Multiple-Fault Diagnosis Based on System Decomposition and Dynamic PLS

Abstract

This study suggests a hybrid method of signed digraph and partial least squares for the fault diagnosis of chemical processes. Using the local qualitative relationships of each variable in a signed digraph, a process is decomposed into subprocesses to allow for the reliable diagnosis of multiple faults. A partial least-squares model is built for the estimation of each measured variable in each decomposed subprocess. To handle the process dynamics accurately, the input of the partial least-squares model uses both the current values and the past values of the process variables. In each case, the measured and estimated values are compared to diagnose the fault. A case study of a continuously stirred tank reactor illustrates the conclusion that the proposed method greatly improves the accuracy, resolution, and reliability of the diagnosis.

Advances and new directions in plant-wide disturbance detection and diagnosis

“…Qualitative models include signed digraphs (SDG) [Venkatasubramanian, Rengaswamy & Kavuri, 2003;Maurya, Rengaswamy & Venkatasubramanian, 2004;Srinivasan, Maurya & Rengaswamy, 2006], Multilevel Flow Modelling [Petersen, 2000] and Bayesian belief networks [Weidl, Madsen & Israelson, 2005]. Chiang and Braatz (2003)] and also Lee, Song & Yoon (2003) showed enhanced diagnosis using signal based analysis if a qualitative model is available.…”

Cause-and-effect analysis in chemical processes utilizing XML, plant connectivity and quantitative process history

“…Norvilas et al (2000) combined multivariate statistical analysis and an expert system for the same purpose, while Leung and Romagnoli (2002) integrated a multivariate statistical analysis method with cause and effect map of a process, which was set up manually, to help in the diagnosis of faults. Lee et al, (2003) also combined SDGs with multivariate statistical analysis for enhanced diagnosis. Maurya et.…”

“…Several authors (Chiang and Braatz, 2003;Lee et al, 2003) have suggested it is possible to develop better techniques of fault diagnosis by combining a qualitative method with a quantitative one. This paper presents an approach that generates hypotheses regarding the possible root cause of a plant-wide disturbance using statistical measurement-based analysis, and tests the hypotheses by automated examination of the connectivity of the process using graph traversal.…”

System Diagnosis Using Fuzzy Cognitive Maps

“…This type of analysis will be left for future research. (Lee et al, 2003;Pelaez & Bowles, 1995;Pelaez & Bowles, 1996) The basic process of diagnosis in a fuzzy cognitive map is to produce and manipulate the reachability matrix for the map. (Ndousse & Okuda, 1996) The reachability matrix is built from the adjacency matrix for the map, J, which captures the cause/effect relationships in the map.…”