Christopher Clarc Reinartz
Modern industrial plants are highly interconnected systems. As a result, disturbances in specific system parts can propagate through the plant. The currently implemented control-architecture is not designed to cope with plant-wide disturbances and often fails in such scenarios. Instead, human operators are relied upon to recover the plant form abnormal states. The efficiency of these recovery operations is dependent on the skill and expertise of the operators as well as the quality of information provided to them.
Developing suitable action-plans in emergency scenarios becomes more difficult as the degrees of automation and plant complexity increase. Decision support systems can aid the operators in reasoning about causes of disturbances and help them to evaluate potential counteractions regarding their consequences and likelihood of success. The decision support system developed is based on Multilevel Flow Modelling, which is a functional modelling language capable of capturing the causal relations within industrial plants.
The PhD project aims at the development of methods for automated generation of counteraction-plans based on a functional representation of the plant. The goal is to increase the overall level of efficiency of operations meant to recover plants from abnormal states.