LAUSR

This paper presents an online methodology to detect and diagnose the inductor degradation in LCL filter integrated with Active Front End Rectifier (AFER) for electric drives in marine propulsion system (MPS). The proposed methodology monitors the leakage fluxes of converter-side inductor and compares it with the same under healthy condition to detect and diagnose the inductance degradation. The magnitude of leakage flux's fundamental component is extracted using the Fast Fourier Transform (FFT) and is used for detection of inductance degradation. Hence, inexpensive and low bandwidth magnetic flux sensors along with a simple FFT algorithm are used to detect and diagnose the inductance degradation. The leakage fluxes of the inductor during degradation are analyzed using the developed analytical model and validated on a laboratory prototype of 11.3-kW MPS test-rig. The validation was carried out for different degrees of converter-side inductor degradation from 3% to 19% under 33%, 73%, and 100% of full-load capacity of the MPS test-rig. Both estimated results using the analytical model and experimental results obtained are found to be in good agreement. The proposed online technique can detect and estimate the degraded inductance as low as 3% and is superior to the technique using grid currents in detecting the inductor abnormality.