Applying wavelet-based hidden Markov tree to enhancing performance of process monitoring

Chen, Junghui; Chang, Wei-Ching

doi:10.1016/j.ces.2005.03.061

Cited by 14 publications

(3 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further research is required to optimize the MSPC procedure to various types of data. As suggested by the anonymous referee, one possible research direction is to look at the literature of wavelet-based control charts, which also involve hidden trees 48 .…”

Section: Discussionmentioning

confidence: 99%

The funnel experiment: The Markov‐based SPC approach

Singer

Ben‐Gal

2007

Quality & Reliability Eng

View full text Add to dashboard Cite

The classical funnel experiment was used by Deming to promote the idea of statistical process control (SPC). The popular example illustrates that the implementation of simple feedback rules to stationary processes violates the independence assumption and prevents the implementation of conventional SPC. However, Deming did not indicate how to implement SPC in the presence of such feedback rules. This pedagogical gap is addressed here by introducing a simple feedback rule to the funnel example that results in a nonlinear process to which the traditional SPC methods cannot be applied. The proposed method of Markov-based SPC, which is a simplified version of the context-based SPC method, is shown to monitor the modified process well.

show abstract

Section: Discussionmentioning

confidence: 99%

The funnel experiment: The Markov‐based SPC approach

Singer

Ben‐Gal

2007

Quality & Reliability Eng

View full text Add to dashboard Cite

show abstract

“…Recently, the broader community of process systems engineering has witnessed a growing interest in HMM technique for various application fields including process modeling, control, and monitoring. [40][41][42] In practice, the process measurements should be viewed as realizations or observations of underlying stochastic processes. The abrupt changes in operating conditions, dynamic interactions, uncertain drifts, and nonstationary disturbances all pose challenges for conventional monitoring approaches.…”

Section: Introductionmentioning

confidence: 99%

Multiway discrete hidden Markov model‐based approach for dynamic batch process monitoring and fault classification

2011

AIChE Journal

View full text Add to dashboard Cite

A new multiway discrete hidden Markov model (MDHMM)-based approach is proposed in this article for fault detection and classification in complex batch or semibatch process with inherent dynamics and system uncertainty. The probabilistic inference along the state transitions in MDHMM can effectively extract the dynamic and stochastic patterns in the process operation. Furthermore, the used multiway analysis is able to transform the three-dimensional (3-D) data matrices into 2-D measurementstate data sets for hidden Markov model estimation and state path optimization. The proposed MDHMM approach is applied to fed-batch penicillin fermentation process and compared to the conventional multiway principal component analysis (MPCA) and multiway dynamic principal component analysis (MDPCA) methods in three faulty scenarios. The monitoring results demonstrate that the MDHMM approach is superior to both the MPCA and MDPCA methods in terms of fault detection and false alarm rates. In addition, the supervised MDHMM approach is able to classify different types of process faults with high fidelity.

show abstract

“…HMT could effectively characterize the joint statistics of wavelet coefficients across scales. Using HMT, process trend classification was recently proposed to analyze the correlations among variables across the time−frequency plane. , The existing statistical analysis methods were also developed based on the HMT model in order to enhance the capability of the statistical process monitoring and effectively cope with increasingly complex systems in the face of measurements with uncertainty disturbances. , However, in reality, the topology of an HMT (i.e., the number of connectivity) was fixed by the collected data. Although the batch signal exhibited a wide range of dynamic behaviors as well as noise levels, so far most of the HMT results are built based on the complete tree structure.…”

Section: Introductionmentioning

confidence: 99%

Self-Growing Hidden Markov Tree Based Multiway Principle Component Analysis for Enhanced Monitoring of Batch Processes

Chen

Hsu

Jiang

et al. 2007

Ind. Eng. Chem. Res.

Self Cite

View full text Add to dashboard Cite

A novel architecture called the growing wavelet based hidden Markov tree (gHMT) for batch process monitoring is proposed. It characterizes process measurements by the growing and learning procedure. It starts with a small size wavelet based hidden Markov tree (HMT) and successively increments the size of the wavelet tree until the desirable size is reached. Like HMT, this modeling scheme in the wavelet domain can not only analyze the measurements at multiple scales in time and frequency but also capture the statistical behavior of the real-world measurements at different scales. Unlike HMT with a structure covering the whole frequency range, gHMT has the ability to explicitly control over the complexity of the HMT architecture, retaining the smallest possible size and the accuracy of the model without introducing additional computational load. With the smallest possible size of the structure, it still has the capability of representing the underlying data distribution with the desired accuracy. After the gHMT model extracts the past operating information, it can be used to generate simple monitoring charts, easily tracking and monitoring the occurrence of observable upsets. The development and implementation of the gHMT for batch monitoring are illustrated through a fed-batch penicillin cultivation example to help readers delve into the matter. Also, the comparisons between the existing statistical process control (SPC) methods, the conventional HMT method, and the gHMT method are demonstrated to explain the advantages of our newly proposed method.

show abstract

Applying wavelet-based hidden Markov tree to enhancing performance of process monitoring

Cited by 14 publications

References 16 publications

The funnel experiment: The Markov‐based SPC approach

The funnel experiment: The Markov‐based SPC approach

Multiway discrete hidden Markov model‐based approach for dynamic batch process monitoring and fault classification

Self-Growing Hidden Markov Tree Based Multiway Principle Component Analysis for Enhanced Monitoring of Batch Processes

Contact Info

Product

Resources

About