Optimization and reduction of blocking of railway lines using three-layer perceptron neural networks

Today, a huge amount of data is transmitted around the world through the high capacity of optical telecommunication networks. In railways, optical transmission telecommunication networks play a very important role in the transmission of vital rail data and the safety of movement. Reliability stability of telecommunication infrastructure is essential in increasing productivity, maintaining safety and reducing maintenance costs. In this article, the existing reliability level and all relevant parameters (the number of network outages and the duration between network failures, MTBF) of the railway optical transmission telecommunication network to obtain the number of blocked railway lines through the block diagram method (Reliability block diagram model) created and simulated through the Monte Carlo method and then optimized with the aim of reducing line blocking, considering that with each network failure, rail lines are blocked, the number of network outages is equal to line blocking. Therefore, with optimization, the reliability of the line blocking network is reduced. Also, predicting the behavior of the network and obtaining the probability of its failure is done through perceptron three-layer neural networks and its results are presented. The network implemented in this article is an optical telecommunication transmission network. The railway zone of Azerbaijan is 654 km long.