Improvement of DC Railway Simulation Using Iteration History-of-Convergence Calculation

Abstract

We developed an integrated railway simulator that can simulate the collaboration each subsystem of a railroad, such as a train, signalling system, and power-supply system, and also train operation and energy usage on the basis of the collaboration of subsystems in all relevant railroads. The purpose of this simulator is to determine the change in energy consumption of trains and substations depending on the changes in the train characteristics, timetables, power-supply systems, and other such variables. This simulator's targets for calculation are the energy consumption of substations and trains restricted by the conditions of signalling and traffic-control systems. In this simulator, we modeled each subsystem on the basis of its physical characteristics, and we designed an electric circuit on the basis of the subsystem's characteristics. The characteristics of each subsystem are non-linear. Therefore, we used an iteration method to calculate the solution of the electric circuit. However, the iteration method may fail to calculate a solution. To prevent this, we developed a method to select an appropriate model of the subsystem using the convergence history of the iteration method. By applying the proposed method to our simulator, we could eliminate convergence failure and reduce the average number of iterations required for convergence.