• Journal of KIISE:Software and Applications
• /
• v.31 no.3
• /
• pp.361-368
• /
• 2004

The purpose of this study is to propose a new on-line nonlinear PCA(OL-NPCA) method for a nonlinear feature extraction from the incremental data. Kernel PCA(KPCA) is widely used for nonlinear feature extraction, however, it has been pointed out that KPCA has the following problems. First, applying KPCA to N patterns requires storing and finding the eigenvectors of a N${\times}$N kernel matrix, which is infeasible for a large number of data N. Second problem is that in order to update the eigenvectors with an another data, the whole eigenspace should be recomputed. OL-NPCA overcomes these problems by incremental eigenspace update method with a feature mapping function. According to the experimental results, which comes from applying OL-NPCA to a toy and a large data problem, OL-NPCA shows following advantages. First, OL-NPCA is more efficient in memory requirement than KPCA. Second advantage is that OL-NPCA is comparable in performance to KPCA. Furthermore, performance of OL-NPCA can be easily improved by re-learning the data.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.8
• /
• pp.3779-3784
• /
• 2012

Real-time process monitoring and diagnosis of industrial processes is one of important operational tasks for quality and safety reasons. The objective of fault diagnosis or identification is to find process variables responsible for causing a specific fault in the process. This helps process operators to investigate root causes more effectively. This work assesses the applicability of combining a nonlinear statistical technique of kernel Fisher discriminant analysis with a preprocessing method as a tool of on-line fault identification. To compare its performance to existing linear principal component analysis (PCA) identification scheme, a case study on a benchmark process was performed to show that the fault identification scheme produced more reliable diagnosis results than linear method.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10a
• /
• pp.18-18
• /
• 1996

This paper reviews some of the most prominent and promising areas of chemical process control both in relations to batch and continuous processes. These areas include the modeling, optimization, control and monitoring of chemical processes and entire plants. Most of these areas explicitly utilize a model of the process. For this purpose the types of models used are examined in some detail. These types of models are categorized in knowledge-driven and datadriven classes. In the areas of modeling and optimization, attention is paid to batch reactors using the Tendency Modeling approach. These Tendency models consist of data- and knowledge-driven components and are often called Gray or Hybrid models. In the case of continuous processes, emphasis is placed in the closed-loop identification of a state space model and their use in Model Predictive Control nonlinear processes, such as the Fluidized Catalytic Cracking process. The effective monitoring of multivariate process is examined through the use of statistical charts obtained by the use of Principal Component Analysis (PMC). Static and dynamic charts account for the cross and auto-correlation of the substantial number of variables measured on-line. Centralized and de-centralized chart also aim in isolating the source of process disturbances so that they can be eliminated. Even though significant progress has been made during the last decade, the challenges for the next ten years are substantial. Present progress is strongly influenced by the economical benefits industry is deriving from the use of these advanced techniques. Future progress will be further catalyzed from the harmonious collaboration of University and Industrial researchers.