Study On The Compensation And Inversion Method Of UAV Aeromagnetic Measurement Data

Aeromagnetic exploration is a magnetic exploration method to detect the change of the earth's magnetic field by loading the magnetometer on the aircraft.The traditional means of aeromagnetic exploration are mostly fixed-wing UAV aeromagnetic measurement.However,with the miniaturization of aeromagnetic instrument and the development of UAV technology,UAV aeromagnetic measurement plays an increasingly important role in geological exploration,mineral exploration and unexploded ordnance detection due to its advantages of low cost,small size,less restriction and no safety hazard.The ferromagnetic material contained in the UAV itself will produce magnetic interference during aviation flight,which seriously affects the accuracy of UAV aeromagnetic measurement data.At the same time,compared with the traditional ground magnetic measurement,UAV aeromagnetic measurement data has the characteristics of small sampling spacing and large amount of data.The traditional inversion method for geological target interpretation has the problems of large memory consumption and slow calculation speed.This paper focuses on the problems existing in the data processing and interpretation of UAV aeromagnetic measurement,so as to improve the measurement accuracy and inversion efficiency of UAV aeromagnetic data.In this paper,the single-rotor UAV is selected as the platform for airborne magnetometer,and a set of UAV aeromagnetic detection system is integrated and developed.Based on the Tolles-Lawson model,the numerical simulation experiment of magnetic interference in aeromagnetic measurement of UAV is carried out,and the least square method and ridge regression algorithm are used to compensate it,respectively.The effectiveness and robustness of the method are verified.In order to further improve the accuracy and generalization ability of aeromagnetic compensation algorithm,this paper uses neural network algorithm to compensate aeromagnetic data.Through the compensation processing of the classical BP neural network in the simulation data and the actual flight data,it is proved that the neural network has higher compensation accuracy and stronger generalization ability.Then a radial basis function(RBF)neural network method with higher compensation accuracy and stronger generalization ability is proposed for the first time,and it is applied to the aeromagnetic compensation calculation of UAV to further improve the performance of the compensation algorithm.For the collected original UAV aeromagnetic measurement data,the magnetic anomaly data caused by the underground target body are obtained through magnetic compensation and equalizing data processing.In order to achieve the inversion of the physical parameters of the target body,this paper carried out the research on the regularized three-dimensional physical inversion method,and adopted the adaptive sampling method to conduct the sparse sampling of the data.While reducing the amount of data involved in the inversion calculation,the main information of the data is retained as much as possible,which improves the inversion efficiency of the large-scale UAV aeromagnetic measurement data.Finally,the method established in this paper is applied to the UAV aeromagnetic measurement data of Ningwu mining area in Ma ' anshan.Through the processing and interpretation of the original aeromagnetic observation data,the effectiveness of the aeromagnetic compensation method and the physical property inversion method established in this paper is verified,and the application level of UAV aeromagnetic measurement technology in mineral resources exploration is improved.