Korbicz / Koscielny Modeling, Diagnostics and Process Control

Implementation in the DiaSter System
1. Auflage 2010
ISBN: 978-3-642-16653-2
Verlag: Springer
Format: PDF
Kopierschutz: 1 - PDF Watermark

E-Book, Englisch, 384 Seiten

Modern control systems are complex in the sense of implementing numerous functions, such as process variable processing, digital control, process monitoring and alarm indication, graphic visualization of process running, or data exchange with other systems or databases. This book conveys a description of the developed DiaSter system as well as characteristics of advanced original methods of modeling, knowledge discovery, simulator construction, process diagnosis, as well as predictive and supervision control applied in the system. The system allows early recognition of abnormal states of industrial processes along with faults or malfunctions of actuators as well as technological and measuring units. The universality of solutions implemented in DiaSter facilitates its broad application, for example, in the power, chemical, pharmaceutical, metallurgical and food industries. The system is a world-scale unique solution, and due to its open architecture it can be connected practically with any other control systems. The monograph presents theoretical and practical results of research into fault diagnosis and control conducted over many years within the cooperation of Polish research teams from the Warsaw University of Technology, the University of Zielona Góra, the Silesian University of Technology in Gliwice, and the Technical University of Rzeszów. The book will be of great interest to researchers and advanced students in automatic control, technical diagnostics and computer engineering, and to engineers tasked with the development of advanced control systems of complex industrial processes.

Autoren/Hrsg.

Korbicz, Józef

Koscielny, Jan M.

Inhaltsverzeichnis

1;Preface;5
2;Contents;8
3;Introduction;14
3.1;Control System Structures;14
3.2;Trends in the Development of Modern Automatic Control Systems;18
3.3;New Functions of Advanced Automatic Control Systems;20
4;Introduction to the DiaSter System;27
4.1;Introduction;27
4.2;System Structure and Tasks;27
4.2.1;Main Uses of the System;27
4.2.2;System Functionality;29
4.2.3;System Structure;37
4.3;Software Platform;39
4.3.1;Information Model and the System Configuration;41
4.3.2;Central Archival Database and User Databases;44
4.3.3;Data Exchange;49
4.3.4;Modeling Module;51
4.3.5;On-line Calculation Module;56
4.3.6;Visualization Module;60
5;Process Modeling;66
5.1;Introduction;66
5.2;Analytical Models and Modeling;68
5.2.1;Basic Relations for the Description of Balance Dependencies for Modeled Physical Processes;69
5.2.2;Integration Methods and Integration Step Selection for Simulation;73
5.2.3;Pneumatic Cylinder Controlled by a Servo-Valve: A Balance Model of the System;75
5.2.4;Pneumatic Cylinder Controlled by a Servo-Valve: A Block Model of the System;81
5.3;Linear Models: Local Approximation of Dynamic Properties;83
5.3.1;Dynamic Model Linearization;83
5.3.2;Pneumatic Cylinder Controlled by a Servo-Valve: A Linear Model of the System;85
5.3.3;Pneumatic Cylinder Controlled by a Servo-Valve: An Optimized Linear Model of the System;89
5.4;Parametric Models;93
5.4.1;Discrete Linear Parametric Models;94
5.4.2;Identification of the Coefficients of Parametric Models;97
5.4.3;Pneumatic Cylinder Controlled by a Servo-Valve: A Parametric Linear Model of the System;100
5.5;Fuzzy Parametric Models;103
5.5.1;Fuzzy Parametric TSK Models;103
5.5.2;Estimation of Fuzzy TSK Model Coefficients;105
5.5.3;Pneumatic Cylinder Controlled by a Servo-Valve: A TSK Fuzzy Model;107
5.6;Neural Models;110
5.7;Neural Networks with External Dynamics;111
5.7.1;Recurrent Networks;112
5.7.2;State Space Neural Networks;114
5.7.3;Locally Recurrent Networks;115
5.7.4;GMDH Neural Networks;122
5.7.5;Implementation of Neural Models in the DiaSter System;129
6;Knowledge Discovery in Databases;130
6.1;Introduction;130
6.2;Selection of Input Variables of Models;132
6.2.1;Correlation-Based Feature Selection;133
6.2.2;Measures Based on Correlation;134
6.2.3;Searching through the Feature Space;135
6.3;Discovery of Qualitative Dependencies;136
6.4;Discovery of Quantitative Dependencies;139
6.4.1;Support Vector Machines;139
6.4.2;Methods Involving Case-Based Reasoning;145
6.5;Conclusion;163
7;Diagnostic Methods;164
7.1;Introduction;164
7.2;Specificity of the Diagnostics of Industrial Processes;165
7.3;Fault Detection Methods;166
7.4;Robust Fault Diagnosis;171
7.4.1;Robust Neural Model: The Passive Approach;172
7.4.2;Fuzzy Adaptive Threshold: The Passive Approach;175
7.4.3;Robust Dynamic Model: The Active Approach;177
7.4.4;Robust Model Design Examples;180
7.4.5;Implementation of Neural Models in the DiaSter System;186
7.5;Process Fault Isolation with the Use of Fuzzy Logic;190
7.5.1;Forms of Diagnostic Relation Notation;190
7.5.2;Reasoning Algorithm for Single and Multiple Faults;195
7.5.3;Algorithms of Reasoning in a Hierarchical Structure;206
7.6;Application of Belief Networks in Technical Diagnostics;217
7.6.1;Introduction;218
7.6.2;Belief-Network-Based Diagnostic Model;221
7.6.3;Input Data Images;224
7.6.4;Additional Variables and Opportunities for Their Adjustment;230
7.6.5;Belief Networks;233
7.6.6;Model Identification and Tuning;240
7.6.7;Implementation in the DiaSter Environment;242
8;Supervisory Control and Optimization;243
8.1;Predictive Control and Process Set-Point Optimization;244
8.1.1;Principle of Model-Based Predictive Control;245
8.1.2;Dynamic Matrix Control Algorithm;250
8.1.3;Generalized Predictive Control Algorithm;257
8.1.4;Non-linear Predictive Control;260
8.1.5;Optimization of Set-Points;266
8.1.6;Examples;269
8.2;Self-tuning and Adaptation of Control Loops;277
8.2.1;Step Response Method;277
8.2.2;Relay Self-tuning;286
8.2.3;Loop Adaptation;292
8.2.4;Function Blocks;301
9;Application of the DiaSter System;304
9.1;Introduction;304
9.2;System of Automatic Control and Diagnostics;305
9.3;Process Information Model in the DiaSter Platform;307
9.4;Applications of the DiaSter System Packages;311
9.4.1;Process Simulator;312
9.4.2;Self-tuning: Selection of PID Settings;317
9.4.3;Reconstructing Process Variables with TSK Models;322
9.4.4;Process Modeling with Neural Networks;328
9.4.5;Incipient Fault Tracking;335
9.4.6;On-Line Diagnostics with Fuzzy Reasoning;337
9.4.7;Belief Networks in a Diagnostic System;347
9.4.8;Knowledge Discovery in Databases;358
9.4.9;Model Predictive Control with Constraints and Faulty Conditions;371
10;References;377
11;Index;389