Study On Improvement Of Downward Continuation Iterative Algorithm For Magnetic Anomaly Data

All along,navigation technology in all areas have shown an important role.The inertial navigation device has the error and drift,the inertial positioning error accumulates with time,and the periodic error correction is needed.In recent years,optical atomic magnetometer is widely used in the field of aeromagnetic field,using of optical atomic magnetometer can obtain high-precision geomagnetic data,making geomagnetic navigation can provide high-precision navigation information correction source;inertial/geomagnetic integrated navigation for submarines accurate navigation information.One of the key problems in the realization of underwater geomagnetic navigation technology is to establish a high-precision and high-resolution underwater geomagnetic database,and the most direct and effective method of establishing geomagnetic database is field continuation technology.In order to achieve the requirement of establishing high precision and high efficiency geomagnetic database,this paper studied the field continuation technology.Aiming at solving the problems of low continuation depth and poor stability;in this paper,a large scale,fast and stable downward continuation algorithm is proposed.The main work of this paper is to improve the downward continuation iterative algorithm of magnetic anomaly data.In the process of the magnetic anomaly data,fast Fourier transform appears boundary effect when processing the data in the frequency domain.Using iterative method as an example,we analyzed the boundary effect and the edge expansion method,in order to remove the influence of boundary effect in the process of data processing.By using the theoretical model and the measured data,the magnetic anomaly data of different edge expansion methods were continued by iterative method.The experimental results showed that the magnetic anomaly data continued by the method of "expanding edge-processing-shrinkage" has high accuracy.At the same time,this paper also studied the generalized inverse algorithm of magnetic anomaly continuation.The continuation operator of this method is consistent with Tikhonov regularization operator.In the past,the size of the damping factor in the generalized inverse operator is given according to the experience.In this paper,according to the regular theory,the optimal damping factor parameter is selected by the L-curve method.Finally,the theoretical model data and the measured data are used to verify this method,The results show that this method has a higher continuation accuracy.Finally,we combined the iteration method with the generalized inverse algorithm of the potential field downward continuation,and the generalized inverse operator in the wavenumber domain was used to replace the downward projection of the iterative method,which method was called iterative compensation method.Recursive operator in the wavenumber domain of the iterative compensation was deduced to one-step compensation operator which related to the damping factor,continuation depth and compensation number.We pushed to filter factor from one-step compensation operator,and discussed the frequency characteristics of this filter factor.Finally,we also analyzed one-step compensation operator convergence and regularity,proved one-step compensation algorithm was regularization algorithm.The theoretical model data and real data are respectively verified and compared with the generalized inverse algorithm,the iterative compensation algorithm and the one-step compensation algorithm,and indicated the one-step algorithm has higher stability and computational efficiency.