INTEGRATION OF METHODS FOR OPTIMALIZING RESERVOIR CHARACTERIZATION

• Main Author: MALIYAN (NIM : 22309301), AHMAD
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/20847
• Institution: Institut Teknologi Bandung

Despite there are limited data in conducting seismic interpretation, the interpretation of the analysis must be performed. In order to overcome the limitation, integeration of methods can be used so that there will be a higher validity in optimalizing reservoir characteristics. At this stage, the analysis should <br /> <br /> <br /> <br /> not only refers to the magnitude of value, but also have to consider the spread of parameter and geometry. <br /> <br /> <br /> <br /> Focus area of the study was determined by the results of the analysis of gamma ray log, NPHI log and density logs. Scale used to determine the sandstone is 0- <br /> <br /> <br /> <br /> 60% for NPHI and 1.7 - 2.7 g / cc for density. Both log was positioned on the same spot and one of them is inversed. Sandstone is indicated to present on the <br /> <br /> <br /> <br /> intersected area. Clay rocks have have relatively high gamma ray value compared to sandstones. Cross plot analysis determined not only the parameters which are sensitive to changes in lithology but also whether the inversion can be conducted. Gamma ray cut-off is established from cross-plot combination of depth, gamma ray, Vp, density, NPHI, resistivity, and accoustic impedance (AI). Density and AI are the parameter that can distinguish sandstone from clay. Sandstone has density < 2.3 g / cc, AI < 8000 (m/s) (g/cc), and gamma rays < 85 gapi. Meanwhile, parameter Vp has overlapping values that sandstone cut-off value can not be determined. Inversion method can be used to generate a map of AI and density. Model-based inversion method and sparse spike inversion represent methods using wavelet <br /> <br /> <br /> <br /> while coloured inversion methods represent methods with no wavelet. Wavelet type is derived from the extraction wells with a dominant frequency of 25 Hz. The method used in the analysis of inversion is determined by comparing the results of previous inversion and considering its best correlation. Since Linear Programming Sparse Spike (LPSS) method yield the greatest correlation (0.88631), it is selected as the comparison to coloured method. The capability of coloured method is better than LPSS method in the terms of detecting thin horizon. Based on inversion results, sandstone exist between U and V horizon as previously suspected. Multiattribute is used to predict the spread of gamma ray on any location in the seismic volume. In multiattribute process, seismic data is treated as raw data while the inversion results of coloured methods as external attributes. Since there is only one well utilized in this study which causing validation error can not be used on the analysis to establish the number of attribute and operator length, trial and error is used in doing so. Quality control is conducted based on analysis of map generated on multiatributte process. Inversion results are used as comparative data on establishing the the appropriate number of attribute and operator length. Coloured inversion, LPSS, and multiattribute yield a structure of anticline with <br /> <br /> <br /> <br /> two peak. The results confirmed that integration of methods could reduce the ambiguity in recognizing the presence of sandstone that the conclusion to explore <br /> <br /> <br /> <br /> a prospective area can be obtained.