Comparison Of Forward Methods Of Gravity And Gravity Gradient

Due to the fact that the gravity data is single,and the information of the subsurface geological body cannot be fully extracted,there is a disadvantage in characterizing the geometry and position of the subsurface geological body.Researchers have proposed gravity gradient data for positive and negative inversion for this defect.Compared with the gravity data,the gravity gradient data can accurately identify the boundary information of the geological body and can better solve the drift problem caused by gravity data inversion.The computational time spent in gravity and gravity gradient inversion has severely constrained the application of gravity and gravity gradient methods.This situation has caused widespread concern among scholars at home and abroad.Saving the inversion calculation time and improving the efficiency of inversion calculation become the main problems facing the present.In response to this problem,many scholars have proposed a variety of fast forward methods to solve the problem of low inversion calculation efficiency.In order to compare the advantages and disadvantages of various accelerating methods and the scope of use,we selected several typical forward acceleration methods for comparative study.1.As the most important calculation tool for geophysical exploration,the computer has put forward parallel computing methods after deeply studying the working methods and characteristics of computers.To maximize the use of computing equipment to achieve the purpose of improving the efficiency of forward calculations.Based on the characteristics of the computer CPU,this paper uses the parallel forward calculation method of the CPU to achieve the purpose of improving the computational efficiency.The speedup that CPU parallelism can achieve depends primarily on the number of cores of the computer CPU used.2.When three-dimensional inversion calculations are performed,it is often necessary to perform three-dimensional meshing on subsurface geological bodies and source networks.As the meshing becomes more and more detailed,the number of meshing units increases drastically,which increases the time for calculation.Under the premise of not affecting the subterranean geological grid meshing,the efficiency of forward calculation can be improved.Some scholars have proposed a forward modeling method based on grid merging.This method aims to reduce the number of grid cells and reduce the forward calculation time by merging neighboring grid cells with uniform physical properties.The accelerating efficiency of this method technique depends only on the merge ratio(compression ratio)of the meshing unit merge.3.Due to the computational redundancy brought about by the repeated computation of the grid nodes after the grid body was meshed,some scholars converted the block density values of the grid cells to the density values at the grid nodes.Therefore,a forward calculation method based on the density of mesh nodes is constructed.In this thesis,a three-dimensional forward modeling method based on parallel,mesh-based and mesh node density is realized by programming forward programs.The correctness of the program is verified by the forward model in the published article.Through a variety of model tests,there is a reasonable analysis of the range of use of various methods and the acceleration efficiency.