Two empirical strategies for open-loop on-line optimization are developed as alternatives to the use of mechanistic process models. These strategies are based on on-line identification of dynamic multi-input single-output (MISO) and multi-input multi-output (MIMO) models. The steady state gain of these models provides information for steady state optimization. Desirability functions, originally developed for multi-objective optimization, are utilized as objective function modifiers for constrained on-line optimization. The integration of dynamic model identification and desirability functions results in an on-line optimizer which combines fast optimizing speed with the ability to predict future encroachments on constraint boundaries. Corrections to the search direction are based on these predictions, reducing the probability of actual constraint violation.
The optimization strategies are tested by simulation on nonlinear multivariable interacting systems at two levels of complexity: a CSTR supporting a multiple reaction and a fluid catalytic cracker. Both methods were effective in avoiding violation of constraints but the MIMO strategy required fewer steps to reach an optimum and was less prone to generate a nonfeasible optimization step.
Deux strategies empiriques pour l'optimisation en ligne a boucle ouverte sont elaborees pour remplacer les modeles de procedes mecanistiques. Ces strategies sont basees sur l'identification en ligne de modeles «a entree multiple et sortie unique» et «a entree et sortie multiples». Le fonctionnement en regime permanent de ces modeles fournit des donnees pour l'optimisation a l'etat permanent. On utilise des fonctions de desirabilite, a l'origine etablies pour l'optimisation multi-objectif, comme modificateurs des fonctions objectives pour l'optimisation en ligne avec contrainte. L'integration de l'identification des modeles dynamiques et des fonctions de desirabilite permettent d'obtenir un optimiseur en ligne combinant une grande vitesse d'optimisation a la capacite de predire l'empietement futur sur les limites de contrainte. Les corrections de la direction de recherche s'appuient sur ces predictions, reduisant la probabilite d'une transgression reelle des contraintes.
Les strategies d'optimisation sont testees par simulation a partir de systemes interelies multivariables non lineaires a deux niveaux de complexite: un reacteur agite continu entraǐnant une reaction multiple et un craqueur catalytique fluide. Les deux methodes se sont averees efficaces pour eviter la transgression des contraintes, mais le modele a entree et sortie multiples necessite moins d'etapes pour atteindre un optimum et s'est montre moins enclin a generer une etape d'optimisation non realisable.
[1]
Ioan Doré Landau,et al.
On the recursive identification of multi-input, multi-output systems
,
1978,
Autom..
[2]
R. Liu,et al.
Determination of the structure of multivariable stochastic linear systems
,
1977
.
[3]
Babu Joseph,et al.
On-Line Optimization of Chemical Processes
,
1982,
1982 American Control Conference.
[4]
G. P. Rangaiah.
Studies in constrained optimization of chemical process problems
,
1985
.
[5]
F. B. Smith,et al.
An application of quadratic performance synthesis techniques to a fluid cat cracker
,
1971
.
[6]
Theodore J. Williams,et al.
A generalized chemical processing model for the investigation of computer control
,
1960,
Transactions of the American Institute of Electrical Engineers, Part I: Communication and Electronics.
[7]
G. Derringer,et al.
Simultaneous Optimization of Several Response Variables
,
1980
.
[8]
Kwan Wong,et al.
Identification of linear discrete time systems using the instrumental variable method
,
1967,
IEEE Transactions on Automatic Control.
[9]
Manfred Morari,et al.
Optimal operation of integrated processing systems. Part I: Open‐loop on‐line optimizing control
,
1981
.
[10]
D. M. Prett,et al.
Optimization and constrained multivariable control of a catalytic cracking unit
,
1980
.
[11]
M. J. Box.
A New Method of Constrained Optimization and a Comparison With Other Methods
,
1965,
Comput. J..
[12]
George Stephanopoulos,et al.
Optimizing control of industrial chemical processes: State of the art review
,
1980
.
[13]
N. Sinha,et al.
Determination of the Structure of a Canonical Model for the Identification of Linear Multivariable Systems
,
1979
.
[14]
Rolf Isermann,et al.
Adaptive on-line optimization of the steady-state behaviour of slow dynamic processes with process computers
,
1977
.
[15]
Rein Luus,et al.
A direct approach to optimization of a complex system
,
1973
.
[16]
Thomas J. Ward,et al.
Fluidized catalytic cracker control. A structural analysis approach
,
1981
.
[17]
George Stephanopoulos,et al.
Studies in the synthesis of control structures for chemical processes: Part IV. Design of steady‐state optimizing control structures for chemical process units
,
1980
.
[18]
R. Guidorzi.
Canonical structures in the identification of multivariable systems
,
1975,
Autom..