Data-Driven Fault Detection and Reasoning for Industrial Monitoring.
Saved in:
| Superior document: | Intelligent Control and Learning Systems Series ; v.3 |
|---|---|
| : | |
| TeilnehmendeR: | |
| Place / Publishing House: | Singapore : : Springer,, 2022. Ã2022. |
| Year of Publication: | 2022 |
| Edition: | 1st ed. |
| Language: | English |
| Series: | Intelligent Control and Learning Systems Series
|
| Online Access: | |
| Physical Description: | 1 online resource (277 pages) |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Table of Contents:
- Intro
- Preface
- Contents
- Abbreviations
- 1 Background
- 1.1 Introduction
- 1.1.1 Process Monitoring Method
- 1.1.2 Statistical Process Monitoring
- 1.2 Fault Detection Index
- 1.2.1 T2 Statistic
- 1.2.2 Squared Prediction Error
- 1.2.3 Mahalanobis Distance
- 1.2.4 Combined Indices
- 1.2.5 Control Limits in Non-Gaussian Distribution
- References
- 2 Multivariate Statistics in Single Observation Space
- 2.1 Principal Component Analysis
- 2.1.1 Mathematical Principle of PCA
- 2.1.2 PCA Component Extraction Algorithm
- 2.1.3 PCA Base Fault Detection
- 2.2 Fisher Discriminant Analysis
- 2.2.1 Principle of FDA
- 2.2.2 Comparison of FDA and PCA
- References
- 3 Multivariate Statistics Between Two-Observation Spaces
- 3.1 Canonical Correlation Analysis
- 3.1.1 Mathematical Principle of CCA
- 3.1.2 Eigenvalue Decomposition of CCA Algorithm
- 3.1.3 SVD Solution of CCA Algorithm
- 3.1.4 CCA-Based Fault Detection
- 3.2 Partial Least Squares
- 3.2.1 Fundamental of PLS
- 3.2.2 PLS Algorithm
- 3.2.3 Cross-Validation Test
- References
- 4 Simulation Platform for Fault Diagnosis
- 4.1 Tennessee Eastman Process
- 4.2 Fed-Batch Penicillin Fermentation Process
- 4.3 Fault Detection Based on PCA, CCA, and PLS
- 4.4 Fault Classification Based on FDA
- 4.5 Conclusions
- References
- 5 Soft-Transition Sub-PCA Monitoring of Batch Processes
- 5.1 What Is Phase-Based Sub-PCA
- 5.2 SVDD-Based Soft-Transition Sub-PCA
- 5.2.1 Rough Stage-Division Based on Extended Loading Matrix
- 5.2.2 Detailed Stage-Division Based on SVDD
- 5.2.3 PCA Modeling for Transition Stage
- 5.2.4 Monitoring Procedure of Soft-Transition Sub-PCA
- 5.3 Case Study
- 5.3.1 Stage Identification and Modeling
- 5.3.2 Monitoring of Normal Batch
- 5.3.3 Monitoring of Fault Batch
- 5.4 Conclusions
- References.
- 6 Statistics Decomposition and Monitoring in Original Variable Space
- 6.1 Two Statistics Decomposition
- 6.1.1 T2 Statistic Decomposition
- 6.1.2 SPE Statistic Decomposition
- 6.1.3 Fault Diagnosis in Original Variable Space
- 6.2 Combined Index-Based Fault Diagnosis
- 6.2.1 Combined Index Design
- 6.2.2 Control Limit of Combined Index
- 6.3 Case Study
- 6.3.1 Variable Monitoring via Two Statistics Decomposition
- 6.3.2 Combined Index-Based Monitoring
- 6.3.3 Comparative Analysis
- 6.4 Conclusions
- References
- 7 Kernel Fisher Envelope Surface for Pattern Recognition
- 7.1 Process Monitoring Based on Kernel Fisher Envelope Analysis
- 7.1.1 Kernel Fisher Envelope Surface
- 7.1.2 Detection Indicator
- 7.1.3 KFES-PCA-Based Synthetic Diagnosis in Batch Process
- 7.2 Simulation Experiment Based on KFES-PCA
- 7.2.1 Diagnostic Effect on Existing Fault Types
- 7.2.2 Diagnostic Effect on Unknown Fault Types
- 7.3 Conclusions
- References
- 8 Fault Identification Based on Local Feature Correlation
- 8.1 Fault Identification Based on Kernel Discriminant Exponent Analysis
- 8.1.1 Methodology of KEDA
- 8.1.2 Simulation Experiment
- 8.2 Fault Identification Based on LLE and EDA
- 8.2.1 Local Linear Exponential Discriminant Analysis
- 8.2.2 Neighborhood-Preserving Embedding Discriminant Analysis
- 8.2.3 Fault Identification Based on LLEDA and NPEDA
- 8.2.4 Simulation Experiment
- 8.3 Cluster-LLEDA-Based Hybrid Fault Monitoring
- 8.3.1 Hybrid Monitoring Strategy
- 8.3.2 Simulation Study
- 8.4 Conclusion
- Reference
- 9 Global Plus Local Projection to Latent Structures
- 9.1 Fusion Motivation of Global Structure and Local Structure
- 9.2 Mathematical Description of Dimensionality Reduction
- 9.2.1 PLS Optimization Objective
- 9.2.2 LPP and PCA Optimization Objectives
- 9.3 Introduction to the GLPLS.
- 9.4 Basic Principles of GPLPLS
- 9.4.1 The GPLPLS Model
- 9.4.2 Relationship Between GPLPLS Models
- 9.4.3 Principal Components of the GPLPLS Model
- 9.5 GPLPLS-Based Quality Monitoring
- 9.5.1 Process and Quality Monitoring Based on GPLPLS
- 9.5.2 Posterior Monitoring and Evaluation
- 9.6 TE Process Simulation Analysis
- 9.6.1 Model and Discussion
- 9.6.2 Fault Diagnosis Analysis
- 9.6.3 Comparison of Different GPLPLS Models
- 9.7 Conclusions
- References
- 10 Locality-Preserving Partial Least Squares Regression
- 10.1 The Relationship Among PCA, PLS, and LPP
- 10.2 LPPLS Models and LPPLS-Based Fault Detection
- 10.2.1 The LPPLS Models
- 10.2.2 LPPLS for Process and Quality Monitoring
- 10.2.3 Locality-Preserving Capacity Analysis
- 10.3 Case Study
- 10.3.1 PLS, GLPLS and LPPLS Models
- 10.3.2 Quality Monitoring Analysis
- 10.4 Conclusions
- References
- 11 Locally Linear Embedding Orthogonal Projection to Latent Structure
- 11.1 Comparison of GPLPLS, LPPLS, and LLEPLS
- 11.2 A Brief Review of the LLE Method
- 11.3 LLEPLS Models and LLEPLS-Based Fault Detection
- 11.3.1 LLEPLS Models
- 11.3.2 LLEPLS for Process and Quality Monitoring
- 11.4 LLEOPLS Models and LLEOPLS-Based Fault Detection
- 11.5 Case Study
- 11.5.1 Models and Discussion
- 11.5.2 Fault Detection Analysis
- 11.6 Conclusions
- References
- 12 New Robust Projection to Latent Structure
- 12.1 Motivation of Robust L1-PLS
- 12.2 Introduction to RSPCA Method
- 12.3 Basic Principle of L1-PLS
- 12.4 L1-PLS-Based Process Monitoring
- 12.5 TE Simulation Analysis
- 12.5.1 Robustness of Principal Components
- 12.5.2 Robustness of Prediction and Monitoring Performance
- 12.6 Conclusions
- References
- 13 Bayesian Causal Network for Discrete Variables
- 13.1 Construction of Bayesian Causal Network
- 13.1.1 Description of Bayesian Network.
- 13.1.2 Establishing Multivariate Causal Structure
- 13.1.3 Network Parameter Learning
- 13.2 BCN-Based Fault Detection and Inference
- 13.3 Case Study
- 13.3.1 Public Data Sets Experiment
- 13.3.2 TE Process Experiment
- 13.4 Conclusions
- References
- 14 Probabilistic Graphical Model for Continuous Variables
- 14.1 Construction of Probabilistic Graphical Model
- 14.1.1 Multivariate Casual Structure Learning
- 14.1.2 Probability Density Estimation
- 14.1.3 Evaluation Index of Estimation Quality
- 14.2 Dynamic Threshold for the Fault Detection
- 14.3 Forward Fault Diagnosis and Reverse Reasoning
- 14.4 Case Study: Application to TEP
- 14.5 Conclusions
- References.