We are looking for a research assistant (m/f/d) with a Dipl.-Ing. (Univ.) or M.Sc. degree in Computer science, Mechanical engineering, Process Engineering, Mechatronics, Informatics or related field.
To implement an efficient control system within a chemical or biotechnological process, an adequate process model, which describes the relationship of the relevant in and outputs, represents a crucial presupposition. Often analytical approaches are not applicable in bioprocess engineering due to the lack of existing knowledge. Due to the capability of the artificial neural network (ANN) models in representing complex nonlinear processes, it be-came a popular tool for modeling, optimization, and control of different processes. However, the main challenges remain to interpret the network model and usage of the prior-knowledge to develop a reliable and generalized network model.
The aim of this research topic is to introduce a general design approach for the direct integration of prior knowledge into an ANN. On the other side, the black box of the ANN should be open to help to dig into the prob-lems, which are not yet understood well. In the context of this research study, soft (ware) sensors for biotechno-logical processes are to be developed and tested in both industrial and academic environments.