深层碳酸盐岩沉积微相决策树测井识别方法

Log Identification Method of Sedimentary Microfacies in Deep Carbonate Rocks by Decision Tree

深层海相碳酸盐岩储层是油气勘探开发的重要领域,沉积微相识别与划分是开展储层精细描述的基础与关键。为建立以测井资料为基础的复杂礁滩相碳酸盐岩储层沉积微相自动识别方法,以元坝长兴组礁滩相储层为例,结合常规测井、电成像测井及地质资料,明确不同微相的测井响应特征,建立涵盖不同沉积微相岩石物理与测井曲线形态差异的智能识别样本集。针对现有C4.5决策树算法在缺失属性样本分类问题中表现出的建树结构复杂、精度低等问题,提出一种基于贝叶斯原理的适用于缺失值问题的决策树新方法,降低了建树的不确定性,提高了运算效率。经7口井检验,沉积微相识别符合率达到90%以上,新方法丰富并完善了深层海相碳酸盐岩储层勘探开发技术体系,为利用测井资料开展复杂礁滩相碳酸盐岩沉积微相划分提供可靠依据。

Deep marine carbonate reservoirs are important field of oil and gas exploration and development,and the identification and division of sedimentary microfacies is the basis and key for conducting detailed reservoir description.To establish an automatic identification method for sedimentary microfacies in complex reef beach carbonate reservoirs based on logging data,taking the Changxing formation reef beach reservoir in Yuanba as an example.By combining conventional logging,electrical imaging logging,and geological data,the logging response characteristics of different microfacies are clarified,and an intelligent identification sample set covering the differences in rock physics and logging curve morphology of different sedimentary microfacies is established.A new decision tree method based on Bayesian principle is proposed to address the complex structure and low accuracy of the existing C4.5 decision tree algorithm for missing attribute samples,which is suitable for missing value problems.This method reduces the uncertainty of tree construction and improves operational efficiency.Seven wells have been tested and the recognition coincidence rate is over 90%.The new method enriches and improves the exploration and development technology system of deep marine carbonate reservoirs,providing a reliable basis for using logging data to classify complex reef beach carbonate sedimentary microfacies.