【作者】 刘斌；

【导师】 沈晓华； 田家祥；

【作者基本信息】 浙江大学 ， 地质工程（专业学位）， 2015， 硕士

【摘要】 大庆长垣油田近几年积累了大量的测井曲线,如何高效、快速和经济的从诸多的测井曲线中来分析油层沉积环境,并指导油气勘探和油田开发,智能化测井曲线识别无疑是一种很好的方法。对于识别研究区块的沉积环境、沉积相、沉积微相和水淹层的分布情况具有重要意义。本文通过研究分析测井曲线沉积微相和水淹层的特点及测井曲线响应特征之间的关系。结合提取的特征值建立了数据库,对Rs-LVQ算法进行了改进,构建了Rs-Ga-LVQ融合算法。并对数据库中20口取芯资料的340个沉积微相小层和550个薄差水淹层进行识别,结果显示改进的算法识别率最高。其中261个沉积微相小层与真实结果一致；407个水淹层小层与实际结果一致,达到比较理想的效果。并证明了Rs-Ga-LVQ融合算法应用于测井曲线识别的可行性,且优于Rs-LVQ融合算法。更多还原

【Abstract】 Daqing changyuan oilfield has accumulated large numbers of logging curves and it is a good way to efficiently and economically analyze the sedimentary environment of oil reservoir according to the abundant logging curves, and to instruct the exploration and development of oilfield, which is significant in identifying and studying the distribution of sedimentary environments, sedimentary facies, microfacies and water-out layers.This paper studies and analyzes the characteristics of sednimeatary microfacies from logging curves and water-out layers, as well as the relation between logging curve response features. We have established a database connection according to the extracted eigenvalues. We used improved algorithm Rs-Ga-LVQ which is based on Rs-LVQ to respectively identify the 340 sedimentary microfacies and the 550 thin, poor water-out layers. The result shows that the improved algorithm has the highest recognition rate. Results from 261 of the sedimentary microfacies and 407 of the water-out layers are compatible to thoe in the parctice. Proved the adaptation of Rs-Ga-LVQ fusion algorithm for well logging curve identify, which is better than the Rs-LVQ fusion algorithm.更多还原