Contribution à l'étude de la fiabilité des postes de distribution électrique (application des processus semi-markoviens)

Abstract

The aim of this work is to study the reliability of electric feeder systems (E.F.S.). So, we have introduced a stochastic model which is the semi-markovian model for the electric power distribution stations reliability evaluation. Markov analysis provides a means of analyzing the reliability and availability of systems whose components exhibit strong dependencies. Since the world enters to the digital area, our dependency of the electric power quality supply will develop consequently. A few seconds of power breakdown or even a voltage fall could cost a company millions of dollars, and as well as an incalculable loss of reputation and significant number of customers. Respond the hopes of the society for a power supply available everywhere and inexpensive is not simply an option. The electric components reliability knowledge is a significant consideration in the design and the operation of industrials and commercials energy distribution systems. The various components failure characteristics of electric equipments can be partially described by the following basic reliability statistics : Failure rate, often expressed as the failures number per year per component (failures per one year unit); The component repair or substitution downtime after its broke-down in service, expressed in hours (or minutes) by failure; and In some special cases, the starting (or operation) probability is used. Estimated calculations of reliability have the aim of checking the reliability level of a system according to the reliability data of its components and material architecture. The data input result from various sources: experience feedbacks, reliability data banks (IEEE, MIL-HDBK 217, NPRD,…), etc. From an industry brought back statistical data, we could estimate and with a checking test, the exponential character of the electric components failures distribution function for these data. Since Algeria have a lack of the data collection and processing systems, we carried out an expertise study in order to determine the climatic and not-climatic factors which have the most influence on the electric components reliability. This method is used in order to target search on the equipments reliability. In order to ensure a high level reliability of the E.F.S., two sources of power supply are envisaged, one principal and the other of reserve, generally a cold reserve (diesel-electric group). The reliability of an electric power supply general structure was studied, the optimal periodicity aintenance was given as well as the influence of the coupling circuit-breaker interlocking time (which allows the power transfer to the load by the reserve source in case of the principal source failure) on the reliability and the availability of our structure