Comparing full-field data from structural components with complicated geometries

Area of Science:

• Structural analysis
• Data processing
• Numerical methods

Background:

• Structural analysis often involves irregularly shaped datasets with missing data due to measurement limitations or component geometry.
• Existing decomposition techniques like Chebyshev or Zernike may require data interpolation or warping, potentially biasing comparisons.
• Accurate comparison of stress and deformation fields is crucial for understanding component behavior under various loads.

Purpose of the Study:

• Introduce a new decomposition algorithm based on QR factorization for processing and comparing irregularly shaped stress and deformation data.
• Enable direct comparison of 2D data fields with missing information without interpolation or warping.
• Ensure similarity metrics are not biased by incomplete datasets.

Main Methods:

• Developed a decomposition algorithm utilizing QR factorization.
• Applied the algorithm to compare irregularly shaped 2D stress and deformation datasets.
• Validated the technique on data from finite-element analysis, digital image correlation, and thermoelastic stress analysis.

Main Results:

• The QR factorization-based algorithm effectively processes and compares irregularly shaped datasets with missing data.
• The method avoids interpolation and warping, ensuring comparisons are based solely on available data.
• Successful application demonstrated in impact, modal analysis, and fatigue studies.

Conclusions:

• The new decomposition algorithm provides a robust method for analyzing incomplete structural data.
• The technique offers unbiased similarity metrics, crucial for accurate structural assessment.
• Potential applications span various fields requiring the analysis of complex, irregularly shaped datasets.