Multiple permeability predictions using an observational learning algorithm

Reservoir permeability is a critical parameter for the evaluation of hydroc arbon reservoirs, Well log data are frequently available to infer this para meter along drilled wells. Many fundamental problems remain unsolved by mos t predictive models. This paper introduces the use of an improved neural ne twork trained by an observational learning algorithm to provide solutions f or two particular problems: the generation of additional or "virtual" sampl es when the number of training data is insufficient; and the generation of multiple permeability values at the same reservoir depth for reliability an alyses. The methodology is illustrated by a case study in western Australia . Four drilled wells with well logs and core permeability are used in this study. The data from the first two wells are used for training, while the o thers are used as unseen data to test the performance of the model. The res ults show that the proposed method gives smaller error compared to multiple linear regression and other neural networks (simple committee networks and bootstrap aggregating). It also provides valuable information on the relia bility of the permeability predictions which is consistent with the geologi cal studies. (C) 2000 Elsevier Science Ltd. All rights reserved.