Digital Metrology for Nanoindentation: Synthetic Data Generator for Error Identification

Area of Science:

• Metrology
• Materials Science
• Data Science

Background:

• Digital metrology enhances manufacturing accuracy and efficiency using digital tools.
• Digital Twins are key technologies for predictive maintenance and optimized performance.
• Establishing traceability for Digital Twins and algorithms in digital metrology is a significant challenge.

Purpose of the Study:

• To develop a parametric synthetic dataset generator for nanoindentation.
• To incorporate correlation and covariance among simulated quantities for realistic data.
• To enable traceable and physics-informed datasets for digital metrology.

Main Methods:

• A power-law formulation fitted via Orthogonal Distance Regression models indentation responses.
• The generator simulates quasi-static, room-temperature nanoindentation data.
• Uncertainty is associated with simulated results for metrological framework integration.

Main Results:

• The synthetic dataset generator provides traceable and physics-informed data.
• Performance is comparable to non-parametric methods like bootstrapping.
• The generator offers significantly reduced computational cost and improved representativeness.

Conclusions:

• The developed methodology simulates key nanoindentation measurement errors.
• It provides a traceable tool based on synthetic data.
• This tool can train advanced quality control systems for error detection.