Tuning of fault semantic network using Bayesian theory for probabilistic fault diagnosis in process industry

Abstract

Investigating complex interaction patterns among multiple process variables (PVs) is an important task for fault propagation analysis and calculation of final risks. This paper demonstrates a robust method to estimate interaction strengths among process variables. The method is based on dynamic fault semantic networks (FSN) combined with Bayesian belief theory for probabilistic tuning of the fault semantic network. The effectiveness and feasibility of the proposed technique is verified on simulated data emanating from Tennessee Eastman (TE) process. The extracted patterns of interaction structure among PVs aid to uncover the polishing mechanisms and provide more insights to investigate fault propagation scenarios.

Process Fault Prognosis Using Hidden Markov Model–Bayesian Networks Hybrid Model

“…As a further development, a method based on dynamic fault semantic networks was presented by Gabbar and Hossein in 2013. 33 The fault semantic network was probabilistically tuned by combining with Bayesian bielief theory. This can be introduced as a robust method that can estimate interaction strengths among process variables.…”

“…As a further development, a method based on dynamic fault semantic networks was presented by Gabbar and Hossein in 2013 . The fault semantic network was probabilistically tuned by combining with Bayesian bielief theory.…”

Simultaneous Fault Diagnosis using multi class support vector machine in a Dew Point process

“…Most of the studies reported in literature on fault diagnosis using neural networks or support vector machine involve chemical systems with a separator as one of its components (Hussain et al, 2013;Rusinov et al, 2013;Salahshoor et al, 2010). This process is selected because it reveals the most common features appearing in industrial processes.…”

Signal de-noising methods for fault diagnosis and troubleshooting at CANDU ® stations

“…Such predictive and condition monitoring system is being developed at the Energy Safety and Control Laboratory (ESCL) -University of Ontario Institute of Technology (UOIT). It is based on Fault Semantic Network (FSN) methodology (Gabbar, 2007(Gabbar, , 2010Gabbar and Khan, 2010) but in addition to fault detection, diagnosis and prognosis (Hosseini, 2013;Hosseini andGabbar, 2011a, 2011b;Hussain et al, 2013) this system incorporates in itself root-cause and consequence analysis, maintenance work and design change recommendations. The essence of this approach is demonstrated below.…”