Reservoir parameter estimation using a hybrid neural network

The accuracy of an artificial neural network (ANN) algorithm is a crucial i ssue in the estimation of an oil field's reservoir properties from the log and seismic data. This paper demonstrates the use of the k-fold cross valid ation technique to obtain confidence bounds on an ANN's accuracy statistic from a finite sample set. In addition, we also show that an ANN's classific ation accuracy is dramatically improved by transforming the ANN's input fea ture space to a dimensionally smaller, new input space. The new input space represents a feature space that maximizes the linear separation between cl asses. Thus, the ANN's convergence time and accuracy are improved because t he ANN must merely find nonlinear perturbations to the starting linear deci sion boundaries. These techniques for estimating ANN accuracy bounds and fe ature space transformations are demonstrated on the problem of estimating t he sand thickness in an oil field reservoir based only on remotely sensed s eismic data. (C) 2000 Elsevier Science Ltd. All rights reserved.