CCP-Based Plant-Wide Optimization and Application to the Walking-Beam-Type Reheating Furnace

Yan ZHANG  Hongyan MAO  

IEICE TRANSACTIONS on Information and Systems   Vol.E99-D   No.9   pp.2239-2247
Publication Date: 2016/09/01
Online ISSN: 1745-1361
DOI: 10.1587/transinf.2016EDP7171
Type of Manuscript: PAPER
Category: Fundamentals of Information Systems
plant-wide optimization,  chance-constrained programming (CCP),  distributed generalized predictive control (DGPC),  serially connected processes,  walking-beam-type reheating furnace,  

Full Text: PDF>>
Buy this Article

In this paper, the integration of dynamic plant-wide optimization and distributed generalized predictive control (DGPC) is presented for serially connected processes. On the top layer, chance-constrained programming (CCP) is employed in the plant-wide optimization with economic and model uncertainties, in which the constraints containing stochastic parameters are guaranteed to be satisfied at a high level of probability. The deterministic equivalents are derived for linear and nonlinear individual chance constraints, and an algorithm is developed to search for the solution to the joint probability constrained problem. On the lower layer, the distributed GPC method based on neighborhood optimization with one-step delay communication is developed for on-line control of the whole system. Simulation studies for furnace temperature set-points optimization problem of the walking-beam-type reheating furnace are illustrated to verify the effectiveness and practicality of the proposed scheme.