Nonlinear Model-based Process Control : Applications in Petroleum Refining

The series Advances in Industrial Control aims to report and encourage technology transfer in control engineering. The rapid development of control technology has an impact on all areas of the control discipline. New theory, new controllers, actuators, sensors, new industrial processes, computer methods, new applications, new philosophies ... , new challenges. Much of this development work resides in industrial reports, feasibility study papers and the reports of advanced collaborative projects. The series offers an opportunity for researchers to present an extended exposition of such new work in all aspects of industrial control for wider and rapid dissemination. The last decade has seen considerable interest in reviving the fortunes of non­ linear control. In contrast to the approaches of the 60S, 70S and 80S a very pragmatic agenda for non-linear control is being pursued using the model-based predictive control paradigm. This text by R. Ansari and M. Tade gives an excellent synthesis of this new direction. Two strengths emphasized by the text are: (i) four applications found in refinery processes are used to give the text a firm practical continuity; (ii) a non-linear model-based control architecture is used to give the method a coherent theoretical framework.

1 Introduction.- 1.1 Non-linear Model-based Control.- 1.2 Motivation for this Book.- 1.3 Objectives and Contributions.- 1.3.1 Objectives.- 1.3.2 Contributions.- 1.3.2.1 Non-linear Control Theory and Development.- 1.3.2.1 Practical Applications in Industries.- 1.4 Scope of the Book.- 1.5 Book Overview.- 2 Literature Review.- 2.1 Introduction.- 2.1.1 Industrial Background.- 2.1.2 Academic Background.- 2.2 Model-predictive Control.- 2.2.1 Dynamic Matrix Control (DMC).- 2.2.2 Limitations of DMC.- 2.2.3 Model Algorithm Control (MAC).- 2.2.4 Difference Between DMC and MAC.- 2.2.5 Principal Component Analysis (PCA).- 2.3 Internal Model Control (IMC).- 2.3.1 IMC Theoretical Background.- 2.3.2 Comparison of IMC with DMC and MAC.- 2.3.3 Extensions and Variants of IMC.- 2.4 Stability and Robustness of Linear MPC.- 2.5 Non-linear Model-based Control (NMBC).- 2.5.1 Introduction.- 2.5.2 Non-linear Model-based Control Architecture.- 2.5.2.1 NMBC-Model.- 2.5.2.2 NMBC-Model Parameter Update.- 2.5.2.3 NMBC- Controller Configuration/Simulation.- 2.5.3 Non-linear Control System Technique.- 2.5.4 Non-linear Programming Methods.- 2.6 Generic Model Control (GMC).- 2.6.1 Introduction.- 2.6.2 GMC and Internal Model Control (IMC).- 2.6.3 GMC and Model-predictive Control (MPC).- 2.6.3.1 Discrete form of GMC.- 2.6.3.2 Relationship between Discrete GMC and MPC.- 2.7 Stability and Robustness of Non-linear System.- 2.8 Conclusions and Discussion.- 3 Inferential Models In Non-linear Multivariable Control Applications.- 3.1 Introduction.- 3.2 Development of Inferential Models.- 3.2.1 Overview of Models.- 3.2.2 Non-linear Inferential Control Model.- 3.2.3 Correlation-Based Model.- 3.2.4 Verification of Correlation-Based Model.- 3.3 On-line Applications of Inferential Models.- 3.3.1 Naphtha Final Boiling Point (FBP) of Crude Distillation.- 3.3.2 Kerosene Flash Point of Crude Distillation.- 3.3.3 Reid Vapour Pressure (RVP) of Debutanizer Bottom.- 3.3.4 Iso-Pentane of Debutanizer Overhead.- 3.3.5 Octane Inferential Model for Catalytic Reforming.- 3.4 Tuning of Inferential Models.- 3.5 Inferential Models in Non-linear Multivariable Control Applications.- 3.6 Benefits of Inferential Models.- 3.7 Conclusions.- 4 Non-linear Model-based Multivariable Control of a Debutanizer.- 4.1 Introduction.- 4.2 The Debutanizer Control Strategy.- 4.2.1 Objective.- 4.2.2 Process Description.- 4.2.3 Control Proposal.- 4.2.4 Hardware Consideration.- 4.3 The Non-linear GMC Control Law.- 4.4 GMC Application to Debutanizer.- 4.5 Model Development.- 4.5.1 Steady-State Model Considerations.- 4.5.2 Development of Inferential Models.- 4.5.2.1 RVP of Platformer Feed.- 4.5.2.2 Iso-pentane of the Debutanizer Overhead.- 4.6 Controller Implementation.- 4.6.1 Controller-Process Interface.- 4.6.2 Non-linear Controller Tuning.- 4.7 Results and Discussions.- 4.8 Cost/Benefit Analysis.- 4.8.1 Benefits Calculations.- 4.9 Conclusions.- 5 Non-linear Model-based Multivariable Control of a Crude Distillation Process.- 5.1 Introduction.- 5.2 Crude Distillation Process Control Overview.- 5.2.1 Process Description.- 5.2.2 Control Objectives and Constraints.- 5.2.2.1 Control Objectives.- 5.2.2.2 Control Constraints.- 5.2.3 Dynamic Model with Uncertainty.- 5.3 Non-linear Control Algorithm for Fractionator.- 5.4 Model Parameter Update.- 5.5 Model-predictive Control.- 5.5.1 Problem Formulation for Linear Control.- 5.5.2 MATLAB®/SIMULINK® Programme.- 5.6 Results and Discussions.- 5.6.1 Simulation Results.- 5.6.2 Integrating Top End Point Inferential Model.- 5.6.3 Real ―time Implementation Results.- 5.7 Conclusions.- 6 Constrained Non-linear Multivariable Control of a Catalytic Reforming Process.- 6.1 Introduction.- 6.2 Process Constraints Classifications.- 6.3 Constraint Non-linear Multivariable Control.- 6.3.1 Control Theory and Design.- 6.3.2 Selection of Design Parameters.- 6.3.2.1 Selection of K1C and K2C.- 6.3.2.2 Selection of W.- 6.4 Application to Catalytic Reforming Process.- 6.4.1 Dynamic Model of the Process.- 6.4.1.1 Introduction.- 6.4.1.2 Model Development.- 6.4.1.3 Numerical Integration.- 6.4.1.4 Results and Discussion.- 6.4.1.5 Conclusions.- 6.4.2 Non-linear Control Algorithm for Reforming Reactors.- 6.4.2.1 Problem Formulation.- 6.4.2.2 Constrained Non-linear Optimization Problem.- 6.4.3 Non-linear Control Objectives and Strategies.- 6.4.3.1 Control Objectives.- 6.4.3.2 WAIT/Octane Control Strategies.- 6.5 Real-time Implementation.- 6.5.1 Non-linear Controller Tuning.- 6.5.2 Results and Discussions.- 6.6 Conclusions.- 7 Non-linear Multivariable Control of a Fluid Catalytic Cracking Process.- 7.1 Introduction.- 7.2 FCC Process Control Overview.- 7.2.1 Process Description.- 7.2.2 Economic Objectives and Non-linear Control Strategies.- 7.3 Dynamic Model of FCC Process.- 7.3.1 Model Development.- 7.3.1.1 Riser and Reactor Section.- 7.3.1.2 Regenerator Section.- 7.3.2 Results and Discussions.- 7.4 Non-linear Control Algorithm for FCC Reactor-Regenerator System.- 7.5 Dynamic Model Parameter Update.- 7.5.1 Introduction.- 7.5.2 Development of Model Parameter Update System.- 7.5.2.1 Parameter Update Algorithm.- 7.5.2.2 Application to FCC Process.- 7.6 Model-predictive Control.- 7.6.1 Problem Formulation for Linear Control.- 7.6.1.1 Signal Conditioning.- 7.6.1.2 Prediction Trend Correction.- 7.6.1.3 Control Move Calculation.- 7.6.2 Process Identification Tests.- 7.6.2.1 Combustion-air-flow Models.- 7.6.2.2 Feed-flow-rate Models.- 7.6.2.3 Feed-preheat-temperature Models.- 7.6.2.4 Riser-outlet-temperature Models.- 7.7 Real-time Implementation.- 7.7.1 Non-linear Controller Tuning.- 7.7.2 Controller Interface to DCS System.- 7.7.3 Results and Discussions.- 7.7.4 Comparison of Non-linear Control with DMC.- 7.8 Plant Results.- 7.9 Conclusions.- 8 Conclusions and Recommendations.- 8.1 Conclusions.- 8.1.1 Summary of Results Achieved.- 8.1.2 Outline of the Major Contributions of this Research.- 8.2 Recommendations.- 8.2.1 Embedded Optimization for Non-linear Control.- 8.2.2 Non-linear Nonminimum Phase Systems.- 8.2.3 Robust Stability and Performance of Non-linear Systems.- 8.2.4 On-line Parameter Estimation (Model Adaptation).- Appendix A.- A.1 Programme for Pressure-compensated Temperature.- A.2 Programme for Naphtha-final-boiling-point Inferential Model.- A.3 Theory Underlying the Pressure-compensated Temperature.- Appendix B.- B.1 S-B GMC Controller Implementation.- Appendix C.- Constrained Multivariable Control System Programme for Shell Heavy Oil Fractionator.- Appendix D.- D.1 Description and Application of Real-time Optimization (RT-Opt.) Software to Catalytic Reforming Reactor Section.- D.1.1 Description.- D.1.2 Mathematical Algorithm.- D.1.3 Application to Catalytic Reforming Reactor Section.- D.2 Implementation Procedure of Real-time Optimization (RT-Opt.).- Appendix E.- Constrained Multivariable Predictive Control for Fluid Catalytic Cracking (FCC) Process.- References.