Accident; analysis and prevention最新文献

{"title":"Sentence-resampled BERT-CRF model for autonomous vehicle crash causality analysis from large-scale accident narrative text data.","authors":"Ruixu Pan, Quan Yuan, Jiaming Cao, Chonghao Zhang, Chengcheng Yu, Qian Liu, Chao Yang, Xingyu Liang","doi":"10.1016/j.aap.2025.108184","DOIUrl":"10.1016/j.aap.2025.108184","url":null,"abstract":"<p><p>As autonomous vehicles (AVs) have been increasingly used, exploring crash causality mechanisms is critical to improving traffic safety related to AVs use. However, existing studies have primarily employed structured data to analyze such causality, while limited efforts have been made to identify causality from unstructured crash narratives, which are featured by data imbalance and small sample sizes. Original crash narratives contain a wealth of latent information about AV crashes that can further the understanding of AV safety. This study proposes a Sentence-resampled BERT-CRF model combined with a DREAM-inspired hierarchical causal attribution framework to systematically analyze the causality mechanisms of AV crashes based on original crash narratives. First, an annotation scheme combining \"BIO\" and \"C-P-R-D\" tags is designed to capture temporal causal relationships in crash narratives and extract causal movement chain (CMC) by the BERT-CRF model. Meanwhile, the data imbalance problem is mitigated by using the sentence-level resampling method, and the results show that the model is 98.03% accurate on the complete dataset, and maintains 96.14% accuracy with a small sample of 10%. Then, a two-tier causal attribution framework(5 categories and 52 elements) inspired by DREAM theory is developed to identify 16 categories of typical scenarios, with rear-end(48.57%) and lane-change (17.04%) collisions as high-risk scenarios. In-depth analysis shows that rear-end crashes are mostly caused by the coupling of a conventional vehicle (CV) following too close to the AV(B5) and the AV's insufficient decisive decision to slow down (A2), while lane-change crashes are associated with the CV's hazardous lane-change (B2) and the delay of AV's intent recognition. The proposed framework bridges the gap between unstructured narratives data and structured causal inference, revealing human-computer interaction deficiencies, environment perception limitations, and roadway facility impacts as the core causal factors. These findings provide data-driven theoretical support for AV manufacturers to optimize sensing algorithms and traffic authorities to develop corresponding regulations.</p>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"221 ","pages":"108184"},"PeriodicalIF":6.2,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144803243","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Twelve years of evidence: modelling the injury severity of single-vehicle collisions pre- and post-20mph (32 km/h) implementation in Edinburgh and Glasgow.","authors":"Torran Semple, Grigorios Fountas","doi":"10.1016/j.aap.2025.108183","DOIUrl":"10.1016/j.aap.2025.108183","url":null,"abstract":"<p><p>This article presents a comprehensive evaluation framework for assessing the collision severity implications of two competing 20mph schemes in the cities of Edinburgh and Glasgow, UK. To achieve this, road traffic collision severity data are statistically analysed to provide a comprehensive overview of road safety pre- and post-20mph implementation in each case city. Advanced discrete outcome models that account for unobserved heterogeneity, namely, Random Parameters Ordered Probit Models with allowances for Heterogeneity in the Means (RPOPHM) of Random Parameters were estimated to analyse the collision-, casualty- and vehicle-specific determinants of collision severity across different speed limit scenarios: Edinburgh pre- (1) and post-20mph (2) and Glasgow pre- (3) and post-20mph (4). The estimation of four separate models facilitated intracity (in other words, pre- versus post-20mph in each case city) and intercity comparisons of collision severity determinants. In terms of intracity findings, the results suggest that the citywide enforcement of 20mph speed limits, as in Edinburgh, has reduced the risk of vulnerable road users, and especially pedestrians, being involved in serious or fatal collisions, relative to other road users. Conversely, the Glasgow models suggest that the Glasgow 20mph scheme, which was less radical and more targeted, has not significantly altered the disproportionately high risk of pedestrians being involved in severe collisions. Policy recommendations are provided, specifically in terms of how varying 20mph approaches may affect existing road safety inequalities.</p>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"221 ","pages":"108183"},"PeriodicalIF":6.2,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144811606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Personalized prediction of unsafe driving behaviors for drivers of dangerous goods transportation trucks based on an attribute graph interaction model.","authors":"Sixian Li, Dalin Qian, Sida Luo, Pengcheng Li, Xinwu Yuan","doi":"10.1016/j.aap.2025.108179","DOIUrl":"10.1016/j.aap.2025.108179","url":null,"abstract":"<p><p>With the growing deployment of dangerous goods transportation trucks (DGTTs), ensuring driving safety has become increasingly important. Given the high disaster potential and hazardous nature of DGTTs, source-level risk control is essential. To support proactive risk management at the source, we propose a method for predicting unsafe driving behaviors before trips. This method leverages trajectory data and intelligent video collected from legally mandated on-board terminals. We adopt a recommender system (RS) approach for its capacity to capture intricate attribute interactions and provide personalized predictions. Drawing an analogy between RS components and our scenario, drivers correspond to users and alarms to items, with their respective attributes forming two sides of the model. We introduce a Bilateral Graph Interaction-based Collaborative Filtering (BGICF) model, enhanced with Adversarial Graph Dropout (AdvDrop). BGICF models both internal coupling and external interaction between attributes. Furthermore, to address attribute popularity bias and improve interpretability in BGICF, we integrate AdvDrop, which constructs bias-mitigating and bias-aware subgraphs using a bias measurement function and optimizes them through adversarial learning. We collected natural driving data from an active safety platform from 23 DGTT companies in Beijing, China, covering over 58 million trajectory points and 211,157 alarm records. Experimental results showed that BGICF-AdvDrop achieves macro precision, recall, F1-score, and accuracy of 0.8202, 0.8114, 0.8101, and 0.8416, respectively, outperforming other models while providing better interpretability.</p>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"221 ","pages":"108179"},"PeriodicalIF":6.2,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144803242","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Causal relationship discovery for highway crash analysis using semi-data-driven Bayesian network.","authors":"Yifan Wang, Xuesong Wang","doi":"10.1016/j.aap.2025.108181","DOIUrl":"10.1016/j.aap.2025.108181","url":null,"abstract":"<p><p>With the widespread application of advanced machine learning techniques, researchers need a more transparent decision-making process. The data-driven causal relationship discovery techniques often lack interpretability. Therefore, a semi-data-driven Bayesian network structure learning algorithm, the Expert Knowledge Constraint-based (EKC) algorithm, is proposed. By integrating expert knowledge with conditional independence tests, the EKC algorithm constructs a causal Bayesian network with a high level of interpretability. The algorithm was applied to a highway safety scene using crash data collected in 2022 from the HuNing Highway in China. The effects of the Bayesian network on variables were estimated using the Bayesian estimation algorithm, and the most dangerous scenarios were ranked using the variable elimination algorithm. Key findings include: (1) date-related variables do not directly affect crashes; (2) unfavorable temperatures, medium-level traffic volumes, and snowy weather conditions are associated with higher crash probabilities; and (3) the highest crash probability occurs under medium traffic volume, cold temperatures, winter season, cloudy weather, morning hours, and weekdays. The EKC algorithm was compared with the Hill Climbing algorithm, Chow-Liu Trees algorithm, and logistic model, demonstrating significant improvements in interpretability while maintaining good fitting scores. Furthermore, the definition framework of model interpretability in traffic crash analytics was discussed, including causality, trust, heterogeneity, transferability, and stability.</p>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"221 ","pages":"108181"},"PeriodicalIF":6.2,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144803231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"In full-touch HMI mode: How does car-following pressure, task complexity, and speed affect driver's visual distraction characteristics?","authors":"Kunchen Li, Menglu Gu, Wei Yuan, Yisi Lu, George Yannis","doi":"10.1016/j.aap.2025.108264","DOIUrl":"https://doi.org/10.1016/j.aap.2025.108264","url":null,"abstract":"<p><p>With the development of vehicle intelligence and automation, full-touch Human-Machine Interaction (HMI) is becoming a vital bridge for transferring substantial information between the driver and the system. Prior research suggests that the full-touch HMI mode may consume more visual resources from the driver, yielding in potential visual distraction. This study aims to investigate the effects of interaction tasks and traffic situations on drivers' visual distraction characteristics in full-touch HMI mode. Under safe conditions, a novel method was employed to create a visually realistic car-following situation. A total of 50 distinct participants were recruited: 30 took part in a real-road experiment, and 30 participated in a driving simulator experiment. Ten participants took part in both experiments, contributing to a total of 60 participant-sessions. Air volume control, temperature control, and call a contact are selected as typical tasks in each experiment. Total off-road glance duration, the number of off-road glances, and the mean off-road glance duration are selected as indicators for visual distraction characteristics. The results show that the mean off-road glance duration is influenced by the car-following pressure but not by the task. The mean glance duration is lower when the driver is following a vehicle, with an average decrease of 21.3%. Complex interaction tasks consume more of the driver's visual attention. Furthermore, higher speeds lead to a decrease in the total off-road glance duration, where participants tend to reduce the duration of each off-road glance while increasing the number of glances to compensate for the increased risk. The study findings can differentiate the visual demands of various HMI tasks and clarify how drivers adapt their gaze behaviors as driving demands change (e.g., car-following pressure), offering safety-related recommendations for drivers.</p>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"223 ","pages":"108264"},"PeriodicalIF":6.2,"publicationDate":"2025-09-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145197466","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Does brain connectivity hold the key to safer roads? EEG-based fatigue detection in young drivers using interpretable deep learning","authors":"Yongjiang Zhou ,&nbsp;Yang Cao ,&nbsp;Hyungchul Chung ,&nbsp;Hanying Guo ,&nbsp;N.N. Sze ,&nbsp;Tiantian Chen","doi":"10.1016/j.aap.2025.108251","DOIUrl":"10.1016/j.aap.2025.108251","url":null,"abstract":"<div><div>Mental fatigue is a significant risk factor for fatal road accidents among young drivers, but its underlying neural mechanisms are still poorly understood. To fill this gap, we explored the neurophysiological basis of driver fatigue using electroencephalography (EEG)-based brain connectivity analysis and designed an accurate, interpretable detection model specifically for young drivers. We collected EEG data from 32 young drivers on real roads and compared them with data obtained in a simulated laboratory environment to verify their reliability. The EEG signals were processed to construct brain functional networks characterised by topological features such as the small-world attribute and node strength. To capture the complex spatial–temporal dynamics of neural activity associated with fatigue, we designed a deep learning model integrating multi-head self-attention with long short-term memory (MHSA-xLSTM). We used the Shapley Additive exPlanation method to analyse the contribution of individual features to driver fatigue recognition, increasing our model’s interpretability. The novel MHSA-xLSTM model achieved an accuracy of 94.39 % (±2.52 %) in detecting mental fatigue amongst young drivers. The small-world attribute and node strength significantly influenced the model’s performance in recognising fatigue. In addition, we found that the brain’s self-regulatory capabilities can mitigate fatigue-related impairments. Young drivers who accumulate driving experience can enhance their driving performance, reducing the likelihood of fatigue-induced impairments and the associated risk of accidents. The findings highlight the potential of EEG-based brain network analysis and advanced deep learning models to enable accurate real-time detection of driver fatigue, informing targeted interventions to reduce accident risks among young drivers.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"223 ","pages":"Article 108251"},"PeriodicalIF":6.2,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145157341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Estimate traffic cyclist crashes using Poisson-Tweedie models","authors":"Ana Karina de Barros Christ ,&nbsp;Carlos Roque ,&nbsp;Filipe Moura","doi":"10.1016/j.aap.2025.108256","DOIUrl":"10.1016/j.aap.2025.108256","url":null,"abstract":"<div><div>Cyclist safety remains a critical issue in urban transportation, where infrastructure configuration and spatial dynamics play a key role in crash occurrence. This study estimates cyclist crash frequencies in Lisbon between 2015 and 2019 using Poisson-Tweedie models, which are well-suited for overdispersed count data. A total of 541 cyclist crashes were analyzed, spatially structured into 250 × 250 meter grid cells and supplemented with covariates such as road length, intersection types, and various cycling infrastructure elements. Two models were developed: a base model with aggregated variables and a disaggregated model distinguishing road types, intersection forms, and cycleway categories. Both models incorporated spatial autocorrelation to account for neighboring effects. The key findings indicate that intersection density and road length are strongly associated with crash frequency, while cycleway length has a more modest yet significant effect. The disaggregated model offers greater interpretability but does not outperform the base model in predictive accuracy or goodness-of-fit, suggesting that a simpler specification may be more effective for policy applications. Elasticity analysis revealed that intersections have the greatest influence on crash risk, followed by road length and cycleways. Spatial predictions aligned with observed crash clusters and highlighted latent high-risk zones, reinforcing the model’s utility for proactive safety planning. The study concludes that improving intersection design is likely to yield greater safety benefits than merely increasing cycling infrastructure length. These results provide actionable insights for data-driven urban mobility planning and emphasize the value of predictive modeling tools for cyclist safety management.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"223 ","pages":"Article 108256"},"PeriodicalIF":6.2,"publicationDate":"2025-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145157340","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trends and disparities in motor vehicle collision injuries in Washington, DC","authors":"Ryan S.D. Calder ,&nbsp;Claire Summa ,&nbsp;Rachel Clark","doi":"10.1016/j.aap.2025.108243","DOIUrl":"10.1016/j.aap.2025.108243","url":null,"abstract":"<div><div>Nonfatal traffic injuries are ~40 times more frequent than traffic fatalities in the United States, but little is known about racial or ethnic disparities in injury-only collisions because commonly used databases report racial/ethnic data only for fatalities. Crash data from police departments (e.g., Vision Zero) are subject to error and bias arising from changing patterns of police intervention and increased use of alternative or automated traffic enforcement. Here, we leverage Trauma Registry data to quantify racial/ethnic, temporal, and spatial patterns of trauma injuries from motor vehicle collisions among adults in Washington, D.C. and compare results to the commonly used Vision Zero database. We report results by year (2019–2023), road user type (motorists, pedestrians, cyclists, and other vulnerable road users), and ZIP code tabulation area (ZCTA) to identify primary contributors to total injury rates and racial/ethnic disparities. Between 2019 and 2023, the overall incidence rate (IR) rose from 69 to 132 per 100,000 persons per year and increased among all road user types and races/ethnicities. Compared to white people, the incidence rate ratio (IRR) was ≥4.3 among Black/African American people and ≥2.9 among Hispanic/Latino people. The IRR between Black/African American vs. white motorists is ≥9.9. Disparities were observed across 21 of 26 ZCTAs, revealing that disparities cannot be explained by solely by higher minority populations in ZCTAs with more hazardous infrastructure. The commonly used Vision Zero dashboard suggests a downward trend in injury-only crashes, but our analysis suggests that this trend is the result of a bias from reduced police intervention.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"223 ","pages":"Article 108243"},"PeriodicalIF":6.2,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145157346","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Bayesian hierarchical non-stationary hybrid modeling for threshold estimation in peak over threshold approach","authors":"Quansheng Yue ,&nbsp;Yanyong Guo ,&nbsp;Tarek Sayed ,&nbsp;Lai Zheng ,&nbsp;Hao Lyu ,&nbsp;Pan Liu","doi":"10.1016/j.aap.2025.108249","DOIUrl":"10.1016/j.aap.2025.108249","url":null,"abstract":"<div><div>The peak over threshold (POT) approach in extreme value theory is widely used for crash risk estimation, but the reliability is often undermined by the subjective and arbitrary selection of the conflict threshold, which can lead to biased outcomes. This study advances the hybrid modeling method for objective threshold determination by developing a non-stationary framework and comprehensively comparing five distinct model structures. The framework allows the threshold to vary with real-time traffic covariates, while the comparison identifies the optimal distribution for general conflicts. The Bayesian hierarchical structure is used to combine traffic conflicts from different sites, incorporating covariates and site-specific unobserved heterogeneity. Five non-stationary BHHM models, including Normal-GPD, Cauchy-GPD, Logistic-GPD, Gamma-GPD, and Lognormal-GPD models, were developed and compared. Traditional graphical diagnostic and quantile regression approaches were also used for comparison. Traffic conflicts collected from three signalized intersections in the city of Surrey, British Columbia were used for the study. The Bayesian approach is employed to estimate the threshold and other parameters in the non-stationary BHHM models. The results show that the proposed BHHM approach could estimate the threshold parameter objectively. The non-stationary BHHM models capture how the threshold varies dynamically across signal cycles in response to changing traffic status. The Lognormal-GPD model is superior to the other four BHHM models in terms of crash estimation accuracy and model fit. The crash estimates using the threshold determined by the BHHM outperform those estimated based on the graphical diagnostic and quantile regression approaches, indicating the superiority of the proposed threshold determination approach. The findings of this study contribute to enhancing the existing EVT methods for providing a threshold determination approach as well as producing reliable crash estimations.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"223 ","pages":"Article 108249"},"PeriodicalIF":6.2,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145147341","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Insights from psychophysiological workload analysis of human-driven vehicle drivers in interactions with autonomous vehicles","authors":"Hoseon Kim ,&nbsp;Jieun Ko ,&nbsp;Cheol Oh ,&nbsp;Hyeonseok Jin","doi":"10.1016/j.aap.2025.108252","DOIUrl":"10.1016/j.aap.2025.108252","url":null,"abstract":"<div><div>This study develops a methodology to evaluate autonomous vehicle (AV) behavior in mixed traffic by incorporating the psychophysiological workload of manually driven vehicle (MV) drivers during vehicle-to-vehicle interactions. The framework is applied to unprotected left turns at unsignalized intersections, analyzing interactions between left-turning AVs and oncoming MVs—situations central to urban safety and mobility. A multi-agent driving simulation (MADS) platform synchronized time and space across two interconnected simulators, enabling real-time analysis of AV–MV trajectories. Electroencephalogram (EEG) signals from MV drivers were used to derive an Anxiety and Nervousness Index (ANI), based on the beta-to-alpha power ratio, to quantify stress and discomfort. Statistical modeling revealed a robust inverse relationship between ANI and post-encroachment time (PET), which represents the temporal separation at the projected conflict point and serves as a surrogate measure of crash potential: driver anxiety declined as PET increased. The rate of decline diminished beyond a PET of 2.7 s, defined as the marginal improvement point (MIP). Guided by this threshold, we propose AV decision protocols: accelerate when PET &gt; 2.7 s to improve flow, and decelerate or yield when PET ≤ 2.7 s to protect human comfort. These findings underscore that AV behavior should integrate human cognitive and psychological responses alongside technical performance. The proposed methodology establishes human-centered behavioral thresholds for AVs in mixed traffic and provides a foundation for improving reliability and promoting safer AV–MV interactions at urban intersections.</div></div>","PeriodicalId":6926,"journal":{"name":"Accident; analysis and prevention","volume":"223 ","pages":"Article 108252"},"PeriodicalIF":6.2,"publicationDate":"2025-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145147337","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}