Fault detection and diagnosis strategy for wind turbine system using partial least square technique

Process monitoring has been used in industry for detecting and diagnosing abnormal behaviour of processes which abundantly improves the overall efficiency of such processes. Recently, the wind turbines’ installed power generation capacity is rising worldwide, so, it is important to achieve reliability and safety on such systems. The major task behind this study is to provide an inclusive framework for the fault detection and diagnosis of the wind turbine system. According to the proposed methodology, the faults are first detected by two fault detection indices using the partial least squares (PLS) approach. Then, the diagonal contribution plot (DC) method based on PLS approach (DC-PLS) is used for fault diagnosis. Data collected from a benchmark of a wind turbine system is utilized to confirm the execution of the proposed methodology. The results proved the efficacy of the proposed methodology from the point of view the fault detection and diagnosis.