The Study On Xujiaweizi Deep Volcanic Gas Reservoir Inversion

With the advances in exploration technology, a growing number of natural gas pools were discovered in the formation of deep volcanic rocks, so more and more attention was paid to the research on natural gas pool. The seismic prospecting is an important method to locate and recognize the lithologic hydrocarbon reservoir, and seismic inversion of seismic exploration technology has been a core technology, The deep volcanic lithology and special geological conditions decide the volcanic reservoir inversion will encounter a lot of technical problems.Xujiaweizi fault depression is an important target for Songliao Basin natural gas exploration in northern Daqing Oilfield, it have a very important strategic significance for the sustainable development of oil fields.Seismic inversion can combine the seismic data, well logging data with the geological interpretation, it can study effectively the change of physical property of reservoir, get the distribution of physical parameters in space, and can guide oil and gas exploration and development.At the basis of the study and analysis of deep volcanic rock samples (Xujiaweizi region), the purpose of this paper is to identify gas-bearing ones from water-bearing rock samples at petrophysical parameters of dual-property crossploting. And the identification methods adopts cluster analysis. First of all, the paper improves the traditional K-means algorithm, and adopts density-sensitive semi-supervised spectral clustering algorithm(DS-SSC) to do Laplacian spectrum transform on the data sets of V p /Vs versus the density, and then to identify the clustering by using of the improved K-means clustering algorithm. It analizes the advantages and disadvantages of the algorithm at the V p /Vs versus density petrophysical parameter crossploting after identified, Finally, to verify further the stability and reliability in the V p /Vs and Z p crossploting, it is a new method for inversion of deep volcanic reservoir identification...