Research Of Magnetic Target Small Sub-domain Localization Algorithm

In recent years,with the constant improvement of detection accuracy and speed of the geomagnetic sensor,mass geomagnetic data with high precision can be obtained,which makes the geomagnetic detection has a wider application prospect.In various application fields,the top priority is to determine the position of the target,which is the precondition of the follow-up work.The intrinsic magnetic field and the induced magnetic field of a ferromagnetic target will superimposeto the geomagnetic field and lead to the change of spatial distribution of geomagnetic field.Thus,it is a efficient way to obtain the location of the target through the analysis and inversion of the magnetic anomaly.Therefore,the research about how to utilize the magnetic anomaly to locate magnetic targets has important academic and practical significance.Many magnetic detection instruments with high accuracy(optically pumped magnetometer,superconducting quantum interferometer,etc.)have been widely used at home and abroad.Those instruments have high measurement precision,stability,and the ability to continuous measurement,which provide the necessary guarantee to obtain reliable geomagnetic data.However,there are still some important technical problems in the localization of magnetic target based on magnetic anomaly,such as:extraction of the target magnetic anomaly,spatial interpolation of the magnetic map,noise suppression method in measurement,and localization algorithm,etc.The above issues are studied,the related research work mainly are focused on the following aspects:Extraction of magnetic anomaly is an essential task in the localization.Geomagnetic field can be decompsed into two parts:the geomagnetic normal field and thegeomagnetic anomalousfield.The geomagnetic anomalous field is relative to thegeomagnetic normal field.Due to the different selection of the geomagnetic normal field,the calculated anomalous field could be not entirely the same.In order to obtain accurate magnetic anomaly,it is necessary to build a reasonable model of the geomagnetic normal field.In practice,the geomagnetic normal fieldcan be divided into the global model and local model.Compared with the global model,the local model can provide more detailed descriptions of the geomagnetic field distribution of the area.Two models are presented to build up the local geomagnetic normal field:polynomial model and spline model,and then analyzed the problems in the two models,finally proposed some measures to improve the accuracy.Meantime,in order to understand the spatial distribution of the anomalous magnetic field generated by magnetic objects,the anomalous magnetic field distribution for some common objects is presented.Geomagnetic map is a useful tool and method to describe the magnetic anomaly,and at some time it is also the key for localization.In practice,it is difficult to realize the large-scale geomagnetic measuring conditions.Therefore,spatial interpolation method can be used to improve the density of measurement data and improve the density of measurement data.Kriging interpolation method has a high degree of approximation in describing the properties'spatial correlations.Thus,we used Kriging interpolation methodto draw the geomagnetic map.The mathematical principles of theKriging interpolation method are analyzed.In Kriging interpolation,the variation function can reflect the change of the characteristics in a local region and along some specific directions.Therefore,the precision of the interpolation is determined with the variation function.The variation function in interpolation is the fitting result from experimental and theoretical variation functions.To improve the fitting precision of the variation function,a variation function fitting method based on particle swarm optimization(pso)algorithm is proposed.The experimental results show that the fitting variation function based on the particle swarm algorithm has a good performance in the process of interpolation,leading to a high accuracy of the interpolated geomagnetic map.In practice,a noise will be disturb the measurement data and affect the accuracy of measuring data.In order to improve the quality of the data,it is necessary to eliminate the noise from the geomagnetic signal.Due to the wide bandwidth of the geomagnetic noise,it is difficult to separate the noise from the geomagnetic signal according to frequency separation.However,there are different between signal and noise in wavelet domain.And wavelet denoising can achive the goal of denoise.Aiming to the shortcomings in traditional threshold function of wavelet denoising method,the threshold function is improved.At the same time,aiming to the problem of how to choose the appropriate threshold,the method is provided.The simulation results show that the improved wavelet denoising method can effectively remove the noise from the geomagnetic signal,and improve the precision of the measurement data.The research on target localization algorithm.The magnetic target can lead to change the spatial distribution of geomagnetic field and generate the magnetic anomaly.Thus,we can effectively locate the target by the magnetic anomaly.The present localization algorithm needs to know the type of the target in advance,but this requirement is hard to satisfy when the target is unknow.Therefore,a locating method based on subdomain is proposed.This positioning method is derived and improved from the subdomain filtering algorithm.In the locating method,the localization of the target can be obtained via automatically compressing the magnetic anomaly,without any known conditions about the target.The simulation results show that the localization of the target can be precisely obtained by the proposed method.Magnetic target locating experiment.To test and verify the correctness and the effectiveness of the above theoretical research,a localization experiment was carried out.The experiment results show that the improved wavelet denoising method can improve the SNR of geomagnetic signal,and improve the precision of the geomagnetic data;a high-precision magnetic anomaly is carried out by using the denoised geomagnetic data to construct the Taylor polynomial model of the studied region's geomagnetic field;the spatial distribution of the magnetic anomaly is obtained via the improved kriging interpolation method;and the magnetic target is located via processing the spacial magnetic anomaly data with subdomain localization algorithm.