Open AccessCasing Damage Classification Method Using Random Forest Algorithms
Author(s) -
Jingling Xue
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1437/1/012131
Subject(s) - casing , random forest , generalization , computer science , reliability (semiconductor) , decision tree , tree (set theory) , feature (linguistics) , algorithm , data mining , artificial intelligence , pattern recognition (psychology) , engineering , mathematics , petroleum engineering , mathematical analysis , power (physics) , linguistics , physics , philosophy , quantum mechanics
The classification of casing damage is one of the important links in casing damage detection. Due to the large number of feature parameters of casing damage, the classification accuracy and the generalization performance of decision tree is poor. Therefore, for improving the prediction accuracy of casing damage classification, considering the data type of oil and gas well casing damage, the paper proposed a casing damage classification method based on random forest algorithms to predict the casing damage classification. The experiment results show that, the prediction accuracy of the method is 95%, which proves the accuracy and reliability of the casing damage classification method using random forest algorithms.