Modified Fuzzy-Q-Learning (MFQL)-Based Mechanical Fault Diagnosis for Direct-Drive Wind Turbines Using Electrical Signals

In this paper, a self-learning multi-class intelligent model for wind turbine fault diagnosis is proposed by using MFQL (Modified-Fuzzy-Q-Learning) technique. The adaptive in nature and extension of fuzzy-Q-learning method where look-up table Q-learning conquered fuzzy based approximation strategy to reduce the curse dimensionality Q-learning. classifier diagnoses mechanical imbalance faults without sensors. Proposed methodology addressed with relying on PMSG (Permanent Magnet Synchronous Generator) stator current signals, which already being used protection system turbines. According aforementioned description, non-stationary signals have been pre-processed extract input features empirical mode decomposition followed J48 algorithm most relevant feature selection. For one-step ahead performance demonstration approach, results compared neural network, support vector machines, logic, conventional Fuzzy-Q-Learning techniques. Demonstrated outperform capability approach. Moreover, developed first time implement area WTGS literature.