Naslov
Koristi upotrebe neuronske mreže u procjeni poroznosti (na primjeru polja Beničanci)
(Some benefits of the neural approach in porosity prediction (Case study from Beničanci field))

Autori
Malvić, Tomislav ; Prskalo, Smiljan

Izvornik
Nafta : exploration, production, processing, petrochemistry (0027-755X) 58 (2007), 9; 455-467

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, prethodno priopćenje, znanstveni

Ključne riječi
seizmički atributi; neuronska mreža; poroznost; aluvijalna lepeza; polje Beničanci;
(Seismic attributes; neural network; porosity; alluvial fan; Beničanci field; Drava depression)

Sažetak
The Beničanci oil field, located in the eastern part of the Drava depression is still one of five main hydrocarbon reservoirs in Croatia. That makes it very meaningful to plan and perform a whole new set of geological reinterpretations and improvements of field geological model. The application of the neural network approach in seismic attribute processing and finally reservoir porosity prediction is presented in the paper. Three seismic attributes were interpreted – amplitude, phase and frequencies making the 3D seismic cube. These attributes were interpolated at 14 well locations, averaged and compared by the mean porosities. It made the network training. The network was of the backpropagation type. It was fitted through 10 000 iterations, searching for the lowest value of correlation between attribute(s) and porosities and minimal convergence. The best training was reached using all three attributes together, which indicated the tendency that neural networks like numerous inputs. The obtained results were compared by previously interpolated geostatistical porosity maps (done by the Kriging and Cokriging approaches). The Cokriging approach, interestingly included only reflection strength (derivation of amplitude) as the secondary seismic source of information (compared by neural inputs of three attributes). It very clearly indicated on position of carefully and geologically meaningful selection of the network inputs for any reservoir analysis. Relatively smooth map, and rarely reaching of measured porosity minimum and maximum, strongly indicates on conclusion that neural estimation is more precisely than previously interpolations.

Izvorni jezik
Engleski

Znanstvena područja
Geologija

Napomena
Rad je tiskan dvojezično, na hrvatskom i engleskom jeziku.

POVEZANOST RADA