Support Vector Machine Model: A New Methodology for Stuck Pipe Prediction

Abstract

Stuck pipe is a common worldwide drilling problem, resulting significant increases in non-productive time and overall well cost. Many oil and gas reservoirs are mature and are becoming increasingly depleted of hydrocarbons which make stuck pipe more severe risks. This is due to the fact that decreasing pore pressure increases the chance of stuck pipe. Minimizing the risks of stuck pipe while drilling has been the goal of many operators recently. This paper describes a robust support vector regression (SVR) methodology that offers superior performance for stuck pipe prediction either mechanically or differentially using available drilling parameters. A new model is developed using drilling parameters such as measured depth, mud weight, plastic viscosity, yield point, gel strengths, PH and solid percent from different wells. The method incorporates hybrid least square support vector regression and Coupled Simulated Annealing (CSA) optimization technique (LSSVM-CSA) for efficient tuning of SVR hyper parameters. The algorithm is applied to classify the stuck types, i.e., differential stuck or mechanical stuck. Performance analysis shows that LSSVM classifier has high accuracy. Using intelligent system would help drilling industry to reduce Non-Productive Time (NPT) during operation in complex zones.

Modeling viscosity of alcohols based on the CPA-EoS + ƒ-theory

“…[10][11][12][13][14][15][16][17][18], or molecular simulations, for modeling the fluid's viscosity. The later models are out of the scope of the present study.…”

Machine learning algorithm for prediction of stuck pipe incidents using statistical data: case study in middle east oil fields

“…Underfitting happens when there are not enough neurons in the hidden layers. Underfitting occurs when there are not enough neurons in the hidden layers to identify the signals in a complicated dataset (Chamkalani et al 2013). Having too many neurons in the hidden layers might cause several problems.…”

Utilization of support vector machine to calculate gas compressibility factor

“…The support vector machine (SVM) is a well-accepted mathematical strategy to obtain an accurate relationship between the parameters of a certain mathematical problem [48][49][50]. Suykens and Vandewalle proposed a modified version of SVM, called least squares-SVM (LS-SVM) [51].…”