An architecture of a knowledge-based process modeling simulation tool

Mathematical process modeling and simulation are the two key requirements for a routine application of model based techniques in the design and operation of complex chemical plants. Efficient support for both issues can only be guaranteed by an integrated environment for process modeling and simulation. This contribution introduces the architecture of a knowledge-based modelling and simulation tool. The tool consists of a temporary and persistent storage, a physical properties server connected to physical property databases, a computer algebra system, a code generator, the process simulator DIVA and an user interface. The design of the architecture is motivated by the blackboard concept of artificial intelligence. The primary advantage of the proposed architecture is based on the fact that self-contained state of the art tools are combined via clear cut software interfaces in a modular fashion to yield an integrated modeling and simulation tool that promises great potential and flexibility. (23 References