摘要
为了对机场集疏运路网可靠性进行评估,通过分析其节点重要度差异及道路等级划分明显特性,得知机场专用道和衔接道的路网组合形式区别于城市路网。为此,在传统复杂网络特征指标算法上赋予对应的边权重,并重新定义其指标及计算方法,从而提出了改进的机场集疏运复杂网络模型。通过ArcGIS、模拟实验等方法结合北京首都机场集疏运路网实例,验证了该模型的可行性。其结果表明:传统复杂网络度分布服从泊松分布,而改进的复杂网络节点强度分布服从幂律分布,能较好的反映出机场集疏运路网的无标度特性,科学性、准确性较强;其次,改进的复杂网络模型加权集聚系数为0.015,仅占传统复杂网络模型集聚系数的20%,平均加权距离10.38,高于平均路径长度1.15,与机场集疏运道路等级划分特性明显更契合,但无标度特性并不明显。同时,通过模拟攻击实验发现节点强度对路网可靠性的影响比节点重要度大,通过对该类节点加以重点保护能有效提高机场集疏运路网的可靠性。
In order to evaluate the reliability of airport collection and distribution,the differences of node importance and road classes are analyzed,which shows that the connections of the network of airport designated roads and service roads differs from their urban counterpart.Therefore,edge weights are given based on feature index algorithm of the traditional complex network,and an improved complex network model is proposed.The model is verified by through simulation tests,combining with a case study for the road network of Beijing Capital Airport.The results show that the degree distribution of traditional complex network model obeys the Poisson distribution,while the improved complex network model follows the power-law distribution,which fits the scale-free property of airport road network better.Moreover,the cluster coefficient of the improved complex network model is 0.015,only 20% of the traditional complex network model,and the average weighted length is 10.38,which is 1.15 longer than the average path length.It fits the functional classification of airport roads,but the scale-free property is not significant.It is also found through the attacking simulations that node strength is more important for network reliability than the node degree,and the reliability of airport road network can be effectively improved by enhancing the protection of this type of nodes.
出处
《交通信息与安全》
2015年第5期49-54,共6页
Journal of Transport Information and Safety
基金
国家自然科学基金项目(批准号:51508274)
江苏省自然科学基金项目(批准号:BK20140821)资助
关键词
交通管理
路网可靠性
机场集疏运
复杂网络理论
无标度特性
traffic management
road network reliability
airport collection and distribution
complex network theory
scale-free property