Research On Seismic Weak Signal Enhancement Method

In recent years,with the further exploration of oil exploration,the exploration targets have been transferred from the conventional oil and gas reservoirs to subtle reservoirs,such as thin interbed,small sand body,micro fracture and so on.The study of seismic reflection weak signal is inevitable,and the research on the enhancement of seismic weak signal has become a new topicThe study of seismic reflection weak signal is inevitable,and the research on the enhancement of seismic weak signal has become a new topic.In this paper,three kinds of seismic weak signal enhancement methods are proposed.In the first method,by combining the independent variable analysis theory and the Kalman optimal estimation theory,the reflection coefficient of seismic data is extracted and the resolution of seismic data is improved;In the second method,the theory of compressive sensing is used to obtain the reflection coefficient and the weak signal is enhanced;The third method is to enhance the weak signal by calculating the instantaneous attributes.The first method in this paper combines two important theories: independent variable analysis theory and Kalman optimal estimation theory.The independent variable analysis theory can be used to decompose the mixed signal into independent source signals.Kalman optimal estimation theory can be used to calculate the estimated value,so that the estimated value is close to the actual value.The application effect of data in different work areas is significantIn this paper,second methods are applied to the field of weak signal processing in seismic exploration.As a new sampling theorem,compressed sensing theory has its unique advantages in signal processing compared with Nyquist's theorem.Firstly,the signal is sparsely represented by four parameter B spline wavelet bases,and then the nonsingular observation matrix is designed.Finally,the signal reconstruction is realized by the norm calculation.The method is applied to model and field data and its application results show the effectiveness of the proposed method.Empirical mode decomposition(EMD)is a signal decomposition algorithm,which can decompose the signal into the form of the superposition of the intrinsic mode function.The instantaneous attributes are obtained by the orthogonal transformation and the normal shear transformation of the intrinsic modefunctions.Compared with the traditional Hilbert transform,the normalized direct orthogonal method based on empirical mode decomposition(EMD)avoids the limitation of the Hilbert transform method,and obtains more accurate seismic instantaneous attributes.