基于统计学的火山岩储层流体识别新技术

A NEW TECHNOLOGY OF FLUID IDENTIFICATION FOR VOLCANIC RESERVOIR BASED ON STATISTIC

目前对于火山岩储层流体的识别技术还处于不断探索的阶段,特别是对于气藏的流体识别,更是目前开发的一个热点。常用的流体识别方法有图版法、曲线重叠法、三孔隙度组合法、比值法和参数法等。虽然每种方法都能从不同的侧面反映流体的性质,但是由于火山岩储层岩石成份复杂多变,利用上述五种方法分别会出现试采数据点少,曲线刻度单一,岩石骨架参数求取不准确,测井序列响应不敏感和数据单一等诸多问题。因此这里以松辽盆地南部英台气田为例,避开火山岩气层岩石成份复杂的问题,通过统计学的原理放大流体性质,提出一种新的针对火山岩气层流体识别技术,从而对流体进行有效识别。

At present, the fluid identification technology of volcanic reservoir still on the develop stage, even more gas identification is a hotspot especially for during exploitation. There are 5 common fluid identification methods such as chart, curve overlap, 3 porosity combination, ratio and parameter etc. Although each method could reflect fluid property from different aspects, there are still many problems using these methods, for example, few producing test data, simplex scale, inaccurate frame parameters, not sensitive for log response, and simple data, etc because of the diversification and complexity of volcanic rock component. This paper will focus on how to identify fluid effectively using enlarge principle of statistic which avoid the complex rock component and put forward a new fluid identification skill for volcanic gas reservoir take Yingtai gas field for example.