Stochastic reservoir modeling is being increasingly used for modeling
reservoir heterogeneity. This paper describes a methodology to model
the distribution of reservoir properties using well data and soft
geological knowledge in the form of sedimentary and diagenetic patterns.
The technique, developed based on a combined use of radial basis
function (RBF) neural networks and geostatistical kriging, is demonstrated
with an application to interpolating porosity in the A'nan Oilfield,
located onshore north China. The integrated technique first uses
neural networks to estimate the porosity trends from high-dimensional
geological patterns. Optimization of the network performance is done
by variogram analysis of the residuals at the conditioning points.
Gaussian simulation of the residuals is then performed, and the resulting
residual maps are combined with the porosity trends obtained from
neural networks. From the case study, the results are realistic and
honor the geological rules of the oilfield. The technique is fast
and straightforward, and provides a computational framework for conditional
simulation.