Edinburgh Napier University

  The diagnostic of electrical machines is of paramount importance, particularly with the Motor Current Signature Analysis (MCSA) method. MCSA stands out as a potent and non-intrusive tool for evaluating the condition of electric motors. This paper introduces an innovative approach for the diagnosis of rotor asymmetries, broken rotor bars, in induction machines operating in steady state directly fed from the grid, emphasising the decomposition of line currents within specific rotating frames. A novel methodology is proposed, and its results are meticulously compared with those obtained through the conventional Fast Fourier Transform (FFT) diagnostic technique. In particular, the newly introduced methodology exhibits enhanced sensitivity, surpassing the capabilities of conventional FFT. Additionally, the innovative approach is proven to be capable of quantifying the increase in fault harmonics in correlation with the load on the machine, aligning with the results observed through conventional FFT analysis. The findings strongly indicate that this pioneering technique offers a more robust and reliable means of diagnosing induction machines when compared to the traditional FFT method, by enhancing sensitivity and providing a nuanced understanding of fault harmonic variations with load changes.