Research on Evaluation Model of Urban Road Scratch Accident Based on SVR
Abstract
In order to effectively evaluate urban road scraping accidents, a road scraping accident evaluation model is established based on historical accident database and environmental parameter indexes. Through variance analysis, four environmental characteristic parameters including the location of the accident area, weather conditions, road conditions and the location of the relationship with the road are extracted as the input of the evaluation model. The functional relationship between the four indicators and the severity of the accident is expressed by the monthly traffic accident index, which is used as the output of the evaluation model. Particle swarm optimization algorithm, CV-K and genetic algorithm are used to calibrate the learning parameters of SVR to obtain the optimal model parameters. On this basis, an urban road scratch accident evaluation model based on SVR regression algorithm is constructed. The model is tested with the collected actual road scraping accident data. The results show that the model has certain accuracy and can be used for risk assessment of urban road scraping accidents based on the above four indicators.
Full Text:
PDFReferences
Ying Yang, Zhu Yun, Wang Lixiang. Diagnostic Model of Situation Awareness for Road Traffic Safety [J]. china safety science journal, 2016, 26(7): 52-57.
CHEN F, WANG J, DENG Y. Road safety risk evaluation by means of improved entropy TOPSIS–RSR. [J] Safety Science, 2015, 79: 39-54.
SUWARTO F, BASUKI K H. The application of traffic conflict technique as a road safety evaluation method: a case study of hasselt intersection[J]. Applied Mechanics&Materials, 2016, 845: 394-403.
WANG J, HUANG H, Road network safety evaluation using Bayesian hierarchical joint model[J]. Accident Analysis &Prevention, 2016, 90: 152-158.
Li Juan, Shao Chunfu. Prediction Model of Traffic Accidents Based on BP Neural Network [J]. Traffic and Computer. 2006, 2(24): 34-37.
Serious, Yang Tianjun, Guan Hongzhi. Decision Support System for Urban Road Traffic Safety Management [J]. Journal of Chang 'an University: Natural Science Edition. 2008, 2(28): 84-88.
Zhang Baoping. Study on Emergency Rescue System for Urban Road Traffic Accidents [D]. Xi’an: Chang’an University, 2012.
Zhao Xuegang. Research on Early Warning and Control of Urban Road Traffic Safety Comprehensive Risk [J]. china safety science journal, 2016, 26(2): 158-163.
He Fangfang, Fang Guoliang, Wu Jianping, et al. Study on the Relationship between Adverse Weather Conditions and Traffic Accidents in Shanghai Area [J]. journal of applied meteorological science, 2004, 15(1): 126-128.
Liu Taian, Zhang Xuping, Wei Guangcun, et al. SVR-based model for underground water level pre-measurement in coal mines [J]. Micro-computer information, 2008, 24(16): 304-306.
Sun Han, Yang Purong, Jin Chenghua. Energy demand forecasting model based on Matlab support vector regression machine [J]. System engineering theory and practice, 2011, 31(10): 2001-2007.
LIANG Y L, REYES M L, LEE J D. Real-time detection of driver cognitive distractionusing support vector machines[J] IEEE Transactions on Intelligent TransportationSystems, 2007, 8(2): 340-350.
Gu Yanping, Zhao Wenjie, Wu Zhansong. Research on Least Squares Support Vector Machine [J]. Qing Hua Da Xue Bao: Ziran Ke Xue Edition, 2010,50 (7): 1063-1066, 1071. 34-37.
DOI: http://dx.doi.org/10.18686/mt.v1i1.1355
Refbacks
- There are currently no refbacks.
