| Controlled source electromagnetic method is an important technical tool in geophysical exploration,which is widely used in solid mineral and oil and gas resources exploration,environmental engineering monitoring,geological disaster prediction and other fields.As the demand for electromagnetic fine structure detection in the national"deep earth" strategy increases and the electromagnetic exploration environment becomes more complex,the research and development of controlled-source electromagnetic instruments and data acquisition technology,forward and inverse theory and data interpretation methods have raised higher requirements.However,compared with the actual demand,the 3D forward and inverse algorithm of controlledsource electromagnetic method in complex media environment is not yet mature,and the accuracy of simulation and inversion of undulating terrain and complex subsurface structures is limited,and the research on 3D forward and inverse of controlled-source electromagnetic method under undulating terrain conditions is urgently needed.In addition,at this stage,the interpretation of controlled-source electromagnetic data processing is usually based on single source and single component,which is easy to cause data waste and interpretation error,while multi-information can increase the amount of data and multi-directional constraints,improve the inversion effect,and recover the location morphology and electrical information of anomalies more accurately,so it is important to carry out the research of controlled-source electromagnetic method 3D forward and inverse under undulating terrain.In this context,this paper achieves accurate fitting of undulating terrain and complex geological anomalies by using unstructured tetrahedral grid dissection,and at the same time,the grid is encrypted near the emission source,reception point and anomalies,and the grid is sparse far from the calculation area,so as to effectively control the number of grids of the forward model and balance the problems of accuracy and efficiency of the forward calculation caused by the poor fitting of the undulating terrain.The vector finite cell method is used for the forward calculation,and the direct solution method is used for the finite element equations in the algorithm to effectively reduce the error effects of the missing scattering conditions and the poor condition number caused by the grid.The calculation of the frequency domain electromagnetic field of the horizontal laminar medium model verifies the correctness and timeliness of the forward algorithm in this paper,and then numerical simulations are conducted by constructing various types of undulating terrain models to analyze and compare the influence characteristics of the undulating terrain on each component of the electromagnetic field,and to discuss the forward shadow effect of the controlled source electromagnetic method and the effectiveness of the multi-source emission model to overcome this effect.Based on the realization of high-efficiency and high-precision three-dimensional forward,the inversion adopts the nonlinear conjugate gradient strategy,and uses the accompanying forward method to calculate and store the correlation terms of the sensitivity matrix to obtain the data and model gradients,so as to realize the threedimensional high-efficiency inversion of controlled-source electromagnetic method for undulating terrain based on nonstructural grid.The effectiveness of the inversion algorithm in detecting the structure of subsurface anomalies is demonstrated by the trial calculation of the inversion of a simple anomalous body model in uniform media.The iterative results of multiple undulating terrain models and the initial model of inversion without considering the undulating terrain show the necessity of undulating terrain for the analysis and interpretation of inversion data.Based on the metallic deposit model of a deposit in western Hunan Province,a 3D inversion of the forward synthetic data is carried out and good results are obtained.By comparing and analyzing the inversion results of single-source and multi-source on the same model,and the inversion results of single-component and multi-component on the same model,the advantages of multisource model and multi-component inversion are determined.All the above theoretical models illustrate the stability and reliability of the inversion algorithm in this paper,which can adapt to the interpretation of 3D inversion in complex detection environment and provide the theoretical basis for the subsequent development of the processing and interpretation of 3D real measurement data. |
