Comparison of Six Implicit Real-Time Optimization Schemes

Real-time optimization (RTO) is a class of methods that use m easurements to reject the effect of uncertainty on optimal performance. This articlecompares six implicit RTO schemes, that is, schemes that implement optimality not through nume rical optimization but rather via the control of appropriate variables. For unconstrained pr ocesses, the ideal controlled variable is the cost gradient. It is shown that, because of their struc tural differences, model-free and model-based techniques exhibit different features in term s of required excitation, convergence, scalability with the number of inputs and rejection of uncer tainty. This comparison is illustrated through a simulated CSTR. RESUME.L'optimsation en temps reel (RTO) est une classe de methodesou les mesures sont util- isees pour rejeter l'effet de l'incertitude. Cet article compare six techniques de RTO implicites qui optimisent un procede en controlant certaines variable s. En l'absence de contraintes, la grandeur commandee ideale est le gradient de la fonction cou t. A cause de leurs differences structurelles, les methodes sans modele et les methodes bas ees sur le modele se comportent differemment en termes de besoin d'excitation, de temps de c onvergence, de capacite de mise a l'echelle et d'aptitude a rejeter l'effet d'incertitudes. Cette comparaison est illustree en simu- lation au moyen d'un reacteur chimique a marche continue.

[1]  Xiaorong He,et al.  Application of steady-state detection method based on wavelet transform , 2003, Comput. Chem. Eng..

[2]  B. Srinivasan,et al.  Real‐time optimization of dynamic systems using multiple units , 2007 .

[3]  Dominique Bonvin,et al.  Optimizing control based on output feedback , 2009, Comput. Chem. Eng..

[4]  Sigurd Skogestad,et al.  Plantwide control: the search for the self-optimizing control structure , 2000 .

[5]  Kartik B. Ariyur,et al.  Real-Time Optimization by Extremum-Seeking Control , 2003 .

[6]  Michel Perrier,et al.  Improved performance in the multi-unit optimization method with non-identical units , 2009 .

[7]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[8]  George E. P. Box,et al.  Empirical Model‐Building and Response Surfaces , 1988 .

[9]  Miroslav Krstic,et al.  Stability of extremum seeking feedback for general nonlinear dynamic systems , 2000, Autom..

[10]  Vinay Kumar Kariwala Self-optimizing control of complex processes , 2010 .

[11]  Dominique Bonvin,et al.  Use of measurements for enforcing the necessary conditions of optimality in the presence of constraints and uncertainty , 2005 .

[12]  Lennart Ljung,et al.  System identification (2nd ed.): theory for the user , 1999 .

[13]  Dominique Bonvin,et al.  Adaptation strategies for real-time optimization , 2009, Comput. Chem. Eng..

[14]  A. C. Zanin,et al.  Multivariable control and real-time optimization- : an industrial practical view , 2005 .

[15]  Sigurd Skogestad,et al.  NULL SPACE METHOD FOR SELECTING OPTIMAL MEASUREMENT COMBINATIONS AS CONTROLLED VARIABLES , 2007 .

[16]  Lennart Ljung,et al.  System Identification: Theory for the User , 1987 .

[17]  M. Perrier,et al.  Influence of Differences in System Dynamics in the context of Multi-unit Optimization , 2009 .

[18]  Sigurd Skogestad,et al.  Self-Optimizing Control and NCO Tracking in the Context of Real-Time Optimization , 2010 .

[19]  Dominique Bonvin,et al.  Modifier-adaptation methodology for real-time optimization , 2009 .

[20]  Sigurd Skogestad,et al.  Extended nullspace method for selecting measurement combinations as controlled variables for optimal steady-state operation , 2007 .