Master Thesis Data-Efficient Hybrid Machine Learning for Robust Vibration System Prediction
Core
Develop robust predictive models for vibration-loaded technical systems by integrating limited real-world measurement data with simulation data using advanced machine learning techniques.
Role type
Master Thesis Researcher (Data-Efficient Hybrid Machine Learning)
Builds
A benchmark integrating simulated and real-world test bench data to predict dynamic behavior of nonlinear coupled vibration systems.
Domain
Mechanical Engineering / Machine Learning / Vibration Analysis
Deliverable
production ML models
Required skills
Python (PyTorch, Pandas, Numpy), fundamental machine learning concepts and algorithms (regression), dynamics (mechanical vibrations) / mechanics
Preferred skills
None stated
Technologies
PyTorch, Pandas, Numpy
Responsibilities
Research and apply advanced machine learning techniques to integrate limited measurement data into model training; develop a benchmark integrating simulated and real-world data; evaluate and compare model performance (accuracy and robustness) against simulation-only trained models; communicate ideas and contributions to colleagues and experts.
Seniority
Master Thesis Researcher