A New Pore Pressure Prediction Method—Back Propagation Artificial Neural Network

Accurate pore pressure prediction is a necessary requirement to well structure optimizing, drilling difficulty minimizing, drilling accidents preventing. It plays a very important role in the economically and efficiently well operating. Previous methods for pore prediction have their own hypothesizes and cannot take into consideration factors that indicate or influence the pore pressure, so their applicability is limited to one or several regions and overpressure causes. This paper presents a new kind of pore pressure prediction BP neural network making use of the powerful learning capability of the neural network. Our model has three layers: two hidden layers and the output layer. The inputs of the first hidden layer include gamma ray and formation density. The result of the first hidden layer is 4 * input to the second hidden layer which additionally has depth, interval transit time and formation density as inputs. The optimized numbers of the neurons of each layer are 2, 5, and 1 respectively. The activation function of the first layer is hard-limit function, and the second and the third layer both has hyperbolic function as activation function. The feasibility and accuracy of our network is verified by the successful applications in a normal pressure field and an overpressure field. The average error is 4.61%. Compared with normal-structure pore pressure prediction neural network, our model doubled the accuracy of the prediction result.