Committee neural networks for porosity and permeability prediction from well logs

Abstract

Neural computing has moved beyond simple demonstration to more significant applications. Encouraged by recent developments in artificial neural network (ANN) modelling techniques, we have developed committee machine (CM) networks for converting well logs to porosity and permeability, and have applied the networks to real well data from the North Sea. Simple three‐layer back‐propagation ANNs constitute the blocks of a modular system where the porosity ANN uses sonic, density and resistivity logs for input. The permeability ANN is slightly more complex, with four inputs (density, gamma ray, neutron porosity and sonic). The optimum size of the hidden layer, the number of training data required, and alternative training techniques have been investigated using synthetic logs. For both networks an optimal number of neurons in the hidden layer is in the range 8–10. With a lower number of hidden units the network fails to represent the problem, and for higher complexity overfitting becomes a problem when data are noisy. A sufficient number of training samples for the porosity ANN is around 150, while the permeability ANN requires twice as many in order to keep network errors well below the errors in core data. For the porosity ANN the overtraining strategy is the suitable technique for bias reduction and an unconstrained optimal linear combination (OLC) is the best method of combining the CM output. For permeability, on the other hand, the combination of overtraining and OLC does not work. Error reduction by validation, simple averaging combined with range‐splitting provides the required accuracy. The accuracy of the resulting CM is restricted only by the accuracy of the real data. The ANN approach is shown to be superior to multiple linear regression techniques even with minor non‐linearity in the background model.

Combining the advantages of neural networks using the concept of committee machine in the groundwater salinity prediction

“…Therefore the performance of the CNN model could be better than any individual neural networks (Bagheripour 2014). Fundamental of committee machine networks were described by Bhatt and Helle (2002), Lim (2005) and Chen and Lin (2006), Kadkhodaie-Ilkhchi et al (2009) and Ghiasi-Freez et al (2012). The assumption is that, there are N trained ANNs with output vector O i , which are used to predict target vector T: The prediction error could be written as:…”

Prediction of Shear Wave Velocity Using Artificial Neural Network Technique, Multiple Regression and Petrophysical Data: A Case Study in Asmari Reservoir (SW Iran)

“…Artificial neurons are similar to biological neurons of the human brain. They are parallel processing systems that are used to detect very complex patterns among data and have learning, training and remembering ability and capability of generalizing the results [11]. In this study, initially the available data were processed and inappropriate data were omitted, since they have a negative effect on network training and testing.…”

Integration of Adaptive Neuro-Fuzzy Inference System, Neural Networks and Geostatistical Methods for Fracture Density Modeling

“…The error is then propagated backward through the net and the weights are adjusted during a number of iterations, named epochs. The training stops when the calculated output values best approximate the desired values (Bhatt and Helle, 2002;Labani et al, 2010). NN are used widely in petroleum geoscience to predict different target parameters and also to model a parameter all over a field.…”