Zhihuan Song

Statistical process monitoring using improved PCA with optimized sensor locations

Abstract The emphasis of most PCA process monitoring approaches is mainly on procedures to perform fault detection and diagnosis given a set of sensors. Little attention is paid to the actual sensor locations to efficiently perform these tasks. In this paper, graph-based techniques are used to optimize sensor locations to ensure the observability of faults, as well as the fault resolution to a maximum possible extent. Meanwhile, an improved PCA that uses two new statistics of PVR and CVR to replace the Q index in conventional PCA is introduced. The improved PCA can efficiently detect weak process changes, and give an insight to the root cause about the process malfunction. Simulation results of a CSTR process show that the improved PCA with optimized sensor locations is superior to conventional methods in fault resolution and sensibility.

HMM-Driven Robust Probabilistic Principal Component Analyzer for Dynamic Process Fault Classification

In this paper, a novel hidden Markov model (HMM)-driven robust latent variable model (LVM) is proposed for fault classification in dynamic industrial processes. A robust probabilistic model with Students t mixture output is designed for tolerating outliers. Based on the robust LVM, the probabilistic structure is further developed into a classifier form so as to incorporate various types of process information during model acquisition. After that, the robust probabilistic classifier is extended within the HMM framework so as to characterize the time-domain stochastic uncertainties. The model parameters are derived through the expectation-maximization algorithm. For performance validation, the developed model is tested on the Tennessee Eastman benchmark process.

Data Mining and Analytics in the Process Industry: The Role of Machine Learning

Data mining and analytics have played an important role in knowledge discovery and decision making/supports in the process industry over the past several decades. As a computational engine to data mining and analytics, machine learning serves as basic tools for information extraction, data pattern recognition and predictions. From the perspective of machine learning, this paper provides a review on existing data mining and analytics applications in the process industry over the past several decades. The state-of-the-art of data mining and analytics are reviewed through eight unsupervised learning and ten supervised learning algorithms, as well as the application status of semi-supervised learning algorithms. Several perspectives are highlighted and discussed for future researches on data mining and analytics in the process industry.

Distributed Parallel PCA for Modeling and Monitoring of Large-Scale Plant-Wide Processes With Big Data

In order to deal with the modeling and monitoring issue of large-scale industrial processes with big data, a distributed and parallel designed principal component analysis approach is proposed. To handle the high-dimensional process variables, the large-scale process is first decomposed into distributed blocks with a priori process knowledge. Afterward, in order to solve the modeling issue with large-scale data chunks in each block, a distributed and parallel data processing strategy is proposed based on the framework of MapReduce and then principal components are further extracted for each distributed block. With all these steps, statistical modeling of large-scale processes with big data can be established. Finally, a systematic fault detection and isolation scheme is designed so that the whole large-scale process can be hierarchically monitored from the plant-wide level, unit block level, and variable level. The effectiveness of the proposed method is evaluated through the Tennessee Eastman benchmark process.

Autoregressive Dynamic Latent Variable Models for Process Monitoring

In most industrial processes, both autocorrelations and cross correlations in the data need to be extracted for the purpose of process monitoring and diagnosis. However, traditional dynamic modeling methods focus on the dynamic relationship while the cross correlations are at best implicit. In this brief, a new autoregressive dynamic latent variable model is proposed to capture both dynamic and static relationships simultaneously. The proposed method is a rather general dynamic model which can improve the performance of modeling and process monitoring. The Kalman filter and smoother are employed for inference while model parameters are estimated with an expectation-maximization algorithm. The corresponding fault detection method is also developed and a numerical example and the Tennessee Eastman benchmark process are used to evaluate the performance of the proposed model.

Parametric Mismatch Detection and Isolation in Model Predictive Control System1

Abstract Predictor that is built from the plant model, plays an important role in model predictive control system. The predictor should be updated timely to maintain certain calculation accuracy and performance optimality. In order to avoid unnecessary interruptions to production, however, updating should only be done when serious mismatch between the process and model appears. A novel method based on subspace approach is proposed to detect the mismatches using closed-loop operation data. The channels with mismatches in multi input multi output system are isolated. And some combinations of the mismatched parameters that have physical significance can be detected. These results provide useful information for the maintenance of model predictive control system. Simulations on a distillation process demonstrate the efficacy of the methodology.

Recursive Subspace Model Identification Based On Vector Autoregressive Modelling

Abstract Recursive subspace model identification (RSMI) has been developed for a decade. Most of RSMIs are only applied for open loop data. In this paper, we propose a new recursive subspace model identification which can be applied under open loop and closed loop data. The key technique of this derivation of the proposed algorithm is to bring the Vector Auto Regressive with eXogenous input (VARX) models into RSMI. Numerical studies on a closed loop identification problem show the effectiveness of the proposed algorithm.

A Probabilistic Just-in-Time Learning Framework for Soft Sensor Development With Missing Data

Just-in-time learning (JITL) is one of the most widely used strategies for soft sensor modeling in nonlinear processes. However, traditional JITL methods have difficulty in dealing with data samples that contain missing values. Meanwhile, data noises and uncertainties have not been taken into consideration for relevant sample selection in existing JITL approaches. To overcome these problems, a new probabilistic JITL (P-JITL) framework is proposed in this brief. In P-JITL, variational Bayesian principal component analysis is first utilized to handle missing values and extract Gaussian posterior distributions of latent variables. Then, symmetric Kullback–Leibler divergence is creatively employed to measure the dissimilarity of two distributions for relevant sample selection in the JITL framework. Finally, a nonlinear regression model, Gaussian process regression, is carried out to model the nonlinear relationship between the output and the extracted latent variables. In this way, the proposed probabilistic JITL (P-JITL) is able to deal with missing data and select relevant samples more accurately. To evaluate the effectiveness and flexibility of P-JITL, comparative studies between P-JITL and traditional deterministic JITL (D-JITL) are carried out on a numerical example and an industrial application example, in which missing data are simulated with percentages from 0% to 50%. The results show that P-JITL can provide more accurate prediction accuracy than D-JITL in each scenario considered.

Dynamic Least Squares Support Vector Machine

Based on narrating the theory of least squares support vector machine (LS-SVM), dynamic LS-SVM (DLS-SVM) is presented in this paper. DLS-SVM is suitable for real time system recognition and time series prediction. Whenever a new example is obtained, the method gets rid of the first vector and replaces it with the new input vector. That is, this algorithm can adjust the model to track the dynamics of the nonlinear time-varying system. Time series prediction can be a very useful tool to forecast and to study the behavior of key process parameters in time. This creates the possibility to give early warnings of possible process malfunctioning. In this paper, DLS-SVM is applied to predict the concentration of 4-carboxybenzaldchyde (4-CBA) in purified terephthalic acid (PTA) oxidation process. Results indicate that the proposed method is effective

Soft Sensor Modeling of Nonlinear Industrial Processes Based on Weighted Probabilistic Projection Regression

Probabilistic principal component regression (PPCR) has been introduced for soft sensor modeling as a probabilistic projection regression method, which is effective in handling data collinearity and random noises. However, the linear limitation of data relationships may cause its performance deterioration when applied to nonlinear processes. Therefore, a novel weighted PPCR (WPPCR) algorithm is proposed in this paper for soft sensing of nonlinear processes. In WPPCR, by including the most relevant samples for local modeling, different weights will be assigned to these samples according to their similarities with the testing sample. Then, a weighted log-likelihood function is constructed, and expectation-maximization algorithm can be carried out iteratively to obtain the optimal model parameters. In this way, the nonlinear data relationship can be locally approximated by WPPCR. The effectiveness and flexibility of the proposed method are validated on a numerical example and an industrial process.