Topology Design and Optimization of Train Communication Network Based on Industrial Ethernet

Abstract

With increasing requirements for the amount of data transmitted by train communication networks (TCN), the industrial Ethernet has become a solution with large transmission rates and high compatibility. To ensure safety and real-time communication, the delay of data transmission must be strictly limited in TCN. However, the delay caused by data transmission conflicts is still a problem in the current Ethernet research, which cannot meet the demands of real-time transmission between network devices. To solve the time-delay problem, an improved adaptive genetic algorithm (IAGA) for topology optimization is proposed, in which a new optimization model is established based on the train Ethernet topology. Compared with two traditional algorithms in solving the optimization model, the IAGA achieved better performance in terms of convergence speed and solution accuracy. To verify the effect of the optimized model, the theoretical calculation, simulation, and experiment based on a communication platform are carried out for the real-time index. Experimental results indicated that the end-to-end delay of the optimized network is significantly reduced. In designing and solving the topology optimization problem of TCN, the proposed model and algorithm effectively realized real-time improvement of the data transmission.