Research On Extracting Rayleigh Wave Dispersion Curve Based On S-transform And Cross-correlation

Rayleigh surface wave exploration is a commonly used shear wave exploration method in geophysical exploration,which has the characteristics of quickness,convenience,wide applicability,non-destructive,and strong energy.The dispersion curve,as the feature of the dispersion properties of Rayleigh surface waves and the basis for inverting,is the most important information in Rayleigh surface wave exploration.Therefore,the extraction of the actual dispersion curve has become a key link in Rayleigh surface wave exploration.Currently,the research on dispersion curves mainly focuses on low frequency,high mode,and dispersion curves of weakly energetic Rayleigh wave in high noise environments(referred to as weak Rayleigh waves in the following).The extraction of dispersion curves actually involves two steps,one is dispersion energy imaging,and the other is picking up dispersion curves.This article will combine S transform,common middle point idea,and cross correlation MSC(Multi-channel Signal Comparison)method from two aspects,and attempt to extract Rayleigh wave dispersion curves from weak Rayleigh wave seismic records.The main research work and conclusions completed in this article are as follows:(1)Based on Rayleigh surface wave theory,the forward modeling of Rayleigh surface wave theoretical dispersion curves through fast vector algorithm,and seismic records through theoretical dispersion curves and IFFT is realized.(2)Implements conventional the τ-p transform method and the phase shift method,and introduce the S-transform and common center point idea to improve the cross correlation MSC method,and the improved method is abbreviated as the S-transform cross correlation method.The feasibility of the S-transform cross correlation method was verified using a uniform medium model,and the dispersion curve extraction effects of the above three methods were compared using a simple layered medium model.The τ-p transform method,the phase shift method,and the S-transform cross correlation method can accurately perform fundamental dispersive energy imaging.The phase shift method has the best effect in the absence of noise,but its anti noise ability is poor,while the S-transform cross correlation method exhibits good anti noise ability.(3)Based on the actual seismic records in an alluvial fan Gobi region,this paper further compares the frequency dispersion curve extraction effects of the above three methods.For small and linear array observation systems of ground source,Rayleigh surface waves are often well developed.Among the three methods,the phase shift method has the best continuity of dispersion energy,the highest resolution,and the best extraction effect of dispersion curves.However,for large and nonlinear array observation systems of underground source,Rayleigh surface waves are very weak,and the anti noise advantage of the S-transform cross correlation method is embodied.It can still pick up the dispersion curve of the high-frequency part under high noise conditions,and accurately inversive the shear wave velocity of the surface layer.(4)In the process of picking up the dispersion curve from the dispersion energy diagram,this paper uses prior information such as refraction waves to pick up the dispersion curve,which effectively avoids some regular interference and abrupt points,and reduces the error of picking up the dispersion curve.