Selection Of Seismic Inversion Methods In Deep-water Reservoir Prediction

The deep water reservoir is one of the main targets of oil and gas exploration at present and the deep water oil and gas reserves account for about 30% of the total oil and gas reserves in China.Due to the large reserves of deep water oil and gas resources,the success of deep water oil and gas exploration will provide great support for economic and social development.Under the influence of many factors,the exploration level of deep water oil and gas resources is limited at present in China and the research on deep water reservoir prediction is not deep enough,which needs to be further studied.Seismic inversion technology plays an important role in reservoir prediction and it is an important reservoir prediction method.There are many seismic inversion methods,such as post-stack wave impedance inversion? prestack elastic impedance inversion? nonlinear AVO inversion and geostatistical inversion.Different inversion methods have their own advantages,disadvantages and applicability.For different deep-water reservoir study areas,appropriate inversion methods should be selected.In this paper,the W deep water area of the Pearl River Delta is taken as the research target area.Firstly,the common seismic inversion methods and their applicable areas are introduced and analyzed.Then,according to the actual situation of the target area,the constrained sparse pulse inversion and the random seismic inversion of geological statistics are selected to be applied to the research.Finally,the advantages and disadvantages of the two inversion methods are compared and analyzed.And the following understandings are obtained:1)In the traditional post-stack wave impedance inversion method,on the one hand,the recursive inversion and trace integral inversion are simple and practical,which can reduce the multiple solutions in the model-based inversion method,but it is not suitable for lithology identification in the complex area of deep water.On the other hand,the inversion method based on the model can predict the transverse distribution of reservoir in deep water area by establishing an accurate initial model,but it requires high well data and is prone to multiple solutions.Also,it has strong dependence on the model.2)The lithology of deep water reservoir is complex and the accurate identification of deep water area reservoir can not be realized by relying solely on P-wave impedance data.As a result,we must take wave impedance data body as inter-well constraint condition,by establishing the covariant difference function between vertical wave impedance and GR data.Then,getting the GR inversion data body by means of associated simulation so as to achieve the purpose of fine description of thin sand body with more distribution in deep water area.3)Constrained sparse pulse inversion compensates for the lack of low frequency information of deep water seismic data affected by ghost by establishing an accurate low frequency model,which can solve the problem of heterogeneity in deep water reservoir.The random seismic inversion based on random modeling technique effectively synthesizes geological,logging and three-dimensional seismic data,greatly improves the vertical resolution of inversion results,and can accurately depict the distribution of sand bodies in sand and mudstone thin interbedded reservoirs in the target area,which can provide a reliable basis for the identification of deep water reservoirs.