| KEYWORD |
Modelling and optimization of industrial processes parameters through machine learning techniques
keywords ARTIFICIAL NEURAL NETWORKS, INDUSTRY 4.0, MACHINE LEARNING
Reference persons ANDREA ACQUAVIVA, ENRICO MACII
Research Groups ELECTRONIC DESIGN AUTOMATION - EDA
Thesis type EXPERIMENTAL, IN COMPANY
Description The thesis focuses on the analysis of the parameters for controlling the process of aluminum casting machines in order to determine the correlation with the quality and quantity of pieces produced. It will have to perform a correlation analysis between already available monitored parameters and environmental parameters. The experiment will be to optimize the parameters using sensitivity tests using sample products as benchmarks. Modeling and learning techniques based on process data and correlation between process parameters and product quality will need to be developed. The outcome will be a "black box" model that can be used to predict quality as a function of process parameters.
Required skills machine learning (basics), programming
Notes tesi in collaborazione con azienda
Deadline 22/06/2019
PROPONI LA TUA CANDIDATURA