A Companion Model Approach to Modelling and Simulation of Industrial Processes: Dissertation

Modelling and simulation represent tremendous possibilities if extensively taken up by engineers as a working method (IMTI 2003). However, if detailed writing of equations and programming is required in addition to inverse engineering efforts for finding the necessary design data, it will only remain attractive to mathematicians, physicists and computer programmers. When considering the use of modelling and simulation tools in an engineering design project, there is no time to write equations, to consult suppliers' experts, or to manually transfer data from one repository to another. Available commercial modelling and simulation software for process and automation interoperability studies, the advances in process and automation design methodologies, evolving specification and communication standards, as well as applicable computer software architectures, have recently been dealt with in VTT Publications (Pasanen 2001; Karhela 2002). Internet based service repositories have developed rapidly, making it possible for equipment manufacturers to supply "extended products", including design data needed by engineers engaged in process and automation integration. A request has been recognized to make the modelling and simulation software suited for relevant structured design information as it is provided for in product and design databases. Initiatives involving automated model generation have been encouraged (IMTI 2002). The companion model approach for specification and solution of process simulation models, as presented herein, is developed from the above premises. The focus is on how to tackle real world processes, which from modelling point of view are dynamic, very stiff, very nonlinear and only piecewise continuous, without extensive manual interventions of human experts. An additional challenge, to solve the arising equations fast and reliably, is dealt with, as well.