A Reservoir Lithology Identification Method Based on Prestack Multi-parameter Dimensionality Reduction

Abstract

The invention provides a reservoir lithology identification method based on pre-stack multi-parameter dimensionality reduction, and belongs to the field of oil exploration. The reservoir lithology identification method comprises the following steps: performing pre-stack seismic inversion under the restriction of a sedimentary facies belt to obtain an elastic parameter inversion data body; optimizing the elastic parameter to obtain an optimized sensitive parameter; performing dimensionality reduction on the optimized sensitive parameter to obtain a function relationship between a new parameter after reducing dimensionality and the original parameter; applying the function relationship between the new parameter after reducing dimensionality and the original parameter to a corresponding pre-stack inversion data body to obtain a lithology identification data body; judging by means of the lithology identification data, to reach the purpose of identifying the reservoir lithology. The reservoir lithology identification method based on pre-stack multi-parameter dimensionality reduction obviously improves the reservoir predicting precision, so that the reservoir calculation result is higher in precision, the calculation result is more reliable, and the exploration risk is effectively reduced.

Description

technical field [0001] The invention belongs to the field of petroleum exploration, and in particular relates to a reservoir lithology identification method based on pre-stack multi-parameter dimension reduction. Background technique [0002] At present, a variety of reservoir prediction methods have been developed at home and abroad, such as seismic facies analysis, seismic attribute analysis and seismic inversion based on post-stack seismic data, which are the most widely used exploration techniques in reservoir description. For a long time, the post-stack inversion method has achieved good results in practical applications. For example, the patent application number 201110171068.2 discloses a method for predicting reservoirs from seismic inversion data, which realizes the wave impedance obtained by model inversion. It is a new method for lithology or reservoir prediction based on or velocity data, but only for the situation when the reservoir near the wellbore and the adj...

AI Technical Summary

Problems solved by technology

[0003] (1) The lithology of the target layer is complex, including mudstone, quartz sandstone, lithic sandstone, conglomerate and calcareous sandstone. The geophysical response characteristics of calcic sandstone reservoirs are not significantly different from those of other lithologic rocks. The effect of reservoir prediction using seismic attribute analysis technology is not ideal;

[0004] (2) There is little difference in wave impedance between different lithologies, and although the elastic parameters of different lithologies (including Poisson's ratio, bulk modulus, Lamé constant and density product, elastic wave impedance and other parameters describing rock properties) The overlapping range is smaller than that of wave impedance, but it still does not meet the requirements of high-precision reservoir lithology identification;

[0005] (3) The porosity of the target layer is relatively low, and the average porosity of the middle subsection of the target layer is only 3.2%, which belongs to tight sandstone reservoirs. At present, no effective petrophysical interpretation template for describing tight sandstones has been developed. Theoretically, the physical interpretation template for multi-parameter lithology identification is not fully mature, and the quantitative seismic amplitude interpretation of tight sandstone lacks sufficient application examples

Images