Study On Complex Near Surface First Arrival Tomography Method With Multiple Information Constraints

The focus of onshore oil and gas exploration is turning to the western and southern piedmont zones with complex surface and underground geological conditions.The quality of seismic data collected in mountainous areas is poor due to the large relief of near-surface topography,complex surface structure,serious variation of transverse velocity,and poor acquisition conditions.Therefore,obtaining accurate near-surface velocity has become a key problem to solve seismic imaging in near-surface areas.As for the method to obtain near-surface velocity,first-arrival traveltime tomography inversion is a good choice,but the complex and variable model space velocity makes the distribution of ray path very messy,which will greatly increase the discomfort of the solution process,thus affecting the resolution and reliability of tomographic imaging.Therefore,it is of great significance to further study travel time tomography inversion with higher accuracy,more robust inversion process and higher efficiency.First of all,in order to obtain a ray tracing method with relatively high computational efficiency and accuracy,which is suitable for complex near-surface ray tracing,a new shortest path ray tracing method with bi-directional Angle restriction is proposed based on previous studies.The core idea is to select the desired nodes by Angle restriction when calculating the nodes near a sub-focal point,so as to avoid a large number of useless nodes.Similarly,when calculating the travel time of a certain grid node,Angle restriction is also used to increase its upper level node used to calculate the travel time to improve the calculation accuracy.The ray path of the algorithm is obtained by recording the focal point information in the process of calculating travel time,and then reverse tracking.Secondly,based on the derivation of the basic formula of the perturbation traveltime integral,the basic formulas of different iterative optimization algorithms are derived,including ART method,SIRT method,CG method and LSQR method.On the basis of solving the tomographic inversion equations with LSQR method,the inversion process is given,and the algorithm is verified by different models.The application effect of the tomographic algorithm and its shortcomings are analyzed,and some conclusions are made.Finally,the article will surface measurement method combined with primary wave traveltime tomography,this paper expounds the basic principle of micro logging method and example verification,analyzes the refraction method applied to the calculation of the horizontal layered model and roll model adaptability,chromatography and constraint is the initial velocity model,the purpose is to use more information constraints,improve the accuracy and precision of the tomography,And ensure its adaptability to complex surfaces.The validity of multi-information constraints is verified by different models.