Skip to main content
Disciplines
Automotive Business IT + Informatics Construction + Real Estate Electrical Engineering + Electronics Energy + Sustainability Insurance + Risk Finance + Banking Management + Leadership Marketing + Sales Mechanical Engineering + Materials
Events
DE
EN
Books
Journals
Topic Page
Marketing
Start single access now
Access for companies
EXTENDED SEARCH
Log in
Top

2018 | Book

Introduction Read chapter Read first chapter

Dynamic Modeling of Complex Industrial Processes: Data-driven Methods and Application Research

Author: Chao Shang

Publisher: Springer Singapore

Book Series : Springer Theses

Part of: Springer Professional "Wirtschaft+Technik" , Springer Professional "Technik"

Login to get access
insite
SEARCH

About this book

This thesis develops a systematic, data-based dynamic modeling framework for industrial processes in keeping with the slowness principle. Using said framework as a point of departure, it then proposes novel strategies for dealing with control monitoring and quality prediction problems in industrial production contexts.

The thesis reveals the slowly varying nature of industrial production processes under feedback control, and integrates it with process data analytics to offer powerful prior knowledge that gives rise to statistical methods tailored to industrial data. It addresses several issues of immediate interest in industrial practice, including process monitoring, control performance assessment and diagnosis, monitoring system design, and product quality prediction. In particular, it proposes a holistic and pragmatic design framework for industrial monitoring systems, which delivers effective elimination of false alarms, as well as intelligent self-running by fully utilizing the information underlying the data. One of the strengths of this thesis is its integration of insights from statistics, machine learning, control theory and engineering to provide a new scheme for industrial process modeling in the era of big data.

Table of Contents

Frontmatter
Chapter 1. Introduction
Abstract
This chapter provides an overview of the entire thesis. First, the research background of process data analytics for modeling, monitoring and fault diagnosis in industrial processes is introduced. Then an extensive review of existing research on multivariate statistical process monitoring and soft sensor modeling is given. Third, the opportunities and challenges in advancing industrial applications of process data analytics are highlighted. Finally, the main contents and the layout of this thesis are provided.
Chao Shang
Chapter 2. Monitoring of Operating Condition and Process Dynamics with Slow Feature Analysis
Abstract
Latent variable (LV) models have been widely used in multivariate statistical process monitoring. However, whatever deviation from nominal operating condition is detected, an alarm is triggered based on classical monitoring methods. Consequently, they cannot distinguish real faults incurring dynamics anomalies from normal deviations in operating conditions. In this chapter, a new process monitoring strategy based on slow feature analysis (SFA) is proposed for the concurrent monitoring of operating point deviations and process dynamics anomalies. Slow features as LVs are developed to describe slowly varying dynamics, yielding improved physical interpretation. In addition to classical statistics for monitoring deviation from design conditions, two novel indices are proposed to detect anomalies in process dynamics through the slowness of LVs. The proposed approach can distinguish whether normal changes in operating conditions or real faults occur. Two case studies show the validity of the SFA-based process monitoring approach.
Chao Shang
Chapter 3. Control Performance Monitoring and Diagnosis Based on SFA and Contribution Plot
Abstract
In this chapter, we further investigate the implication of SFA in control performance monitoring and diagnosis. It will be clarified that the process dynamics described by SFA essentially is related to control performance. Then we propose a new data-driven control performance monitoring approach based on SFA. In addition, by employing the contribution plot techniques, we further develop a new control performance diagnosis method to locate potential root cause variables that make dominant contributions to control performance changes, which is instructive for making further maintenance actions. In comparison with existing methods, the proposed method enables real-time diagnosis of control performance changes, and can make use of all kinds of process variables. Its efficacy is validated based on a simulated example and the Tennessee Eastman benchmark process.
Chao Shang
Chapter 4. Recursive SFA Algorithm and Adaptive Monitoring System Design
Abstract
The common occurrence of slow drifts in industrial processes ask for the need of adaptive monitoring. In this chapter, a recursive slow feature analysis algorithm for adaptive process monitoring is developed to handle time-varying processes. An algebraic property of slow feature analysis is first revealed. We then show that such a property may be violated in the presence of online updating, and we give an effective remedy. A novel algorithm based on rank-one modification and orthogonal iteration procedure is developed to recursively adjust the solution to the generalized eigenvalue problem, model parameters, and associated monitoring statistics conveniently. In addition, an improved stopping principle for model updating is proposed based on statistics related to process dynamics, which provides an intelligent maintenance mechanism of monitoring systems. The efficacy of the proposed schema is finally evaluated on a industrial crude heating furnace system.
Chao Shang
Chapter 5. Probabilistic Slow Feature Regression for Dynamic Soft Sensing
Abstract
Latent variable (LV) models provide explicit representations of underlying driving forces of process variations and retain the dominant information of process data. In this chapter, we develop a new soft sensor model called probabilistic slow feature regression (PSFR). Slow features as temporally correlated LVs are first derived using probabilistic slow feature analysis (PSFA). Probabilistic slow features that evolve in a state-space form effectively represent nominal variations of processes, some of which may be potentially correlated to quality variables and hence help improving the prediction performance of soft sensors when used as inputs. An efficient expectation maximization algorithm is proposed to estimate parameters of the PSFA model, and two alternative criteria are put forward to select quality-relevant SFs in the PSFR model. The validity and advantages of the proposed method are demonstrated via two case studies.
Chao Shang
Chapter 6. Enhanced Dynamic PLS with Temporal Smoothness for Soft Sensing
Abstract
Traditional data-driven static soft sensors only utilize single snapshots of process samples for quality prediction, thereby falling short of addressing process dynamics appropriately. Hence, a series of limitations in practice are incurred, such as sensitivity to temporal noises and inadequate descriptions to process dynamics. Because of these concerns, static models have been extended to dynamic counterparts such dynamic partial least squares (DPLS) with lagged input measurements to describe process dynamics. However, the dimension of model inputs may be remarkably larger than those of static models, thereby resulting in the over-fitting phenomenon. In this chapter, we seek to enhance DPLS-based dynamic soft sensor modeling approach by imposing temporal smoothness on model parameters. It assumes that historical measurements exert smoothly varying influence on latent variables in DPLS, which is used as a valid prior knowledge. In this manner, abrupt changes in model dynamics are properly penalized and the DPLS-based soft sensors enjoy better generalizations and interpretations. A numerical example and the Tennessee Eastman process case study are provided to show the feasibility as well as effectiveness of our method.
Chao Shang
Chapter 7. Nonlinear Dynamic Soft Sensing Based on Bayesian Inference
Abstract
In this chapter, we put forward a new nonlinear dynamic soft sensing model based on finite impulse response (FIR) and support vector machine (SVM). The whole model has a Wiener structure, in which nonlinearity and dynamics are described separately. In addition, a novel four-level Bayesian framework is developed to probabilistically illustrate and iteratively optimize the proposed model, which helps alleviating the over-fitting phenomenon automatically. Case studies based on a numerical example and an industrial application in propylene melt index prediction are presented to demonstrate the advantages of the proposed method over classic dynamic soft sensing models.
Chao Shang
Chapter 8. Conclusions and Recommendations
Abstract
This chapter summarizes the whole work of this thesis, and some challenges on data-driven modeling methodologies and industrial applications are pointed out as the future work.
Chao Shang
Metadata
Title
Dynamic Modeling of Complex Industrial Processes: Data-driven Methods and Application Research
Author
Chao Shang
Copyright Year
2018
Publisher
Springer Singapore
Electronic ISBN
978-981-10-6677-1
Print ISBN
978-981-10-6676-4
DOI
https://doi.org/10.1007/978-981-10-6677-1