Application of an analytic hierarchy process to hydrocarbon accumulation coefficient estimation

The hydrocarbon accumulation coefficient is a key parameter in resources evaluation by genetic techniques. Methods of obtaining its value scientifically have always been an important factor influencing evaluation credibility. In this paper, the hydrocarbon accumulation system is evaluated quantitatively by establishing a hierarchy structure model based on an analytical hierarchy process. The hydrocarbon accumulation system of a higher exploration degree is selected as a calibration area and its hydrocarbon accumulation coefficient can be calculated using methods of hydrocarbon generation potential and reservoir-scale sequence. The hydrocarbon accumulation coefficient of a petroleum accumulation system can be gained by analogy of reservoir forming comprehensive evaluation results with the calibration area. The hydrocarbon accumulation coefficient of each petroleum accumulation system in the upper reservoir-forming combination of the Liaohe Western Sag can be obtained with this method. Practice shows that using the analytical hierarchy process to quantitatively evaluate the hydrocarbon accumulation system and then quantitatively predict the hydrocarbon accumulation coefficient decreases the influence of human factors in resources evaluation, and makes the resources assessment more objective and closer to the actual geological condition.

Comparison of different spectral decompositions for non-marine deep water sandstone reservoir prediction in the Xingma area

It is difficult to identify and predict non-marine deep water sandstone reservoir facies and thickness,using routine seismic analyses in the Xingma area of the western Liaohe sag,due to low dominant frequencies,low signal-to-noise ratios,rapid lateral changes and high frequencies of layered inter-bedding.Targeting this problem,four types of frequency spectral decomposition techniques were tested for reservoir prediction.Among these,the non-orthogonal Gabor-Morlet wavelet frequency decomposition method proved to be the best,was implemented directly in our frequency analysis and proved to be adaptable to non-stationary signals as well.The method can overcome the limitations of regular spectral decomposition techniques and highlights local features of reservoir signals.The results are found to be in good agreement with well data.Using this method and a 3-D visualization technology, the distribution of non-marine deep water sandstone reservoirs can be precisely predicted.