Research And Application Of The Extraction Method Of Seismic Attributes

With the further and deepening exploration as well as development of oil and gas,as a geophysical technology which is rapidly developed,reservoir prediction methods are becoming more and more important and significant,and the methods of seismic attribute analysis are one of the most common methods.Seismic properties can be regard as the tools for inferring geological conditions and lithological conditions from seismic reflection data through some mathematical processing methods or signal processing methods.The seismic reflected wave is a kind of seismic wave that is back to the ground when the seismic wave is transmitted through the ground and encounters an interface with wave impedance.After the received reflected wave is processed,we can read the situation of the underground stratum structure and the changing through it.When the horizon or structure changes,the characteristics of the reflected wave and the extracted seismic attributes will also show the corresponding trends.The purpose of our study of seismic attributes is to study the characteristic information contained in subsurface formations and structures through seismic attributes,which is also the physical basis for studying favorable reservoirs.To a certain extent seismic attributes help geologists interpret seismic data and describe oil and gas reservoirs by quantifying the characteristics of seismic data reflecting from the ground below,revealing hidden features of subsurface formations,identifying similar patterns,and quantifying special attributes as well.Seismic attribute analysis is a significant branch of petroleum exploration reflection seismology,which has been widely used in anomaly identification,extracting the seismic attributes and predicting the hiding possible reservoirs.In recent years,with the rapid development of signal processing technology,some new and high-precision signal processing methods have been introduced into the data processing of seismic exploration,such as Hilbert-Huang transform,wavelet transform,generalized S transform,etc.Adopting these methods greatly improves the signal-to-noise ratio of seismic signals and improves the accuracy and reliability of seismic attribute analysis.This paper is based on the technical requirements for improving the accuracy of seismic attribute analysis in the project named “Predicting the dense reservoir dessert of the Fengcheng Formation in the West Slope of the Mahu Depression”.By studying and comparing the effects of different signal processing methods in seismic attribute analysis,it is considered that the seismic attributes extracted by other two kinds of signal processing methods,Hilbert-Huang transform and generalized S transform,are fitted perfectly the needs of the study area.The seismic spectral-decomposition properties and the sweetness properties were applied to the study area dense reservoir dessert prediction,and good results were obtained.Based on the analysis of integrated seismic spectral-decomposition attributes and the results of inversion,the high-quality reservoir distribution in the study area was predicted,which provided a scientific basis and authorized method and technical support for the exploration and development of dense reservoir dessert prediction in the area.