Contribution To Advanced Faults Diagnosis in Renewable Energy Conversion Chain

Abstract

Wind energy plays an increasingly crucial role in electricity production, requiring proactive involvement of wind farms in managing the electrical grid. Additionally, diagnosing potential faults in the wind energy chain and detecting failures are major priorities for both industry and research. This thesis aims to enhance the energy quality of variable-speed wind turbines using a doubly-fed induction generator (DFIG) by developing a comprehensive control method. The primary objective is to regulate active and reactive power through field-oriented control (FOC) to meet the requirements of the wind farm control system. To achieve this, three types of controllers were investigated: proportional-integral (PI) controller, fuzzy logic controller (FLC), and neural network-based controller (NNC). Simulations have shown the superiority of NNC in terms of dynamic responsiveness and precise tracking of power reference values. Simultaneously, a diagnostic method using fuzzy logic was devised to monitor and detect ITSC and open-phase circuits in the stator windings of the DFIG within the wind system. This method solely utilizes phase currents to detect and locate faults in real-time. Furthermore, to address converter failure issues, a fault detection technique for pulse width modulation (PWM) inverters was studied. This technique combines fuzzy logic and neural networks to accurately identify short-circuit and opencircuit faults in the wind generator inverter