Noise Emission Curves for Train in Arrival, In-front and Departure Mode

Abstract

Rail transport has long been a cornerstone of societal development, contributing significantly to economic growth by facilitating the movement of goods and people. However, it also presents environmental challenges, one of the most prominent being noise pollution, affecting both urban and rural populations. This paper addresses the critical issue of railway noise emissions by identifying and analyzing the various noise sources produced during train operations. Noise emission curves were developed for trains in three distinct operational modes-arrival, in-front and departure. Analysis was done on a dataset of 180 noise measurements for trains traveling at different speeds using a Class 1 sound level meter, kept at the height of 1.2m from the ground, to simulate the location of human ears at a safe distance from the center of the track. The overall LAeq of trains passing a point, encompassing the arrival, in-front, and departure phases, ranged from 53.3 to 104.0 dB(A), with a mean value of 87.05 dB(A). Detailed analysis indicated that a logarithmic curve best fits the noise data across all modes and vehicle categories, supported by an R² value of about 0.85 during both the training and testing stages. This suggests the logarithmic model effectively represents the measured noise levels of concurrently moving noise sources loosely bound together. R² values indicate that the developed equations are reliable for predicting train noise emissions under various conditions. Notably, the mean actual and root mean square error for the arrival stage were both under 1 dB(A), which is lower than that for other stages. These predictive models can be helpful in noise mitigation efforts, offering a tool for railway planners and engineers to forecast noise levels and implement appropriate control measures. © 2025 Seventh Sense Research Group®.