Bayesian regularization BP Neural Network model for predicting oil-gas drilling cost

Oil-gas drilling cost is an important indicator which reflects the economic benefit of oilfield enterprise. Following taking the characteristics of oil-gas drilling cost which belongs to subsidiary of CNPC (China National Petroleum Corporation) into account, determinants concerning oil-gas drilling cost are identified. Bayesian Regularization Back Propagation Neural Network (BRBPNN) is proposed to predict oil-gas drilling cost. Through comparing with Levenberg-Marquardt Back Propagation, Momentum Back Propagation, Variable Learning Rate Back Propagation models in terms of prediction precision, convergence rate and generalization ability, the results exhibit that BRBPNN has better comprehensive performances. Meanwhile, results also exhibit that BRBP model has the automated regularization parameter selection capability and may ensure the excellent adaptability and robustness. Thus, this study lays the foundation for the application of BRBPNN in the analysis of oil-gas drilling cost prediction.