Premium
Smart plant operations: Vision, progress and challenges
Author(s) -
Christofides Panagiotis D.,
Davis James F.,
ElFarra Nael H.,
Clark Don,
Harris Kevin R. D.,
Gipson Jerry N.
Publication year - 2007
Publication title -
aiche journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.958
H-Index - 167
eISSN - 1547-5905
pISSN - 0001-1541
DOI - 10.1002/aic.11320
Subject(s) - dept , engineering , library science , chemistry , computer science , stereochemistry
T he process industry is involved with the conversion of raw materials, through a series of chemical processing steps, to valued products and is a key economic sector in the U.S. and globally. The global market share and business performance of the process industry is heavily based on the value that can be generated from its assets which are comprised of process sites, people and materials, as well as intellectual property in the form of product knowledge, process expertise and physical properties of materials. While the range of valuable assets is large, nearly all the economic value in terms of operating profit in the process industry is a direct result of plant operations. This realization has motivated extensive research, over the last 40 years, on the development of advanced operation and control strategies to achieve economically optimal plant operation by regulating process variables at appropriate values. Figure 1 depicts the existing paradigm that couples plant management with process feedback control. This paradigm has been widely adopted by the process industries and extensively studied by the process systems engineering community. This paradigm features two distinct levels of plant operations: a plant management level and a process control level. At the plant management level, an optimization problem is solved on the basis of a (typically) steady state model of the plant to compute the economically optimal values for the process variables, while in the process control level, feedback control systems are used to regulate the process variables at the specified values. Additionally, the important task of plant monitoring — that is the determination of abnormal, potentially faulty plant behavior by proper analysis of plant sensor data — and the incorporation of plant operator input take place at the plant management level. While the paradigm of Figure 1 has undoubtedly been a successful one, over the last few years there have been numerous calls (e.g.,) for expanding this paradigm in a number of directions. Specifically, while economic prosperity has always Perspective