Selected projects: AI and Decision Support in Energy and Automation Systems
MORE-KIBA
MORE-KIBA: Human-understandable, optimal resource and energy management for complex, grid-integrated, biogenic production plants (2025-2028)
The objective of this study is to utilise artificial intelligence (AI) to elucidate the algorithmic decisions that result in sustainable operation. This will be achieved by employing a biorefinery and an energy grid simulator as exemplars. In periods of limited resources, the optimal functioning of production facilities is imperative. It is evident that proposals for the enhancement of technical processes frequently entail the utilisation of mathematical algorithms. However, these algorithms can often be challenging for operators to comprehension, primarily due to the intricacy of the industrial facilities and the complexity of the algorithmic representations employed. Consequently, the potential for sustainable operation is often overlooked, or technologies are even discarded despite the fact that their optimised operation could lead to considerable short-term savings. The objective of this junior research group is to facilitate the transparency and comprehensibility of algorithmic decisions for the operators of technically complex production systems utilising AI. The approach will be demonstrated using the example of a biochemical production facility integrated with an energy grid simulator. The project is co-financed by tax revenues based on the budget approved by the members of the Saxon State Parliament and by the European Union.
CUBES Circle
CUBES Circle: Agricultural systems of the future: A closed symbiotic cycle system of modular units with the aim of resource-efficient food production (2019-2028)
In the project CUBES Circle (closed urban modular energy- and resource-efficient agricultural systems), three agricultural production systems - aquaculture, insect production and horticultural plant production - are linked together as a closed-loop system. The organisms utilize the residual materials from the respective other production processes. In this way, the residual materials from one production step become valuable materials again in the next. The CUBES production systems are also digitally networked in order to control and optimize the circulation system.
Apfel4NULL
Apfel4NULL: Use of sensors for sustainable apple production and storage (2020-2023)
The storage of fruit, especially long-term storage, is of fundamental importance today for price-stable marketing, but also for supplying consumers in Europe across seasons. Robust prediction methods for apple diseases do not generally exist. Methods of data analysis and machine learning show great potential to classify such complex processes. Based on spectral measurements, weather data and storage measurements, apples are classified according to fruit quality. Further information can be found on this website.
SensCEA
SensCEA: Sensor-based efficiency enhancement in controlled environment agriculture (2023-2026)
The aim of the project is to develop a digital twin for the optimization and automation of cultivation processes for agriculture in controlled environments (greenhouses, vertical farms). The core of the digital twin is a growth model that describes plant growth as a function of the material and energy flows, which are influenced by the energy-intensive climate and light control, among other things. The modeling is intended to deepen the process understanding of the biotechnical greenhouse system so that energy requirements can be reduced. This is to be achieved, for example, by shortening the lighting time with artificial light (LED lighting) through increased CO2 fertilization. In addition, the potential of optimal control algorithms based on model predictions is to be exploited (predictive control). For modeling, data is collected in experiments on GreenResearcher systems from greenhub, the data transfer and the development of a graphical user interface is carried out by mewedo.