#TITLE_ALTERNATIVE#

• Main Author: MAULANA (NIM: 12304016), IMAM
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/20188
• Institution: Institut Teknologi Bandung

This study presents an application of Multiattribute seismic analysis to predict p-wave, porosity and gamma rayas as an effort to characterize reservoir in the area of an existing gas and oil field in South Sumatra Basin. The target was Batu Raja Formation, consists of reef carbonate and ramp carbonate. The 3-D seismic data were used to interpret the location of major stratigraphic markers between wells, statistic wavelet and these seismic horizons were used to invert 3D seismic into AI cube. Crossplot indicates that p-wave could better distinguish reservoir and non reservoir than Acoustic Impedance (AI), but these parameters and also density could not distinguish fluid in reservoir, hence pseudo OWC and pseudo GWC from wells marker were used as reference of fluid contact. The AI cube, derived from sparse spike inversion, showed good corelation with AI well and then utilized as external attribute. Stepwise regression and crossvalidation were used to combine seismic attributes to predict p-wave, porosity and gamma ray in wells where all of these parameters were known from the well logs. The results of multivariate linear regression p-wave model showed good correlation (0.906) between four seismic attributes and the observed p-wave logs at 10 wells in the study area. <br /> <br /> <br /> <br /> <br /> <br /> <br /> <br /> The results of multivariate linear regression porosity model showed good correlation (0.858) between two seismic attributes and the observed porosity logs at 10 wells in the study area. The results of multivariate linear regression gamma-ray model showed less correlation (0.65) between five seismic attributes and the observed porosity logs at 10 wells in the study area. A probabilistic neural network was then trained to look for a nonlinear relationship between the input data and the observed gamma-ray at the 10 wells and the result was then used for second iteration of a linear regression gamma-ray model, the correlation was better (0.836). Combination analysis of p-wave, integrated map, isochore and time structure map was prepared to give suggestion the best location of next well.