Monitoring and control of information-poor systems: an approach based on fuzzy relational models

The monitoring and control of a system whose behaviour is highly uncertain isan important and challenging practical problem. Methods of solution based on fuzzy techniques have generated considerable interest, but very little of the existing literature considers explicit ways of taking uncertainties into account. This book describes an approach to the monitoring and control of information-poor systems that is based on fuzzy relational models which generate fuzzy outputs.The first part of Monitoring and Control of Information-Poor Systems aims to clarify why design decisions must take account of the uncertainty associated with optimal choices, and to explain how a fuzzy relational model can beused to generate a fuzzy output, which reflects the uncertainties associated with its predictions. Part two gives a brief introduction to fuzzy decision-making and shows how it can be used to design a predictive control scheme that is suitable for controlling information-poor systems using inaccurate measurements. Part three describes different ways in which fuzzy relational models can be generated online and explains the practical issues associated with their identification and application. The final part of the book provides examples of the use of the previously described techniques in real applications.Key features:Describes techniques applicable to a wide range of engineering, environmental, medical, financial and economic applicationsUses simple examples to help explain the basic techniques for dealing with uncertaintyDescribes a novel design approach based on the use of fuzzy relational modelsConsiders practical issues associated with applying the techniques to real systemsMonitoring and Control of Information-Poor Systems forms an invaluable resource for a wide range of graduate students, and is also a comprehensive reference for researchers and practitioners working on problems involving mathematical modelling and control. INDICE: Preface xiAbout the Author xvAcknowledgements xviiI ANALYSING THE BEHAVIOUR OF INFORMATION-POOR SYSTEMS1 Characteristics of Information-Poor Systems 31.1 Introduction to Information-Poor Systems 31.1.1 Blast Furnaces 31.1.2 Container Cranes 31.1.3 Cooperative Control Systems 41.1.4 Distillation Columns 41.1.5 Drug Administration 41.1.6 Electrical Power Generation and Distribution 41.1.7 Environmental Risk Assessment Systems 41.1.8 Financial Investment and Portfolio Selection 51.1.9 Health Care Systems 51.1.10 Indoor Climate Control 51.1.11 NOx Emissions from Gas Turbines and Internal Combustion Engines 61.1.12 Penicillin Production Plant 61.1.13 Polymerization Reactors 61.1.14 Rotary Kilns 61.1.15 Solar Power Plant 71.1.16 Wastewater Treatment Plant 71.1.17 Wood Pulp Production Plant 71.2 Main Causes of Uncertainty 71.2.1 Sources of Modelling Errors 81.2.2 Sources of Measurement Errors 81.2.3 Reasons for PoorlyDefined Objectives and Constraints 91.3 Design in the Face of Uncertainty 9References 92 Describing and Propagating Uncertainty 132.1 Methods of DescribingUncertainty 132.1.1 Uncertainty Intervals and Probability Distributions 132.1.2 Fuzzy Sets and Fuzzy Numbers 142.2 Methods of Propagating Uncertainty 152.2.1 Interval Arithmetic 152.2.2 Statistical Methods 162.2.3 Monte Carlo Methods162.2.4 Fuzzy Arithmetic 172.3 Fuzzy Arithmetic Using α-Cut Sets and Interval Arithmetic 182.4 Fuzzy Arithmetic Based on the Extension Principle 212.5 Representing and Propagating Uncertainty Using Pseudo-Triangular Membership Functions 242.6 Summary 27References 273 Accounting for Measurement Uncertainty 293.1 Measurement Errors 293.2 Introduction to Fuzzy Random Variables 293.2.1 Definition of a Fuzzy Random Variable 303.2.2 Generating Fuzzy Random Variables from a Knowledge of the Random and Systematic Errors 303.3 A Hybrid Approach tothe Propagation of Uncertainty 323.4 Fuzzy Sensor Fusion Based on the Extension Principle 343.5 Fuzzy Sensors 383.6 Summary 39References 394 Accounting forModelling Errors in Fuzzy Models 414.1 An Introduction to Rule-Based Models 414.2 Linguistic Fuzzy Models 414.2.1 Fuzzy Rules 414.2.2 Fuzzy Inferencing 424.2.3 Compositional Rules of Inference 434.3 Functional Fuzzy Models 474.4 Fuzzy Neural Networks 484.5 Methods of Generating Fuzzy Models 504.5.1 Modifying Expert Rules to Take Account of Uncertainty 504.5.2 Identifying Fuzzy Rules from Data 564.6 Defuzzification 584.7 Summary 60References 605 Fuzzy Relational Models 635.1 Introduction to Fuzzy Relations and Fuzzy Relational Models 635.2 Fuzzy FRMs 655.3 Methods of Estimating Rule Confidences from Data 675.4 Estimating Probability Density Functions from Data 705.4.1 Probabilistic Interpretation of RSK Fuzzy Identification 715.4.2 Effect of Structural Errors on the Output of a Fuzzy FRM 785.4.3 Estimation Based on Limited Amounts of Training Data 835.5 Generic Fuzzy Models 865.5.1 Identification of Generic Fuzzy Models 875.5.2 Reducing the Time Required to Generate the Training Data 915.6 Summary 92References 92II CONTROL OF INFORMATION-POOR SYSTEMS6 Fuzzy Decision-Making 976.1 Risk Assessment in Information-Poor Systems 976.2 Fuzzy Optimization in Information-Poor Systems 996.2.1 Fuzzy Goals and Fuzzy Constraints 996.2.2 FuzzyAggregation Operators 996.2.3 Fuzzy Ranking 1006.3 Multi-Stage Decision-Making 1016.3.1 Fuzzy Dynamic Programming 1026.3.2 Branch and Bound 1036.3.3 Genetic Algorithms 1066.4 Fuzzy Decision-Making Based on Intuitionistic Fuzzy Sets 1066.4.1 Definition of an Intuitionistic Fuzzy Set 1066.4.2 Multi-Attribute Decision-Making Using Intuitionistic Fuzzy Numbers 1076.5 Summary 108References 1087 Predictive Control in Uncertain Systems 1117.1 Model-Based Predictive Control 1117.2 Fuzzy Approaches to Model-Based Control of Uncertain Systems 1127.2.1 Inverse Control of Fuzzy Interval Systems 1127.2.2 Fuzzy Model-Based Predictive Control 1147.3 Practical Issues Associated with Multi-Step Fuzzy Decision-Making 1157.3.1 Limiting the Accumulation of Uncertainty 1157.3.2 Avoiding Excessive Computational Demands When Using Enumerative Search Optimization 1157.3.3 Avoiding Excessive Computational Demands When Using Evolutionary Algorithms 1167.3.4 Handling Infeasibility 1177.3.5 Choosing the Weighting in Multi-Criteria Cost Functions 1177.3.6 Dealing with Hard Constraints 1187.4 A Simplified Approach to Fuzzy FRM-Based Predictive Control 1187.4.1 The Fuzzy Decision-Maker 1197.4.2 Conditional Defuzzification 1207.5 FMPC of an Uncertain Dynamic System Based on a Generic Fuzzy FRM 1227.6 Summary 127References 1288 Incorporating Fuzzy Inputs 1298.1 Fuzzy Setpoints and Fuzzy Measurements 1298.1.1 Fuzzy Setpoints 1298.1.2 Fuzzy Measurements 1298.2 Fuzzy Measures of the Tracking Error and its Derivative 1318.3 Inference with Fuzzy Inputs 1368.4 Fuzzy Output Neural Networks 1388.5 Modelling Input Uncertainty Using a Fuzzy FRM 1408.6 Summary 151References 1519 Disturbance Rejection in Information-Poor Systems 1539.1 Rejecting Unmeasured Disturbances in Uncertain Systems 1549.1.1 Robust Fuzzy Control 1549.1.2 Feedback Linearization Using a Fuzzy Disturbance Observer 1559.1.3 Fuzzy Model-Based Internal Model Control 1559.2 Fuzzy IMC Based on a Fuzzy Output FRM 1579.3 Rejecting Measured Disturbances in Non-Linear Uncertain Systems 1619.4 Fuzzy MPC with Feedforward 1629.5 Summary 166References 166III ONLINE LEARNING IN INFORMATION-POOR SYSTEMS10 Online Model Identification in Information-Poor Environments 17110.1 Online Fuzzy Identification Schemes 17110.1.1 Recursive Fuzzy Least-Squares 17110.1.2 Recursive Forms of the RSK Algorithm 17210.2 Effect of Poor-Quality and Incomplete Training Data 17610.3 Ways of Reducing the Computational Demands 17710.3.1 Evolving Fuzzy Models 17710.3.2 Hierarchical Fuzzy Models 18110.4 Summary 185References 18511 Adaptive Model-Based Control of Information-Poor Systems 18711.1 Robust Adaptive Fuzzy Control 18711.2 Adaptive Fuzzy FRM-Based Predictive Control 18811.3 Commissioning the Controller 18911.3.1 Methods of Incorporating Prior Knowledge 18911.3.2 Initialization Using a Generic Fuzzy FRM 18911.4 Generating an Optimal ControlSignal Using a Partially Trained Model 19211.4.1 Taking the Amount of Training into Account 19211.4.2 Incorporating a Secondary Controller 19411.4.3 Combining the Fuzzy Predictions Generated by More than One Model 20111.5 Dealing with the Effects of Disturbances 20211.5.1 Adaptive Feedforward Control Based on an Inaccurate Disturbance Measurement 20311.6 Summary 209References 20912 Adaptive Model-Free Control of Information-Poor Systems 21112.1 Introduction to Model-Free Adaptive Control of Non-Linear Systems 21112.2 Fuzzy FRM-Based DirectAdaptive Control 21112.3 Behaviour in the Presence of a Noisy Measurement of the Plant Output 21312.4 Behaviour in the Presence of an Unmeasured Disturbance 21812.5 Accounting for Uncertainty Arising from a Measured Disturbance 22212.6 Summary 227References 22713 Fault Diagnosis in Information-Poor Systems 22913.1 Introduction to Fault Detection and Isolation in Non-Linear Uncertain Systems 22913.1.1 Model-Based Methods for Non-Linear Systems 23013.1.2 Ways of Accounting for Uncertainty 23213.2 A Fuzzy FRM-Based Fault Diagnosis Scheme 23313.2.1 Measuring the Similarity of FRMs 23413.2.2 Accumulating Evidence of Fault-Free or Faulty Operation 23613.2.3 Generating Robust Generic Models of Faulty Operation 23913.2.4 Multi-Step Fault Diagnosis 23913.3 Summary 242References243IV SOME EXAMPLE APPLICATIONS14 Control of Thermal Comfort 24714.1 Main Sources of Uncertainty and Practical Considerations 24814.2 Review of Approaches Suggested for Dealing with the Uncertainty 24914.3 Design of the Fuzzy FRM-Based Control System 24914.3.1 The Fuzzy FRM 25014.3.2 The Fuzzy Cost Functions 25214.3.3 The Fuzzy Goals 25214.3.4 The Fuzzy Decision-Maker 25414.3.5 The Conditional Defuzzifier 25414.4 Performance of the Thermal Comfort Controller 25414.5 Concluding Remarks 258References 25915 Identification of Faults in Air-Conditioning Systems 26115.1 Main Sources of Uncertainty and Practical Considerations 26115.2 Design of a Fuzzy FRM-Based Monitoring System for a Cooling Coil Subsystem 26315.3 Diagnosis of Known Faults in a Simulated Cooling Coil Subsystem 26415.3.1 Fault-Free Operation 26415.3.2 Leaky Valve 26415.3.3 Fouled Coil26515.3.4 Valve Stuck in the Fully Closed Position 26615.3.5 Valve Stuck in the Midway Position 26715.3.6 Valve Stuck in the Fully Open Position 26815.4 Commissioning of Air-Handling Units 26915.5 Concluding Remarks 272References 27216 Control of Heat Exchangers 27516.1 Main Sources of Uncertainty and Practical Considerations 27516.2 Design of a Fuzzy FRM-Based Predictive Controller 27616.3 Design of a Fuzzy FRM-Based Internal Model Control Scheme 28316.4 Concluding Remarks 290References 29017 Measurement of Spatially Distributed Quantities 29317.1 Review of Approaches Suggested for Dealing with Sensor Bias 29317.2 An Example Application 29417.2.1 Air Temperature Estimation Using a Single-Point Sensor with Bias Correction 29417.2.2 Air Temperature Estimation Based on Mass and Energy Balances 29917.3 Using Bias Estimation and Fuzzy Data Fusion toImprove Automated Commissioning in Air-Handling Units 30217.3.1 Diagnosis When the Measurement Bias is Estimated Accurately 30317.3.2 Diagnosis When the Estimate of the Measurement Bias is Inaccurate 30317.4 Concluding Remarks 305References 306Index 309

• ISBN: 978-0-470-68869-4
• Editorial: John Wiley & Sons
• Encuadernacion: Cartoné
• Páginas: 336
• Fecha Publicación: 13/04/2012
• Nº Volúmenes: 1
• Idioma: Inglés