Unaprijeđeni model procjene neuronskim algoritmom miocenskih litofacijesa u polju Ladislavci

master's thesis

• Title: Unaprijeđeni model procjene neuronskim algoritmom miocenskih litofacijesa u polju Ladislavci : diplomski rad
• Title (english): Improved neural network model for lithofacies interpretion in miocene layers of the Ladislavci field : master’s thesis
• Author: Andrija Varenina
• Mentor: Tomislav Malvić (mentor)
• Committee member: Tomislav Malvić (predsjednik povjerenstva)
• Committee member: Josipa Velić (član povjerenstva)
• Committee member: Rajna Rajić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Mining, Geology and Petroleum Engineering; (Department of Geology and Geological Engineering); Zagreb
• Defense date and country: 2017-09-27, Croatia
• Scientific / art field, discipline and subdiscipline: NATURAL SCIENCES; Geology; Geology and Paleontology

Abstract

Naftno-plinsko polje Ladislavci nalazi se u istočnoj Slavoniji na području sela Ladislavci koje je udaljeno desetak kilometara od Donjeg Miholjca. Na polju Ladislavci danas postoji 21 bušotina. Ležište nafte je u miocenskim naslagama vukovarske formacije, a plinsko u osječkim pješčenjacima (serija B). Ležište nafte podijeljeno je u tri ležišna sloja (El, F1a i F1b) i radi se kombiniranoj zamci strukturno-stratigrafskog tipa. Ležišne stijene su vapnenačke breče, klastični vapnenci i brečo-konglomerati. Nakon opisa osnovnih svojstava umjetnih neuronskih mreža i povijesti njihovog razvoja, prikazane su još i elektrokarotažne metode čiji su rezultati korišteni za interpretaciju ležišnih stijena te kao ulazni podatci za uvježbavanje neuronskih mreža. Neuronske mreže su uvježbavane na podatcima iz bušotina (A i B) i to u intervalu od EK-markera Rs7 do dna bušotine. Na tim mjestima kao ulazni podatci izdvojene su vrijednosti spontanog potencijala (SP) i električne otpornosti (R16 i R64). Iz takvih ulaza predviđan je litološki sastav uz pomoć višeslojne perceptronske mreže (skr. VPM).

Abstract (english)

The Ladislavci Oilfield is located in eastern Slavonia, in the Ladislavci village, about ten kilometers from Donji Miholjac. There are 21 boreholes in the Ladislavci Oilfield today. The oil reservoir is in the Miocene deposits of Vukovar formation and gas reservoir is in Osijek sandstones (series B). The oil reservoir is divided into three sections (El, F1a and F1b) and it is a combined structural-stratigraphic trap. The reservoir rocks are limestone breccia’s, clastic limestone and breccia conglomerates. After the description of basic properties of artificial neural networks, well logs have been presented, whose results are used for the interpretation of the bedrock and as input data for the training of neural networks. Neuronal networks are trained on the data from wells (A and B) from the interval of the EK marker Rs7 to the bottom of the well. The analysis input data where log data (curves SP, R16 and R64) and from such data the lithological composition is foreseen with the help of multilayer perceptron network (MLP).

• Keywords: Neuronske mreže; Ladislavci; srednji miocen; litologija; Hrvatska
• Keywords (english): Neural networks; Ladislavci; middle miocene; lithological composition; Croatia
• Language: croatian
• URN:NBN: urn:nbn:hr:169:942478
• Study programme: Title: Geology; specializations in: Geology of mineral resources and geophysical explorations; Course: Geology of mineral resources and geophysical explorations; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra geologije (magistar/magistra geologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2018-01-10 13:51:48