Model on Information Revising of Supervisory and Control Systems

Sie sind hier: Homepage > Suche

Model on Information Revising of Supervisory and Control Systems

Xu, Jie / Cheng, Xiaoqing / Zhou, Huijuan / Jia, Limin

To provide effective support for the functioning of the transportation channel, the overall supervisory and control systems architecture of railway transportation systems must be capable of linking or coordinating the information systems of the individual parties into a cohesive whole. After analyzing the features of supervisory and control systems of railway transportation system, issues of data noise were raised regarding the information revising algorithm for extracting appropriate data from massive and heterogeneous data sources. Bayesian Networks (BN) provides a robust probabilistic method of reasoning under uncertainty. In order to eliminate the noise, a new method is presented for extracting appropriate data from control data sources, which is based on the Bayes decision-making theory. Prior specification and stochastic search were two important components of this approach. First, Bayesian networks and the Monte Carlo method were introduced briefly. Secondly, the combination of prior probability and data samples induced a posterior-distribution. Then, Bayesian networks were used to investigate the performance of Gibbs sampling algorithms for inference with WinBugs. Finally, the experimental results were presented and discussed.