Detecting Incipient Inter-Turn Short Circuit Faults in DFIGs Based on Magnetic Flux Leakage Theory and MPJ-SVDD Analytical Method

This manuscript presents the Magnetic Flux Leakage (MFL) theory for incipient inter-turn short circuit (ITSC) faults in the Doubly Fed Induction Generators (DFIGs), which involves the simulation and modeling of the failure mechanism, the quantitative and qualitative description analysis, the health indicator construction, and early weak fault detection and state evaluation based on MFL theory. Firstly, the fault state and the MFL model of the wind generator were constructed from virtual simulation using Matlab/Simulink, due to the fault mechanism of the DFIGs is complex and not straightforward under varying operating conditions and operation loads, therefore the simulations are used in order to obtain datasets and conduct theoretical analysis, including the synthetic failure datasets, feature information representation with feature-based harmonic, feature-based statistical, feature-based entropy, metrics-based distance, feature importance ranking and fusion, development of health indicator and performance analysis based on Multi-Parameter Joint Analysis and Support Vector Data Description (MPJ-SVDD). Finally, the magnetic flux leakage theory for ITSC faults and MPJ-SVDD analytical method is verified by simulation experiments and obtains a better effect of the quantitative and qualitative description.