Fault Diagnosis of Event-Driven Systems based on Timed Markov Model with Maximum Entropy Principle

Abstract

This paper presents a new fault diagnosis technology for event-driven controlled systems like PLC. The controlled plant is modeled by means of the Timed Markov Model, which regards the time interval between successive two events as a random variable. In order to estimate the probability density functions of the randomized time intervals, the maximum entropy principle is introduced, which can estimate probability density functions so as to maximize the uniformity with satisfying the constraints caused by measured data. Then, the fault diagnosis algorithm, which returns the probabilistic diagnosis results, is developed. Finally, the usefulness of the proposed strategy is verified through some experimental results.