Titulo Estágio
Predictive Model for Detection of Defects in Sheet Metal Components for Automotive Industry
Áreas de especialidade
Engenharia de Software
Sistemas Inteligentes
Local do Estágio
Laboratório de Computação Adaptativa (LARN)
Enquadramento
Sheet metal forming is a high production rate process widely used in the automotive sector, for the production of chassis and body components. The design of the tools used in the process involves the virtual (finite element based simulation) and the experimental try-out. However, there is no absolute control over the actual conditions of a forming process, i.e. there are always unavoidable sources of variability (e.g. geometric, material and process), which may lead to defects in the components. In this context, it is important to construct a database with relevant information regarding the main geometrical, material and process features, in order to enable its analysis and the development of predictive models.
Objetivo
The main objective of this proposal is to develop a predictive model for detection of defects in metal forming sheets used for business intelligence in the scope of automotive industry. The first goal is to develop a data warehouse of metal forming processes. The database should contain information on:
(i) Metal materials (e.g. steel and aluminum alloys), in particular supplier data and mechanical and metallurgical properties;
(ii) Process parameters, namely those of sheet and bulk metal forming processes;
(iii) Geometric features of components;
(iv) Type and frequency of defects occurring in a particular forming process.
The second main goal is to develop a model able to predict the occurrence of defects during metal forming processes, using machine learning and pattern recognition techniques.
(v) Construct the data set for the predictive model;
(vi) Perform Data Pre-Processing, Normalization and Scaling;
(vii) Select appropriate ML algorithms for building the Predictive Model;
(viii) Perform Sampling and Model Evaluation;
(ix) Validate the overall Model with real data.
Plano de Trabalhos - Semestre 1
•Overview metal forming processes, including process parameters, geometric features of components and type of defects that may occur;
•Define database framework;
•Select relevant metal materials and mechanical and metallurgical properties;
•Writing of intermediate report.
Plano de Trabalhos - Semestre 2
•Select, prepare, and preprocess a collection of large datasets for experiments;
•Study, and select, machine learning (ML) algorithms and feature selection (FS) algorithms for building the the predictive model for detecting the occurrence of defects in metal components;
•Analyze experimental results: e.g., study parameter values; compare performance of the reduced datasets vs. previous results, etc.;
•Writing of scientific article;
•Writing of the thesis.
Condições
This work will be carried out in the Laboratory of Neural Networks (LARN) of CISUC, where there will be a regular supervision and feedback on the behalf of the supervisor.
Familiarity with machine learning and data mining algorithms and software tools are essential. Participating students will acquire valuable knowledge and experience with model building and data science by mining massive datasets, which skills are currently in high demand for various technology employers due to the relevance to various applications.
Observações
Opportunities may become available during internship to short-term paid visit to the ToolPress to collaborate on the project
Logistics @Laboratory of Neural Networks (LARN)
DEI-FCTUC
Orientador
Bernardete Ribeiro, Pedro Prates (bribeiro@dei.uc.pt, pedro.prates@dem.uc.pt)
bribeiro@dei.uc.pt 📩