Microstructure-Based Finite-Element Prediction of Surface Integrity with Bayesian-Informed Wear States

Beschreibung

Surface integrity plays a decisive role in the functional performance, fatigue life, and reliability of machined components. This dissertation presents a physics-informed and uncertainty-aware framework for predicting wear-dependent surface integrity during longitudinal turning of AISI 4140 steel. The approach combines tool-integrated thin-film sensors, experimentally validated two- and three-dimensional finite-element simulations, microstructure-based modelling of dynamic recrystallisation, and Bayesian inference for tool-wear estimation. By integrating sensing, physics-based simulation, surrogate modelling, and probabilistic inference, the proposed framework enables uncertainty-aware prediction of thermomechanical loads, microstructural evolution, and ultrafine-grained surface-layer formation. The resulting hybrid soft-sensor architecture establishes a foundation for future process-accompanying monitoring, digital twins, and closed-loop machining applications. The dissertation is intended for researchers and engineers working in manufacturing science, machining, finite-element modelling, surface integrity, and intelligent production systems.

Autor*in

Germán González Fernández

Themen in »Microstructure-Based Finite-Element Prediction of Surface Integrity with Bayesian-Informed Wear States«

Turning Machining Surface Integrity FEM

Stimmen zu »Microstructure-Based Finite-Element Prediction of Surface Integrity with Bayesian-Informed Wear States«

Details

Link teilen