Chemical Process Performance Evaluation

by Cinar; Ali

• ISBN13: 9780849338069
• ISBN10: 0849338069
• Format: Hardcover
• Copyright: 2007-01-11
• Publisher: CRC Press

Summary

The latest advances in process monitoring, data analysis, and control systems are increasingly useful for maintaining the safety, flexibility, and environmental compliance of industrial manufacturing operations.Focusing on continuous, multivariate processes, Chemical Process Performance Evaluation introduces statistical methods and modeling techniques for process monitoring, performance evaluation, and fault diagnosis.This book introduces practical multivariate statistical methods and empirical modeling development techniques, such as principal components regression, partial least squares regression, input-output modeling, state-space modeling, and modeling process signals for trend analysis. Then the authors examine fault diagnosis techniques based on episodes, hidden Markov models, contribution plots, discriminant analysis, and support vector machines. They address controller process evaluation and sensor failure detection, including methods for differentiating between sensor failures and process upset. The book concludes with an extensive discussion on the use of data analysis techniques for the special case of web and sheet processes. Case studies illustrate the implementation of methods presented throughout the book.Emphasizing the balance between practice and theory, Chemical Process Performance Evaluation is an excellent tool for comparing alternative techniques for process monitoring, signal modeling, and process diagnosis. The unique integration of process and controller monitoring and fault diagnosis facilitates the practical implementation of unified and automated monitoring and diagnosis technologies.

Table of Contents

Nomenclature
Introduction	p. 1
Motivation and Historical Perspective	p. 2
Outline	p. 4
Univariate Statistical Monitoring Techniques	p. 7
Statistics Concepts	p. 8
Univariate SPM Techniques	p. 11
Shewhart Control Charts	p. 11
Cumulative Sum (CUSUM) Charts	p. 18
Moving Average Monitoring Charts for Individual Measurements	p. 19
Exponentially Weighted Moving Average Chart	p. 22
Monitoring Tools for Autocorreleated Data	p. 22
Monitoring with Charts of Residuals	p. 26
Monitoring with Detecting Changes in Model Parameters	p. 27
Limitations of Univariate Monitoring Techniques	p. 32
Summary	p. 35
Multivariate Statistical Monitoring Techniques	p. 37
Principal Components Analysis	p. 37
Canonical Variates Analysis	p. 43
Independent Component Analysis	p. 43
Contribution Plots	p. 46
Linear Methods for Diagnosis	p. 48
Clustering	p. 48
Discriminant Analysis	p. 50
Fisher's Discriminant Analysis	p. 53
Nonlinear Methods for Diagnosis	p. 58
Neural Networks	p. 58
Kernel-Based Techniques	p. 64
Support Vector Machines	p. 66
Summary	p. 69
Empirical Model Development	p. 73
Regression Models	p. 75
PCA Models	p. 78
PLS Regression Models	p. 79
Input-Output Models of Dynamic Processes	p. 83
State-Space Models	p. 89
Summary	p. 97
Monitoring of Multivariate Processes	p. 99
SPM Methods Based on PCA	p. 100
SPM Methods Based on PLS	p. 105
SPM Using Dynamic Process Models	p. 108
Other MSPM Techniques	p. 112
Summary	p. 114
Characterization of Process Signals	p. 115
Wavelets	p. 115
Fourier Transform	p. 116
Continuous Wavelet Transform	p. 119
Discrete Wavelet Transform	p. 123
Filtering and Outlier Detection	p. 127
Simple Filters	p. 128
Wavelet Filters	p. 131
Robust Filter	p. 133
Signal Representation by Fuzzy Triangular Episodes	p. 135
Development of Markovian Models	p. 138
Markov Chains	p. 139
Hidden Markov Models	p. 141
Wavelet-Domain Hidden Markov Models	p. 145
Summary	p. 147
Process Fault Diagnosis	p. 149
Fault Diagnosis Using Triangular Episodes and HMMs	p. 149
CSTR Simulation	p. 152
Vacuum Column	p. 155
Fault Diagnosis Using Wavelet-Domain HMMs	p. 157
pH Neutralization Simulation	p. 161
CSTR Simulation	p. 164
Fault Diagnosis Using HMMs	p. 166
Case Study of HTST Pasteurization Process	p. 167
Fault Diagnosis Using Contribution Plots	p. 174
Fault Diagnosis with Statistical Methods	p. 179
Fault Diagnosis Using SVM	p. 191
Fault Diagnosis with Robust Techniques	p. 192
Robust Monitoring Strategy	p. 192
Pilot-Scale Distillation Column	p. 198
Summary	p. 202
Sensor Failure Detection and Diagnosis	p. 203
Sensor FDD Using PLS and CVSS Models	p. 204
Real-Time Sensor FDD Using PCA-Based Techniques	p. 215
Methodology	p. 218
Case Study	p. 224
Summary	p. 230
Controller Performance Monitoring	p. 231
Single-Loop Controller Performance Monitoring	p. 233
Multivariable Controller Performance Monitoring	p. 237
CPM for MPC	p. 238
Summary	p. 248
Web and Sheet Processes	p. 251
Traditional Data Analysis	p. 252
MD/CD Decomposition	p. 252
Time Dependent Structure of Profile Data	p. 256
Orthogonal Decomposition of Profile Data	p. 257
Gram Polynomials	p. 259
Principal Components Analysis	p. 262
Flatness of Scanner Data	p. 264
Controller Performance	p. 268
MD Control Performance	p. 269
Model-Based CD Control Performance	p. 271
Summary	p. 274
Bibliography	p. 277
Index	p. 305
Table of Contents provided by Ingram. All Rights Reserved.