Research On Interpretation Method Of Gravity Gradiometry Data Based On Correlation Coefficient

Gravity exploration plays an important role in geophysical exploration due to its advantages of high work efficiency,low exploration cost,and easy operation,and it is widely used in the fields of geological structure survey and mineral resource exploration.Aiming at the problem of low resolution in current boundary recognition and rapid imaging methods,this thesis has carried out research on the interpretation method of potential field data,and proposed a comprehensive interpretation method of gravity gradiometry data based on correlation coefficients,which was verified in theoretical models and measured data.Field source boundary recognition is an indispensable link in the interpretation of potential field data,accurately extracting boundary information can enhance the understanding of geological structure.The shortcomings of conventional boundary recognition methods are insufficient detection accuracy,sensitive to noise,and may produce false anomalies,resulting in the loss of boundary details.To solve this problem,this thesis proposed a boundary recognition method based on correlation coefficient based on the EDT.This method does not require gradient calculation,is simple and easy to implement,and has strong anti-interference ability.By establishing different models and comparing them with several commonly used boundary recognition methods,the results show that the boundary recognized by this method is clearer,more stable,accurate,and has good noise resistance.For the window size problem involved in the method,this thesis designed different windows to compare and analyze the effect of boundary recognition,and got the best window size.Each independent gradient component of full tensor gravity gradiometry(FTG)contains different geological information,and combining different components can improve the accuracy of geological interpretation.Therefore,this thesis obtained the best combination of gradient components Gxz|Gyz|Gzz by combining different gradient components,it is proved that this component combination has the best effect in the model test;Fast imaging has developed rapidly due to its advantages of high computational efficiency,no need for any prior information,and stable results,this article mainly focuses on the related imaging methods.The shortcoming of related imaging lies in the low vertical resolution.To solve this problem,this thesis introduces a depth weighting function based on prior information,the vertical resolution of the target body and the ability to distinguish between adjacent geological bodies are further improved by dividing the research area.Fast imaging obtains the equivalent physical parameters of the target body,not the true density or susceptibility.Therefore,this thesis carried out 3D density inversion based on related imaging,improved the inversion algorithm,including:prior density constraints,gradient component combination,and depth weighting function,etc.,and the feasibility of the inversion method was verified by model tests.Finally,on the basis of boundary recognition,related imaging and 3D density inversion,this thesis proposed a comprehensive interpretation method of gravity gradiometry data based on correlation coefficients,and summarized the specific process of the method.Applying it to the measured airborne gravity gradient data of the Vinton Salt Hill,the interpretation results in the x,y,and z directions were consistent with the geological data,logging information and other scholars’research results in the area.The measured data confirmed the reliability and practicability of the method.In summary,this thesis proposed a comprehensive interpretation method of gravity gradiometry data based on correlation coefficients in view of the shortcomings of conventional boundary recognition and rapid imaging.This method has the advantages of simple and efficient calculation,accurate identification of anomalous objects,etc.,and provides a theoretical basis for future high-precision potential field exploration data processing and interpretation.