of open-phase faults.