The Study And Application On The Gravity And Magnetic Joint Inversion Based On The Smoothed L0 Norm (SL0) Constraint Of Compactly Supported Conjugate Gradient Algorithm

The results of geophysical inversion are affected by many factors.Firstly,the detection capabilities are not the same for different methods,thereby leading to the difference of vertical and horizontal resolutions for each method.Secondly,the goal of inversion is using the measuring data to detect underground structure.However,the data are affected by many factors,such as human interference,natural conditions,and acquisition instrument errors.In addition,there would be the unique solutions according to the inversion theory.That is,it may be the same response for the different subsurface models.In order to solve these problems,despite the improvement of the algorithm of single inversion,the joint inversion of different methods or different data of the same method have become a research hotspot.Through the research and improvement of the algorithm,the ideal results could be gotten.With the prior geological information or borehole data,the geological problems could be analyzed by the inversion methods.In this paper,it will mainly studied and discussed on the single inversion and joint inversion method.First,the formulas of the kernel function are given.By setting the spatial distribution of the three-dimensional density and the magnetic susceptibility,the forward responses are calculated.Second,the inversion theory is briefly summarized,and the least squares inversion method is used,while the results show that it is not applicable.Furthermore,the discussion of depth weighting and the theory of inversion based on the smoothed L0 norm(SL0)constraint of compactly supported conjugate gradient algorithm are introduced,and the calculation of 2D and3 D examples are given.On the basis of this method,the objective function and the calculation flow of joint inversion of gravity and magnetic methods are established by introducing the cross gradient.Then,similarly 2D and 3D examples are analyzed.Finally,this method is applied to the satellite gravity data of Shuanghu area,and the joint inversion method is applied to the inversion calculation of gravity and magnetic data in Sichuan Basin and Gonghe basin with discussions.Through the works mentioned above,the following results and conclusions could be got.Firstly,due to the deep attenuation and the invalid problem.There will be the obvious skin effect and multi-solution results when using the least squares inversion.The regularization inversion method and the depth of the weight function can be helpful to reduce the impact of these factors.As one of the regularization inversion methods,the compactly supported method can highlight the position of the abnormal body.After introducing the L0 norm,the focus degree and the calculation efficiency of the single data are improved.For the joint inversion of different types of data,it is necessary to introduce the calculation of the cross gradient value.The crossed gradient value can show the change of the different attribute values of the abnormal bodies,so as to use the coupling relation of the geological body to realize the joint inversion of the different physical parameters.The combination of gravity and magnetic inversion of the model shows the advantages of high computational efficiency,relatively small error,and more model information.The inversion of the gravity anomaly in the Shuanghu area shows the distribution of different density rock mass.As for the Sichuan Basin,the results show that the results of inversion of gravity and magnetic data reflect the distribution of density stratification and different magnetic susceptibility regions,which are helpful to delineate the basal lithology inference.The joint inversion results of gravity and magnetic data in Gonghe basin show that geological distribution of information,corresponding with the related geological information well.This study shows that there is certain advantages of using gravity and magnetic joint inversion methods based on SL0 constraint of compactly supported conjugate gradient algorithm.Through the discussion of the actual data,the effectiveness and the value of the method are explained.This would provide a reference for the future multi-data joint inversion and the application and discussion of more practical data.