改进近邻函数准则下煤尘爆炸危险等级评价研究

发布时间：2018-02-23 17:27

本文关键词： 煤尘爆炸危险等级评价火焰长度近邻函数邻近系数　出处：《中国安全科学学报》2015年12期 　论文类型：期刊论文

【摘要】：为快速有效评价煤尘爆炸危险等级,进行激波管道改造,实现40 m范围内火焰长度测试,并以火焰长度划分爆炸危险等级。选取10项煤尘爆炸危险性影响指标,改进近邻函数准则评价算法,采用欧氏距离法构造隶属关系矩阵,以隶属关系矩阵替代近邻函数准则算法中的距离矩阵,增强煤尘爆炸危险性影响指标间的关联性。计算影响指标近邻函数矩阵与类内类间连接损失量,得到10种待评价煤样爆炸危险等级。结果表明,评价结果与实际情况基本吻合,表明改进算法适用于爆炸危险等级评价。
[Abstract]:In order to evaluate coal dust explosion hazard grade quickly and effectively, to carry out shock wave pipeline transformation, to realize flame length test within 40 m, and to divide explosion hazard grade according to flame length, 10 items of coal dust explosion hazard influence index are selected. The Euclidean distance method is used to construct the membership relation matrix, and the distance matrix in the nearest neighbor function criterion algorithm is replaced by the membership relation matrix. The correlation between the influence indexes of coal dust explosion hazard is enhanced. The loss between the nearest neighbor function matrix of the impact index and the inter-class connection is calculated, and the explosion hazard grade of 10 kinds of coal samples to be evaluated is obtained. The results show that the evaluation results are basically consistent with the actual situation. It is shown that the improved algorithm is suitable for the evaluation of explosion hazard grade.
【作者单位】：辽宁工程技术大学安全科学与工程学院;矿山热动力灾害与防治教育部重点实验室;
【基金】：国家自然科学基金资助(51204089,51274115) 辽宁省教育厅科研项目(L2014131)
【分类号】：TD714.5

【参考文献】