Student/Thesis Project: Bearing and Screw Diagnosis with Smart Washer and Edge AI

Fraunhofer-Gesellschaft · Stuttgart

ThesisGerman needed: not said

Still open on 11 Nov

You apply on the company's own website.

Key facts

Pay: Pay based on qualifications
Language needed: German and English
Subject you study: Automation Technology, Computer Science, Cybernetics, Aerospace Engineering, Mechanical Engineering, Mechatronics

Simulation: Modeling and simulation of typical rolling bearing damage signals in MATLAB including a physical sensor model of the force-measuring washer taking into account screw calculation according to VDI 2230
Test bench and construction: Mechanical and measurement integration of the sensor washer into an existing e-motor/bearing test bench (incl. minor design adjustments) and execution of measurement campaigns with pre-damaged bearings and varying screw preload forces
Model development: Design of Physics-Informed Machine Learning models that utilize physical prior knowledge and synthetic simulation data, and their transfer to real sensor data of the test bench
Edge integration: Porting and optimization of trained models for an energy-efficient STM32 microcontroller and systematic evaluation regarding memory requirements, latency and energy consumption

Valid enrollment in a master's program in Mechanical Engineering, Mechatronics, Electrical Engineering, Computer Science or a comparable course of study within Germany/EU
Solid fundamentals in digital signal processing and in the field of Machine Learning with Python and/or MATLAB
Enjoyment of practical experimental work on the test bench and basic understanding of structural relationships and machine elements (e.g. screw connections according to VDI 2230, rolling bearings)
Structured and independent working style as well as good German or English skills

Versatile tasks in applied research and active participation in an ongoing collaborative project with industrial partners
Excellent technical equipment with direct access to an e-motor/bearing test bench, modern measurement technology, a powerful GPU workstation and microcontroller boards
Close professional supervision by scientific staff with the clear goal of a joint scientific publication
Friendly and committed team

Simulation: Modeling and simulation of typical rolling bearing damage signals in MATLAB including a physical sensor model of the force-measuring washer taking into account screw calculation according to VDI 2230
Test bench and construction: Mechanical and measurement integration of the sensor washer into an existing e-motor/bearing test bench (incl. minor design adjustments) and execution of measurement campaigns with pre-damaged bearings and varying screw preload forces
Model development: Design of Physics-Informed Machine Learning models that utilize physical prior knowledge and synthetic simulation data, and their transfer to real sensor data of the test bench
Edge integration: Porting and optimization of trained models for an energy-efficient STM32 microcontroller and systematic evaluation regarding memory requirements, latency and energy consumption

Valid enrollment in a master's program in Mechanical Engineering, Mechatronics, Electrical Engineering, Computer Science or a comparable course of study within Germany/EU
Solid fundamentals in digital signal processing and in the field of Machine Learning with Python and/or MATLAB
Enjoyment of practical experimental work on the test bench and basic understanding of structural relationships and machine elements (e.g. screw connections according to VDI 2230, rolling bearings)
Structured and independent working style as well as good German or English skills

Versatile tasks in applied research and active participation in an ongoing collaborative project with industrial partners
Excellent technical equipment with direct access to an e-motor/bearing test bench, modern measurement technology, a powerful GPU workstation and microcontroller boards
Close professional supervision by scientific staff with the clear goal of a joint scientific publication
Friendly and committed team