Thesis defense of Andrés Sela
Thesis defense of Andrés Sela
Thesis defense of Andrés Sela
Title of the thesis: "Advanced measurement techniques to improve predictive modelling of cutting processes by using inverse simulation". Obtained the SOBRESALIENTE CUM LAUDE qualification and has received the ‘Doctor Internacional’ mention.
- Title of the thesis: "Advanced measurement techniques to improve predictive modelling of cutting processes by using inverse simulation".
- Court:
- President: Jose Carlos Outeiro (Arts et Metiers Institute of Technology)
- Vocal: Juan Asensio Lozano (Universidad de Oviedo)
- Vocal: François Ducobu (Université de Mons)
- Vocal: Idriss Tiba (Arts et Metiers Institute of Technology)
- Secretary: Iñaki Mirena Arrieta Galdos (Mondragon Unibertsitatea)
Abstract
Machining is one of the most widely employed manufacturing operations, contributing from 3 to 10% of GDP in developed countries. Despite this prevalence, tools and cutting conditions are often chosen on a trial-and-error basis which is costly and time consuming. In recent years, predictive models have emerged as one of the most promising approaches to address this issue. To function correctly however, predictive models require accurate data related to appropriate material properties (e.g., constitutive models, damage law).
Thermomechanical tests are usually not representative of the extreme conditions found in machining processes and thus, the material laws obtained through these approaches might not accurately reflect real industrial processes. Inverse identification has the potential to address this problem, as material properties are directly obtained from machining outcomes. However, acquiring information about some of these outputs, such as workpiece temperature, strain, and strain rate, remains a challenge.
The present study, therefore, presents improved techniques to measure temperature, plastic strain and strain rate to be used as input in an inverse approach. These techniques were applied to the thermomechanical characterization and orthogonal machining of a Ti6Al4V alloy. This alloy is widely used because of its good mechanical properties --high strength-to-weight ratio, high stiffness and toughness, and fatigue and corrosion resistance-- and its ability to maintain them at elevated temperatures. However, it is also classified as a difficult-to-cut material, due to the generation of segmented chips.
As a first approach, thermomechanical tests were carried out to acquire a more in-depth knowledge of the material behaviour and to set-up a methodology to measure (i) temperature with infrared filming, and (ii) plastic strain with Digital Image Correlation. In addition, a methodology to measure the adiabatic self-heating based on a thermodynamic analysis of a 3D volume was developed.
Thermal measurements were carried out under orthogonal cutting conditions. These measurements provided validation of the ductile failure law. Chip segmentation frequency was then used as an input in an inverse approach to optimize this law. The prediction error was reduced from more than 100\% employing the initial material law to less than 10% in chip segmentation frequency predictions, while maintaining the accuracy in the other outcomes.
A novel grid distortion based methodology able to measure plastic strain in the shear zone under plane strain conditions was also developed and validated. This grid method can have significant application in industry, as it proved to be an appropriate technique to measure subsurface damage, providing accurate results and reducing measurement subjectivity. The method was applied to analyze the effect of inputs such as workpiece initial microstructure, tool radius, and tool coating.