Entity

Time filter

Source Type


Malvic T.,Sector for Geology and Reservoir Management | Velic J.,University of Zagreb | Horvath J.,University of Szeged | Cvetkovic M.,University of Zagreb
Central European Geology | Year: 2010

Three examples of the use of neural networks in analyses of geologic data from hydrocarbon reservoirs are presented. All networks are trained with data originating from clastic reservoirs of Neogene age located in the Croatian part of the Pannonian Basin. Training always included similar reservoir variables, i.e. electric logs (resistivity, spontaneous potential) and lithology determined from cores or logs and described as sandstone or marl, with categorical values in intervals. Selected variables also include hydrocarbon saturation, also represented by a categorical variable, average reservoir porosity calculated from interpreted well logs, and seismic attributes. In all three neural models some of the mentioned inputs were used for analyzing data collected from three different oil fields in the Croatian part of the Pannonian Basin. It is shown that selection of geologically and physically linked variables play a key role in the process of network training, validating and processing. The aim of this study was to establish relationships between log-derived data, core data, and seismic attributes. Three case studies are described in this paper to illustrate the use of neural network prediction of sandstone-marl facies (Case Study # 1, Okoli Field), prediction of carbonate breccia porosity (Case Study # 2, Beničanci Field), and prediction of lithology and saturation (Case Study # 3, Kloštar Field). The results of these studies indicate that this method is capable of providing better understanding of some clastic Neogene reservoirs in the Croatian part of the Pannonian Basin. Source


Zelenika K.N.,Sector for Geology and Reservoir Management | Cvetkovic M.,University of Zagreb | Malvic T.,Sector for Geology and Reservoir Management | Malvic T.,University of Zagreb | And 2 more authors.
Journal of Maps | Year: 2013

Data from selected Lower Pontian sandstone reservoir in the Kloštar Field, situated in the Sava Depression (Northern Croatia), were used for mapping with Sequential Indicator Simulations rather than using a classical approach. Such approaches offer better insight in distribution of geological variables or zonal uncertainties in the cases with larger datasets (15 points or more). Obtained maps of porosity and reservoir thickness are presented here along with probability maps of certain selected cut of values of petrophysical parameters. Maps showed distinct sedimentological features that can clearly be observed on the both sets of maps. © 2013 Copyright Marko Cvetković. Source


Zelenika K.N.,Sector for Geology and Reservoir Management | Malvic T.,Sector for Geology and Reservoir Management
Geologia Croatica | Year: 2011

The research presented herein is the fi rst attempt to perform geostatistical simulations on three geological variables, porosity, thickness, and depth to reservoir, in the Croatian part of the Pannonian Basin. The data were collected from a reservoir of Lower Pontian age in the Kloštar Field, located in the western part of the Sava Depression. All three variables were analyzed using sequential Gaussian simulations (SGS). Information regarding present-day depth, thickness, and locations of areas with higher porosity values were used to reconstruct palaeo-depositional environments and the distribution of different lithotypes, ranging from medium-grained, to mostly clean sandstones and to pure, basin marls. Estimates of present-day thickness and depth can help to defi ne areas of gross tectonic displacement and the role of major faults that have been mapped in the fi eld. However, since mapping of the raw data (including porosities) does not allow the reconstruction of palaeo-depositional environments, sequential indicator simulations (SIS) were applied as a secondary analytical tool. For this purpose, several cut-off values for thickness were defi ned in an effort to distinguish the orientation of depositional channels (main and transitional). This was accomplished by transforming porosities to indicator values (0 and 1) and by applying a non-linear indicator kriging technique such as the "zero" map for obtaining numerous indicator realizations by SIS. In the SGS and SIS approaches, the simulations encompassed 100 realizations. A representative realization was then selected using purely statistical criteria, i.e., two realizations were almost always chosen in accordance with the order of the calculation. The 1st and 100th realizations were selected for each variable in the SGS and SIS and fi ve indicator kriging maps were chosen for the thicknesses cut-offs. Source

Discover hidden collaborations