自动识别多期断层擦痕的一种应力反演算法

由于地质历史上构造应力场的演变,多期断层擦痕数据的存在是应力反演所面临的普遍性问题.以往提出处理多期断层擦痕的应力反演算法都基于硬划分,忽视了数据自身的不确定性,并且一些只是传统的、处理一期断层擦痕的算法的简单延拓.在Fry (1999)的sigma空间里,同期断层擦痕向量具有统一的线性分布趋势,多期断层擦痕向量具有不同的线性分布趋势.在此基础上,本文提出利用模糊线性聚类法来识别多期断层擦痕向量的线性结构.这种算法不仅可以弥补以往算法的上述缺陷,还具有自动、直接、有效,且计算量也较小的优点.

A Stress Inversion Procedure for Automatic Recognition of Polyphase Fault/Slip Data Sets

The presence of polyphase fault/slip data caused by the variability of tectonic stress fields in the geological history is a general problem in stress inversion. Algorithms previously presented for separation of polyphase fault/slip data sets are based on hard subdivision and underestimate the intrinsic nondeterminacy of data. In some of these algorithms, the classic algorithm for one phase fault/slip data is embedded. In Fry's (1999) sigma space, the vectors of one phase fault/slip data must have a linear tendency whe reas the vectors of polyphase data have multiple linear tendencies. The authors herein apply modern fuzzy clustering analysis to detec ting the linear structures of fault/slip data. The algorithm used here considers the nondeterminacy of data and hence can overcome the shortcomings of existing algorithms. It is automatic, direct and effective,and needs less running time.