Reference : UMR5295-MEHCHE-003
Workplace : TALENCE
Date of publication : Monday, November 21, 2022
Scientific Responsible name : CHERIF
Type of Contract : PhD Student contract / Thesis offer
Contract Period : 36 months
Start date of the thesis : 16 January 2023
Proportion of work : Full time
Remuneration : 2 135,00 € gross monthly
Description of the thesis topic
This thesis is part of the ANR SHAIR project led by ICA Toulouse, which brings together the LS2N (Nantes), ICA (Toulouse), and LGCO (Toulouse) laboratories. This thesis will deal with the prediction and control of the quality of drilling in multi-material assemblies through the development of process and machine digital twins.
In the aeronautical field, the assembly of structural elements (wings, fuselage, rear rudder, etc.) requires the installation of fasteners (rivets, bolts), and therefore the realization of the mounting housings by drilling. This drilling operation is often carried out using automatic drilling machines (UPA), which clamp onto a drilling grid. These historically pneumatic drilling units are now electric. This major development allows a much more precise control of the operating conditions and makes it possible to develop smart drilling strategies.
During the final stages of assembly, drilling operations are carried out at the end of the production cycle, on parts with high added value. It is therefore essential to master these critical operations because the quality of the bores has a major impact on the service life of the assemblies. The complex geometry of the cutting tools, the wear phenomena and the coupled thermomechanical phenomena involved make the modelling of the process very complex. In addition, there are issues related to multi-material stacking, traditionally composite-aluminium-titanium. It is therefore necessary to develop process monitoring strategies. The main objective of the thesis is to develop a pair of digital twins allowing decision support through intelligent process and machine monitoring (e.g. recommendation for additional quality control, change of operating parameters, tool change, machine maintenance, start of a recovery operation for holes out of tolerance, etc.).
A major originality is to investigate and guide the technological choices and the decision level of Smart drilling in terms of its integration in the industrial operational context, in view of the acceptability and decision levels taken by the operator. To achieve this goal, a digital twin of the drilling process will be developed to predict the quality of the drilled hole. This twin will be based on defining and learning a behaviour model based on signals from process-integrated instrumentation and model simulations.
The processing of signals and their transformation into quality indicators (KPIs) is based on the integration of manufacturing knowledge and the exploitation of databases of signals collected in laboratory but also in industrial production. The data collected will also allow the monitoring of the state of the production equipment and will feed a digital twin of the drilling unit, apprehending its wear, and the effects on the quality of the parts produced.
This thesis will be carried out in the MPI (Material-Processes-Interaction) department at I2M Bordeaux. The Institute of Mechanics and Engineering covers the entire spectrum of solid, fluid and energy mechanics. Theoretical approaches and experimental methodologies are conducted at different observation scales. The development of numerical methods for intensive computing also proposed. The activities of the MPI department focus on the fields of manufacturing focusing on the study of interactions between the process and the material. They aim at understanding the physical and physico-chemical mechanisms involved in manufacturing processes through both experimental and numerical approaches complemented by physico-chemical and mechanical characterizations of the materials developed. The different topics are developped in a multidisciplinary approach in which all phenomena and their couplings are studied using multi-scale approaches.
Constraints and risks
We talk about it on Twitter!