| The Halahatang area has favorable background for hydrocarbon accumulation,which is located on Lunnan low bulge of Tabei uplift.In particular,the successful drilling of Well Manshen 1 has achieved a breakthrough in oil and gas exploration in the pomace between Tabei Uplift and Tazhong Uplift,and further improved the geological theory understanding of continuous oil bearing in Tabei and Tazhong Uplift.Large thick igneous rocks exist in the Permian in the study area.this paper aims at the complex lithology of the igneous rocks in the study area,and makes comprehensive use of logging and seismic data,combined with seismic attributes,seismic facies and seismic attribute fusion and clustering based on deep learning for lithological distribution.According to the characteristics of the working area,constrained sparse pulse inversion and geostatistical inversion are carried out to develop the lithologic identification and velocity modeling of Permian igneous rocks.Firstly,undertake the overall grasp to the geological situation of the working area,This article has mainly carried on the following work: 1)Aiming at the characteristics of disorderly and abrupt distribution of igneous rocks,Collect and analyze the thin section data of geological logging to grasp the overall distribution of lithology in working area.2)Seismic attributes and seismic facies of igneous rocks were extracted according to seismic data,and the attributes with high correlation,such as arc length attribute root mean square attribute absolute amplitude attribute reflection intensity attribute and energy attribute,were selected to study the distribution range and morphological characteristics of igneous rocks in the working area.3)Based on the theory of deep learning and image fusion clustering analysis,OpenCV library is used to fuse the preferred information of different attributes and Kmeans clustering analysis is used to classify the lithology in the study area.After attribute fusion,the lithology distribution law in the work area can be described more precisely.When the classification number is 5,it can well reflect the “arc” distribution characteristics and basalt distribution.4)Aiming at the problem that it is difficult to identify the igneous channels in this area,the spatial distribution characteristics and geological understanding rules of the igneous rocks in this area are analyzed by using coherence technology and combining with the formation mechanism of the area,which is mainly composed of mantle pillar support and melt overflow.It lays a good foundation for velocity modeling of Permian igneous rock.Secondly,in view of the velocity modeling problem of igneous rocks,the wavelet interpolation method with different frequencies and the small layer control are compared by using the constrained sparse pulse inversion method.In this area,the wavelet frequency of igneous rock inversion is 25 Hz.The interpolation method adopts the natural neighborhood interpolation method and combines with the characteristics of the wide distribution of basalt in the working area,adding the basalt layer inside the igneous rock to carry out fine modeling inversion.The comparative analysis shows that the inversion results of adding basalt are better in describing the fine structure.Finally,on the basis of CSSI inversion,determine the geostatistical inversion study function and parameters such as SNR,which achieves fine reservoir characterization,precise velocity modeling error control 2.12%,improved the further understanding of igneous rock lithology.The fine characterization and velocity modeling of deep underground igneous rock can improve the accuracy of oil and gas exploration in the study area. |
PDF Full Text Request