Method Research On Reliability Design Of Catenary System Based On RBFNN

In this dissertation, regarding Bao- Cheng railway line as testing base, main failure model and systemic reliability of catenary were analyzed and researched by the methods of artificial intellective technology, reliability project, probability theory, mathematical statistics and stochastic process. At the same, reliability analysis software was programmed by VISUAL BASIC6.0 and MATLAB.Probability distribution is base of reliability analysis . Intellective recognition model of probability distribution on reliability analysis was established by RBF neural network basis on the characteristic of self-orgnization, self-adaptive and strong containable wrong ability in the paper. Its total recognitive rate reaches 93.75%. At the same time, probability distribution on fault time of catenary orientation setting was recognized, its result is exponential distribution. It provides bases for Markov reliability analysis of catenary system.Aiming at the difficulty of wear characteristic establishment, nonlinear simulative relative of wear characteristic was obtained by radial basis function neural network for the first time in this paper. Variational law of wear rate along with exterior factors was analyzed by the relative formula. As strong approach ability, improved...