Regime-Aware Bayesian Identification of the Viscoplastic Dissipation Potential in Heritage Paint Films from Nanoindentation

Standard nanoindentation analysis suppresses the time-dependent response of a tested material, discarding the property that governs how a heritage paint film survives climatic loading: its viscoplastic dissipation potential. We recover it, together with the elastic modulus, from a single displacement-controlled relaxation indent by Bayesian inverse identification through a Gaussian-process surrogate of a finite-element contact model. The central difficulty is that the constitutive law is adequate during relaxation but structurally violated during loading, where the particulate film deforms by granular compaction. Overweighting the loading forces the optimiser to absorb this model-form discrepancy into the constitutive parameters, leaving a deceptively good global fit. We expose this failure mode through a three-way audit of weighting schemes: expert-encoded physics, Fisher-information optimality, and a data-driven Empirical Bayes filter. Information-optimal weighting under model-form error produces the most biased parameters with an equally tight posterior, an effect we term Constitutive Overfitting. A polycarbonate control, for which the model is uniformly adequate, confirms that the three schemes converge, establishing the audit as a diagnostic of model inadequacy rather than a pipeline artefact. This work has two outcomes. The first is the complete viscoplastic dissipation potential of a heritage paint film, the rate-dependent descriptor that fracture and digital-twin models require, from a single high-information indent. The second is the audit: an unaudited inversion would report a tight posterior on a biased modulus, whereas the three-way comparison separates reliable parameters from artefacts wherever constitutive fidelity varies within a single experiment. Population-level variability is deferred to a forthcoming multi-indent study.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.23307930
Primary Topic
Metal and Thin Film Mechanics
Type
preprint

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Regime-Aware Bayesian Identification of the Viscoplastic Dissipation Potential in Heritage Paint Films from Nanoindentation

Naoki Fujisawa, Michał Łukomski, Lorenzo Riparbelli
Zenodo (CERN European Organization for Nuclear Research)
Metal and Thin Film Mechanics
preprint

Regime-Aware Bayesian Identification of the Viscoplastic Dissipation Potential in Heritage Paint Films from Nanoindentation

Naoki Fujisawa, Michał Łukomski, Lorenzo Riparbelli
preprint en

Abstract

Standard nanoindentation analysis suppresses the time-dependent response of a tested material, discarding the property that governs how a heritage paint film survives climatic loading: its viscoplastic dissipation potential. We recover it, together with the elastic modulus, from a single displacement-controlled relaxation indent by Bayesian inverse identification through a Gaussian-process surrogate of a finite-element contact model. The central difficulty is that the constitutive law is adequate during relaxation but structurally violated during loading, where the particulate film deforms by granular compaction. Overweighting the loading forces the optimiser to absorb this model-form discrepancy into the constitutive parameters, leaving a deceptively good global fit. We expose this failure mode through a three-way audit of weighting schemes: expert-encoded physics, Fisher-information optimality, and a data-driven Empirical Bayes filter. Information-optimal weighting under model-form error produces the most biased parameters with an equally tight posterior, an effect we term Constitutive Overfitting. A polycarbonate control, for which the model is uniformly adequate, confirms that the three schemes converge, establishing the audit as a diagnostic of model inadequacy rather than a pipeline artefact. This work has two outcomes. The first is the complete viscoplastic dissipation potential of a heritage paint film, the rate-dependent descriptor that fracture and digital-twin models require, from a single high-information indent. The second is the audit: an unaudited inversion would report a tight posterior on a biased modulus, whereas the three-way comparison separates reliable parameters from artefacts wherever constitutive fidelity varies within a single experiment. Population-level variability is deferred to a forthcoming multi-indent study.

Zenodo (CERN European Organization for Nuclear Research)
Getty Conservation Institute (US), University of Florence (IT)
European Commission
Metal and Thin Film Mechanics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon