A Hybrid Model for Fault Diagnosis Using Model Based Approaches and Support Vector Machine
-
- Lee Chang Jun
- School of Chemical and Biological Engineering, Seoul National University
-
- Lee Gibaek
- Department of Chemical and Biological Engineering, Chungju National University
-
- Han Chonghun
- School of Chemical and Biological Engineering, Seoul National University
-
- Yoon and En Sup
- School of Chemical and Biological Engineering, Seoul National University
Bibliographic Information
- Other Title
-
- Hybrid Model for Fault Diagnosis Using Model Based Approaches and Support Vector Machine
Search this article
Abstract
The procedure followed in chemical processes can be expressed in simple terms such as the flow of events from the raw materials to the product. To obtain the best final product, chemical engineers have to consider many factors including environmental effects, stability, economic considerations, and so on. In particular, when considering the stability if the process and the purity of the product, it is very important to detect any faults in the chemical process immediately.<BR>In this paper, a hybrid fault diagnosis model based on the signed digraph (SDG) and support vector machine (SVM) is proposed. By means of the system decomposition based on SDG, the local models of each measured variable are constructed and more accurate and fast models are using an SVM, which has no loss of information and shows good performance, in order to obtain the estimated value of the variable, which is then compared with the measured value in order to diagnose the fault. To verify the performance of the proposed model, the Tennessee Eastman (TE) Process was studied and the proposed method was found to demonstrate a good diagnosis capability compared with previous statistical methods.
Journal
-
- JOURNAL OF CHEMICAL ENGINEERING OF JAPAN
-
JOURNAL OF CHEMICAL ENGINEERING OF JAPAN 39 (10), 1085-1095, 2006
The Society of Chemical Engineers, Japan
- Tweet
Keywords
Details 詳細情報について
-
- CRID
- 1390282679546868224
-
- NII Article ID
- 10018326867
-
- NII Book ID
- AA00709658
-
- ISSN
- 18811299
- 00219592
-
- NDL BIB ID
- 8513364
-
- Text Lang
- en
-
- Data Source
-
- JaLC
- NDL
- Crossref
- CiNii Articles
-
- Abstract License Flag
- Disallowed