首页|New Machine Learning Study Findings Recently Were Reported by Researchers at Ind ian School of Mines (Estimation of Vertical Permeability of Hugin Sandstone From Petrophysical Well Logs Using Ensemble Methods: an Enhanced Machine Learning .. .)
New Machine Learning Study Findings Recently Were Reported by Researchers at Ind ian School of Mines (Estimation of Vertical Permeability of Hugin Sandstone From Petrophysical Well Logs Using Ensemble Methods: an Enhanced Machine Learning .. .)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
By a News Reporter-Staff News Editor at Robotics & Machine Learning Daily News Daily News – Research findings on Machine Learning are discussed in a new report. According to news reporting originating in Dhanba d, India, by NewsRx journalists, research stated, “Understanding the directional distribution of permeability is crucial for oil and gas exploration. In this st udy, well log and core data were used to predict continuous vertical permeabilit y (K-v) logs.” The news reporters obtained a quote from the research from the Indian School of Mines, “The predictions were performed for two wells drilled through the Hugin s andstone in the Volve field. Conventional well log and corresponding vertical co re permeability measurements were utilized for training machine learning (ML) mo dels. Data enhancement techniques, such as log smoothing, data transformation, a nd outlier removal were used. Thereafter, ensemble models, such as Random forest (RF), gradient boosting (GB), extreme gradient boosting (XGBoost), and adaptive boosting (AdaBoost) were used with decision trees, which are the base learners. Multilinear regression models were used for comparison of K-v values from ML al gorithms and core measurements. Hyperparameter tuning was performed using grid s earch to obtain the best learning parameters for model optimization. The trained predictive models were used to predict K-v in two wells. Four metric scores, i. e., coefficient of determination (R-2), mean squared error (MSE), mean absolute error, and root mean square error were used for model evaluation. The obtained R -2 of RF and AdaBoost models are 0.94 and 0.92, respectively. Moreover, the R-2 of 0.99 for GB and XGBoost are overestimates confirmed by their high MSE values. The standard deviation of R-2’s obtained from four-fold cross-validation indica tes that GB and XGBoost are less stable compared to RF and AdaBoost. The RF ense mble model outperformed others.”
DhanbadIndiaAsiaCyborgsEmerging TechnologiesMachine LearningIndian School of Mines
NETL
NSTL
