Student Mechanical Engineering, Computer Science, Safety Engineering or similar (f/m/x)

Master Thesis

Enter the fascinating world of the German Aerospace Center (Deutsches Zentrum für Luft- und Raumfahrt; DLR) and help shape the future through research and innovation! We offer an exciting and inspiring working environment driven by the expertise and curiosity of our 11,000 employees from 100 nations and our unique infrastructure. Together, we develop sustainable technologies and thus contribute to finding solutions to global challenges. Would you like to join us in addressing this major future challenge? Then this is your place!

For our Institut of Flight Systems in Braunschweig we are looking for a

Student Mechanical Engineering, Computer Science, Safety Engineering or similar (f/m/x)

Explainable AI for Deep Learning-based health indicator construction for RUL prediction on ball bearings

What to expect:

Condition monitoring systems for safety-critical flight control components are being developed at DLR's Institute of Flight Systems in Braunschweig. To predict the Remaining Useful Life (RUL) of electromechanical flight control actuators (EMA) it is necessary to monitor mechanical components such as the ball bearings, e.g. by using acceleration measurements on the EMA housing. Due to the continuous flight control surfaces adjustments combined with excessive loads during flight operation, the degradation behavior of the ball bearings becomes apparent in the monitored data. This degradation can then be modelled using deep learning-based health indicators as a basis for the RUL prediction. However, due to the black box characteristics of deep learning models, the trustworthiness of the results is limited. Improving this trustworthiness is particularly relevant for safety-critical systems.

As part of a master's thesis, explainability approaches are therefore to be investigated in order to improve the trustworthiness of the results for deep learning-based health indicator construction methods. A literature review of existing deep learning-based health indicator construction methods for ball bearings and applicable explainable AI methods shall be carried out first. Subsequently, the explainable AI methods shall be implemented in Python on run-to-failure data sets for rotating ball bearings and the results shall be visualized. The aim is to better understand the degradation behavior of the ball bearings and the inherent uncertainties and to increase the accuracy of the modelled health indicator.

In our department, you will be part of a dynamic and scientifically highly innovative team. You will benefit from the existing expertise and infrastructure and contribute to its continuous development in the course of your work. In addition to your thesis, employment on a part-time basis is possible. Do you have the necessary degree of personal responsibility and do you share our high standards for the scientific quality of your work? We offer you the ideal environment for personal and professional development at an internationally high level.

What we expect from you:

• current enrollment in a Master’s program in Mechanical Engineering, Computer Science, Safety Engineering or a related field
• good programming skills in Python
• good knowledge in the field of artificial intelligence
• very good English language skills

What we offer:

DLR stands for diversity, appreciation and equality for all people. We promote independent work and the individual development of our employees both personally and professionally. To this end, we offer numerous training and development opportunities. Equal opportunities are of particular importance to us, which is why we want to increase the proportion of women in science and management in particular. Applicants with severe disabilities will be given preference if they are qualified.

Further information:

Starting date: 01.11.2024
Duration of contract: 6 months
Type of employment: Full-time (part-time possible)

Vacancy-ID: 97495

Contact:
Lauri Bodenröder

Institut für Flugsystemtechnik
E-Mail auf connecticum.de ansehen