水上交通事故组合预测模型的构建及应用
|
摘要
为提高水上交通事故的预测精度,考虑到多个相关因素对事故发生的影响作用,选用背景值优化的MGM(1,n)模型以及SARIMA模型两种单预测模型进行基于IOWA算子的组合,应用该模型拟合分析2008-2015年天津水域水上交通事故的统计数据,并以2016-2017年的数据验证其有效性,通过对比各预测模型的评价结果,验证组合预测模型的适用性和准确性.对天津水域的事故数进行预测,得到近3年水域内事故发生趋势.
To improve the prediction accuracy of water traffic accidents,considering the influence of many related factors on the occurrence of accidents,two single prediction models,MGM(1,n)model optimized by background value and SARIMA model,were selected for combination based on IOWA operator.The proposed model was applied to fit and analyze the statistical data of water traffic accidents in Tianjin waters from 2008 to 2015,and the data from 2016 to 2017 were used to verify its effectiveness.The applicability and accuracy of the combined prediction model were verified by comparing the evaluation results of various prediction models.Finally,the number of accidents in Tianjin waters was predicted,and the trend of accidents in the past three years was obtained.
To improve the prediction accuracy of water traffic accidents,considering the influence of many related factors on the occurrence of accidents,two single prediction models,MGM(1,n)model optimized by background value and SARIMA model,were selected for combination based on IOWA operator.The proposed model was applied to fit and analyze the statistical data of water traffic accidents in Tianjin waters from 2008 to 2015,and the data from 2016 to 2017 were used to verify its effectiveness.The applicability and accuracy of the combined prediction model were verified by comparing the evaluation results of various prediction models.Finally,the number of accidents in Tianjin waters was predicted,and the trend of accidents in the past three years was obtained.
引文
[1]孟永富.天津水域水上交通事故预测和对策研究[D].大连:大连海事大学,2014.
[2]吕晓夫,杨亚东.回归分析方法在船舶交通事故预测中的应用[J].武汉理工大学学报(交通科学与工程版),2006,30(3):546-548.
[3]DEBNATH A K,CHIN H C.Navigational Traffic conflict technique:aproactive approach to quantitative measurement of collision risks in port waters[J].The Journal of Navigation,2010,63(1):137-152.
[4]RAHMAN S.Introduction of bayesian network in risk analysis of maritime accidents in Bangladesh[J].Proceedings of the 1st International Conference on Mechanical Engineering and Applied Science,2017(3):589-593.
[5]李铃铃,仇蕾.基于灰色神经网络组合模型的水上交通事故预测[J].交通信息与安全,2014,32(1):111-113.
[6]赵佳妮,吴兆麟.基于灰色马尔可夫模型的水上交通事故预测[J].大连海事大学学报,2005(4):17-20.
[7]李晓磊,肖进丽,刘明俊.基于SARIMA模型的船舶交通流量预测研究[J].武汉理工大学学报(交通科学与工程版),2017,41(2):329-332,337.
[8]刘寒冰,向一鸣,阮有兴.背景值优化的多变量灰色模型在路基沉降预测中的应用[J].岩土力学,2013,34(1):173-181.
[9]王晓,刘兮,陈华友,等.基于IOWA算子的区间组合预测方法[J].武汉理工大学学报(信息与管理工程版),2010,32(2):221-225.
[10]刘家军,王明军,薛美娟,等.基于Theil不等系数的IOWA算子和马尔科夫链的电量组合模型研究[J].电力系统保护与控制,2011,39(19):30-36,43.
[11]刘香云.基于灰色关联度的道路交通事故组合预测方法研究[D].北京:北京交通大学,2016.
[12]袁宏俊,陈华友.基于改进灰色关联度的IOWA算子最优组合预测模型[J].数学的实践与认识,2010,40(11):145-151.
[13]陆梦.基于DEMATEL与ISM集成方法的水上交通系统脆性影响因素研究[J].上海海事大学学报,2014,35(1):18-22.
[2]吕晓夫,杨亚东.回归分析方法在船舶交通事故预测中的应用[J].武汉理工大学学报(交通科学与工程版),2006,30(3):546-548.
[3]DEBNATH A K,CHIN H C.Navigational Traffic conflict technique:aproactive approach to quantitative measurement of collision risks in port waters[J].The Journal of Navigation,2010,63(1):137-152.
[4]RAHMAN S.Introduction of bayesian network in risk analysis of maritime accidents in Bangladesh[J].Proceedings of the 1st International Conference on Mechanical Engineering and Applied Science,2017(3):589-593.
[5]李铃铃,仇蕾.基于灰色神经网络组合模型的水上交通事故预测[J].交通信息与安全,2014,32(1):111-113.
[6]赵佳妮,吴兆麟.基于灰色马尔可夫模型的水上交通事故预测[J].大连海事大学学报,2005(4):17-20.
[7]李晓磊,肖进丽,刘明俊.基于SARIMA模型的船舶交通流量预测研究[J].武汉理工大学学报(交通科学与工程版),2017,41(2):329-332,337.
[8]刘寒冰,向一鸣,阮有兴.背景值优化的多变量灰色模型在路基沉降预测中的应用[J].岩土力学,2013,34(1):173-181.
[9]王晓,刘兮,陈华友,等.基于IOWA算子的区间组合预测方法[J].武汉理工大学学报(信息与管理工程版),2010,32(2):221-225.
[10]刘家军,王明军,薛美娟,等.基于Theil不等系数的IOWA算子和马尔科夫链的电量组合模型研究[J].电力系统保护与控制,2011,39(19):30-36,43.
[11]刘香云.基于灰色关联度的道路交通事故组合预测方法研究[D].北京:北京交通大学,2016.
[12]袁宏俊,陈华友.基于改进灰色关联度的IOWA算子最优组合预测模型[J].数学的实践与认识,2010,40(11):145-151.
[13]陆梦.基于DEMATEL与ISM集成方法的水上交通系统脆性影响因素研究[J].上海海事大学学报,2014,35(1):18-22.