Data Mining and Knowledge Discovery for Process Monitoring and Control

摘要: 1 Introduction.- 1.1 Current Approaches to Process Monitoring, Diagnosis and Control.- 1.2 Monitoring Charts for Statistical Quality 1.3 The Operating Window.- 1.4 State Space Based 1.5 Characteristics of Operational Data.- 1.6 System Requirement Architecture.- 1.7 Outline the Book.- 2 Data Mining Knowledge Discovery - an Overview.- 2.1 Definition Development.- 2.2 KDD Process.- 2.3 Techniques.- 2.4 Feature Selection with Mining.- 2.5 Final Remarks Additional Resources.- 3 Pre-processing Extraction, Dimension Reduction Concept Formation.- 3.1 Pre-processing.- 3.2 Use Principal Component Analysis.- 3.3 Wavelet 3.4 Episode Approach.- 3.5 Summary.- 4 Multivariate Analysis 4.1 PCA Identification Monitoring.- 4.2 Partial Least Squares (PLS).- 4.3 Variable Contribution Plots.- 4.4 Multiblock PLS.- 4.5 Batch Using Multiway PCA.- 4.6 Nonlinear 4.7 Strategy Development Product Design Industrial Case Study.- 4.8 General Observations.- 5 Supervised Learning Support.- 5.1 Feedforward Neural Networks.- 5.2 Extraction FFNN Inputs.- 5.3 Model Validation Confidence Bounds.- 5.4 Application Fault Diagnosis.- 5.5 Fuzzy 5.6 Set Covering Method.- 5.7 Signed Digraphs.- 5.8 Studies.- 5.9 6 Unsupervised Identification.- 6.1 vs. Learning.- 6.2 Adaptive Resonance Theory.- 6.3 A Framework Integrating ART2.- 6.4 ARTnet FCC 6.5 Bayesian Automatic Classification.- 6.6 AutoClass 6.7 Comments.- 7 Inductive Conceptual Clustering Real-time 7.1 7.2 IL from Averaged 7.3 7.4 Refinery MTBE 7.5 Review.- 8 Rules 8.1 Generation Operation.- 8.2 8.3 Rough 8.4 Network Method Extraction.- 8.5 Discussion.- 9 Inferential Models Software Sensors.- 9.1 Networks as 9.2 Training / Test Retraining.- 9.3 An 9.4 Input Variables.- 9.5 Dynamic Models.- 9.6 10 Concluding Remarks.- Appendix Continuous Stirred Tank Reactor (CSTR).- B Residue Fluid Catalytic Cracking (R-FCC) C Methyl Tertiary Butyl Ether (MTBE) References.