Thesis (D.Eng.)--Chulalongkorn University, 2008

The combined mathematic and heuristic based approach is proposed for establishing the plantwide control structure in this dissertation. The proposed approach takes advantages of both heuristic based and mathematic based approaches to develop the appropriate plantwide control structures. In heuristic part, sets of controlled variables and manipulated variables are selected followed by establishing of the obvious control loops. In mathematic part, the dynamic performance-based optimization is proposed for establishing the plantwide control structures. The optimization problem is formulated as a mixed integer nonlinear programming (MINLP). The objective function is presented as an integral of the time-weighted absolute error (ITAE) of all measurements and manipulated variables in the face of disturbances. A discrete state-space model is used as the process model in the optimization problem. The proposed approach is investigated on the Tennessee Eastman (TE) process. The obtained plantwide control structures are compared with the earlier work given by Luyben et al. (1999). In dynamic simulation, the performance of the plantwide control structures is evaluated in the face of disturbances and setpoint changing. The simulation results show that the obtained plantwide control structures give the appropriate responses compares with those of Luyben et al. (1999).