« Förderinformationen
Sie verwenden einen sehr veralteten Browser und können Funktionen dieser Seite nur sehr eingeschränkt nutzen. Bitte aktualisieren Sie Ihren Browser. http://www.browser-update.org/de/update.html
Priority Programme "Machine Learning in Chemical Engineering. Knowledge Meets Data: Interpretability, Extrapolation, Reliability, Trust" (SPP2331)
Deutsche Forschungsgemeinschaft (DFG)
In 2020 the Senate of the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) established the Priority Programme "Machine Learning in Chemical Engineering. Knowledge Meets Data: Interpretability, Extrapolation, Reliability, Trust" (SPP 2331). The programme brings together the chemical engineering (CE) and machine learning (ML) communities. By teaming chemical engineers with mathematicians and/or computer scientists, progress in all disciplines is expected. The programme is designed to run for six years.

The present call invites tandem proposals for the first three-year funding period. Each proposal must operate at the interface of CE and ML and have at least two applicants with corresponding expertise. The projects shall consider at least one of six areas: 1. optimal decision making, 2. introducing/enforcing physical laws in ML models, 3. heterogeneity of data, 4. information and knowledge representation, 5. safety and trust in ML applications, and 6. creativity. The projects will be organised in a matrix between the areas of CE and the ML tasks. Data, models, and methods will be shared among all participants of the programme in an internal platform. The organisation matrix and further information can be found on the homepage of the Priority Programme (see below).

The projects are expected to open up new methods for CE, formulate new types of problems for ML, and jointly generate advances for methods in both ML and CE. Since ML has been used within CE since several years (e.g., surrogate models for complex unit operations), projects shall go well beyond this state-of-the-art. Under the umbrella of the six areas, the collaborative projects shall have promise for progress in process synthesis (especially regarding feedstock transformation), process flexibility, material selection, generation of alternatives, and uncovering hidden information. Projects should address at least one CE area and one ML area, i.e., one of the nine collaboration fields in the matrix, and clearly state why it does, and how it will achieve progress in at least one of the areas 1. to 6. Projects investigating and comparing different methods from ML for the same field of the collaboration matrix are particularly encouraged. Similarly, projects are encouraged where outcomes are transferable from the matrix field considered to other fields in the same row.

The focus of the programme shall be on the field of fluid processes with or without chemical reactions. Examples or products from other fields could be included, in case the fluid process remains the focus. Reflecting the scientific challenges and needs of fluid processes, relatively broad CE methods are allowed, ranging from molecular modelling, thermodynamic calculations, reactor development, and the prediction of fluid properties up to methods dedicated to operation, synthesis, and design of whole processes (including control and optimisation, uncertainty quantification and optimal experimental design). Projects may be purely computational and/or have ML methods directly applied on experimental CE. Topics reaching beyond this scope may be included, in case they contain sufficient work on the methods above.

For scientific enquiries please contact the Priority Programme coordinator:
Prof. Alexander Mitsos, Ph.D.
Rheinisch-Westfälische Technische Hochschule Aachen
Fakultät für Maschinenwesen
Aachener Verfahrenstechnik - Systemverfahrenstechnik (SVT)
Forckenbeckstr. 51
52074 Aachen
phone +49 241 8094704

Questions on the DFG proposal process can be directed to:
Programme contact:
Dr. Simon Jörres
phone +49 228 885-2971

Administrative contact:
Silke Stieber
phone +49 228 885-2687

Further Information: