Application of Boosting Tree algorithm in logging lithology interpretation
Kai Jiang, Shoudong Wang, Cui Baosheng, Shizhao Pu, Hang Duan · International Geophysical Conference, Beijing, China, 24-27 April 2018 · 2018
The lithology interpretationby well logs is essentially a highly nonlinear mapping function problem. Boosting Tree algorithm that combines multiple decision tree classifiers to complete classification tasks is one of the most widely used methods for dealing with nonlinear classification problems. This study builds a lithology identification model by utilizing Boosting Tree algorithm on the basis of mud logging data and welllogsderived from GR (natural gamma ray), SP (spontaneous potential), RXO (resistivity of flushed zone), RI(resistivity of invaded zone), RT (true formation resistivity), DEN (density), CNL (compensated neutron log) and AC (acoustic interval travel time). In the experiments, the model is used toidentify the lithology of the target layer of No. 6 well in Mabei Oilfield and achievesanaccuracy of 89.1% that is superior to the traditional machine learning methods such as Decision Tree and support vector machine (SVM). Using Boosting Tree algorithm to identify lithology also provides a new way for logging interpretation.