PREDIKSI POROSITAS MENGGUNAKAN METODE NEURAL NETWORK PADA LAPANGAN OZZA, CEKUNGAN SUMATRA TENGAH

Abstract

Information from geological structures that are considered to contain hydrocarbons may not necessarily contain economical hydrocarbons, so additional analysis is needed to determine the position of new wells. Seismic and log methods can be used to determine areas considered prospective for oil and gas exploration. Seismic analysis method developed to be able to integrate seismic data and log data is a neural network. Neural network is a data processing to get a non-linear approach of the statistical relationship of the input data to the output data, then distributed to all seismic volumes. The results of the study of sand reservoir characteristics in the Ozza Field have a porosity value of more than or equal to 20%, and for shale it has a porosity value of less than 20%. The correlation between the original porosity value and predictive porosity is that the higher the porosity value in the original log the higher the value of the neural network porosity, and vice versa. The porosity distribution map in the prospect area has a higher porosity value than the surrounding area. The prospect zone for new exploration is in the southwest area of the study area.

Keywords

• hydrocarbons
• reservoirs
• seismic
• well logging