Konstantin E. Zakrevskiy
|JSC "NK "Rosneft"
Viktor L. Popov
Synthetic geological models are used to obtain the results which are presented in the paper. The methodology of creation of that geological models and variation of its source data are described. The influence of variation of thesource data (bodies’ size and number) on theresults of statistical estimation of reservoir parameters was analyzed.
Materials and methods
Numerical calculations in synthetic geological models, object modeling, indicator modeling, multi-point statistics.
It was found that variogram range can besmaller than bodies’ size due to cutting bodiesby model boundaries. This effect is significant in a model when bodies size exceed half of a model size in that direction.On the other hand, variogram range can be larger due to infl e of clusterization of bodies in a model. This eff more signifi ant in models with low number of bodies (low NTG). Variogram curve shape is greately affected by model NTG.It should be noted that model size has significant effect on the variability of statistical parameters. It is a result of discrete model nature. When the bodies size is comparable to cell size it results in upward bias in variability.
Variogram range is strongly related with size of bodies in shallow marine facies which can berepresented as cubes and spheres. The body size estimation error is approximately 25%. It permit using in indicator modeling the data of the statistical analysis of the bodies sizes which are obtained from outcrops. Statistical parameters which are related with channel geometry (sinuosity, wave amplitude, wave length) were founded.It should be noted that this results were obtained with zero level of statistical noise. The influence of statistical noise will be studied in future work.