沃尔什数据离散化方法及在油藏描述中的应用

Walsh data discretization method and its application in reservoir description

（1.中国石油大学石油天然气成藏机理教育部重点实验室，北京102249；2.中国石油大学(北京)资源与信息学院，北京102249；3.中国石油大港油田集团有限公司测井公司，天津300280）

Key Laboratory of Education Ministry for Accumulation Mechanism of Oil & Gas,China University of Petroleum (Beijing),Beijing 102249,China

在油藏描述中必须对具有时序特征的连续型测井数据进行数据离散化，以往的一些离散方法需要人为地规定划分维数，或者需要预先给定一个参数。针对此，提出了一种在沃尔什变换基础上的连续数据离散化方法。该方法不需要给定划分维数，将连续数据经过沃尔什低通滤波后划分成离散区间，再用整数和符号标示。在应用沃尔什变换实现测井数据的地层分层及离散化数据中，分层的精细度可以通过调整截止序率以及截止值来控制。实际应用结果证实，该方法简单，易于实现，计算效率高。

It is very important to discretize continuous well logging data with time sequence characteristic in reservoir description.In some conventional discretization methods,dimensions need to be divided by human or need a given parameter.Aiming at this situation,the continuous data discretization method based on Walsh transform was proposed.Without a given dimension, this method divides the data filtered by Walsh low-pass filtering into discretization intervals and then marks the data by integers and symbols.In realizing stratigraphic classification and discretizing data through logging data by using Walsh transform,the layering fineness can be controlled by adjusting cut-off order and cut-off value. The actual application shows that the method is simple in operation with high computation efficiency.