RESUMEN

Size-fractionated particulate matter (PM2.5 and PM>2.5) was collected at a traffic site in Kanazawa, Japan in a seasonal sampling work in 2020. Nine polycyclic aromatic hydrocarbons (4- to 6-ring PAHs) were determined in fine and coarse particles. The gas/particle partitioning coefficients (Kp) of the PAHs were calculated from the supercooled liquid vapour pressure and octanol-air partitioning coefficient based on the relationships obtained in previous traffic pollution-related studies. Gaseous PAHs were estimated by Kp and the concentrations of PM and particulate PAHs. The concentrations of total PAHs were 32.5, 320.1 and 5646.2 pg/m3 in the PM>2.5, PM2.5 and gas phases, respectively. Significant seasonal trends in PAHs were observed (particle phase: lowest in summer, gas phase: lowest in spring, particle and gas phase: lowest in spring). Compared to 2019, the total PAH concentrations (in particles) decreased in 2020, especially in spring and summer, which might be due to reduced traffic trips during the COVID-19 outbreak. The incremental lifetime cancer risk (ILCR) calculated from the toxic equivalent concentrations relative to benzo[a]pyrene (BaPeq) was lower than the acceptable limit issued by the US Environmental Protection Agency, indicating a low cancer risk in long-term exposure to current PAH levels. It is notable that gaseous PAHs considerably contributed to BaPeq and ILCR (over 50%), which highlighted the significance of gaseous PAH monitoring for public health protection. This low-cost estimation method for gaseous PAHs can be expected to reliably and conveniently obtain PAH concentrations as a surrogate for traditional sampling in the future work.

Asunto(s)

RESUMEN

Purpose: To investigate the visual function correlates of self-reported vision-related night driving difficulties among drivers. Methods: One hundred and seven drivers (age: 46.06 ± 8.24, visual acuity [VA] of 0.2logMAR or better) were included in the study. A standard vision and night driving questionnaire (VND-Q) was administered. VA and contrast sensitivity were measured under photopic and mesopic conditions. Mesopic VA was remeasured after introducing a peripheral glare source into the participants' field of view to enable computation of disability glare index. Regression analyses were used to assess the associations between VND-Q scores, and visual function measures. Results: The mean VND-Q score was -3.96±1.95 logit (interval scale score: 2.46±1.28). Simple linear regression models for photopic contrast sensitivity, mesopic VA, mesopic contrast sensitivity, and disability index significantly predicted VND-Q score (P<0.05), with mesopic VA and disability glare index accounting for the greatest variation (21 %) in VND-Q scores followed by photopic contrast sensitivity (19 %), and mesopic contrast sensitivity (15 %). A multiple regression model to determine the association between the predictors (photopic contrast sensitivity, mesopic VA, mesopic contrast sensitivity, and disability index) and VND-Q score yielded significant results, F (4, 102) = 8.58, P < 0.001, adj. R2 = 0.2224. Seeing dark-colored cars was the most challenging vision task. Conclusion: Changes in mesopic visual acuity, photopic and mesopic contrast sensitivity, as well as disability glare index are associated with and explain night driving-related visual difficulties. It is recommended to incorporate measurement of these visual functions into assessments related to driving performance.(AU)

Asunto(s)

RESUMEN

BACKGROUND: There is increasing evidence that air pollution and noise may have detrimental psychological impacts, but there are few studies evaluating adolescents, ground-level ozone exposure, multi-exposure models, or metrics beyond outdoor residential exposure. This study aimed to address these gaps. METHODS: Annual air pollution and traffic noise exposure at home and school were modelled for adolescents in the Greater London SCAMP cohort (N=7555). Indoor, outdoor and hybrid environments were modelled for air pollution. Cognitive and mental health measures were self-completed at two timepoints (baseline aged 11-12 and follow-up aged 13-15). Associations were modelled using multi-level multivariate linear or ordinal logistic regression. RESULTS: This is the first study to investigate ground-level ozone exposure in relation to adolescent executive functioning, finding that a 1 interquartile range increase in outdoor ozone corresponded to -0.06 (p < 0.001) z-score between baseline and follow-up, 38 % less improvement than average (median development + 0.16). Exposure to nitrogen dioxide (NO2), 24-hour traffic noise, and particulate matter < 10 µg/m3 (PM10) were also significantly associated with slower executive functioning development when adjusting for ozone. In two-pollutant models, particulate matter and ozone were associated with increased externalising problems. Daytime and evening noise were associated with higher anxiety symptoms, and 24-hour noise with worse speech-in-noise perception (auditory processing). Adjusting for air pollutants, 24-hour noise was also associated with higher anxiety symptoms and slower fluid intelligence development. CONCLUSIONS: Ozone's potentially detrimental effects on adolescent cognition have been overlooked in the literature. Our findings also suggest harmful impacts of other air pollutants and noise on mental health. Further research should attempt to replicate these findings and use mechanistic enquiry to enhance causal inference. Policy makers should carefully consider how to manage the public health impacts of ozone, as efforts to reduce other air pollutants such as NO2 can increase ozone levels, as will the progression of climate change.

RESUMEN

Most current methods use spatial-temporal graph neural networks (STGNNs) to analyze complex spatial-temporal information from traffic data collected from hundreds of sensors. STGNNs combine graph neural networks (GNNs) and sequence models to create hybrid structures that allow for the two networks to collaborate. However, this collaboration has made the model increasingly complex. This study proposes a framework that relies solely on original Transformer architecture and carefully designs embeddings to efficiently extract spatial-temporal dependencies in traffic flow. Additionally, we used pre-trained language models to enhance forecasting performance. We compared our new framework with current state-of-the-art STGNNs and Transformer-based models using four real-world traffic datasets: PEMS04, PEMS08, METR-LA, and PEMS-BAY. The experimental results demonstrate that our framework outperforms the other models in most metrics.

RESUMEN

Accurate and timely forecasting of traffic on local road networks is crucial for deploying effective dynamic traffic control, advanced route planning, and navigation services. This task is particularly challenging due to complex spatio-temporal dependencies arising from non-Euclidean spatial relations in road networks and non-linear temporal dynamics influenced by changing road conditions. This paper introduces the spatio-temporal network embedding (STNE) model, a novel deep learning framework tailored for learning and forecasting graph-structured traffic data over extended input sequences. Unlike traditional convolutional neural networks (CNNs), the model employs graph convolutional networks (GCNs) to capture the spatial characteristics of local road network topologies. Moreover, the segmentation of very long input traffic data into multiple sub-sequences, based on significant temporal properties such as closeness, periodicity, and trend, is performed. Multi-dimensional long short-term memory neural networks (MDLSTM) are utilized to flexibly access multi-dimensional context. Experimental results demonstrate that the STNE model surpasses state-of-the-art traffic forecasting benchmarks on two large-scale real-world traffic datasets.

RESUMEN

Within the context of smart transportation and new infrastructure, Vehicle-to-Everything (V2X) communication has entered a new stage, introducing the concept of holographic intersection. This concept requires roadside sensors to achieve collaborative perception, collaborative decision-making, and control. To meet the high-level requirements of V2X, it is essential to obtain precise, rapid, and accurate roadside information data. This study proposes an automated vehicle distance detection and warning scheme based on camera video streams. It utilizes edge computing units for intelligent processing and employs neural network models for object recognition. Distance estimation is performed based on the principle of similar triangles, providing safety recommendations. Experimental validation shows that this scheme can achieve centimeter-level distance detection accuracy, enhancing traffic safety. This approach has the potential to become a crucial tool in the field of traffic safety, providing intersection traffic target information for intelligent connected vehicles (ICVs) and autonomous vehicles, thereby enabling V2X driving at holographic intersections.

RESUMEN

As a fundamental element of the transportation system, traffic signs are widely used to guide traffic behaviors. In recent years, drones have emerged as an important tool for monitoring the conditions of traffic signs. However, the existing image processing technique is heavily reliant on image annotations. It is time consuming to build a high-quality dataset with diverse training images and human annotations. In this paper, we introduce the utilization of Vision-language Models (VLMs) in the traffic sign detection task. Without the need for discrete image labels, the rapid deployment is fulfilled by the multi-modal learning and large-scale pretrained networks. First, we compile a keyword dictionary to explain traffic signs. The Chinese national standard is used to suggest the shape and color information. Our program conducts Bootstrapping Language-image Pretraining v2 (BLIPv2) to translate representative images into text descriptions. Second, a Contrastive Language-image Pretraining (CLIP) framework is applied to characterize not only drone images but also text descriptions. Our method utilizes the pretrained encoder network to create visual features and word embeddings. Third, the category of each traffic sign is predicted according to the similarity between drone images and keywords. Cosine distance and softmax function are performed to calculate the class probability distribution. To evaluate the performance, we apply the proposed method in a practical application. The drone images captured from Guyuan, China, are employed to record the conditions of traffic signs. Further experiments include two widely used public datasets. The calculation results indicate that our vision-language model-based method has an acceptable prediction accuracy and low training cost.

RESUMEN

Introduction: Road traffic injuries stands as a major concern globally, as they result in significant loss of life, economic impact, and erode trust in government and societal safety. The influence of weather on road traffic safety is undeniable, impacting road conditions, individuals, and vehicles. However, the specific influence of weather on road traffic casualties has seldom been explored. Method: This study assesses the effect of weather factors on road traffic casualties in China from 2006 to 2021. Vector error correction models (VECM) were utilized to determine the Granger causality between weather factors and covariates. Furthermore, panel autoregressive distribution lag models (ARDL) were applied to quantify the association between weather factors and road traffic casualties. Results: The findings indicate that rainfall and temperature exert a short-term negative impact on casualty risk, which intriguingly becomes positive in the long term. A standout discovery is the significant role of health investments, which are shown to reduce casualty numbers in both the short and long-terms. In the long run, the gross domestic product significantly enhances casualties, while expressway mileage notably decreases them. Conclusions: These results demonstrate the significant influence of weather on road traffic casualties and highlight the critical roles played by factors such as gross domestic product, health investment, and expressway mileage. The evidence presented in the study underscores the urgent need for more effective strategies to mitigate road traffic casualties. Thus, some effective measures are proposed to reduce road traffic casualties. This study is conducive to the improvement of traffic in severe weather in China and provides guidance for traffic management departments.

RESUMEN

Currently, one of the serious problems in the world is transport-related air pollution. Air pollution from vehicles was not considered properly in the plan, design, and management system of the roads in Addis Ababa city, to solve the problem of traffic congestion. It is influenced by different factors such as road geometry, road surface type, traffic congestion, traffic characteristics, fuel type, and meteorological parameters. This study was aimed to investigate the contribution of congested traffic flow conditions on air pollution at intersection points in Addis Ababa city. To carry out the study, seven intersection points were selected in the city randomly. For the measurement technique of carbon monoxide (CO) and sulfur dioxide (SO2) gases, a portable Aeroqual series 500 instrument was exercised, and for the particular matter (PM2.5 and PM10), Light Amplification by Stimulated Emission of Radiation (LASER) was adapted. To count the number of traffic on the site, a video camera method was utilized. For the data analysis techniques, a multiple linear regression method was practiced, which is very powerful for multiple independent variables with a single dependent one. The study revealed that the level of traffic congestion at each selected intersection is at a severe level. During the congested time, as a base of the steady condition, the pollutants concentration of CO, SO2, PM2.5, and PM10 were increased by an average of 19.10, 51.61, 33.83, and 29.07 % respectively. Based on the analyzed results, when the concentration of pollutants increases, the traffic volume, percentage of heavy vehicles, green time, and approach grade are also increased. On the other hand, the concentration of pollutants decreased when the lane width and wind speed increased. Therefore, any concerned bodies in and around Addis Ababa city should take remedial measures to decrease the emission of the concentration of the pollutants. This investigation is very important for policymakers, municipalities, and planners to do any activities in the city. Researchers can also use it as a reference, for further investigations and remedial measurements.

RESUMEN

INTRODUCTION: Traffic-related air and noise pollution are important public health issues. The aim of this study was to estimate their effects on allergic/respiratory outcomes in adult and elderly subjects. MATERIALS AND METHODS: Six hundred and forty-five subjects living in Pisa (Tuscany, Italy) were investigated through a questionnaire on allergic/respiratory symptoms and diseases. Traffic-related air pollution and noise exposures were assessed at residential address by questionnaire, modelled annual mean NO2 concentrations (1 km and 200 m resolution), and noise level over a 24-h period (Lden). Exposure effects were assessed through logistic regression models stratified by age group (18-64 years, ≥65 years), and adjusted for sex, educational level, occupational exposure, and smoking habits. RESULTS: 63.6% of the subjects reported traffic exposure near home. Mean exposure levels were: 28.24 (±3.26 SD) and 27.23 (±3.16 SD) µg/m3 for NO2 at 200 m and 1 km of resolution, respectively; 57.79 dB(A) (±6.12 SD) for Lden. Exposure to vehicular traffic (by questionnaire) and to high noise levels [Lden ≥ 60 dB(A)] were significantly associated with higher odds of allergic rhinitis (OR 2.01, 95%CI 1.09-3.70, and OR 1.99, 95%CI 1.18-3.36, respectively) and borderline with rhino-conjunctivitis (OR 2.20, 95%CI 0.95-5.10, and OR 1.76, 95%CI 0.91-3.42, respectively) only in the elderly. No significant result emerged for NO2. CONCLUSIONS: Our findings highlighted the need to better assess the effect of traffic-related exposure in the elderly, considering the increasing trend in the future global population's ageing.

Global population is ageing.Allergic diseases are globally widespread even on adult population.The susceptibility due to ageing may increase the impact of air pollution on the elderly.Traffic-related air and noise pollution affects allergic status of the elderly.

Asunto(s)

RESUMEN

To explore the effect of interchange spacing on drivers' visual characteristics in the merging areas of interchange, a high-density group of five interchanges on the expressway of Chongqing, China, was selected as the test site. An naturalistic driving test was conducted with 47 participants, and the Tobii Glasses II portable eye tracker was used to collect gaze data during driving. The drivers' fixation field was divided into six areas by applying a K-means dynamic clustering algorithm combined with the actual scenario. Markov chains were used to calculate the drivers' gaze transition probability matrices under different driving conditions, and the analysis of gaze transition behaviors was directed at common spacing interchanges, small spacing interchanges, and composite interchanges. Under the ramp-mainline condition, drivers' fixations were primarily concentrated on the near ahead and the left side areas, with higher rates of repeated fixations on the left rearview mirror and left-side line areas. The average value of fixation duration, saccade distance, and saccade speed of small spacing interchange is higher than common spacing interchange. Additionally, under the mainline condition, the probability of one-step transition and repeated fixation rates significantly increased for the right-side lane areas, and the average values of fixation index and saccade index of small spacing interchange are lower than those of common spacing interchange. The results show that the highest probabilities of repeated fixation by drivers occurred in the near ahead and far ahead areas in the interchange merging areas. Insufficient spacing resulted in more frequent occurrences of zero values in one-step transition probability matrices. The research conclusions provide theoretical support for the optimal design and safe operation of the merging area of high-density interchange group of urban expressway.

RESUMEN

Vehicle trajectory data are invaluable for driving behaviour and traffic flow modelling studies, especially at the microscopic level. However, existing public vehicle trajectory datasets only provide data with inherent errors and lack the corresponding ground truth. This study presents a comprehensive vehicle trajectory dataset obtained using both drone and high-precision Global Navigation Satellite System (GNSS) receiver technologies with an error of less than 5 cm. The dataset contains 70 complete trajectories with a total of 10,840 data points and an average length of 48.4 m. This includes 27 left-turn trajectories, 27 through trajectories and 16 right-turn trajectories. The trajectories collected by the centimetre-level precision GNSS receiver can be regarded as the ground truth of the trajectories extracted by the drone video. Researchers can use these two trajectory datasets to analyse driving behaviour at interactive scenarios, validate and calibrate microscopic traffic flow models, and validate trajectory reconstruction methods.

RESUMEN

This article describes a dataset comprising 16,426 real-world urban photographs, capturing vehicles, cyclists, motorbikes, and pedestrians across Morning, Evening, and Night scenes. The dataset is valuable for machine learning tasks in traffic analysis, urban planning, and public safety. It enables the development and validation of algorithms for pedestrian detection, traffic flow analysis, and infrastructure optimization. Our main goal is to assist academics, urban planners, and decision-makers in creating sophisticated models for pedestrian safety, traffic control, and accident avoidance. This dataset is a useful resource for training and verifying algorithms targeted at boosting real-time traffic monitoring systems, optimizing urban infrastructure, and raising overall road safety because of its high variability and significant volume. This dataset represents a major advancement for smart city projects and the creation of intelligent transportation systems.

RESUMEN

Extensive experimental analyses concerned with Adaptive Cruise Control (ACC) have clearly shown that such systems have failed to deliver the promise of safe and traffic-flow effective car-following. On the contrary, large reaction times and poor string stability performances characterize commercial ACCs. While a huge research line is investigating the introduction of communication among vehicles to overcome the mentioned limitation, market adoption of connectivity-enhanced vehicles is struggling. In this context, an alternative approach based on multiple vehicle anticipation using RADAR only has emerged. Multianticipation is definitely not a new concept within the transportation community. However, until now, it was mainly associated with human driving. In the present manuscript, we demonstrate instead how, at least, one vehicle manufacturer has implemented multianticipation on a commercial vehicle. Following an in-house carried out testing campaign, we give an experimental characterization of the functioning of such a system including the potential impact on the flow and safety using a state-of-the-art fuzzy-logic safety performance model. The first results demonstrate that the vehicle under test reacted to one additional vehicle in front of the leader vehicle. Moreover, the actual realization appears to mainly target safety applications whereas there is only a marginal benefit on the string stability characteristics of the system. While we recorded a marginal string stability improvement (about 10 %), the minimum TTC was twice as large when multianticipation occurred with respect to the cases when that was not activated. Relevant Fuzzy Surrogate Safety Metrics further supported the safety argument.

RESUMEN

The study investigates the sources of metals in urban road dusts using elemental concentration and Pb isotopic ratios. The elemental concentrations are also utilized to determine the present heavy metal emissions as well as projected emissions till 2045. Bayesian mixing model for source apportionment highlights the significant contributions of both exhaust and non-exhaust sources to the metal-enriched urban road dusts, with each contributing approximately 40 %. Emission analysis reveals that India's projected electric vehicle (EV) penetration may not be sufficient to suppress the metal emissions from vehicular exhausts. Further challenge is posed by high metal concentrations in the non-exhaust sources, that dominates the emission of some metals compared to exhaust sources. If the metal concentrations remain unchanged, the emission analysis predicts alarming increases in total emissions from all the exhaust and non-exhaust sources by 174 %, 176 %, 163 % and 184 % for Ni, Cu, Zn and Pb, respectively, from 2022 to 2045. Thus, it is crucial to reduce the metal concentrations in traffic emission sources and also impose better regulatory measures to improve the urban metal pollution scenario.

RESUMEN

Objectives: This study aimed to determine the impact of standing electric scooters on maxillofacial on the Italian territory. Methods: The authors analyzed the epidemiology of the injuries to define electric mobility's impact on maxillofacial surgery practice. For this retrospective cohort study, data were collected by unifying the standing e-scooter-related fractures database from 10 Italian maxillofacial surgery departments. The reference period considered was from January 2020 to December 2023. The main data considered included age, gender, type of access, time slot of admission, type of admission, alcohol level, helmet use, dynamics of the accident, and area of the fracture. Results: A total of 79 patients were enrolled. The average age of the participants was approximately 31 years. The blood alcohol level was found to be above the Italian norm in 15 cases (19%). Only one patient wore a helmet. The most affected facial third was the middle one with 36 cases (45.5%), followed by the lower one (31, 39.3%). The most recurrent patterns were fractures of the orbito-malar-zygomatic complex (15, 19%), followed by multifocal (bifocal, trifocal) fractures of the mandible (14, 17.5%). Conclusions: This study demonstrated how maxillofacial fractures related to the use of electric scooters are associated with complex patterns, associated with a high rate of post-surgical aftermaths.

RESUMEN

In response to growing public health concerns, governments worldwide have implemented various nutrition labelling schemes to promote healthier eating habits. This study aimed to assess the consistency and effectiveness of these labels in an out-of-home context, specifically focusing on restaurant, hospitality, and institutional food service settings. In total, 178 different dishes from Spain were analysed using labels from the Mazocco method, the UK's traffic light system, the Health Star Rating (Australia), Nutri-Score (France), multiple traffic lights (Ecuador), and warning labels (Chile and Uruguay). The results demonstrated a generally low level of agreement among these labels (K < 0.40), indicating notable variability and a lack of consensus, which could hinder consumers' ability to make informed food choices in out-of-home settings. Nutri-Score classified the highest number of dishes as unhealthy (38%). This study underscores the need for an easy-to-understand labelling system tailored to each country's culinary and socio-cultural contexts to improve consumer decision-making in various dining environments. Future research should focus on developing and testing qualitative methods to more accurately gauge the nutritional quality of cooked dishes in diverse out-of-home settings, thereby enhancing public health outcomes. By addressing the specific needs of the home, restaurants, hospitality, and institutional food services, tailored labelling schemes could significantly improve consumers' ability to make healthier food choices.

Asunto(s)

RESUMEN

Taxis are essential to economic growth due to the ease and comfort they offer passengers. This is evident as most cities, especially in Africa, are dominated by taxis providing door-to-door services. However, their susceptibility to road traffic accidents (RTA) raises serious concerns due to their risky driving behaviours. In contrast, studies on taxi driver involvement in RTA due to their risky driving behaviours are sparse, especially in African countries. Consequently, the study examined the relationship between risky driving behaviour and traffic accident involvement among Nigerian commercial taxi drivers using the structural equation modeling (SEM) approach. Prior to the structural model analysis, the modified driver behaviour questionnaire (DBQ) was valid. This was assessed through the measurement model, and the results showed that the composite reliability, average variance extracted, and discriminant validity were greater than 0.7, greater than 0.5, and less than 0.90, respectively. Furthermore, the structural equation modeling results show that the driving violation and driving error constructs influenced road traffic accidents among taxi drivers, while inattention error was insignificant (p > 0.05). Although driving violations and errors significantly increase the chances of RTA among taxi drivers, driving violations had a more substantial influence than driving errors. Also, the regression coefficient indicates the risky driving behaviour of commercial taxi drivers accounts for 5.2 % of the RTAs in Nigeria. This research contributed to validating the DBQ for commercial taxi drivers in Nigeria, examining the influence of their driving violations, driving errors, and inattention errors on accident involvement and that inattention error may not necessarily influence accidents, which will aid policymakers in formulating mitigative strategies for RTA reductions. Moreso, it will guide driver trainers in curriculum development for specific commercial taxi driver training.

RESUMEN

In the real world, the correct recognition of traffic signs plays a crucial role in vehicle autonomous driving and traffic monitoring. The research on its adversarial attack can test the security of vehicle autonomous driving system and provide enlightenment for improving the recognition algorithm. However, with the development of transportation infrastructure, new traffic signs may be introduced. The adversarial attack model for traffic signs needs to adapt to the addition of new types. Based on this, class incremental learning for traffic sign adversarial attacks has become an interesting research field. We propose a class incremental learning method for adversarial attacks on traffic signs. First, this method uses Pinpoint Region Probability Estimation Network (PRPEN) to predict the probability of each pixel being attacked in old samples. It helps to identify the high attack probability regions of the samples. Subsequently, based on the size of high probability pixel concentration area, the replay sample set is constructed. Old samples with smaller concentration areas receive higher priority and are prioritized for incremental learning. The experimental results show that compared with other sample selection methods, our method selects more representative samples and can train PRPEN more effectively to generate probability maps, thereby better generating adversarial attacks on traffic signs.

RESUMEN

The vehicle noise source strength prediction model is a crucial component in the field of traffic noise prediction. Despite the establishment of noise source strength localized models in various countries, the theoretical underpinnings of the sound power level models within these frameworks remains unclear. This study addresses this gap by analyzing the correlation between vehicle noise and energy consumption. An energy-based source strength model framework (E-SSIM) is proposed, focusing on developing nonlinear models for basic noise level. E-SSIM is built on acoustical principles and the energy flow of vehicles, integrating noise and energy consumption through the application of multivariate regression theory, characterized by a transient or simplified mathematical framework. Furthermore, sensitivity analysis and road experiments are conducted to validate the proposed framework. The findings reveal that E-SSIM effectively integrates vehicle energy flow and principles of acoustics, thereby providing a theoretical foundation for the logarithmic mathematical structure in classical noise source strength models. The study reveals that in low-speed driving conditions (17-40 km/h), the sensitivity of noise energy to aerodynamic drag energy consumption reaches its peak. Specifically, the sensitivity of E-SSIM, as assessed by the A-weighted sound level, progressively decreases with increasing speed. On the contrary, for the Z-weighted sound level, the sensitivity initially decreases before rising again, reaching its peak stability and robustness at a speed of 23.8 km/h. E-SSIM exhibits superior precision in predicting A/Z-weighted sound pressure levels. Compared to classic logarithmic structural prediction models, the mean absolute percentage error of E-SSIM was reduced by 4.19% and 0.07%. Compared to typical models such as ASJ developed by the Acoustical Society of Japan and CNOSSOS-EU used by the European Commission, E-SSIM yielded a mean absolute percentage error reduction of 68% and 67%. Interestingly, as vehicle internal energy consumption increases, the prediction deviations of E-SSIM, ASJ, and CNOSSOS-EU gradually decrease, possibly because vehicle operating conditions approach stability. E-SSIM can utilize abundant vehicle data to develop generic models, promoting the advancement of noise prediction.

Asunto(s)