Dynamic modeling of traffic noise in both indoor and outdoor environments by using a ray tracing method

Abstract

The complexity of the urban environment and traffic characteristics give the dynamic simulation of traffic noise practical significances. A new dynamic modeling of traffic noise in both indoor and outdoor environments was proposed. First, considering multiple reflections and diffractions, 3D ray tracing method subordinated to space partitioning was used to simulate sound propagations. Then, microscopic traffic simulation which provided time-variant parameters for the vehicle emission model, was used to gain spatial and temporal outputs. Space partitioning improved the computational efficiency by transforming the global search into a local one in complex sound propagation simulation. A case was validated the accuracy, as the indoor and outdoor average errors of equivalent sound levels (Leq) were respectively 1.68 dB and 1.65 dB. Moreover, less than 1.5 dB errors of statistical indicators (L90, L50 and L10) and well coincidence of statistical distributions of the instantaneous sound pressure levels (Leq,1s) illustrated a good performance on dynamic simulation. Finally, the effects of vehicle speed, proportion of heavy vehicles and signal phase on traffic noise were discussed. The results showed that 1) Indoor/outdoor Leq had logarithmic relationships with vehicle speed. 2) It presented quadratic polynomial relationships between Leq and proportion of heavy vehicles. 3) Dynamic results showed that noise variations were consistent with the changes of the signal period, and it revealed different noise distributions with various vehicle statuses; upstream average Leq was 3.3 dB louder than the downstream one with vehicle acceleration; average Leq,1s during the red light was 20 dB softer than it during the green light.