A strain reversal method for estimating extension from fragmented rigid inclusions
A strain reversal method for estimating extension from fragmented rigid inclusions
复制标题
用于估计碎片刚性夹杂物延伸的应变反转方法
DOI：
10.1016/0040-1951(81)90108-6
复制
发表时间：
1981
期刊：
影响因子：
2.9
通讯作者：
C. Ferguson
中科院分区：
文献类型：
--
作者：
C. Ferguson
在使用破碎的刚性包裹体（如“拉伸”菱铁矿）进行应变分析的标准方法中，通过比较包裹体的初始长度与其最终（拉伸）长度来估计扩展。这种方法没有认识到，即使在变形开始时出现交叉断裂，也不是所有的伸展都是由碎片的分离记录的。本文提出了一种替代方法，其中拉伸应变被逆转，碎片间的间隙被逐渐关闭，直到包裹体恢复到其原始状态。这允许对“未记录的”扩展（由包裹体附近的不均匀基质流动容纳）以及片段分离记录的扩展进行估计。在模型实验中，碎片包裹体由设置在延性矩阵中的木块表示，从而可以将应变反转方法与标准方法进行比较。前者的平均相对误差（10.2%）远小于后者的平均相对误差（37.2%）。应变反转法的主要缺点是需要收集更多的数据（按正确顺序的片段长度和间隙长度），这些数据必须由计算机处理（提供了一个简短的BASIC程序）。虽然从两种方法得到的应变估计之间没有一对一的对应关系，但从一种方法到另一种方法的近似数值转换是可能的。一个这样的转换（基于对自然边界的研究）已经应用于实验数据，并将标准方法估计中的平均相对误差百分比降低到11.5%。
In the standard method of strain analysis using fragmented rigid inclusions (such as “stretched” belemnites) extension is estimated by comparing the initial length of the inclusion with its final (stretched) length. This approach fails to recognise that not all of the total extension is recorded by separation of the fragments, even when cross-fractures are present at the onset of deformation. This paper presents an alternative approach in which the extensional strain is reversed, the inter-fragment gaps being incrementally closed until the inclusion is restored to its original state. This allows an estimate of the “unrecorded” extension (accommodated by inhomogeneous matrix flow near the inclusion) as well as that recorded by fragment separation.Model experiments, in which the fragmented inclusion is represented by wooden blocks set in a ductile matrix, allow the strain reversal method to be compared with the standard method. The mean percentage relative error is much smaller in the former (10.2%) than in the latter (37.2%). The main disadvantage of the strain reversal method is that it requires the collection of more data (fragment lengths and gap lengths in correct sequence), which must be processed by computer (a short BASIC program is provided). Although there is no one-to-one correspondence between strain estimates derived from the two methods, an approximate numerical transformation from one to the other may be possible. One such transformation (based on a study of natural boudinage) has been applied to the experimental data and reduces the mean percentage relative error in the standard method estimates to 11.5%.