Simultaneous solution strategies for inclusion of input saturation in the optimal design of dynamically operable plants (Q1826350)

Recent work has considered the inclusion of input saturation in optimization-based integer plant and control system design. Young and Swartz gave two models using slack variables and involving respectively mixed-integer and bilinear constraints. Both of these formulations are relatively simple and conform to standard mathematical programming structures, but both have disadvantages in the numerical sense in solving the problems. The general integrated design and control problem is discussed here. It is represented as a minimization problem with complementarity constraints and solved by an initial point path-following algorithm. Convergence cannot be proven. The application study illustrates the viability of the approach, and it works in a reasonable amount of time.