SWT and Statistical Hypothesis Testing for Power Inverter FDI

 F.Charfi¹, S.Lesecq², F. Sellami^1*

Abstract
In this paper, a monitoring approach for fault detection in power system drives is presented. Due to its critical position in the power train the diagnosis of the two level three phase voltage inverter is of a great concern. Fault scenarios with single open-switch are considered because they are the most likely to occur. Several signals are analysed simultaneously in order to perform the diagnosis. The fault occurrence is revealed by a change in the mean value of a subset of the analysed signals. The diagnosis is realised in three steps. In the first step, signals are filtered using the SWT performed with the DB4 wavelet to extract the detail and approximation coefficients up to level 6. In the second step the approximation at level 6 is examined to detect changes in the mean. This is achieved with statistical hypothesis techniques. A Neyman-Pearson change in the mean detection test is used. Then at the third step, a signature table allows to isolate the faulty switch. The whole diagnostic procedure can perform on line because of its low computational cost. Real data recorded from a benchmark feed the proposed diagnostic tool. The results Presented here confirm the effectiveness of the proposed methodology.

Keywords: Fault Detection and Isolation (FDI), Stationary Wavelet Transform (SWT), Neyman-Pearson detection test, power converter, hypothesis testing.

(P1110907642, 1 MB)