The mathematical statistics method and extrema variance clustering method can be used to visible automatic classifying and reading of logging curves . the lithofacies classifying program basing on multi - mineral model analysis presents a new method to analyze logging - facies and more accuracy and visualized logging facies section can be reached by using this method . in addition , it supplies reliable lithologic layering reference for search and evaluation of oil / gas caprock and it also made up for the high cost of core - drilling and the inaccuration between lithic fragment description and depth 采用基于“數(shù)理統(tǒng)計(jì)-極值方差聚類法”的面向?qū)ο罂梢暬僮鞣椒捎行У亟鉀Q測(cè)井曲線的可視化自動(dòng)分層取值問題;而基于多礦物模型分析的巖相劃分程序又提供了一種新的測(cè)井相分析方法,能得到更為準(zhǔn)確直觀的測(cè)井相剖面,為尋找和評(píng)價(jià)油氣蓋層提供了可靠的巖性分層依據(jù),同時(shí)彌補(bǔ)了鉆井取心費(fèi)用高和錄井巖屑描述與深度有誤差的缺陷。