Ultrasound Flowmeter Reliability Research Based On Stochastic Petri Nets And Monte Carlo Simulation

The ultrasonic flowmeter has developed rapidly in recent years and is a widelyused class of flowmeter. There are some defects in its practical application, such as;failure can not be found in time, failure data collection unreasonable, not serviceuntil the system can not work. Therefore, the reliability research of the ultrasonicflowmeters is important for manufacturers and users. The work of this paper is tobuild the system model of the ultrasonic flowmeter, to solve the reliability index ofthe system, to develop preventive maintenance strategies and so on. The purpose ofthis subject is to to develop a system of preventive maintenance strategies byreliability index, to solve the problems as the collection of failure data which is gray,system working with potential trouble.The main content of this paper is as follows: firstly, the paper proves thestochastic Petri net model is reasonable by comparing the several research methodsof dynamic reliability. Then the paper obtains the dynamic changing process of thesystem by building the stochastic Petri nets model according to the failure types ofultrasonic flowmeters. Secondly, the paper make Monte Carlo simulation accordingto status flags of the System failure obtained by stochastic Petri nets and the randomnumber and programming generation methods of a variety of distribution functionsare given during the simulation. After simulation, the dynamic reliability indexes ofthe ultrasonic flowmeter are obtained by calculating. Finally, the preventivemaintenance cycle of the system is determined by using dynamic reliability index ofthe system and RCM to develop a preventive maintenance strategy for the system,then improve the theory of the paper is feasible through the analysis of the field data.The preventive maintenance cycle of the system is3200h by using the method of thepaper, when the availability is not less than0.95. The paper sets the ultrasonicflowmeter as a carrier, not only gives the stochastic Petri net model of ultrasonic flowmeter which is a parallel system, but aslo introduces the stochastic Petri netmodel for voting systems and standby systems. The conclusion of the paper showsthat an appropriate preventive maintenance cycle of large and complex systems canbe obtained by the method used in this paper.The paper ignores the correlation between the various units of the system andintroduces the correlation between the various units into the the Monte Carlosimulation of system dynamic reliability, when using Monte Carlo simulation tosolve the reliability index of the system. The problem should be further researched.