Research on Drilling Frictional Resistance Prediction Technology Based on SVR
暂无分享,去创建一个
In petroleum exploitation, the frictional resistance has a great influence on the cost and speed of horizontal drilling. Support Vector Machine(SVM) is an efficient supervised machine learning method, and Support Vector Regression(SVR) is the application of SVM in function regression. This paper designed friction predictive model by analyzing data of horizontal drilling. The support vector regression is used to analyze the data of the horizontal drilling and find out the main parameters and the degree affecting the frictional resistance. Compared with the predictive results of existing nonlinear multivariate models, the frictional resistance prediction based on SVR has a significant advantage in accuracy. This research has guiding significance in actual production when we evaluate and reduce the costs of construction and accelerate construction progress.
[1] Chih-Jen Lin,et al. LIBSVM: A library for support vector machines , 2011, TIST.
[2] Andrew K. Wojtanowicz,et al. Minimum-cost casing design for vertical and directional wells , 1987 .
[3] Alexander J. Smola,et al. Support Vector Regression Machines , 1996, NIPS.
[4] Andrew K. Wojtanowicz,et al. A field method for assessing borehole friction for directional well casing , 1988 .