A review of simulation models for railway systems

With the advances in computer hardware and software\uddevelopment techniques in the past 25 years, digital\udcomputer simulation of train movement and traction\udsystems has been widely adopted as a standard\udcomputer-aided engineering tool [1] during the design\udand development stages of existing and new railway\udsystems. Simulators of different approaches and scales\udare used extensively to investigate various kinds of\udsystem studies. Simulation is now proven to be the\udcheapest means to carry out performance predication\udand system behaviour characterisation.\udWhen computers were first used to study railway\udsystems, they were mainly employed to perform\udrepetitive but time-consuming computational tasks, such\udas matrix manipulations for power network solution and\udexhaustive searches for optimal braking trajectories.\udWith only simple high-level programming languages\udavailable at the time, full advantage of the computing\udhardware could not be taken. Hence, structured\udsimulations of the whole railway system were not very\udcommon. Most applications focused on isolated parts of\udthe railway system. It is more appropriate to regard\udthose applications as primarily mechanised calculations\udrather than simulations.\udHowever, a railway system consists of a number of\udsubsystems, such as train movement, power supply and\udtraction drives, which inevitably contains many\udcomplexities and diversities. These subsystems interact\udfrequently with each other while the trains are moving;\udand they have their special features in different railway\udsystems. To further complicate the simulation\udrequirements, constraints like track geometry, speed\udrestrictions and friction have to be considered, not to\udmention possible non-linearities and uncertainties in the\udsystem.\udIn order to provide a comprehensive and accurate\udaccount of system behaviour through simulation, a large\udamount of data has to be organised systematically to\udensure easy access and efficient representation; the\udinteractions and relationships among the subsystems\udshould be defined explicitly. These requirements call\udfor sophisticated and effective simulation models for\udeach component of the system. The software\uddevelopment techniques available nowadays allow the\udevolution of such simulation models. Not only can the\udapplicability of the simulators be largely enhanced by advanced software design, maintainability and\udmodularity for easy understanding and further\uddevelopment, and portability for various hardware\udplatforms are also encouraged.\udThe objective of this paper is to review the development\udof a number of approaches to simulation models.\udAttention is, in particular, given to models for train\udmovement, power supply systems and traction drives.\udThese models have been successfully used to enable\udvarious ‘what-if’ issues to be resolved effectively in a\udwide range of applications, such as speed profiles,\udenergy consumption, run times etc