Muutke küpsiste eelistusi

Statistical Process Monitoring Using Advanced Data-Driven and Deep Learning Approaches: Theory and Practical Applications [Pehme köide]

4.00/5 (2 hinnangut Goodreads-ist)
(Department of Chemical Enginee), (King Abdullah University of Science and Technology, Saudi Arabia), (Baylor University, Dept of Statistical Sciences, Waco, Texas, USA), , (King Abdullah University of Science and Technology, Saudi Arabia)
  • Formaat: Paperback / softback, 328 pages, kõrgus x laius: 229x152 mm, kaal: 520 g
  • Ilmumisaeg: 04-Jul-2020
  • Kirjastus: Elsevier Science Publishing Co Inc
  • ISBN-10: 0128193654
  • ISBN-13: 9780128193655
Teised raamatud teemal:
Statistical Process Monitoring Using Advanced Data-Driven and Deep Learning Approaches: Theory and Practical ApplicationsSuurem pilt
  • Formaat: Paperback / softback, 328 pages, kõrgus x laius: 229x152 mm, kaal: 520 g
  • Ilmumisaeg: 04-Jul-2020
  • Kirjastus: Elsevier Science Publishing Co Inc
  • ISBN-10: 0128193654
  • ISBN-13: 9780128193655
Teised raamatud teemal:
Püsilink: https://www.kriso.ee/db/9780128193655.html
Märksõnad:
Statistical Process Monitoring Using Advanced Data-Driven and Deep Learning Approaches tackles multivariate challenges in process monitoring by merging the advantages of univariate and traditional multivariate techniques to enhance their performance and widen their practical applicability. The book proceeds with merging the desirable properties of shallow learning approaches – such as a one-class support vector machine and k-nearest neighbours and unsupervised deep learning approaches – to develop more sophisticated and efficient monitoring techniques.

Finally, the developed approaches are applied to monitor many processes, such as waste-water treatment plants, detection of obstacles in driving environments for autonomous robots and vehicles, robot swarm, chemical processes (continuous stirred tank reactor, plug flow rector, and distillation columns), ozone pollution, road traffic congestion, and solar photovoltaic systems.

  • Uses a data-driven based approach to fault detection and attribution
  • Provides an in-depth understanding of fault detection and attribution in complex and multivariate systems
  • Familiarises you with the most suitable data-driven based techniques including multivariate statistical techniques and deep learning-based methods
  • Includes case studies and comparison of different methods
Preface ix
Acknowledgments xi
1 Introduction
1.1 Introduction
1(5)
1.1.1 Motivation: why process monitoring
1(1)
1.1.2 Types of faults
2(2)
1.1.3 Process monitoring
4(1)
1.1.4 Physical redundancy vs analytical redundancy
5(1)
1.2 Process monitoring methods
6(7)
1.2.1 Model-based methods
7(2)
1.2.2 Knowledge-based methods
9(1)
1.2.3 Data-based monitoring methods
9(4)
1.3 Fault detection metrics
13(1)
1.4 Conclusion
14(5)
References
15(4)
2 Linear latent variable regression (LVR)-based process monitoring
2.1 Introduction
19(1)
2.2 Development of linear LVR models
20(10)
2.2.1 Full rank methods
21(1)
2.2.2 Latent variable regression (LVR) models
22(8)
2.3 Dynamic LVR models
30(2)
2.4 Process monitoring methods
32(15)
2.4.1 Univariate chart for process monitoring
32(7)
2.4.2 Distribution-based process monitoring schemes
39(5)
2.4.3 Multivariate process monitoring schemes with parametric and nonparametric thresholds
44(3)
2.5 Linear LVR-based process monitoring strategies
47(6)
2.5.1 Conventional LVR monitoring statistics
47(3)
2.5.2 Fault isolation
50(3)
2.6 Cases studies
53(10)
2.6.1 Simulated example
53(2)
2.6.2 Monitoring influent measurements at water resource recovery facilities
55(8)
2.7 Discussion
63(8)
References
63(8)
3 Fault isolation
3.1 Introduction
71(8)
3.1.1 Pitfalls of standardizing data
72(5)
3.1.2 Shortcomings of contribution plots/scores
77(2)
3.2 Fault isolation
79(20)
3.2.1 Variable thinning
79(1)
3.2.2 Iterative traditional isolation
80(3)
3.2.3 Variable selection methods
83(16)
3.3 Fault classification
99(1)
3.4 Fault isolation metrics
100(3)
3.4.1 Fault isolation errors
101(1)
3.4.2 Precision and recall
102(1)
3.4.3 Phase I FI metrics
102(1)
3.4.4 Discussion
103(1)
3.5 Case studies
103(8)
3.5.1 Retrospective fault isolation
104(4)
3.5.2 Real-time fault isolation
108(3)
3.6 Further reading
111(8)
References
112(7)
4 Nonlinear latent variable regression methods
4.1 Introduction
119(2)
4.2 Limitations of linear LVR methods for process monitoring
121(2)
4.3 Developing nonlinear LVR methods for process monitoring
123(15)
4.3.1 Nonlinear partial least squares
123(4)
4.3.2 ANFIS-PLS modeling framework
127(4)
4.3.3 Kernel PGA
131(1)
4.3.4 Kernel principal components analysis (KPCA) model
131(4)
4.3.5 KPCA-based fault detection procedures
135(3)
4.4 Cases study: monitoring WWTP
138(4)
4.4.1 Anomaly detection using KPCA-OCSVM method
139(3)
4.5 Simulated synthetic data
142(7)
4.5.1 Application of plug flow reactor
143(6)
4.6 Discussion
149(6)
References
151(4)
5 Multiscale latent variable regression-based process monitoring methods
5.1 Introduction
155(3)
5.2 Theoretical background of wavelet-based data representation
158(9)
5.2.1 Wavelet transform
159(1)
5.2.2 Multiscale representation of data using wavelets
159(5)
5.2.3 Advantages of multiscale representation
164(3)
5.3 Multiscale filtering using wavelets
167(3)
5.3.1 Single scale filter method
167(1)
5.3.2 Multiscale filtering methods
168(1)
5.3.3 Advantages of multiscale denoising
169(1)
5.4 Wavelet-based multiscale univariate monitoring techniques
170(6)
5.4.1 An illustrative example
172(4)
5.5 Multiscale LVR modeling
176(1)
5.5.1 Benefits of multiscale denoising in LVR modeling
176(1)
5.6 Multiscale LVR modeling
177(3)
5.7 Results and discussions
180(6)
5.7.1 Application with synthetic data
180(3)
5.7.2 Application of monitoring distillation column
183(3)
5.8 Discussion
186(7)
References
188(5)
6 Unsupervised deep learning-based process monitoring methods
6.1 Introduction
193(2)
6.2 Clustering
195(7)
6.2.1 Partition-based clustering techniques
196(1)
6.2.2 Hierarchy-based clustering techniques
197(1)
6.2.3 Density-based approach
198(3)
6.2.4 Expectation maximization
201(1)
6.3 One-class classification
202(4)
6.3.1 One-class SVM
202(1)
6.3.2 Support vector data description (SVDD)
203(3)
6.4 Deep learning models
206(11)
6.4.1 Autoencoders
206(4)
6.4.2 Probabilistic models
210(3)
6.4.3 Deep neural networks
213(2)
6.4.4 Deep Boltzmann machine
215(2)
6.5 Deep learning-based clustering schemes for process monitoring
217(1)
6.6 Discussion
218(7)
References
219(6)
7 Unsupervised recurrent deep learning scheme for process monitoring
7.1 Introduction
225(2)
7.2 Recurrent neural networks approach
227(8)
7.2.1 Basics of recurrent neural networks
227(2)
7.2.2 Long short-term memory
229(5)
7.2.3 Gated recurrent neural networks
234(1)
7.3 Hybrid deep models
235(6)
7.3.1 RNN-RBM
236(1)
7.3.2 RNN-RBM method
237(1)
7.3.3 LSTM-RBM model
238(1)
7.3.4 LSTM-DBN
239(2)
7.4 Recurrent deep learning-based process monitoring
241(3)
7.4.1 Residuals-based process monitoring approaches
242(1)
7.4.2 Recurrent deep learning-based clustering schemes for process monitoring
243(1)
7.5 Applications: monitoring influent conditions at WWTP
244(6)
7.6 Discussion
250(5)
References
251(4)
8 Case studies
8.1 Introduction
255(3)
8.2 Stereovision
258(16)
8.2.1 Deep stacked autoencoder-based KNN approach
261(5)
8.2.2 Data description
266(1)
8.2.3 Results and discussion
266(1)
8.2.4 Model trained using data with no obstacles
267(2)
8.2.5 Evaluation of performance for busy scenes
269(2)
8.2.6 Obstacle detection using the Bahnhof dataset
271(3)
8.3 Detecting abnormal ozone measurements using deep learning
274(14)
8.3.1 Introduction
274(2)
8.3.2 Data description
276(2)
8.3.3 Ozone monitoring based on deep learning approaches
278(6)
8.3.4 Detection results
284(4)
8.4 Monitoring of a wastewater treatment plant using deep learning
288(20)
8.4.1 Introduction
288(2)
8.4.2 Proposed DBN-based kNN, OCSVM, and k-means algorithms
290(1)
8.4.3 Real data application: monitoring a decentralized wastewater treatment plant in Golden, CO, USA
291(6)
8.4.4 Conclusion
297(1)
References
297(11)
9 Conclusion and further research directions
References
308(3)
Index 311
Fouzi Harrou received the M.Sc. degree in telecommunications and networking from the University of Paris VI, France, and the Ph.D. degree in systems optimization and security from the University of Technology of Troyes (UTT), France. He was an Assistant Professor with UTT for one year and with the Institute of Automotive and Transport Engineering, Nevers, France, for one year. He was also a Postdoctoral Research Associate with the Systems Modeling and Dependability Laboratory, UTT, for one year. He was a Research Scientist with the Chemical Engineering Department, Texas A&M University at Qatar, Doha, Qatar, for three years. He is actually a Research Scientist with the Division of Computer, Electrical and Mathematical Sciences and Engineering, King Abdullah University of Science and Technology. He is the author of more than 150 refereed journals and conference publications and book chapters. He is co-author of the book "Statistical Process Monitoring Using Advanced Data-Driven and Deep Learning Approaches: Theory and Practical Applications" (Elsevier, 2020). Dr. Harrous research interests are in the area of statistical anomaly detection and process monitoring with a particular emphasis on data-driven, machine learning/deep learning methods. The algorithms developed in Dr. Harrous research are utilized in many applications to improve the operation of various environmental, chemical, and electrical systems. Professor Ying Sun received her Ph.D. in Statistics from Texas A&M in 2011 followed by a two-year postdoctoral research position at the Statistical and Applied Mathematical Sciences Institute and at the University of Chicago. She was an Assistant Professor at the Ohio State University for a year before joining KAUST in 2014. At KAUST, Professor Sun established and leads the Environmental Statistics research group which works on developing statistical models and methods for complex data to address important environmental problems. She has made original contributions to environmental statistics, in particular in the areas of spatio-temporal statistics, functional data analysis, visualization, computational statistics, with an exceptionally broad array of applications. Professor Sun won two prestigious awards: the Early Investigator Award in Environmental Statistics presented by the American Statistical Association, and the Abdel El-Shaarawi Young Research Award from the International Environmetrics Society Professor Amanda Hering obtained her Ph.D. from Texas A&M University in Statistics in 2009. She joined the Department of Applied Mathematics and Statistics at Colorado School of Mines in Golden, Colorado in 2009 as an Assistant Professor and was promoted to Associate Professor in 2016. She joined the Department of Statistical Science at Baylor University in the fall of 2016 as an Associate Professor. Her research interests are in modeling big, multivariate, spatial datasets; developing methods for categorical spatial data; and detecting outliers and faults for process and data control. She works with researchers whose data structures generate new statistical methodologies because either the goals or the size of the data presents a new challenge. She is an Associate Editor of Technometrics, Environmetrics, and Stat. She received the American Statistical Associations Section on Statistics in the Environment Early Investigator Award in 2017. Muddu Madakyaru is an Associate professor of Department of Chemical Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, India. He received B.E. degree in Chemical Engineering and M.Tech. in Chemical plant Design from the R.V.C.E and National Institute of Technology Karnataka, India respectively. In the year 2010 he obtained his Ph.D degree in process control from Indian Institute of Technology, Bombay, India. Later he was involved in post-doctoral research at Texas A&M University, Doha, Qatar for four years. His research interests are in advanced process control, including, system identification, Fault detection and diagnosis, model predictive control and latent variable regression modeling using wavelets. He has published more than 20 papers in peer reviewed journals and 10 international conference proceedings papers. He is fellow of Institution of Engineers (India), Life Member of Indian Society for Technical Education and Indian Society of Systems for Science and Engineering (ISSE). Dr. Abdelkader Dairi received the Engineer degree in computer science from the University of Oran 1 Ahmed Ben Bella, Algeria, in 2003. He also received the Magister degree in computer science from the National Polytechnic School of Oran, Algeria, in 2006. From 2007 to 2013 he was a senior Oracle database administrator (DBA) and enterprise resource planning (ERP) manager. He has over 20 years of programming experience in different languages and environments. In 2018 he received the Ph.D. degree in computer sciences from Ben Bella Oran1 University. His research interests include deep learning approach for autonomous robot navigation, computer vision, image processing, and mobile robotics.