Search

This article demonstrates the implementation of recurrent neural network (RNN) model in optimizing amplitude versus offset (AVO) attributes for indicating lithology and hydrocarbon zone on seismic data. Several drawbacks exist in the conventional implementation of AVO attributes for hydrocarbon exploration, including ambiguous AVO amplitude response as direct hydrocarbon indicators (DHIs), high cost of conventional seismic inversion and ambiguity from a non-absolute range of AVO scale of quality factor of P-wave (AVO SQp) and AVO scale of quality factor of S-wave (AVO SQs) attributes. Hence, this study aimed to optimize the application of AVO attributes by implementing an RNN model that solves nonlinear approximation with a faster, more economical, and reliable approach, supplied with well data as the data control. Sixteen features were extracted from seismic data as input with two targets of SQp and SQs from wells as output in the Angsi field. The model consisted of the improved algorithm of standard RNN called gated recurrent unit layers, followed by a simple RNN layer, and fully connected dense layers to predict SQp and SQs as lithology and hydrocarbon indicators, respectively. The model managed to predict the I-35 hydrocarbon reservoir zone anomalies, characterized by low SQp (sand-prone interval) and high SQs (hydrocarbon-bearing interval). The results indicated that the proposed RNN model performed efficiently as an alternative approach to bypass the conventional seismic inversion method and to optimize the application of AVO attributes in indicating lithology and hydrocarbon zone in seismic data. [ABSTRACT FROM AUTHOR]

During the past four decades, numerous amplitude variation with offset (AVO) attributes has been proposed and implemented to detect seismic expressions associated with hydrocarbon-charged sediments. For the above purpose, the principal AVO attributes, intercept (I) and gradient (G), have been used in a combination to produce another AVO attribute or cross-plotted against each other to detect hydrocarbons. Some of the very common combinations are the product I*G, energy-weighted AVO (En-AVO), fluid factors, and scaled Poisson reflectivity (SPR). The latter attribute is proven sensitive to changes in the Poisson ratio between sediments. Irrespective of sand-shale impedance contrast, the SPR is found always negative for shale over gas-saturated reservoir reflections. On the other hand, gas-saturated sands regardless of their age, depth, and consolidation are commonly characterized by their negative gradient (G) which is also influenced by changes in the Poisson ratio. In order to benefit from the dependence of the two attributes on the Poisson ratio, the product of SPR and G is applied on well and seismic data from North Western Australia to detect hydrocarbon-saturated reservoirs. In addition, a cross-plot of SPR vs gradient is used instead of the commonly used intercept-gradient. The product of SPR and G could help distinguish the gas-saturated reservoirs from their surrounding shale and siltstone. Furthermore, the SPR-G cross-plot shows that hydrocarbon-saturated sandstones can be easily identified in the G-SPR domain. Results from the application of the SPR*G and from the cross-plot are calibrated with well data and found very consistent with information from wells. [ABSTRACT FROM AUTHOR]

Featured Application: This article demonstrates SQp and SQs methods as the new AVO attributes that sensitive to determine facies and fluid analysis. Amplitude versus offset (AVO) analysis integration to well log analysis is considered one of the advanced techniques to improve the understanding of facies and fluid analysis. Generating AVO attributes are one solution to give an accurate result in facies and fluid characterization. This study is focused on a field of Northern Malay basin, which is associated with a fluvial-deltaic environment, where this system has high heterogeneity, whether it is vertically or horizontally. This research is aimed to demonstrate an application of the scale of quality factor of P-wave (SQp) and the scale of quality factor of S-wave (SQs) AVO attributes for facies and fluid types separation in field scale. These methods are supposed to be more sensitive to predict the hydrocarbons and give less ambiguity. SQp and SQs are the new AVO attributes, which derived from AVO analysis and created according to the intercept product (A) and gradient (B). These new attributes have also been compared to the common method, which is the Scaled Poisson's Ratio attribute. By comparing with the Scaled Poisson's Ratio attribute, SQp and SQs attributes are more accurate in determining facies and hydrocarbon. SQp and SQs AVO attributes are integrated with well log data and considered as the best technique to determine facies and fluid distribution. They are interpreted by using angle-stack seismic data based on amplitude contrast on interfaces. Well log data, e.g., density and sonic logs, are used to generate synthetic seismogram and well tie requirements. The volume of shale, volume of coal, porosity, and water saturation logs are used to identify facies and fluid in well log scale. This analysis includes AVO gradient analysis and AVO cross plot to identify the fluid class. Gassmann's fluid substitution modeling is also generated in the well logs and AVO synthetics for in situ, pure brine, and pure gas cases. The application of the SQp and SQs attributes successfully interpreted facies and fluids distributions in the Northern Malay Basin. [ABSTRACT FROM AUTHOR]

Amplitude versus offset (AVO) analysis integration to well log analysis is considered one of the advanced techniques to improve the understanding of facies and fluid analysis. Generating AVO attributes are one solution to give an accurate result in facies and fluid characterization. This study is focused on a field of Northern Malay basin, which is associated with a fluvial-deltaic environment, where this system has high heterogeneity, whether it is vertically or horizontally. This research is aimed to demonstrate an application of the scale of quality factor of P-wave (SQp) and the scale of quality factor of S-wave (SQs) AVO attributes for facies and fluid types separation in field scale. These methods are supposed to be more sensitive to predict the hydrocarbons and give less ambiguity. SQp and SQs are the new AVO attributes, which derived from AVO analysis and created according to the intercept product (A) and gradient (B). These new attributes have also been compared to the common method, which is the Scaled Poisson’s Ratio attribute. By comparing with the Scaled Poisson’s Ratio attribute, SQp and SQs attributes are more accurate in determining facies and hydrocarbon. SQp and SQs AVO attributes are integrated with well log data and considered as the best technique to determine facies and fluid distribution. They are interpreted by using angle-stack seismic data based on amplitude contrast on interfaces. Well log data, e.g., density and sonic logs, are used to generate synthetic seismogram and well tie requirements. The volume of shale, volume of coal, porosity, and water saturation logs are used to identify facies and fluid in well log scale. This analysis includes AVO gradient analysis and AVO cross plot to identify the fluid class. Gassmann’s fluid substitution modeling is also generated in the well logs and AVO synthetics for in situ, pure brine, and pure gas cases. The application of the SQp and SQs attributes successfully interpreted facies and fluids distributions in the Northern Malay Basin.

An application of classifier fusion technique is presented to improve the performance of automated reservoir facies identification system. The algorithm presented in this study uses three well-known classifiers, namely Bayesian, k-nearest neighbor (kNN), and support vector machine (SVM) to automatically identify four defined facies of Asmari Formation from log-derived amplitude versus offset (AVO) attributes. Fuzzy Sugeno integral (FSI) method is then employed to combine the outputs of three investigated classifiers and increase the consistency of reservoir facies identification process. The experimental results obtained from applying the presented algorithm on data related to three wells drilled in Asmari Formation provide evidence of the effectiveness of the proposed algorithm regarding true positive (TP), false positive (FP), and classification accuracy criteria. [ABSTRACT FROM AUTHOR]

Currently the seismic exploration technique is mainly used to reveal the structure of coal beds, and cannot be used to evaluate the distribution of deformed coal, the permeability of coal beds and their roof/bottom, which are related to gas outburst. In other words, it cannot evaluate the lithology of coal beds and rock strata. In recent years, some geophysicists suggest to investigate the lithology of coal beds and rock strata, and define the deformed coal distribution and gas concentrated area with AVO technique. In this paper, a 3D seismic cube of deformed coal model was initially obtained using seismic simulation. Subsequently, AVO inversions for the 3D seismic cube were calculated using different Zoeppritz approximate equations, and many AVO attributes were obtained. Through the analysis of AVO attributes, it was proposed that the B-value cube inversed with Aki & Richards approximate equation, and the G-value cube and Δσ cube inversed with Shuey approximate equation are used as the basis of AVO interpretation. Finally, AVO inversions for real 3D coal seismic data were performed to define the distribution of deformed coal and gas concentrated areas using AVO attributes. [ABSTRACT FROM AUTHOR]