Thesis "Machine Learning-Based Kinetic Modeling of Methanol Synthesis"

As one of the world’s largest solar research institutes, the Fraunhofer Institute for Solar Energy Systems ISE makes a significant contribution to a sustainable, economical, secure and socially equitable energy supply worldwide. Our goal is to drive the energy transition forward with practically applicable technological solutions – through excellent research results, successful industry collaborations and spin‑offs. With around 1,300 employees, we conduct research in four main areas: energy supply, energy distribution, energy storage and energy use. The state‑of‑the‑art R&D infrastructure of Fraunhofer ISE with 22,300 m² of laboratory space enables cutting‑edge research at an international level.

Be part of change

You want to actively help shape the energy transition and gain practical experience during your studies? With us, you will work on making this goal a reality. As part of the transformation towards a sustainable and CO₂‑neutral economic and energy system, Power‑to‑X processes, in which sustainably produced hydrogen (H₂) and carbon dioxide (CO₂) from biomass or the atmosphere are converted into base chemicals and synthetic fuels, will play an essential role. Methanol will be of particular importance due to its broad range of applications. However, because of the complexity of the reaction network, the exact kinetic modelling of methanol synthesis using conventional approaches is a challenging task. The aim of the thesis is to fit a machine‑learning (ML)‑based kinetic model of methanol synthesis to an extensive experimental data set. The necessary balance equations and a detailed physical model are already available and validated. Accordingly, the focus of the thesis will be less on the chemical processes in the reactor and more on the use and adaptation of suitable ML and optimisation algorithms in order to achieve a further improved description of the chemical reaction.

• Familiarise yourself with the fundamentals of methanol synthesis and reactor modelling using relevant scientific literature.
• Familiarise yourself with our MATLAB‑based simulation platform with our guidance.
• Develop concepts for embedding ML models into our simulation platform.
• Using simplified fitting campaigns, select the most promising approach and then use it to fit all available measurement data.
• Through a detailed comparison between the results of the existing conventional models and the ML‑based models you have developed, analyse the strengths and weaknesses of both approaches.
• Present your results to our team.
• Summarise your findings in the form of a thesis.

What You Contribute

• You study Data Science, Environmental Sciences, Chemical Engineering or a comparable field.
• In the course of your studies, you have already gained initial experience in the field of machine learning, for example in implementing and using neural networks.
• In the course of your studies or through your own projects, you have built up expertise in programming and are familiar with using MATLAB or a comparable programming language.
• Experience with different optimisation algorithms is an advantage, but not a requirement.
• You plan upcoming work steps independently and proactively, set priorities and ensure appropriate time management.
• It is important to you to contribute to your team and to achieve goals together, including in an interdisciplinary environment.
• You prepare and deliver presentations with confidence, have a confident manner and are able to convince others.
• You have very good English skills.

What We Offer

• Exclusive insight: By working together with the scientists in our work unit, you will gain an insight into everyday research and development at a research institute.
• Research mix: We give you the opportunity to combine experimental work with theory and thus apply and expand your knowledge from your studies.
• Supervision: You will be supervised in your work by scientists and receive feedback on your progress.
• Teamwork: You will gain experience of working in a team by interacting with academic and student staff and will be able to contribute the experience you have already gained.
• Working hours and location: We offer you the opportunity to flexibly adapt your working hours to your needs by arrangement and to work from home from time to time.
• Equal opportunities: We value equal opportunities and make room for diversity.
• After work: Celebrate yourself and your colleagues at after‑work events or our annual employee parties.

In addition to the thesis, a contract as a student assistant is agreed. Remuneration is based on the level of your university degree.

We value and promote the diversity of our employees' skills and therefore welcome all applications – regardless of age, gender, nationality, ethnic and social origin, religion, ideology, disability, sexual orientation and identity. Severely disabled persons are given preference in the event of equal suitability.

Ready for change? Apply now with your detailed application documents (including CV, cover letter, employment references/certificates and/or academic transcript) and make a difference! After we receive your online application, you will automatically receive a confirmation of receipt. We will then contact you as soon as possible to let you know the next steps.

Questions about this position will be answered by:
Florian Nestler

Fraunhofer Institute for Solar Energy Systems ISE
Number: 83331