Résumé

As a generalization of the successful hidden Markov models, Dynamic Bayesian Networks (DBNs) are a natural basis for the general temporal action interpretation task. This document provides a conditional probabilistic approach to analyze the energy availability in electrical distribution networks by using Bayesian networks (BN). Firstly a static BN modelling is presented to show the influence of the switch behaviour on the energy availability. Then, the dynamic behaviour of the switch is cared by switch reliability modelling using DBN which permits to predict the energy availability. The prediction by DBNs discussed in the case study of this paper gives a strong contribution on electrical network supervisory control and it can also be applied to transportation networks.

Information

Item Type:	Journal
Divisions:	» Faculté des Sciences et de la Technologie
ePrint ID:	395
Date Deposited:	2015-07-30
Further Information:	Google Scholar
URI:	https://www.univ-soukahras.dz/fr/publication/article/395

BibTex