Optimal control and AI-based fault diagnosis for modern grid-connected wind turbines

Due to the depletion of fossil fuel resources and their environmental impact, the global energy market is transitioning toward renewable technology, with wind turbines playing an important role. However, due to their unpredictable nature, the functioning of the grid is not reliable. This paper proposes a new control technique for wind-based connected energy conversion systems using an online multi-objective PSO algorithm, which enhances the robustness and stability of the system with greater efficiency in maximum power point tracking (MPPT). It also optimizes dynamic performance by fine-tuning the proportional integral (PI) regulator parameters. In addition, a machine learning (ML)-based fault detection framework has been implemented to improve reliability by recognizing and diagnosing turbine defects. The simulation findings show that integration of the multi-objective PSO algorithm in the PI controller ensures optimal system control, whereas ML-driven diagnostic tools successfully categorize defects and contribute to improved operational performance.