Hybride Prozess- und Systemmodellierung

Project team: 

 Kevin von Bargen (Ph.D. Student), Prof. Dr. Tim Römer (UOS), Dr. Sharvari Raut (ATB), Prof. Dr. Barbara Sturm (ATB)

The optimization of food production processes is crucial for a sustainable, resilient and human centric transformation. The main goal of this research project is to optimise the process control based on both initial matrix and desired product properties with special focus on drying food products. To reach that objective, innovative hybrid methods based on data- and physical-modelling have high potential. The key idea is to combine process understanding and its representation in the form of corresponding mechanistic models so that machine learning models based on high-dimensional sensor and process data can optimize the corresponding complex processes as well as the process control.

Project team: 

 Jonas Schmidinger (Ph.D. Student), Prof. Dr. Martin Atzmüller(UOS), Dr. Sebastian Vogel (ATB)

Soils are heterogenous in space. This means, that the quality of a soil may change from barren to fertile in just a few meters. Ideally, farmers should spatially adapt their farming practices given the local soil conditions. For example, when the in-field variability of soil nutrients is known, it allows us to spatially optimize our fertilization rate to maximize crop production, while minimizing fertilizer input. This of course requires that we have soil maps. We can use soil sensors to create cost-efficient and high-resolution soil maps. Unfortunately, these soil maps are never perfectly accurate, meaning we have to deal with a certain amount of uncertainty. In this PhD project, I am tackling the following research questions:

• How to increase the accuracy of soil predictions, by means of different soil sensors and statistical tools.
• How to obtain accurate soil predictions with limited resources i.e., low training sample sizes or low-cost sensors.
• How to quantify and assess the uncertainty of the soil maps with different probabilistic machine learning techniques.
• How to take into account the uncertainty in the agronomical decision-making process.

Project team: 

 Tim Wollschläger (Ph.D. Student), Prof. Dr. Andreas Focks (UOS), Prof. Dr. Tim Römer (UOS), Dr. Susanne Theuerl (ATB), Dr. Christiane Herrmann (ATB)

The conversion of biogenic residues to biogas as sustainable energy source is one major challenge in the context of the energy transition from fossil fuels to renewable resources. In that context, this research project aims in general at the understanding, prediction and control of biomass conversion dynamics to biogas by a combination of process engineering, microbiological analyses and mathematical modelling. To reach this goal, an in-depth knowledge of the influence of feedstock
composition and quality on product quantity and quality based on physical, chemical and microbiological process parameters, and the formulation of respective process models is required.
Resulting models should allow to control demand-driven product formation and to avoid instabilities/disturbances of the transformation processes. Process modeling is in this project thought to be based on combining model equations for material transformations (based on chemical process parameters) and microbial diversity dynamics using differential equations. The microbial diversity is intended to be considered for dynamic simulations/analyses by using concepts from (microbial) ecology such as e.g. traits-based approaches. Examples for relevant modelling approaches are the Anaerobic Digestion Model No. 1 (Batstone et al. 2002) or the Anaerobic Microbial Degradation Model (Dalby et al. (2021). The combination of this project with other KIDS methods is possible, e.g. for process optimization/control or for data-driven model approaches. This project is related to overarching themes digital twins/process control.