Bayesian Network Method for Fault Diagnosis in a Continuous Tubular Reactor

Haoran Liu (1) , Jean-Christophe Buvat (1) , Lionel Estel (1) , Isabelle Polaert (1)

1 LSPC - Laboratoire de Sécurité des Procédés Chimiques

The aim is this paper is to study fault diagnosis in a continuous chemical process. An experimental system is built to be the research base and a model is proposed and trained to carry out the fault diagnosis in the process. A Bayesian network model with two-layer nodes structure is designed and Maximum likelihood estimation (MLE) is used to amend the conditional probability table (CPT) given by expert knowledge. Then a Monte Carlo method is applied to simplify the inference rules and the data samples collected from the experimental system has been used to test the model.

Keywords: continuous chemical reactor; Bayesian network model; fault diagnosis; maximum likelihood estimation

Domains: Engineering Sciences [physics]; Chemical and Process Engineering

Haoran Liu, Jean-Christophe Buvat, Lionel Estel, Isabelle Polaert. Bayesian Network Method for Fault Diagnosis in a Continuous Tubular Reactor. Chemical Product and Process Modeling, 2010, 5 (1), pp.Article 28. ⟨10.2202/1934-2659.1491⟩. ⟨hal-02409976⟩