A novel fault diagnostic method in power converters for wind power generation system

In order to decrease the downtime, enhance maintenance efficiency, and reduce the management cost of the wind power system, fault diagnosis technologies are considered as powerful tools for its good operation and maintenance. For example, high fault diagnosis accuracy can ensure effective fault-tolerant control and improve the durability of wind turbines under fault conditions. However, fault diagnosis accuracy can be easily affected by nonlinear and noise of measured signals under different working conditions. In this paper, a novel fault diagnostic method of power converters is proposed for the wind power generation system. In the proposed method, the measured output voltage is firstly processed by ensemble empirical mode decomposition (EEMD), and a series of intrinsic mode functions (IMF) can be obtained. Norm entropy (NE) is then calculated based on the statistical characteristics of the IMFs, and the extracted IMF-NE information is used to describe the diagnostic features. The effectiveness and reliability of the proposed method are then validated in a simulated 1.5 MW doubly-fed wind power system. The results show that the final diagnostic accuracy of 22 fault modes is 99.57% for wind speed variation, and the diagnosis accuracy can be maintained above around 70% for different signal to noise ratio. Compared with the other advanced fault diagnosis methods, the proposed method shows outstanding performance in terms of robustness, high accuracy, and simple implementation without complex parameter tuning.