Damage Feature Extraction and Parameter Characterization of Large Generator Stator Insulation Based on Lamb Waves Detection Method

Abstract

Insulation degradation assessment plays a vital role in the life management of power equipment. Due to the merits of Lamb wave to Structural Health Monitoring (SHM) of composite, the Lamb wave-based damage detection method shows a potential solution for the evaluation of the insulation condition of stator windings in high-voltage rotating machines. The main objective of this paper is to study damage characteristic parameters of large generator stator insulation based on Lamb wave detection method. The fundamental asymmetric mode A0 wave packet of Lamb wave was used to detect the stator insulation damage. Firstly, the damage characteristic parameters are extracted from time domain, frequency domain and fractal dimension respectively. Hilbert transform (HT), power spectral density (PSD) analysis, fast Fourier transform (FFT) and wavelet fractal dimension (WFD) methods were used to extract features of Lamb wave signals received in stator insulation damage detection and the change rule of damage characteristics were explored. Then, the correlation between characteristic parameters and damage size were concluded based on the partial least square (PLS) multivariate statistical analysis approach. Finally, detection of four typical insulation damages, i.e. void, delamination, puncture, and crack are carried out by FEM simulations and experiments for validation. These damage feature extraction methods and characteristic parameters of stator insulation can provide a valuable reference information for the condition assessment of large generator stator insulation furtherly.