- 无标题文档
查看论文信息

论文中文题名:

 基于链路预测的煤矿瓦斯灾害网络研究    

姓名:

 刘柁    

学号:

 201312692    

保密级别:

 公开    

学科代码:

 0837    

学科名称:

 安全科学与工程    

学生类型:

 硕士    

学位年度:

 2016    

院系:

 能源学院    

专业:

 安全科学与工程    

第一导师姓名:

 魏引尚    

第一导师单位:

 西安科技大学    

论文外文题名:

 Study of Coal Mine Gas Disaster Network based on Link Prediction Theroy    

论文中文关键词:

 网络 ; 瓦斯灾害 ; 节点重要性 ; 链路预测    

论文外文关键词:

 network ; gas disaster ; node importance ; link prediction    

论文中文摘要:
煤矿瓦斯灾害是发生在矿井生产系统内部的一种破坏性很强的灾害,矿井生产系统主要是由井下巷道系统组成,再加上通风条件,整个系统形成瓦斯灾害网络。煤矿瓦斯灾害网络就是用网络的方式来描述煤矿瓦斯灾害,其加深对煤矿瓦斯灾害形成规律的认识。 本文在查阅大量煤矿井下瓦斯事故调查报告和以往对瓦斯事故的研究结果的基础上,使用系统工程的方法对这些资料进行分析总结,构建瓦斯灾害网络。通过对煤矿瓦斯灾害网络进行结构分析:(1)煤矿瓦斯灾害网络的度分布分析表明其内部具有明显的不均匀分布的现象,煤矿瓦斯灾害网络具备了无标度性质;(2)由网络的密度可知:其整体紧密程度较低,结构较为分散,(3)对网络的中心性进行分析:网络中内部联系紧密;网络的重要节点对网络的影响很大。 节点相似性指标可知:网络中具有大度的节点与其它节点更容易有高的相似性;煤矿瓦斯灾害网络跟BA无标度网络有很高的相似性。网络内部具有新加入节点倾向于和度大的节点相连接的有限连接机制,它们更倾向于连接。高阶路径对于整个网络的影响很小;煤矿瓦斯灾害网络的任意两个节点之间走到对方的可能性很高,煤矿瓦斯灾害网络的任意两个节点之间的通勤时间很短。煤矿瓦斯灾害网络局部节点之间的交流程度明显。删除重要节点的连边后,网络的偏好连接性质并没有改变,其他节点之间的连接倾向依然存在。在突出直接连接重要性的同时,也考虑了节点对网络的整体影响,节点重要性在网络中的体现就是节点边的权重。通过实例对网络进行赋权,加权后预测精度明显提高。
论文外文摘要:
Gas disaster is a devastating disaster in mine production system. mine production system is mainly composed of underground tunnel system, adding conditions of ventilation and this is a typical gas disaster network. The network of coal mine gas disaster is describing the coal mine gas disasters in a network way and it deepens the understanding of the law of the formation coal mine gas disaster. Based on a large amount of coal mine investigation report on gas accident and previous gas accident researches in this paper, the gas disaster network is constructed by system engineering method. The network of coal mine gas disaster is describing the coal mine gas disasters in a network way and it deepens the understanding of the law of the formation coal mine gas disaster. Through the network topology analysis of coal mine gas disaster network:(1)It shows the phenomenon of the degree's distribution is uneven, so the gas disaster network own a nature of non-degree.(2)It can be known from the density of the network: its overall compactness is relatively low and the structure is more dispersed . (3)Analysis of the centrality of the network: the important nodes of network have a great impact on the network ,the internal links closely; Node similarity index shows that large degree nodes are more likely to have high similarity than other nodes; coal mine gas disaster network with BA scale free network has a high degree of similarity. The network of gas disaster in coal mine has a high similarity with the BA scale free network. The new join nodes in the network are more likely to be connected to the nodes with large degree in the law of limited connect. It also shows that high order path has a small influence on the whole network. any two nodes in the coal mine disaster network can go to the other side with a high possibility. and the commute time of two nodes is very short. The communications of local nodes in the coal mine gas disaster network are obvious. The preference connection nature of network is not changed and the connection tendency between the nodes still exists after deleting the important nodes 'sides. While highlighting the importance of direct connection, but also consider the overall impact of the node on the whole network. the importance of nodes in the network embodies in the weight of the node- side. The network is weighted by an example and the accuracy of prediction is improved obviously.
中图分类号:

 TD712    

开放日期:

 2016-06-21    

无标题文档

   建议浏览器: 谷歌 火狐 360请用极速模式,双核浏览器请用极速模式