An experimental study from NASA Ames presents a multi-feature integration method for detecting and quantifying fatigue cracks in aircraft lap joints. Embedded PZT sensors perform in situ non-destructive evaluation during cyclical loading, correlating crack length with signal correlation, amplitude, and phase changes. The model was trained on data from five specimens and validated on joints from different manufacturers under varied loading conditions.
Use Cases
- Training predictive models for fatigue crack length estimation based on correlation coefficient, amplitude change, and phase change features.
- Developing multi-feature integration methods for damage quantification using second-order multivariate regression analysis.
- Benchmarking sensor-based non-destructive evaluation techniques for riveted joints under different loading conditions.
- Studying the variability in Lamb wave response across different structural specimens.
Strengths
- Experimental data from a NASA study, indicating a high standard of collection.
- Methodology validated on several lap joint specimens from different manufacturers and under different loading conditions.
- Model parameters derived from training datasets from five specimens.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- The dataset description focuses on methodology; the exact data format and features require inspection.
Provenance
- Source
- National Aeronautics and Space Administration (NASA)
- Collection Method
- Experimental study using embedded PZT sensors to perform in situ non-destructive evaluation during fatigue cyclical loading.
- Time Range
- null
- Freshness
- Last updated 2026-03-13 20:35:57.570473; freshness should be verified.
- Geography
- null