Dynamics and control simulation of railway virtual coupling

This paper developed a parallel computing architecture for high-fidelity virtual coupling simulations. Multi-body train dynamics mod-els considered various nonlinear components including wheel-railcontact, suspensions, and inter-vehicle connections. A virtual cou-pling controller was developed which can be implemented undervarious train-to-train communication topologies. The controller alsoallows existing trains to leave the platoon and new trains to mergeinto the platoon without re-designing the controller. The parallelcomputing architecture is also scalable and not limited by: the num-ber of vehicles in each train; the number of trains in each trainplatoon and the topology of train-to-train communications. A casestudy by simulating a three-train (18 vehicles in total) platoon on areal-world track section was conducted. The results show that, byusing 19 computer cores, parallel computing speed is nearly twiceas fast as real-time. Parallel computing is about 17 times faster thanserial computing. The results also show that the maximum spacingerrors of the follower trains were about 0.22 m. Dynamics results suchas wheel-rail contact forces, suspension forces, carbody vibrationsand inter-vehicle forces were obtained; these results can be usedto conduct system assessments in terms of passenger ride comfort, mechanical wear etc