Open AccessCluster Discriminant Prediction of Oil Well Production Based on Mahalanobis Distance
Author(s) -
Bo Huang,
Yunwei Kang,
Ang Chen,
Zixi Guo,
Yicheng Sun
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 - 10.1088/1742-6596/1437/1/012089
Subject(s) - mahalanobis distance , cluster analysis , selection (genetic algorithm) , data mining , discriminant , linear discriminant analysis , data set , oil field , set (abstract data type) , computer science , field (mathematics) , feature selection , artificial intelligence , statistics , pattern recognition (psychology) , engineering , petroleum engineering , mathematics , pure mathematics , programming language
The cost of oil field drilling engineering is very high. Effective well selection and formation selection will avoid waste of energy and resources. Aiming at the problem of well selection and reservoir selection in oilfield, this paper presents a clustering discriminant prediction method for oil well production based on Mahalanobis distance. Based on the field data of Xinjiang oilfield, the main control factors with greater influence are selected after data pretreatment in this paper. Then we randomly divide the data into training data set and prediction data set, and establish a discriminant model of oil well classification based on Mahalanobis distance. Finally, the model is used to predict the production of oil wells in the forecasting data set, and the prediction accuracy reaches 87.5%.