Title: Modified stem cells algorithm-based neural network applied to bottom hole circulating pressure in underbalanced drilling
Authors: Reza Taherdangkoo; Mohammad Taherdangkoo
Department of Petroleum Engineering, Omidiyeh Branch, Islamic Azad University, Omidiyeh, Iran
Department of Artificial Intelligence, Tehran Business School, 2532 Valiasr Street, Tehran, Iran
Abstract: In this paper, we address the problem of inaccuracy in evaluating bottom hole circulating pressure (BHCP) in the petroleum industry by proposing a new predicting scheme. This scheme utilises a modified version of the stem cells algorithm (MSCA), recently introduced as a powerful meta-heuristic optimisation method, along with the back propagation (BP) training strategy to build a three-layer artificial neural network (ANN) as a predictive scheme. This new method is able to predict the complex relationship between inputs and outputs of a highly nonlinear system such as BHCP more accurately. The results by applying the proposed method compared with those by applying previous predicting methods used for BHCP demonstrate the superiority of the proposed method in terms of accuracy and time consumption.
Keywords: modified stem cells algorithm; artificial neural networks; ANNs; bottom hole circulating pressure; BHCP; underbalanced drilling; petroleum engineering; oil industry; metaheuristics; optimisation.
Int. J. of Petroleum Engineering, 2015 Vol.1, No.3, pp.178 - 188
Date of acceptance: 12 Mar 2015
Available online: 10 Aug 2015