In order to estimate meteorological conditions on a time-scale and spatial-scale sufficient to prevent meteorological disasters on the railway system by interpolating data collected from anemometers or rain gauges, meteorological conditions were estimated using numerical simulations. Numerical simulation results were compared with recorded data for strong winds, heavy rainfall, and heavy snowfall. As a result, although there were cases where meteorological conditions were underestimated in the numerical simulations, it was possible to reproduce meteorological phenomena qualitatively.

No abstract

This paper proposes a method for evaluating train running safety in strong wind conditions using the probabilities of strong wind occurrence according to wind direction. The Weibull coefficients c d and k d for 16 wind directions were calculated on the basis of data from 10-min maximum instantaneous wind velocities for each wind direction at 772 Automated Meteorological Data Acquisition System (AMeDAS) stations. Probabilities P x of the occurrence of strong winds exceeding the critical wind speeds for overturning were estimated for a train in a virtual railway section by using c d and k d calculated on the basis of the wind data observed at a windy AMeDAS station. P x values were found to vary from 6.2×10-6 to 8.6×10-5 depending on the angle between the traveling direction of the train and each wind direction. The values of P x ranged from 1.4% to 20% of the P y values, which were calculated out of consideration of strong wind occurrence according to wind direction.