Qualitative approximation methodology for modeling and simulation of large dynamic systems: Applications to a marine steam power plant

Human operators of large complex dynamic systems function as system managers with primary responsibility for monitoring system states and taking actions when operating conditions change or problems arise. Use of simulators is an efficient cost-effective means of providing the training necessary to acquire the skills needed for such a supervisory control operation. A qualitative approximation methodology used to design a training simulator is described. Qualitative approximation is based on a hierarchical description of dynamic systems, where the primitives that approximate the functions of the components are at the lowest level. The primitives provide qualitative states that describe the evolution of the system. The details of the modeling methodology are discussed. The model definition is preceded by detailed discussions of qualitative modeling of dynamic systems and qualitative reasoning. A marine power plant simulator that has been developed using qualitative approximation is described. The simulator provides access to system states via an interactive graphical interface and is being used in training programs for marine engineers with varying amounts of expertise.