Fault diagnosis of the TE process based on discrete hidden Markov model

It turns out that a prerequisite to widespread deployment of condition-based maintenance (CBM) technology and practice in industry is effective fault diagnosis and prognosis. Diagnosis is an assessment about the current health of a system based on observed symptoms. Prognosis is to predict the progression of a fault condition to system failure and estimate the remaining useful life (RUL) of the system. Consequently diagnosis is essential to security maintenance of the industry process. This paper presents a method based on discrete hidden Markov model (DHMM) for carrying out diagnosis. The proposed method was validated on a chemical process-the Tennessee Eastman process. The result indicates the effectiveness of this method.