| Geophysical exploration is one of the basic means to solve the geological problems.With the development of the earth science and computer science,it's necessary to propose more stable and effective methods to process the potential-field data for geological interpretation.Based on the hotpots and focus issues in the potential-field data,this paper studies and discusses three aspects potential-field data processing methods,including the conversion of the high-order derivatives in the potential field,the potential field edge detection and the interface inversion methods,which aim to improve the stability and accuracy of the calculation.This paper is based on theoretical research and model tests,combined with the real potential field data processing,aiming to support the interpretation of relevant geological issues.In the potential field data processing,the vertical derivatives have an important physical meaning.Its calculation directly affects the accuracy of some forward and inversion methods(Euler deconvolution,normalized full gradient of gravity anomalies,etc.).It is also possible to divide the overlay anomalies produced by different depths and size anomaly sources.The higher the order of the derivative,the stronger the capability of distinguishing the superimposed abnormalities,but the conversion of the high-order derivatives is not stable.In this paper,we calculate the higher-order derivatives in the potential field based on Tikhonov regularization and use the iterative method to successively approach the approximation.The Tikhonov regularization iterative method which calculates the high-order derivatives of the potential field in the wavenumber domain is proposed,and the recursion formula of Tikhonov regularization iteration method is deduced.Through the analysis of the filter characteristics of this method,we can see that the calculation of high-order vertical derivative of the potential field by this method has a certain stability and amplitude preservation.The model test and actual data processing show that the method has higher stability and practical value.Edge detection is an important task in the processing and interpretation of potential field data.The operators of many existing edge detection methods are composed of horizontal and vertical derivatives of the potential field data.The geological edges identified by these methods are often larger than their true edges.When the depth of the field source is different,the edge will sometimes become blurred,and it is difficult to accurately identify the edges of anomalous bodies at different depths.Aiming at this case,a correlation coefficient method based on the average vertical derivative and its standard deviation in a sliding window is proposed.After the explanation and analysis of its basic principles,we conducted the 2D and 3D model tests of gravity and magnetic data respectively.Compared with other widely used edge detection methods,it can be seen that the algorithm of the proposed method is simple,the field sources of different depths can be detected in the same degree,and its sensitivity to noise is weak.The method uses the zero line of the correlation coefficient to detect the edge of the anomaly body,which makes the edges continuous and clear,and it can be displayed more accurately.When the sliding window is small,more geological details can be displayed,which has practical value for real potential field data processing.The classic Parker-Oldenburg interface inversion method can quickly invert the fluctuation of the interface,which plays a certain reference role in the study of regional structure and oil and gas exploration.However,the classic Parker-Oldenburg interface inversion method has divergence problems.In this paper,on the basis of the classic Parker-Oldenburg interface inversion method,an improved Parker-Oldenburg interface inversion method referring to the regularization-integral iteration method for downward continuation of potential filed is proposed.This method is analogous to the regularization-integral iteration method that changes the divergent downward continuation to the convergent upward continuation,which is different from the divergence of the classic Parker-Oldenburg interface inversion method.In this paper,the method uses the iterative methods of convergent forward to avoid the direct calculation of the exponential factor in inversion.The method can suppress the noise very well and make the calculation stable while iterative convergence.As one of the emerging geothermal energy sources,dry hot rock has the advantages of practicality,environmental protection,low operating costs,safety,and reliability.The Gonghe Basin in Qinghai Province is located in the Cenozoic faulted basin in the ?Qin Kun Fork?.China first discovered that the large-scale available dry hot rock resources are located in the Gonghe Basin in Qinghai Province.Therefore,the study of the geological problems in the Gonghe Basin is of great significance.On the basis of verifying the reliability,stability and feasibility of the data processing methods above,they were applied to the 1:500,000 gravity and aeromagnetic data collected in the Gonghe Basin and the 1:50,000 ground gravity and magnetic data in the Qiaboqia region.Combined with geological information,the characteristics of the gravity and magnetic field of the study area are analyzed;the structural distribution characteristics of the Gonghe Basin and its peripheral faults in the study area were preliminarily inferred;the inversion of the depth of Curie and Moho point in Gonghe Basin and its periphery were completed;the prospect area of the dry hot rock in the Qinboqia area was preliminarily delineated.It is expected to enrich relevant geophysical data and provide some basic support for further research in this area. |
PDF Full Text Request