Course paper / final thesis

Design and analysis of machine learning-based property models for hybrid jet fuel sensors at airports

Starting date


Duration of contract

6 months

Type of employment

Full-time (part-time possible)

The ramp up of sustainable aviation fuel (SAF) usage enables new opportunities for reducing aviation’s climate impact during a flight mission. These novel opportunities include adjusting the flight routing with respect to the available jet fuel or optimizing the refueling strategy. However, currently these potentials cannot be fully exploited since the required fuel property data (e.g. aromatics content, heat of combustion) is not collected systematically at the airport or is only checked using costly and time-consuming laboratory analysis. Therefore, a hybrid sensor concept for fast online monitoring of jet fuel properties is developed within the department Multiphase flows and alternative fuels (MAT) at the DLR Institute of Combustion Technology. This concept combines sensor data and machine learning models and offers the potential of replacing time-consuming laboratory analysis.

In the course of the thesis, ML-based models for the prediction of complex physical properties of jet fuels shall be investigated systematically. For this purpose, a worldwide unique database on conventional and alternative jet fuels is available with the MAT department. As safety is of major concern for aviation fuels, uncertainties in the developed models must be identified and the resulting predictive capability must be assessed critically. Finally, the potential of the models shall be evaluated within the hybrid sensor context.

Your tasks:

  • familiarize with the fields of conventional and sustainable aviation fuels, modeling of jet fuel properties and fuel supply infrastructure at airports
  • data science analysis of the existing fuel property data
  • training and analysis of ML models for selected fuel properties (e.g. aromatics content, heat of combustion)
  • quantification of uncertainties and assessment of the predictive capability of the trained models
  • documentation and presentation of the results (English or German)

Your qualifications:

  • degree in aerospace engineering or comparable field (engineering or natural science)
  • programming experience in Python, ideally with frameworks like scikit-learn, pyTorch or comparable
  • basic theoretical and applied knowledge in data-driven modeling and machine learning
  • interest in sustainable aviation
  • interest in working in a motivated team (hybrid/in person)

Your benefits:

Look forward to a fulfilling job with an employer who appreciates your commitment and supports your personal and professional development. Our unique infrastructure offers you a working environment in which you have unparalleled scope to develop your creative ideas and accomplish your professional objectives. Our human resources policy places great value on a healthy family and work-life-balance as well as equal opportunities for persons of all genders (f/m/x). Individuals with disabilities will be given preferential consideration in the event their qualifications are equivalent to those of other candidates.

  • Apply online now
  • You can send this job advertisement via e-mail and complete your application on a personal computer or laptop.

    We need your digital application documents (PDF). The document upload function is not supported by all mobile devices. Please complete your application on a PC/laptop.

    Complete application on PC

Technical contact

Dr. Benedict Enderle
Institute of Combustion Technology

Phone: +49 711 6862-518

Send message

Vacancy 68898

HR department Stuttgart

Send message

DLR site Stuttgart

To location

DLR Institute of Combustion Technology

To institute