考虑结构面间距的双因素分组方法研究
发布时间：2018-11-27 16:41
【摘要】：同一期地质运动中岩体所产生的结构面间距一般认为服从负指数分布。目前方法多根据产状单一因素进行分组,分组完成后同一组结构面的间距分布可能非常不规则,针对该问题,结合目前结构面采集到的数据质量高且具有空间位置等特点,提出一种新的双因素分组方法 (a new double-factor clustering method,NDCM)。分组时可综合考虑结构面产状与间距信息,该方法分组初值由模糊K均值法(fuzzy K-means)获取,通过调整隶属度较低的结构面分组编码达到修正结构面间距,使其满足负指数分布的目的。为获得NDCM的分组精度,进行了数值模型计算,该模型中结构面分组情况已知,利用模糊K均值法和NDCM法分别对此模型进行结构面分组。对比结果表明,NDCM法分组精度明显优于模糊K均值分组方法;重叠率越高组数越多这种优势越明显;相比于模糊K均值法,NDCM分组精度平均可以提升0.06左右。
[Abstract]:In the same period of geological movement, the structural plane spacing produced by rock mass is generally regarded as a negative exponential distribution. At present, most of the methods are grouped according to a single factor of occurrence, and the spacing distribution of the same structure plane may be very irregular after the completion of the grouping. In view of this problem, combining with the characteristics of high quality and spatial location of the data collected from the structure plane at present, A new two-factor grouping method (a new double-factor clustering method,NDCM) is proposed. When grouping, the occurrence and spacing information of structural surfaces can be considered synthetically. The initial values of the groups are obtained by the fuzzy K-means method (fuzzy K-means), and the modified structural plane spacing is obtained by adjusting the block coding of the structural planes with lower membership degree. It satisfies the purpose of negative exponential distribution. In order to obtain the grouping accuracy of NDCM, the numerical model is calculated. The structural plane grouping is known in the model. The fuzzy K-means method and the NDCM method are used to group the structural surface of the model. The results show that the accuracy of NDCM is better than that of fuzzy K-means, the higher the overlap rate is, the more the number of groups is. Compared with the fuzzy K-means method, the accuracy of NDCM grouping can be improved by 0.06 or so on average.
【作者单位】： 四川大学水力学与山区河流开发保护国家重点实验室;四川大学水利水电学院;
【基金】：国家重点研发计划(No.2016YFC0600702)~~
【分类号】：TU45
本文编号：2361457
[Abstract]:In the same period of geological movement, the structural plane spacing produced by rock mass is generally regarded as a negative exponential distribution. At present, most of the methods are grouped according to a single factor of occurrence, and the spacing distribution of the same structure plane may be very irregular after the completion of the grouping. In view of this problem, combining with the characteristics of high quality and spatial location of the data collected from the structure plane at present, A new two-factor grouping method (a new double-factor clustering method,NDCM) is proposed. When grouping, the occurrence and spacing information of structural surfaces can be considered synthetically. The initial values of the groups are obtained by the fuzzy K-means method (fuzzy K-means), and the modified structural plane spacing is obtained by adjusting the block coding of the structural planes with lower membership degree. It satisfies the purpose of negative exponential distribution. In order to obtain the grouping accuracy of NDCM, the numerical model is calculated. The structural plane grouping is known in the model. The fuzzy K-means method and the NDCM method are used to group the structural surface of the model. The results show that the accuracy of NDCM is better than that of fuzzy K-means, the higher the overlap rate is, the more the number of groups is. Compared with the fuzzy K-means method, the accuracy of NDCM grouping can be improved by 0.06 or so on average.
【作者单位】： 四川大学水力学与山区河流开发保护国家重点实验室;四川大学水利水电学院;
【基金】：国家重点研发计划(No.2016YFC0600702)~~
【分类号】：TU45
【相似文献】
相关期刊论文 前3条
1 闫长斌;王贵军;常福庆;陈艳国;高旭辉;;基于结构面间距标准的层状岩体结构分类方法探讨与应用[J];资源环境与工程;2010年05期
2 于国新;白明洲;许兆义;冯震;;工程岩体结构面间距空间分布的ARIMA模型及应用[J];工程地质学报;2006年06期
3 ;[J];;年期
，本文编号：2361457
本文链接：https://www.wllwen.com/kejilunwen/diqiudizhi/2361457.html
