A road traffic noise pattern simulation model that includes distributions of vehicle sound power levels

De Coensel, Bert; Brown, A.L.; Tomerini, Deanna · 2016 · Crossref

DOI: 10.1016/j.apacoust.2016.04.010

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the limitations of current road traffic noise prediction models, which typically rely on energy-equivalent metrics like $L_{Aeq}$ and assume a single, prototypical emission law for each vehicle category. The authors argue that such models fail to capture the temporal pattern of noise fluctuations, which is critical for assessing health effects like sleep disturbance and annoyance. Real-world measurements show significant variation in noise emissions among vehicles within the same category due to factors like age, maintenance, and driving style. To accurately predict the occurrence and magnitude of individual noise events, the study proposes a refined simulation approach that incorporates measured distributions of vehicle sound power levels rather than average values. The methodology couples a microscopic traffic simulation (Aimsun) with a dynamic noise emission model (Imagine) and a point-to-point sound propagation model (ISO 9613). The key innovation is the "Distribution" model, which applies a unique emission correction ($\Delta L_W$) to each simulated vehicle. These corrections are randomly sampled from empirical distributions derived from over 85,000 vehicle pass-by measurements in Brisbane, Australia. The distributions account for variations in cars, trucks, and motorcycles, normalized to zero mean to ensure energy-equivalent levels remain comparable to standard models. The framework, implemented as a Python plugin called Noysim2, simulates instantaneous sound levels at receiver locations based on individual vehicle positions, speeds, and accelerations. The study evaluates this approach through 36 simulation scenarios on a dual-lane road with free-flow traffic, varying speed limits, traffic volumes, and heavy vehicle percentages. Results indicate that while energy-equivalent levels ($L_{Aeq}$) remain unchanged, the Distribution model significantly alters estimates of noise event indicators. Maximum sound levels ($L_{Amax}$) were predicted to be 12 to 18 dB higher than those from the standard Imagine model, reflecting the impact of rare, high-emission vehicles. Percentile levels ($L_{A01}$ and $L_{A05}$) showed differences ranging from -1.5 dB to +3.8 dB, depending on traffic volume. Median levels ($L_{A50}$) were consistently 1 to 4 dB lower due to the skew of emission distributions toward quieter vehicles. Furthermore, the model demonstrated a larger dynamic range in sound levels, leading to higher counts of detected noise events in high-traffic scenarios compared to the quasi-constant levels predicted by standard models. The significance of this work lies in its demonstration that accounting for vehicle emission distributions is essential for accurately characterizing the temporal structure of road traffic noise. The findings suggest that standard models may misrepresent the frequency and intensity of noise peaks, which are crucial for assessing human health impacts and planning noise mitigation strategies. By providing a freely available software tool that integrates realistic emission variability, the study offers a more robust method for predicting noise events, particularly in contexts where individual noise peaks drive community reaction and sleep disturbance.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• traffic density