基于交会图和贝叶斯聚类分析法的岩性识别方法

Method of Lithologic Identification Based on Crossplot and Bayesian Cluster Analysis Algorithm

将针对造成伊通盆地莫里青断陷测井解释符合率低的主要因素,即复杂的岩性开展研究。根据这一状况,提出了利用交会图及贝叶斯聚类分析算法的岩性识别方法。先对区域资料进行测井响应标准化,选择出合适的样本数据,确定自然伽马、电阻率、声波时差、中子、密度五条测井曲线作为岩性识别的参数,符合率达到82.9%。在对区域内十口探井进行岩性解释,最终符合率达到87.1%,有效地提高该区域的解释精度。

According of Yitong Basin, Carrying to main factor of low coincidence rate of log interpretation in Moliqing Faulted Depression out research a lithologic identification method of using the crossplot and bayesian cluster analysis algorithm is put forward. First, logging response standardization on the regional data, choosing the appro- priate sample data, putting five logging curve of the natural gamma ray, resistivity, interval transit time, neutron, and density as lithologic identification parameters, coincidence rate reached 82. 9%. Through lithology interpreta- tion on ten exploration wells in the area, and ultimately coincidence rate reached 87.1%, and the method effective- ly improves interpretation accuracy in the region.