|
|
 |
| |
|
|
|
Open Switch Fault Diagnosis for Three-Level Inverter based on Wavelet Packet Analysis and Hybrid LSSVM |
|
|
|
PP: 379-384 |
|
|
|
Author(s) |
|
|
|
Zhengwei Li,
Guojun Tan,
|
|
|
|
Abstract |
|
|
| Three-level neutral-point-clamped (NPC) inverter is widely used in the high-power and high-voltage applications. However,
the increasing of the power switch devices improves the system’s fault rate. In this paper, a novel fault diagnosis scheme is proposed
based on wavelet packet analysis (WPA) and hybrid least square support vector machine (LSSVM) for open switch fault diagnosis
of three-level NPC inverter. The waveforms of output phase voltage obtained from the faulty inverter are decomposed by WPA into
wavelet energy coefficient matrices, through which we can obtain energy vectors acted as features of time-series periodic waveforms.
And then a multi-classes classification method based on decision directed acyclic graph (DDAG) is presented to realize hybrid LSSVM
strategy. Through simulation and experimental results, the superiority of the proposed scheme is demonstrated. |
|
|
|
|
|
|
|
