Integration of Petrophysical Log Data with Computational Intelligence forthe Development of a Lithology Predictor

摘要

Wrong manual interpretation from the log data about the formation type and other important information canbe catastrophic for the company-operator.With Machine-Learning(ML)(a branch of Artificial Intelligence)algorithms,the interpretation of formation type from the log data has been addressed.As a result,we havesuccessfully developed a program able to accurately predict the type of formation.Using the conventional Machine Learning technique of splitting the data into training,validation andtest sets,we tried six different ML algorithms to fit with the training part of the data and then verifytheir prediction accuracy with cross-validation scores and cross-validation predictions which tests theperformance of the classifiers(ML algorithms)on the validation set.The three best performing classifierswere selected and further improved by a search of classifier's best hyperparameters.These improvedclassifiers are further tested on unseen data to produce a comparative analysis.Our prediction accuracy with Receiver Operating Characteristic(ROC)scores and ROC-Area Under-the-Curve(ROC-AUC)for each type of formation from the log data lies in the range of 95-99%,exceptfor formations such as shaly sandstone and shale(50% and 84% respectively).The reason for this seemedto be under-fitting i.e.,during the training,the classifiers did not see enough instances of these types offormation to know exactly what characteristics of the data make the type of formation to be shaly sandstoneor shale.The issue of under-fitting was verified by skimming through the data.To resolve this problem,we suggest training classifiers with a larger data with more targets(types of formation).Furthermore,during the data cleaning(prior to classifier training)and data analysis phases we have discovered importantrelationships between well logs and defined relative importance of each well log for different formations.This observation can be investigated further to help eliminate the use of multiple well logs while dealing withsome formations(based on prior geological knowledge)and reduce the cost of the well logging operations.Using our program with a larger well log data consisting of more formation type instances,we can train theclassifiers to accurately predict the formation type irrespectively of differences in formation type.Our program is dynamic in the sense that with different targets,i.e.,type of formation fluid instead oftype of formation or both together,it can successfully predict either or both targets.Increasing the numbers of data instances resulted in a better training and thus,more accurate predictions.Utilization of the programwill make the formation-evaluation process easier,faster,automated and more-precise.