With the increasing complexity and requirements on availability, plant asset management is becoming an important topic in the industry. Process monitoring and fault diagnosis play a central role in the plant asset management by delivering information on the “health” status of sensors, actuators and plant components. Such information allows an early detection of abnormalities in the system and provides the basis for predictive maintenance. In our research group, we develop advanced model and data based fault diagnosis methods to improve the performance of diagnosis systems and reduce engineering efforts.
Contact: M.Sc. Zhihua Zhang
In modern production plants different system components are becoming more and more connected through communication networks. While the flexibility and the reconfigurability improve significantly, the vulnerability to cyber attacks increases as well. In order to better protect production plants from cyber attacks, new algorithms are being developed for early detection of cyber attacks based on the system theory. Moreover, it is investigated how to mitigate the influence of cyber attacks on the system behaviour.
Mathematical modeling of biological systems has received much attention in the recent years. The interpretation of large amount data and the analysis of complex feedback mechanisms are the major challenges. Methods in the control theory can be transferred and further developed in the context of biological systems. Combined with prior biological knowledge, mathematical models of biological systems can be identified and validated from high-throughput gene expression data efficiently. The methods of optimal control and diagnosis can contribute to the optimization of medical treatment and early diagnosis of diseases.
Building automation encompasses many areas such as smart homes, assisted living as well as home automation. We tackle the topic of energy efficiency, which grows in importance in the global energy transition. With the continuous increase in energy consumption, there is great potential of energy saving in buildings. The challenge is to utilize these saving capacities and at the same time guarantee comfort of inhabitants as well as economic efficiency. We develop modern control methods which incorporate knowledge of disturbances and are applicable to large-scale distributed systems.
Contact: Dipl.-Ing. Anna Nehring
In the framework of Indutry 4.0 the plants in the manufacturing industry are becoming significantly more flexible and characterized by variable topology. Such systems can be described by discrete-event systems (for instance, automata, petri-nets) conveniently. Faults in actuators, sensors or plant components lead to a reduction of the overall system performance. With the help of fault tolerant control it is possible to ensure that the system continues its operation and the impact of faults is automatically compensated. In order to improve the plant availability, new methods are being developed for the fault tolerant control of discrete-event systems.
Contact: M.sc. Raphael Fritz
Besides economic aspects, the safety of technical operating units is of supreme interest for plant operators. To protect people and the environment from accidents, automated safety systems are an essential part of many technical plants. They are subject to international standards and need to satisfy stringent requirements on availability. New methods for the description of safety systems can not only significantly reduce the efforts of quantitative performance evaluation, but also open up new possibilities for the optimization of maintenance activities of safety systems.