Automatic Control of Bioprocesses

Denis Dochain is a Professor at the CESAME (Centre for Systems Engineering and Applied Mechanics) of the Université Catholique de Louvain and Honorary Research Director of the FNRS (Fonds National de la Recherche Scientifique, National Fund for Scientific Research), France. He is associate editor of Automatica, the IEEE Transactions on Automatic Control and the Journal of Process Control.

• Chapter 1. What are the Challenges for the Control of Bioprocesses? 11Denis DOCHAIN1.1. Introduction . 111.2. Specific problems of bioprocess control 121.3. A schematic view of monitoring and control of a bioprocess 121.4. Modeling and identification of bioprocesses: some key ideas 131.5. Software sensors: tools for bioprocess monitoring 141.6. Bioprocess control: basic concepts and advanced control151.7. Bioprocess monitoring: the central issue 151.8. Conclusions 161.9. Bibliography 16Chapter 2. Dynamic Models of Biochemical Processes: Properties of Models 17Olivier BERNARD and Isabelle QUEINNEC2.1. Introduction 172.2. Description of biochemical processes 182.2.1. Micro-organisms and their use 182.2.2. Types of bioreactors 192.2.3. Three operating modes 192.3. Mass balance modeling 212.3.1. Introduction 212.3.2. Reaction scheme 212.3.3. Choice of reactions and variables 232.3.4. Example 1 232.4. Mass balance models 242.4.1. Introduction 242.4.2. Example 2 242.4.3. Example 3 252.4.4. Matrix representation 252.4.4.1. Example 2 (continuation) 262.4.4.2. Example 1 (continuation) 262.4.5. Gaseous flow 272.4.6. Electroneutrality and affinity constants 272.4.7. Example 1 (continuation) 282.4.8. Conclusion 292.5. Kinetics 302.5.1. Introduction 302.5.2. Mathematical constraints 302.5.2.1. Positivity of variables 302.5.2.2. Variables necessary for the reaction 312.5.2.3. Example 1 (continuation) 312.5.2.4. Phenomenological knowledge 312.5.3. Specific growth rate 322.5.4. Representation of kinetics by means of a neural network 342.6. Validation of the model 352.6.1. Introduction 352.6.2. Validation of the reaction scheme 352.6.2.1. Mathematical principle 352.6.2.2. Example 4 362.6.3. Qualitative validation of model 372.6.4. Global validation of the model 392.7. Properties of the models 392.7.1. Boundedness and positivity of variables 392.7.2. Equilibrium points and local behavior 402.7.2.1. Introduction 402.8. Conclusion 422.9. Bibliography 43Chapter 3. Identification of Bioprocess Models 47Denis DOCHAIN and Peter VANROLLEGHEM3.1. Introduction 473.2. Structural identifiability 483.2.1. Development in Taylor series 493.2.2. Generating series 503.2.3. Examples for the application of the methods of development in series 503.2.4. Some observations on the methods for testing structural identifiability 513.3. Practical identifiability 523.3.1. Theoretical framework 523.3.2. Confidence interval of the estimated parameters 543.3.3. Sensitivity functions 553.4. Optimum experiment design for parameter estimation (OED/PE) 573.4.1. Introduction 573.4.2. Theoretical basis for the OED/PE 593.4.3. Examples 613.5. Estimation algorithms 633.5.1. Choice of two datasets 633.5.2. Elements of parameter estimation: least squares estimation in the linear case 643.5.3. Overview of the parameter estimation algorithms 653.6. A case study: identification of parameters for a process modeled for anaerobic digestion 683.6.1. The model 693.6.2. Experiment design 703.6.3. Choice of data for calibration and validation 703.6.4. Parameter identification 713.6.5. Analysis of the results 753.7. Bibliography 75Chapter 4. State Estimation for Bioprocesses 79Olivier BERNARD and Jean-Luc GOUZÉ4.1. Introduction 794.2. Notions on system observability 804.2.1. System observability: definitions 804.2.2. General definition of an observer 814.2.3. How to manage the uncertainties in the model or in the output 834.3. Observers for linear systems 844.3.1. Luenberger observer 854.3.2. The linear case up to an output injection 864.3.3. Local observation of a nonlinear system around an equilibrium point 864.3.4. PI observer 874.3.5. Kalman filter 874.3.6. The extended Kalman filter 894.4. High gain observers 894.4.1. Definitions, hypotheses 894.4.2. Change of variable 904.4.3. Fixed gain observer 914.4.4. Variable gain observers (Kalman-like observer) 914.4.5. Example: growth of micro-algae 924.5. Observers for mass balance-based systems 944.5.1. Introduction 944.5.2. Definitions, hypotheses 964.5.3. The asymptotic observer 964.5.4. Example 984.5.5. Improvements 994.6. Interval observers 1014.6.1. Principle 1024.6.2. The linear case up to an output injection 1034.6.3. Interval estimator for an activated sludge process 1054.6.4. Bundle of observers 1074.7. Conclusion 1104.8. Appendix: a comparison theorem 1114.9. Bibliography 112Chapter 5. Recursive Parameter Estimation 115Denis DOCHAIN5.1. Introduction 1155.2. Parameter estimation based on the structure of the observer 1165.2.1. Example: culture of animal cells 1165.2.2. Estimator based on the structure of the observer 1175.2.3. Example: culture of animal cells (continued) 1195.2.4. Calibration of the estimator based on the structure of the observer: theory 1195.2.5. Calibration of the estimator based on the structure of the observer: application to the culture of animal cells 1245.2.6. Experimental results 1275.3. Recursive least squares estimator 1295.4. Adaptive state observer 1335.4.1. Generalization 1385.5. Conclusions 1405.6. Bibliography 141Chapter 6. Basic Concepts of Bioprocess_Control 143Denis DOCHAIN and Jérôme HARMAND6.1. Introduction 1436.2.1. Biological system dynamics 1446.2.2. Sources of uncertainties and disturbances of biological systems 1466.3. Stability of biological processes 1476.3.1. Basic concept of the stability of a dynamic system 1476.3.2. Equilibrium point 1486.3.3. Stability analysis 1496.4. Basic concepts of biological process control 1506.4.1. Regulation and tracking control 1506.4.2. Strategy selection: direct and indirect control 1516.2. Bioprocess control: basic concepts 1446.4.3. Selection of synthesis method 1526.5. Synthesis of biological process control laws 1536.5.1. Representation of systems 1536.5.2. Structure of control laws 1546.6. Advanced control laws 1606.6.1. A nonlinear PI controller 1606.6.2. Robust control 1626.7. Specific approaches 1656.7.1. Pulse control: a dialog with bacteria 1656.7.2. Overall process optimization: towards integrating the control objectives in the initial stage of bioprocess design 1676.8. Conclusions and perspectives 1706.9. Bibliography 170Chapter 7. Adaptive Linearizing Control and Extremum-Seeking Control of Bioprocesses 173Denis DOCHAIN, Martin GUAY, Michel PERRIER and Mariana TITICA7.1. Introduction 1737.2. Adaptive linearizing control of bioprocesses 1747.2.1. Design of the adaptive linearizing controller 1747.2.2. Example 1: anaerobic digestion 1767.2.2.1. Model order reduction 1777.2.2.2. Adaptive linearizing control design 1797.2.3. Example 2: activated sludge process 1837.3. Adaptive extremum-seeking control of bioprocesses 1887.3.1. Fed-batch reactor model 1897.3.2. Estimation and controller design 1917.3.2.1. Estimation equation for the gaseous outflow rate y 1917.3.2.2. Design of the adaptive extremum-seeking controller 1927.3.2.3. Stability and convergence analysis 1957.3.2.4. A note on dither signal design 1967.3.3. Simulation results 1977.4. Appendix: analysis of the parameter convergence 2027.5. Bibliography 207Chapter 8. Tools for Fault Detection and Diagnosis 211Jean-Philippe STEYER, Antoine GÉNOVÉSI and Jérôme HARMAND8.1. Introduction 2118.2. General definitions 2128.2.1. Terminology 2128.2.2. Fault types 2138.3. Fault detection and diagnosis 2148.3.1. Methods based directly on signals 2158.3.1.1. Hardware redundancy 2158.3.1.2. Specific sensors 2168.3.1.3. Comparison of thresholds 2178.3.1.4. Spectral analysis 2178.3.1.5. Statistical approaches 2188.3.2. Model-based methods 2188.3.2.1. Parity space 2198.3.2.2. Observers 2208.3.2.3. Parametric estimation 2218.3.3. Methods based on expertise 2228.3.3.1. AI models 2238.3.3.2. Artificial neural networks 2248.3.3.3. Fuzzy inference systems 2258.3.4. Choice and combined use of diverse methods 2278.4. Application to biological processes 2278.4.1. “Simple” biological processes 2288.4.2. Wastewater treatment processes 2298.5. Conclusion 2318.6. Bibliography 232List of Authors 239Index 241