Signal Path Oriented Approach for Generation of Dynamic Process Models

The article at hand discusses a tool for automatic generation of structured models for complex dynamic processes by means of genetic programming. In contrast to other techniques which use genetic programming to find an appropriate arithmetic expression in order to describe the input-output behaviour of a process, this tool is based on a block oriented approach with a transparent description of signal paths. A short survey on other techniques for computer based system identification is given and the basic concept of SMOG (Structured MOdel Generator) is described. Furthermore latest extensions of the system are presented in detail, including automatically defined sub-models and qualitative fitness criteria.