Research On Seismic Signal Enhance Method Based On Seismic Interferometry

In recent years,with the gradual shift of exploration targets towards complex,hidden and deep exploration areas,the environmental limitation to data acquisition have gradually increased,which makes it difficult for seismic data to meet imaging requirements and ultimately unable to obtain high-precision imaging results.Under complex geological conditions,data acquisition is severely restricted.Obstacles such as mountains,rivers,villages,etc.affect the regular setting of observing system.Equipment failure or human factors may also cause irregular or incomplete data,which may lead to aliasing in imaging and seriously affect imaging quality.Moreover,the recorded dataset under complex conditions is often interferred by noise,including industrial electromagnetic interference,surface waves,etc.These noises have strong amplitudes,which can mask the useful signals in seismic data and make them unrecognized.Furthermore,complex terrain can also lead to mutual interference between different seismic wave fields,resulting in a reduction in the wavelet resolution,an increase in the width of the events,and inability to same-phase stacking,ultimately leading to imaging distortion and even information loss.The existence of these problems means that the impact of seismic data quality on imaging accuracy is increasingly deepening.This prompts geophysicists to propose a seismic data processing method that can effectively improve the quality of seismic data and enhance the effective information to ensure the final seismic imaging quality meet the needs of seismic exploration.Seismic interferometry,as a fully data-driven technique,can generate reflection response between any two seismic traces without the need for detailed prior information.This method is easy-implemented and cost-efficient,which significantly improves the flexibility of seismic exploration.Especially in the presence of complex overlying strata,seismic interferometry technology can avoid the influence of complex overlying strata by reconstructing wave field data,so that the reconstructed seismic response is focused on the target area,thus enabling accurate structural imaging of the target body.The proposal of seismic interference theory and methods has greatly enriched people’s understanding of the laws of seismic wave propagation.It can extract useful information from chaotic seismic signals and has enormous development potential,which is regarded as an important breakthrough in seismic exploration technology.This thesis combines the advantages of seismic interferometry and focuses on the difficulties in seismic data processing.It aims to enhance seismic signals from three aspects: amplitude,undersampling,and resolution,striving to provide high-quality input data for subsequent seismic imaging,and thus providing new ideas for the application of seismic interferometry in seismic data processing.The research content mainly includes: weak signal enhancement methods,interpolation reconstruction of near offset missing data,and ghost suppression of virtual shot gathers.The effectiveness and applicability of the research method were verified through synthetic data and field data test.Firstly,according to the problem of weak signals being difficult to be picked under strong noise conditions,this thesis proposes a signal enhancement method based on Radon domain supervirtual refraction interference to recover seismic weak signals at far offset section.On this basis,this thesis applies the proposed method to the first arrival refraction signal.According to the kinematic characteristics of the first arrival refractions,supervirtual refraction interference is selected in the linear Radon domain,and a model shrinkage iterative algorithm is introduced to highlight the effective signal,suppressing the strong noise interference while enhancing the effective first arrival signal.On this basis,this thesis also explores the effect of missing channel seismic data on the enhancement method.Secondly,in response to the difficulty of reconstructing near offset data in seismic data processing,a framework for reconstructing near offset data based on iterative seismic interferometry is proposed.This framework expands the applicability of data reconstruction methods and provides technical support for subsequent processes such as multiple attenuation,seismic imaging,and data interpretation.Its basic conception is the seismic interference interpolation method,which utilizes the flexibility of the interference interpolation method to convert multiple reflection in marine data into missing near offset wave fields,and constructs a global amplitude balancing operator using the characteristics of Focal transformation to solve the amplitude imbalance problem between the virtual shot gathers and the real trace set generated by the interference.Finally,through iterative updates between the interferometric interpolation and the balancing operators,the error between the final reconstructed data and the observed signal converges,achieving the goal of accurate reconstruction of near offset missing data.Finally,a deghosting algorithm based on nuclear-sparse norm constraints is proposed to address the problems of low resolution and non-physical wave interference in the virtual shot gathers after interferometry.This solves the problem of conventional ghost suppression methods being ineffective due to the large amount of coherent noise caused by seismic interferometry,and eliminates its limitations on the application of seismic interferometry.This method abandons traditional shot by shot operations and adopts the simultaneous inversion of seismic data from multiple shot gathers.The objective function includes two penalty terms: the low rank constraint based on nuclear norm take the stratigraphic continuity as prior constraints,and the sparse constraint based on L1 norm use the amplitude difference between seismic and noise signals as prior constraints for the inversion method,respectively.The proposed method reducing the sensitivity to noise and then improving the stability and accuracy of the ghost suppression algorithm.