Lughofer / Sayed-Mouchaweh Predictive Maintenance in Dynamic Systems

Advanced Methods, Decision Support Tools and Real-World Applications
1. Auflage 2019
ISBN: 978-3-030-05645-2
Verlag: Springer International Publishing
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 564 Seiten

This book provides a complete picture of several decision support tools for predictive maintenance. These include embedding early anomaly/fault detection, diagnosis and reasoning, remaining useful life prediction (fault prognostics), quality prediction and self-reaction, as well as optimization, control and self-healing techniques. It shows recent applications of these techniques within various types of industrial (production/utilities/equipment/plants/smart devices, etc.) systems addressing several challenges in Industry 4.0 and different tasks dealing with Big Data Streams, Internet of Things, specific infrastructures and tools, high system dynamics and non-stationary environments . Applications discussed include production and manufacturing systems, renewable energy production and management, maritime systems, power plants and turbines, conditioning systems, compressor valves, induction motors, flight simulators, railway infrastructures, mobile robots, cyber security and Internet of Things. The contributors go beyond state of the art by placing a specific focus on dynamic systems, where it is of utmost importance to update system and maintenance models on the fly to maintain their predictive power.  

Edwin Lughofer received his PhD-degree from the Johannes Kepler University Linz (JKU) in 2005. He is currently Key Researcher with the Fuzzy Logic Laboratorium Linz / Department of Knowledge-Based Mathematical Systems (JKU) in the Softwarepark Hagenberg. He has participated in several basic and applied research projects on European and national level, with a specific focus on topics of Industry 4.0 and FoF (Factories of the Future). He has published around 200 publications in the fields of evolving fuzzy systems, machine learning and vision, data stream mining, chemometrics, active learning, classification and clustering, fault detection and diagnosis, quality control and predictive maintenance, including 80 journals papers in SCI-expanded impact journals, a monograph on 'Evolving Fuzzy Systems' (Springer) and an edited book on 'Learning in Non-stationary Environments' (Springer). In sum, his publications received 4200 references achieving an h-index of 36. He is associate editor of the international journals Information Sciences, IEEE Transactions on Fuzzy Systems, Evolving Systems, Information Fusion, Soft Computing and Complex and Intelligent Systems, the general chair of the IEEE Conference on EAIS 2014 in Linz, the publication chair of IEEE EAIS 2015, 2016, 2017 and 2018, the program co-chair of the International Conference on Machine Learning and Applications (ICMLA) 2018, the tutorial chair of IEEE SSCI Conference 2018, the publication chair of the 3rd INNS Conference on Big Data and Deep Learning 2018, and the Area chair of the FUZZ-IEEE 2015 conference in Istanbul. He co-organized around 12 special issues and more than 20 special sessions in international journals and conferences. In 2006 he received the best paper award at the International Symposium on Evolving Fuzzy Systems, in 2013 the best paper award at the IFAC conference in Manufacturing Modeling, Management and Control (800 participants) and in 2016 the best paper award at the IEEE Intelligent Systems Conference. Moamar Sayed-Mouchaweh received his Master degree from the University of Technology of Compiegne-France in 1999. Then, he received his PhD degree from the University of Reims-France in December 2002. He was working as Associated Professor in Computer Science, Control and Signal processing at the University of Reims-France in the Research center in Sciences and Technology of the Information and the Communication (CReSTIC). In December 2008, he obtained the Habilitation to Direct Researches (HDR) in Computer science, Control and Signal processing. Since September 2011, he is working as a Full Professor in the High National Engineering School of Mines 'Ecole Nationale Supérieure des Mines de Douai' at the Department of Computer Science and Automatic Control (Informatique & Automatique). He edited the Springer book Learning in Non-Stationary Environments: Methods and Applications, in April 2012 and wrote two Brief Springer books in Electrical and Computer Engineering: Discrete Event Systems: Diagnosis and Diagnosability, and Learning from Data Streams in Dynamic Environments. He was a guest editor of several special issues of international journals. He was IPC Chair of the 12th IEEE International Conference on Machine Learning and Applications (ICMLA'13), the Conference Chair and IPC Chair of IEEE International Conference on Evolving and Adaptive Intelligent Systems (EAIS2015), and the IPC Chair of the 15th IEEE International Conference on Machine Learning and Applications (ICMLA'16). He is working as a member of the Editorial Board of Elsevier Journal Applied Soft Computing and Springer Journals Evolving Systems and Intelligent Industrial Systems.

Autoren/Hrsg.

Lughofer, Edwin

Sayed-Mouchaweh, Moamar

Inhaltsverzeichnis

1;Preface;6
2;Contents;8
3;Contributors;10
4;Prologue: Predictive Maintenance in Dynamic Systems;13
4.1;1 From Predictive to Preventive Maintenance in Dynamic Systems: Motivation, Requirements, and Challenges;13
4.2;2 Components and Methodologies for Predictive Maintenance;16
4.2.1;2.1 Models as Backbone Component;17
4.2.2;2.2 Methods and Strategies to Realize Predictive Maintenance;19
4.3;3 Beyond State-of-the-Art—Contents of the Book;23
4.4;References;32
5;Smart Devices in Production System Maintenance;36
5.1;1 Introduction;36
5.2;2 State of the Art;39
5.2.1;2.1 Definition of Terms;40
5.2.2;2.2 Physical Devices/Hardware;41
5.2.2.1;2.2.1 Smartphones and Tablets;41
5.2.2.2;2.2.2 Smartglasses;41
5.2.2.3;2.2.3 Smartwatches;42
5.2.3;2.3 Market View;43
5.2.4;2.4 Device Selection and Potentials;44
5.3;3 Application Examples in Maintenance;47
5.3.1;3.1 Local Data Analysis and Communication for Condition Monitoring;48
5.3.2;3.2 Remote Expert Solutions;50
5.3.3;3.3 Process Data Visualization for Process Monitoring;52
5.4;4 Limitations and Challenges;54
5.4.1;4.1 Hardware Limitations;54
5.4.2;4.2 User Acceptance;55
5.4.3;4.3 Information Compression on Smart Devices;57
5.4.4;4.4 Legal Aspects;58
5.5;5 Summary;59
5.6;References;60
6;On the Relevance of Preprocessing in Predictive Maintenance for Dynamic Systems;63
6.1;1 Introduction;63
6.2;2 Preprocessing;64
6.2.1;2.1 Taxonomy;65
6.2.2;2.2 Data Cleansing;66
6.2.2.1;2.2.1 Outlier Detection Based on Mahalanobis Distance;67
6.2.2.2;2.2.2 Outlier Detection Based on ?2 Approximations of Q and T2 Statistics;71
6.2.3;2.3 Data Normalization;73
6.2.4;2.4 Data Transformation;74
6.2.4.1;2.4.1 Statistical Transformations;74
6.2.4.2;2.4.2 Signal Processing;75
6.2.5;2.5 Missing Values Treatment;77
6.2.6;2.6 Data Engineering;78
6.2.6.1;2.6.1 Feature Selection;78
6.2.6.2;2.6.2 Feature Extraction;80
6.2.6.3;2.6.3 Feature Discretization;81
6.2.7;2.7 Imbalanced Data Treatment;82
6.2.7.1;2.7.1 Oversampling;83
6.2.7.2;2.7.2 Undersampling;84
6.2.7.3;2.7.3 Mixed Sampling;86
6.2.8;2.8 Models;87
6.2.8.1;2.8.1 Classification;87
6.2.8.2;2.8.2 Regression;88
6.3;3 Experimentation;89
6.3.1;3.1 Datasets;90
6.3.1.1;3.1.1 PHM Challenge 2014;90
6.3.1.2;3.1.2 PHM Challenge 2016;91
6.3.2;3.2 Experimental Schema;92
6.3.3;3.3 Results;93
6.4;4 Conclusions;97
6.5;References;98
7;Part I Anomaly Detection and Localization;104
7.1;A Context-Sensitive Framework for Mining Concept Drifting Data Streams;105
7.1.1;1 Concept Drifting Data Streams;105
7.1.1.1;1.1 Concept Drift;106
7.1.2;2 A Novel Framework for Online Learning in Adaptive Mode;107
7.1.2.1;2.1 Basic Components;107
7.1.2.2;2.2 Optimizing for Stream Volatility and Speed;108
7.1.3;3 Implementation of a Context-Sensitive Staged Learning Framework;108
7.1.3.1;3.1 The Use of the Discrete Fourier Transform in Classification and Concept Encoding;110
7.1.3.2;3.2 Repository Management;114
7.1.3.3;3.3 The Staged Learning Approach;115
7.1.3.3.1;3.3.1 Transition Between Stages;117
7.1.3.4;3.4 Space and Time Complexity of Spectral Learning;120
7.1.4;4 Empirical Study;121
7.1.4.1;4.1 Datasets Used for the Empirical Study;122
7.1.4.1.1;4.1.1 Synthetic Data;122
7.1.4.1.2;4.1.2 Synthetic Data Recurring with Noise;123
7.1.4.1.3;4.1.3 Synthetic Data Recurring with a Progressively Increasing Pattern of Drift;124
7.1.4.1.4;4.1.4 Synthetic Data Recurring with an Oscillating Drift Pattern;124
7.1.4.1.5;4.1.5 Real-World Data;125
7.1.4.2;4.2 Parameter Values;125
7.1.4.3;4.3 Effectiveness of Staged Learning Approach;125
7.1.4.4;4.4 Accuracy Evaluation;128
7.1.4.4.1;4.4.1 ARF vs SOL Accuracy of a Concept;130
7.1.4.5;4.5 Throughput Evaluation;131
7.1.4.6;4.6 Accuracy Versus Throughput Trade-Off;132
7.1.4.7;4.7 Memory Consumption Evaluation;132
7.1.5;5 Sensitivity Analysis;133
7.1.6;6 Conclusion;134
7.1.7;7 Future Research;136
7.1.8;References;136
7.2;Online Time Series Changes Detection Based on Neuro-FuzzyApproach;138
7.2.1;1 Introduction;138
7.2.2;2 Fuzzy Online Segmentation-Clustering;139
7.2.2.1;2.1 Probabilistic Approach;141
7.2.2.2;2.2 Possibilistic Approach;143
7.2.2.3;2.3 Online Combined Approach;145
7.2.2.4;2.4 Robust Approach;148
7.2.3;3 Robust Forecasting and Faults Detection in Nonstationary Time Series;163
7.2.4;4 Conclusions;172
7.2.5;References;172
7.3;Early Fault Detection in Reciprocating Compressor Valves by Means of Vibration and pV Diagram Analysis;174
7.3.1;1 Introduction;174
7.3.2;2 Problem Statement;176
7.3.2.1;2.1 Reciprocating Compressor Operation;176
7.3.2.2;2.2 Problem Statement;179
7.3.3;3 Vibration Analysis;181
7.3.3.1;3.1 Motivation;181
7.3.3.2;3.2 Feature Extraction;187
7.3.3.3;3.3 Feature Space;189
7.3.4;4 Analysis of the pV Diagram;190
7.3.4.1;4.1 Motivation;190
7.3.4.2;4.2 Feature Extraction;192
7.3.4.3;4.3 Feature Space;195
7.3.4.4;4.4 Classification;197
7.3.5;5 Experimental Setup;200
7.3.5.1;5.1 Compressor Test Bench;200
7.3.5.2;5.2 Test Runs;201
7.3.6;6 Results;203
7.3.6.1;6.1 Vibration Analysis;203
7.3.6.2;6.2 pV Diagram Analysis;206
7.3.7;7 Conclusions;209
7.3.8;References;210
7.4;A New Hilbert-Huang Transform Technique for Fault Detection in Rolling Element Bearings;213
7.4.1;1 Introduction;213
7.4.2;2 Minimum Entropy Deconvolution Filter;217
7.4.3;3 The Proposed eHT Technique for Bearing Fault Detection;220
7.4.3.1;3.1 Brief Discussion of Mathematical Morphology Analysis;221
7.4.3.1.1;3.1.1 Structural Element (SE);221
7.4.3.1.2;3.1.2 Dilation and Erosion;221
7.4.3.1.3;3.1.3 Closing and Opening;222
7.4.3.2;3.2 The Proposed Morphological Filter;223
7.4.3.3;3.3 The Proposed eHT Technique;225
7.4.4;4 Application of the Proposed eHT Technique for Bearing Fault Detection;226
7.4.4.1;4.1 Experimental Setup and Instrumentations;226
7.4.4.2;4.2 Performance Evaluation;228
7.4.4.2.1;4.2.1 Validation of Morphological-Based Filtering Technique;228
7.4.4.2.2;4.2.2 Validation of the Normality Measure;228
7.4.4.3;4.3 Evaluation of the Proposed eHT Technique;231
7.4.4.3.1;4.3.1 Condition Monitoring of a Healthy Bearing;231
7.4.4.3.2;4.3.2 Outer Race Fault Detection;232
7.4.4.3.3;4.3.3 Inner Race Fault Detection;232
7.4.4.3.4;4.3.4 Rolling Element Fault Detection;233
7.4.5;5 Conclusion;234
7.4.6;References;234
7.5;Comparison of Genetic and Incremental Learning Methods for Neural Network-Based Electrical Machine Fault Detection;237
7.5.1;1 Introduction;237
7.5.2;2 Electrical Machine Fault Detection;239
7.5.3;3 Genetic Algorithm for Neural Network Learning;243
7.5.3.1;3.1 Initialization and Parameterization;244
7.5.3.2;3.2 Phenotype Representation;245
7.5.3.3;3.3 Recombination Operator;247
7.5.3.3.1;3.3.1 Arithmetic Crossover;247
7.5.3.3.2;3.3.2 Multipoint Crossover;247
7.5.3.3.3;3.3.3 Local Intermediate Crossover;248
7.5.3.4;3.4 Mutation Operator;249
7.5.3.4.1;3.4.1 Gaussian Mutation;249
7.5.3.4.2;3.4.2 Random Mutation;249
7.5.3.4.3;3.4.3 Post-Processing Based on Local Random Mutation;250
7.5.3.5;3.5 Fitness Function;250
7.5.3.6;3.6 Selection Operator;251
7.5.3.6.1;3.6.1 Tournament Selection;251
7.5.3.6.2;3.6.2 Elitism;251
7.5.3.7;3.7 Stopping Criteria;252
7.5.4;4 Incremental Algorithm for Neurofuzzy Network Learning;252
7.5.4.1;4.1 Numerical and Fuzzy Data;253
7.5.4.2;4.2 Network Architecture;253
7.5.4.3;4.3 Fuzzy Neuron;255
7.5.4.3.1;4.3.1 Triangular Norm and Conorm;256
7.5.4.3.2;4.3.2 Neuron Model;256
7.5.4.4;4.4 Granular Region;257
7.5.4.5;4.5 Granularity Adaptation;258
7.5.4.6;4.6 Developing Granules;258
7.5.4.7;4.7 Adapting Connection Weights;260
7.5.4.8;4.8 Learning Algorithm;260
7.5.5;5 Results and Discussion;261
7.5.5.1;5.1 Preliminaries;261
7.5.5.2;5.2 Genetic EANN for Fault Detection;262
7.5.5.3;5.3 Incremental EGNN for Fault Detection;266
7.5.5.4;5.4 Comparative Analyses and Discussion;269
7.5.6;6 Conclusion;271
7.5.7;References;272
7.6;Evolving Fuzzy Model for Fault Detection and Fault Identification of Dynamic Processes;275
7.6.1;1 Introduction;275
7.6.2;2 Evolving Fuzzy Model;277
7.6.2.1;2.1 Fuzzy Cloud-Based Model Structure;277
7.6.2.2;2.2 Evolving Mechanism;279
7.6.3;3 Fault Detection and Identification;280
7.6.3.1;3.1 Learning/Training Phase;280
7.6.3.2;3.2 Fault Detection Phase;280
7.6.3.3;3.3 Fault Identification Phase;281
7.6.4;4 Description of the HVAC Process Model;282
7.6.4.1;4.1 Possible Faults on HVAC System;284
7.6.5;5 Experimental Results;285
7.6.6;6 Conclusion;287
7.6.7;References;289
7.7;An Online RFID Localization in the Manufacturing Shopfloor;292
7.7.1;1 Introduction;292
7.7.2;2 RFID Localization System;295
7.7.3;3 eT2QFNN Architecture;296
7.7.3.1;3.1 Input Layer;298
7.7.3.2;3.2 Quantum Layer;298
7.7.3.3;3.3 Rule Layer;298
7.7.3.4;3.4 Output Processing Layer;299
7.7.3.5;3.5 Output Layer;299
7.7.4;4 eT2QFNN Learning Policy;300
7.7.4.1;4.1 Rule Growing Mechanism;301
7.7.4.2;4.2 Parameter Adjustment;303
7.7.4.2.1;4.2.1 Fuzzy Rule Initialization;304
7.7.4.2.2;4.2.2 Winning Rule Update;305
7.7.5;5 Experiments and Data Analysis;309
7.7.5.1;5.1 Experiment Setup;309
7.7.5.2;5.2 Comparison with Existing Results;310
7.7.6;6 Conclusions;312
7.7.7;References;313
8;Part II Prognostics and Forecasting;315
8.1;Physical Model-Based Prognostics and Health Monitoring to Enable Predictive Maintenance;316
8.1.1;1 Introduction;316
8.1.2;2 Challenges in the Field of Predictive Maintenance;317
8.1.2.1;2.1 Combining Diagnosis and Prognosis;317
8.1.2.2;2.2 System Versus Component Level;318
8.1.2.3;2.3 Monitoring of Usage, Loads, Condition or Health;319
8.1.2.4;2.4 Interpretation of Monitoring Data;320
8.1.2.5;2.5 Data-Driven or Model-Based Prognostics;320
8.1.2.6;2.6 Selection of Most Suitable Approach and Technique;321
8.1.2.7;2.7 Data Quality;322
8.1.3;3 Structural Health and Condition Monitoring;323
8.1.3.1;3.1 Sensors;323
8.1.3.2;3.2 Vibration and Vibration-Based Monitoring;325
8.1.4;4 Physical Model-Based Prognostics;328
8.1.4.1;4.1 Relation Between Usage, Loads and Degradation Rate;329
8.1.4.2;4.2 Developing a Prognostic Method;330
8.1.4.3;4.3 Comparison to Data-Driven Approaches;334
8.1.5;5 Decision Support Tools;335
8.1.5.1;5.1 Guidelines for Selecting Suitable Approach;335
8.1.5.2;5.2 Critical Part Selection;338
8.1.6;6 Case Studies;341
8.1.6.1;6.1 Maritime Systems;341
8.1.6.2;6.2 Railway Infrastructure;344
8.1.6.3;6.3 Wind Turbines;345
8.1.7;7 Conclusions;351
8.1.8;References;353
8.2;On Prognostic Algorithm Design and Fundamental Precision Limits in Long-Term Prediction;357
8.2.1;1 Introduction;357
8.2.2;2 Cramér–Rao Lower Bounds;358
8.2.2.1;2.1 Bayesian Cramér–Rao Lower Bounds;359
8.2.2.2;2.2 BCRLBs for Discrete-Time Dynamical Systems;359
8.2.3;3 Methodology for Prognostic Algorithm Design;361
8.2.3.1;3.1 Conditional Predictive Bayesian Cramér–Rao Lower Bounds;363
8.2.3.2;3.2 Analytic Computation of MCP-BCRLBs;366
8.2.4;4 Case Study: End-of-Discharge Time Prognosis of Lithium-Ion Batteries;367
8.2.4.1;4.1 State-Space Model;367
8.2.4.2;4.2 Prognostic Algorithm;369
8.2.4.3;4.3 Avoiding Monte Carlo Simulations in EoD Prognostic Algorithms;370
8.2.4.4;4.4 Prognostic Algorithm Design: Known Future Operating Profiles;371
8.2.4.5;4.5 Prognostic Algorithm Design: Statistical Characterizations of Future Operating Profiles;377
8.2.5;5 Conclusions;380
8.2.6;Acronyms;380
8.2.7;References;381
8.3;Performance Degradation Monitoring and Quantification: A Wastewater Treatment Plant Case Study;382
8.3.1;1 Introduction;382
8.3.1.1;1.1 Energy Consumption on WWTPs;383
8.3.1.2;1.2 Energy Savings Through Maintenance;384
8.3.2;2 Methodology;386
8.3.3;3 Results;388
8.3.3.1;3.1 External Recirculation Pumping System;388
8.3.3.1.1;3.1.1 Experimental Setup;389
8.3.3.1.2;3.1.2 Application of the Methodology;389
8.3.3.1.3;3.1.3 Results and Discussion;390
8.3.3.2;3.2 Plant Input Pumping System;392
8.3.3.2.1;3.2.1 Experimental Setup;393
8.3.3.2.2;3.2.2 Application of the Methodology;393
8.3.3.2.3;3.2.3 Results and Discussion;394
8.3.3.3;3.3 Aeration System Blowers;396
8.3.3.3.1;3.3.1 Experimental Setup;396
8.3.3.3.2;3.3.2 Application of the Methodology;397
8.3.3.3.3;3.3.3 Results and Discussion;399
8.3.4;4 Conclusions and Future Works;400
8.3.5;References;401
8.4;Fuzzy Rule-Based Modeling for Interval-Valued Data: An Application to High and Low Stock Prices Forecasting;403
8.4.1;1 Introduction;403
8.4.2;2 Interval Arithmetic;407
8.4.3;3 Interval Fuzzy Rule-Based Model;408
8.4.3.1;3.1 Interval Participatory Learning Fuzzy Clustering with Adaptive Distances;409
8.4.3.2;3.2 Rules Consequent Parameters Identification;411
8.4.3.3;3.3 iFRB Identification Procedure;412
8.4.4;4 Computational Experiments;413
8.4.4.1;4.1 Performance Assignment;414
8.4.4.2;4.2 Empirical Results;416
8.4.5;5 Conclusion;421
8.4.6;References;422
9;Part III Diagnosis, Optimization and Control;425
9.1;Reasoning from First Principles for Self-adaptive and Autonomous Systems;426
9.1.1;1 Introduction;426
9.1.2;2 Example;428
9.1.3;3 Model-Based Reasoning;431
9.1.3.1;3.1 Model-Based Diagnosis;433
9.1.3.2;3.2 Abductive Diagnosis;438
9.1.3.3;3.3 Summary on Model-Based Reasoning for Diagnosis;441
9.1.4;4 Modeling for Diagnosis and Repair;442
9.1.5;5 Self-adaptation Using Models;447
9.1.6;6 Related Research;454
9.1.7;7 Conclusions;455
9.1.8;References;456
9.2;Decentralized Modular Approach for Fault Diagnosis of a Class of Hybrid Dynamic Systems: Application to a Multicellular Converter;460
9.2.1;1 Learning from Data Streams;460
9.2.1.1;1.1 Basic Definitions and Motivation;460
9.2.1.2;1.2 State of the Art;461
9.2.1.3;1.3 Contribution of the Proposed Approach;462
9.2.2;2 Proposed Approach;463
9.2.2.1;2.1 System Decomposition;463
9.2.2.2;2.2 Discrete Component Modeling;466
9.2.2.3;2.3 Residual Generation Based on System Continuous Dynamics;468
9.2.2.4;2.4 Enriched Local Models Building;470
9.2.2.5;2.5 Local Hybrid Diagnoser Construction;471
9.2.2.6;2.6 Equivalence Between Centralized and Decentralized Diagnosis Structures;472
9.2.2.7;2.7 Computation Complexity Analysis;474
9.2.3;3 Experimental Results;475
9.2.4;4 Conclusion;479
9.2.5;References;481
9.3;Automated Process Optimization in Manufacturing Systems Based on Static and Dynamic Prediction Models;483
9.3.1;1 Introduction;483
9.3.1.1;1.1 Our Approach;485
9.3.2;2 Problem Statement;486
9.3.2.1;2.1 Process Optimization Based on Parameters;486
9.3.2.2;2.2 Process Optimization Based on Process Values Trends;488
9.3.3;3 Establishment of Predictive Models;492
9.3.3.1;3.1 Iterative Construction of Static Predictive Mappings (Parameters Quality);492
9.3.3.1.1;3.1.1 Expert Knowledge Initialization;493
9.3.3.1.2;3.1.2 Hybrid Design of Experiments (HDoE);493
9.3.3.1.3;3.1.3 Predictive Mapping Models Construction;495
9.3.3.2;3.2 Time-Series-Based Forecast Models (Process Values Quality) Learning and Adaptation;496
9.3.3.2.1;3.2.1 Learning by a Nonlinear (Fuzzy) Version of PLS (PLS-Fuzzy);497
9.3.3.2.2;3.2.2 On-Line Model Adaptation with Increased Flexibility;500
9.3.4;4 Process Optimization with Predictive Models;504
9.3.4.1;4.1 Static Case (Mappings as Surrogates);504
9.3.4.1.1;4.1.1 Evolutionary Algorithms for Solving Many-Objective Optimization Problems;505
9.3.4.1.2;4.1.2 A New Efficient Method for Multi-Objective EA (DECMO2);506
9.3.4.2;4.2 Dynamic Case (Time-Series-Based Forecast Models as Surrogates);507
9.3.4.2.1;4.2.1 Optimization Strategies;507
9.3.4.2.2;4.2.2 Reducing Dimensionality of the Optimization Space;509
9.3.5;5 Some Results from a Chip Production Process;510
9.3.5.1;5.1 Application Scenario;510
9.3.5.2;5.2 Experimental Setup and Data Collection;511
9.3.5.3;5.3 Results;514
9.3.5.3.1;5.3.1 Static Phase (Based on Process Parameter Settings);514
9.3.5.3.2;5.3.2 Dynamic Case (Based on Time-Series of Process Values);519
9.3.6;6 Conclusion and Outlook;524
9.3.7;References;526
9.4;Distributed Chance-Constrained Model Predictive Control for Condition-Based Maintenance Planning for Railway Infrastructures;530
9.4.1;1 Introduction;530
9.4.2;2 Preliminaries;532
9.4.2.1;2.1 Hybrid and Distributed MPC;532
9.4.2.2;2.2 Chance-Constrained MPC;533
9.4.3;3 Problem Formulation;534
9.4.3.1;3.1 Deterioration Model;534
9.4.3.2;3.2 Local Chance-Constrained MPC Problem;535
9.4.3.3;3.3 Two-Stage Robust Scenario-Based Approach;536
9.4.3.4;3.4 MLD-MPC Problem;538
9.4.4;4 Distributed Optimization;538
9.4.4.1;4.1 Dantzig-Wolfe Decomposition;539
9.4.4.2;4.2 Constraint Tightening;540
9.4.5;5 Case Studies;541
9.4.5.1;5.1 Settings;541
9.4.5.2;5.2 Representative Run;543
9.4.5.3;5.3 Computational Comparisons;544
9.4.5.4;5.4 Comparison with Alternative Approaches;545
9.4.6;6 Conclusions and Future Work;547
9.4.7;Appendix;547
9.4.7.1;Parameters for Case Study;547
9.4.7.2;Cyclic Approach;548
9.4.8;References;549
10;Index;552