Covariance‐based hardware selection‐part III: Distributed parameter systems

Abstract

In this article, we propose a new method for the selection of control system hardware (sensors and actuators) for distributed parameter systems. The proposed design scheme seeks to minimize the capital cost of hardware while satisfying predefined performance constraints. Within this minimum capital cost framework, three design scenarios will be discussed; the closedloop full state information actuator selection problem, the open-loop partial state information sensor selection problem, and the closed-loop partial state information simultaneous sensor and actuator selection problem. The proposed method will be illustrated through application to a nonisothermal tubular reactor example.

Covariance‐based hardware selection part IV: Solution using the generalized benders decomposition

“…In the SAS case, Problem Q is found to be: Example Consider the example of simultaneous SAS for the tubular reactor . A finite‐dimensional model, linearized around the steady‐state profiles of Figure is then obtained.…”

“…Both the branch and bound and GBD methods are used to solve the problem, and each obtains optimal solutions (see Table ) that are the same as in Ref. . The comparison of computational efforts for these two methods is presented in Table .…”

“…While these methods guarantee an implementable feedback element, the resulting nonlinear optimization was solved using local search methods. Peng and Chmielewski and Ahmed et al proposed a formulation similar to Refs. and , but subsequently transformed the problem to a MICP.…”

“…The current effort focuses on the MICP formulations of Refs. 8, 15, and , but replaces the branch and bound algorithm with the generalized Benders decomposition (GBD) algorithm. In the remainder of this section the MICP formulation for SND will be reviewed along with the generic GBD approach.…”