基于灰色关联度分析法和主成分分析法对泥页岩储层评价方法的探讨

Evaluation of Shale Reservoirs Based on Grey Relation Analysis and Principal Component Analysis

泥页岩储层的影响因素众多,明确各因素间的内部联系才能更好地评价储层。为对比灰色关联法和主成分分析法在储层评价中的适用性,本文选取鄂尔多斯盆地东缘上古生界Y88井丰富的储层数据进行分析。通过对比两种多元统计方法的原理、算法和结果,认为主成分分析法把各项因子视为离散变量,通过降维后用三个综合评价指标来反映储层的优劣,是对各项指标的概括性分析;灰色关联法将参数分为参考数据列和比较数据列,参考数据列的有机碳含量高低与储层的优劣成正相关,参考数据列对评价对象具有可测性。因此,若是对评价对象的概括性分析,用主成分分析法比较合适;若是对评价对象的可测性分析,如有机碳含量的高低决定能不能形成泥页岩储层,则用灰色关联法评价更合理。

Shale reservoirs are generally affected by various factors,and it is significant to clarify the internal relationship between these factors for reservoir evaluation.This work selected the well Y88 with rich exploration data of Upper Paleozoic in the eastern margin of Ordos Basin to compare the applicability of grey relation analysis with principal component analysis.The comparison of the principles,algorithms and results of the two multivariate statistical methods shows that,the principal component analysis regards various factors as discrete variables,and three comprehensive evaluation indexes are used to reflect the advantages and disadvantages of the reservoir after dimension reduction,which is a general analysis of each index.The grey relation method divides the parameters into reference data series and comparative data series,and the organic carbon content of the reference data series is positively related to the quality of the reservoir,implying that the reference data series is testable for the evaluation objects.Therefore,the principal component analysis is suitable for the general analysis of evaluation objects.For testability analysis of the evaluation objects,for example,the organic carbon content determines whether the shale reservoir can be formed,the grey correlation method is more reasonable.