Surface pressure measurements on train for an on board cross-wind speed detection device

The paper presents the results of a research activity aimed to study the possibility to evaluate the cross wind velocity acting on a running train though surface pressure measurements. A complete design of the system is carried out using a neural network approach and wind tunnel test results. It will be demonstrated that the estimation of the magnitude and the direction of the wind speed can be solved by using only two suitably chosen pressure taps. Three train models, having different geometries and surface roughness, are tested in order to highlight some general features for the identification of the most suitable pressure taps for the considered application.