Digital Metrology for Nanoindentation: Synthetic Data Generator for Error Identification

纳米压痕数字计量:误差识别的合成数据生成器

阅读:1

Abstract

Digital metrology enables precise, real-time measurement and data analysis using digital tools, which enhances accuracy and efficiency in manufacturing and quality control. Among key enabling technologies, Digital Twins allow continuous control, enabling predictive maintenance, faster error detection, and optimised performance of the measurement system. A current challenge is establishing traceability for the Digital Twins and for the data processing algorithms implemented in digital metrology. Nanoindentation is a challenging measurement technique that may be susceptible to several random and systematic measurement errors. This work presents a parametric synthetic dataset generator for quasi-static, room-temperature nanoindentation that incorporates correlation and covariance among simulated quantities. The method models indentation responses through a power-law formulation fitted via Orthogonal Distance Regression, allowing for traceable and physics-informed datasets. The generator enables the association of uncertainty with simulated results, supporting its use within a metrological framework. Its performance is benchmarked against non-parametric methods such as bootstrapping, showing comparable accuracy with significantly reduced computational cost and improved representativeness. Furthermore, the methodology can simulate main measurement errors for advanced material characterisation and develops a traceable tool based on synthetic data which could be used to train advanced quality control tools for the detection of main measurement errors.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。