Rapid and high-precision forward modeling method for fluctuation stratum frequency domain

Abstract

The invention provides a rapid and high-precision forward modeling method for the fluctuation stratum frequency domain. An existing calculation method is improved, and the method comprises the steps that firstly, upper and lower interfaces of a field source are unfolded according to the Taylor series on corresponding mean water levels l1 and l2, the absolute fluctuation values of the upper and lower interfaces are converted into relative fluctuation amounts, and the forward modeling algorithm precision is improved; secondly, when rapid Fourier transformation is performed, Gaussian nodes are added so that the integral region is subdivided, the existing rectangular integral is converted into the Gaussian type integral, the problems appearing in boundary vibration effect, forced periodizationand a series of other frequency domain forward modeling processes are effectively solved, and the frequency domain forward modeling numerical precision is improved. According to the method, due to model experimental verification, the good effect on ensuring both calculation efficiency and forward modeling precision is obtained.

Description

technical field [0001] The invention relates to the technical field of geophysical gravity prospecting, in particular to a frequency domain forward modeling method of gravity potential field under the condition of undulating formation; Taylor expansion, and the introduction of Gaussian algorithm in Fourier calculation, significantly improved the numerical accuracy of forward calculation. Background technique [0002] As one of the most basic geophysical methods, gravity exploration has been widely used in technical fields such as terrain correction, hydrogeological survey, geoid measurement, solid mineral resource exploration, lithosphere crust-mantle structure research, etc. Research can carry out geological work in related fields, such as dividing the inner circle structure of the earth, in-depth research on regional structures, and the exploration of underground mineral deposits such as oil storage structures, etc. Compared with other geophysical prospecting methods, gra...

AI Technical Summary

Problems solved by technology

[0004] The existing gravity forward modeling methods can be divided into two categories according to the calculation domain, the space domain and the frequency domain. Among them, the conventional topographic correction or crustal density correction usually adopts the space domain forward modeling method. Although the forward modeling accuracy of this method is high, the forward modeling method The calculation formula is particularly complicated, the geometric parameters of the source body in the space domain and the forward formula are cumbersome, and the calculation efficiency is very low

In the frequency domain method, due to the use of fast Fourier transform, the forward calculation formula is greatly simplified, and the correspondence between the anomalous spectrum and the geometric characteristics of the field source body is very simple, so that the calculation efficiency is significantly improved; but the disadvantage is that compared with the spatial The calculation accuracy of the domain method is greatly reduced. This is because the fast Fourier algorithm is a numerical method that uses the trapezoidal rule to calculate the continuous Fourier transform, which belongs to the continuous integral calculation, and the continuous Fourier transform belongs to the oscillation integral problem, so The traditional fast Fourier numerical algorithm will inevitably produce a series of forward modeling problems in the frequency domain, such as boundary oscillation effects and forced periodicity, which will affect the numerical accuracy of the forward modeling in the frequency domain

At present, the common solution to the calculation accuracy in the frequency domain is to discretize the integral interval, so that the continuous integral calculation is converted into the rectangular integral calculation, but there is still room for further optimization and improvement

Images