Modeling porosity distribution in the A'nan Oilfield: Use of geological quantification, neural networks, and geostatistics

A'nan Oilfield is located in the northeast of the Erlian Basin in North Chi na. The porosity distribution of the oil-bearing stratum is primarily contr olled by complex distribution patterns of sedimentary lithofacies and diage netic facies. This paper describes a methodology to provide a porosity mode l for the A'nan Oilfield using limited well porosity data, with the incorpo ration of the conceptual reservoir architecture. Neural network residual kr iging or simulation is employed to tackle the problem. The integrated techn ique is developed based on a combined use of radial basis function neural n etworks and geostatistics. It has the flexibility of neural networks in han dling high-dimensional data, the exactitude property of kriging and the abi lity to perform stochastic simulation via the use of kriging variance. The results of this study show that the integrated technique provides a realist ic description of porosity honoring both the well data and the conceptual f ramework of the geological interpretations. The technique is fast, straight forward and does not require any tedious cross-correlation modeling. It is of great benefit to reservoir geologists and engineers.