Karakterisasi reservoir menggunakan metode Seismik Inversi Acoustic Impedance (AI) dan Seismik Multiatribut dengan Probabilistic Neural Network (PNN) pada lapangan Blok F3, North Sea Netherland

Abstract

A 3D seismic acquisition has been carried out for oil and gas exploration in F3 field block of North Sea sector of the Netherland formed between the Jurassic and Cretaceous periods. The presence of hydrocarbons is indicated by the phenomenon of bright spots and gas chimneys below the surface. The data used are 3D post stack time migration seismic data and four wells with well log, checkshot and marker data availability. This study uses two methods in determining reservoir zones, namely the acoustic impedance inversion method and the multi-attribute method with PNN. Both methods integrate seismic data with well data. AI inversion method is used to predict the physical properties of rocks, namely their acoustic impedance values. The multi-attribute method is used to predict well log properties from seismic data. Non-linear multi-attribute transformation is obtained by the process of training neural networks with a type of probabilistic neural network (PNN). In this research, acoustic impedance volume and porosity estimation volume will be made to identify the hydrocarbon reservoir prospect zone. The two methods are then applied to the Netherlands F3 seismic field data, and the results show that there are three sandstone reservoir zones that have an acoustic impedance range between 4100-4800 (m/s)*(gr/cc) and porosity range between 29-35 (% ).