Machine Learning

Why it matters

Predicting fatigue damage is essential to prevent the failure of structures and mechanical components, especially in the aerospace and offshore sectors, where structures are subjected to constant vibrations caused by dynamic loads. Modern sensors allow real-time data acquisition, but considerable effort is still needed to compare these data with the results of numerical simulations.

What you'll do

• Develop automatic pre/post-processing tools to run multiple numerical simulations.
• Develop Machine Learning algorithms for time series analysis.
• Prepare project reports and presentations for engineering companies.

Prerequisites

• Knowledge of FEM simulations from the Computational Mechanics course (or equivalent), to understand the data you will analyse.
• Basics of Data Science and Python.

Tools

• Python, for the Machine Learning and Deep Learning algorithms and for automating the simulations.
• A FEM code, to generate simulation data to compare with real data.

Opportunities

Project in collaboration with M3E (m3eweb.it), a consulting company specialised in numerical simulations and scientific software development.

Supervisor

Prof. Gianluca Mazzucco · gianluca.mazzucco@unipd.it