Development of a regional feature selection-based machine learning system (RFSML v1.0) for air pollution forecasting over China

With the explosive growth of atmospheric data, machine learning models have achieved great success in air pollution forecasting because of their higher computational efficiency than the traditional chemical transport models. However, in previous studies, new prediction algorithms have only been tested at stations or in a small region; a large...

Experimental study and discrete element analysis on lateral resistance of windblown sand railway

The lateral resistance of ballast bed is an important parameter to prevent track expansion and maintain track stability. The invasion of sand particles can cause the change of lateral resistance of ballast bed and affect the stability of track structure, but little attention has been paid to the change characteristics of nonlinear lateral...

Mechanical behaviour and energy evolution of polyurethane-mixed ballast under revised bonding constitutive model

The bonding contact presents complex modes in polyurethane-mixed ballast. The commonly used parallel bond model is revised and four different contact models are developed including Ballast-Ballast Contact Bonding, Ballast-Ballast Noncontact Bonding, Ballast-Sleeper Contact Bonding, and Ballast-Sleeper Noncontact Bonding. The mechanical...

Discrete Element Analysis on Mechanical Properties of Ballast Bed by Tamping in Railway Turnout Areas

Turnout is the key component of the railway tracks for trains to change direction, which is vital to operational safety and passenger comfort. Therefore, it is of great importance to perform a scientific and reasonable tamping operation for turnout areas. In this paper, based on the commercial software EDEM and RECURDYN, a coupled simulation...