This study systematically examines AE signal evolution across different crack propagation stages through controlled experiments. A multi-parameter cross-correlation analysis is introduced to quantify the interdependencies among key AE parameters, offering a more comprehensive assessment than traditional single-parameter methods.

The results reveal that AE amplitude, energy, event count, and duration exhibit distinct variations as cracks grow. Notably, energy, event count, and duration demonstrate strong positive correlations, making them robust indicators for crack propagation pattern recognition. In contrast, rise time and peak count show more scattered distributions, reflecting localized damage characteristics.

Additionally, AE signals from surface cracks exhibit higher amplitude and energy than those from deep-embedded cracks, validating the spatial attenuation effect and providing a quantitative basis for crack depth estimation. This study presents a multi-parameter correlation-based AE signal analysis method, enhancing AE-based damage classification and monitoring accuracy.

The proposed approach strengthens the theoretical foundation for structural health monitoring (SHM) and fatigue damage early warning, while also contributing to the optimization of non-destructive testing (NDT) techniques in engineering applications.

Highlights

• Multi-parameter AE analysis enhances fatigue crack monitoring accuracy.
• AE energy, event count, and duration strongly correlate with crack propagation.
• Surface cracks show higher AE amplitude and energy, aiding depth assessment.
• Cross-correlation analysis improves AE signal classification and damage detection.
• Proposed AE-based monitoring framework supports SHM and fatigue prediction.

Read the full article at Science Direct.