Analysis And Research Of Euler Deconvolution In Gravity And Magnetic Inversion

Euler deconvolution is one of the most popular techniques to estimate the source location in potential field methods.For a few ideal modes,Euler deconvolution solution is reliable,but some interference could influence the credibility of the solution.Up to now,the problems that how to effectively deal with the interfering factors and select best Euler solutions have not been well solved.To deal with the above problems,The influence of the structural index,the background,the singularity of the Euler equation and the window size have been studied and analyzed.1)By deriving formulas for the structural index,we obtain that there is a strong coupling between the structural index and the estimated depths and the structural index has a greater effect on estimated depths than horizontal positions.2)We propose a new method to deal with the influence of the background,which uses reconfiguration Euler equation to obtain inversion solution,and it can effectively improve the accuracy of the inversion solution.3)The damped least square method is used to weaken the singularity of the coefficient matrix in Euler deconvolution.4)And we find that the damping factor could push the Euler solutions to the coordinate origin.Based on this phenomenon,We propose a new method of selecting window size.By increasing the window size,we could pick out the optimum window size with a small degree of compression.Our selection method could greatly improve the solution accuracy.To select the Euler solutions near the source,we do some research on the distinguish technique.We propose two new selection methods,named variable damping factor selection method and variable coordinate reference point selection method,based on the stability of the Euler equation after we study the Euler deconvolution based on the damped least square method.Besides that,We also propose another new method to select the best Euler solutions based on the angle variation after we study the spatial distribution law of the Euler solution and the sliding window.All the proposed algorithms are tested by synthetic and real-data examples and produce favorable results.Our algorithms could provide important information for further geologic interpretation.