Multi-feature Fusion and Damage Identification of Large Generator Stator Insulation Based on Lamb Wave Detection and...

Area of Science:

• Materials Science and Engineering
• Electrical Engineering
• Non-destructive Testing

Background:

• Structural Health Monitoring (SHM) of composite materials, particularly large generator stators, requires effective insulation condition evaluation.
• Existing Lamb wave-based damage detection methods struggle to accurately identify stator insulation damage using single signal features.

Purpose of the Study:

• To develop and validate a multi-feature fusion method for enhanced identification of stator insulation damage using Lamb waves.
• To overcome the limitations of single-feature analysis in Lamb wave-based structural health monitoring.

Main Methods:

• Extraction of diverse damage features from Lamb wave signals in the time domain, frequency domain, and fractal dimension.
• Utilized signal processing techniques including Hilbert transform (HT), power spectral density (PSD), fast Fourier transform (FFT), and wavelet fractal dimension (WFD).
• Employed a Support Vector Machine (SVM) machine learning model for multi-feature fusion, reconstruction, and damage type identification.

Main Results:

• Successful extraction of multiple distinct damage-sensitive features from Lamb wave signals.
• Demonstrated the efficacy of SVM in fusing these multi-features for improved damage identification accuracy.
• Validation of the proposed method through both simulation and experimental testing on typical stator insulation damage scenarios.

Conclusions:

• The proposed Lamb wave multi-feature fusion method offers a significant advancement in stator insulation damage identification for SHM.
• Combining features from different domains enhances the robustness and accuracy of damage detection compared to single-feature approaches.
• This technology holds potential for reliable insulation condition evaluation in large generator systems.