Research On Magnetic Multiple Target Optimal Localization Method Based On Magnetic Anomaly Detection

The magnetic anomaly detection technology can be used to obtain magnetic target information,and based on the obtained target signal and magnetic field model,the spatial position of magnetic targets can be inverse performed,thus realizing the spatial positioning and tracking monitoring of magnetic multiple targets.This technology has been widely used in the fields of geological resources exploration,underwater magnetic target detection and localization,navigation based on magnetic maps,etc.It plays an important role in national economic construction,scientific research and national defense security.This thesis takes magnetic target localization as the research goal,and conducts relevant research on magnetic field environment noise model,magnetic target detection,magnetic detection point array distribution optimization,single magnetic target and multi-magnetic target localization,aiming to improve the stability and accuracy of multitarget localization.The main works are as follows:For the magnetic field environment noise,the geomagnetic field model as well as the ocean magnetic field model are studied,the main factors affecting the ocean magnetic field are analyzed,and the law of seawater motion is verified through simulation,and the seawater motion perturbed magnetic field model is analyzed in combination with seawater motion.For magnetic target detection,the target detection methods based on standard orthogonal basis function decomposition and minimum entropy detector are introduced,and the target detection effects of the two detection methods are verified by using simulation data,and then the accuracy of the algorithm is verified by actual detection experimental data,and the differences between the two detection methods are compared.For the optimization of magnetic detection point array distribution,the particle swarm optimization algorithm is used to optimize the detection point array distribution for a specific monitoring area.Using the area detection coverage as a measure to compare the uniformly distributed array and the optimized array area coverage using simulation data,and the simulation verifies the effectiveness of the optimization algorithm.For magnetic target localization,a magnetic target localization method based on the full-tensor magnetic field gradient tensor is introduced,and the sources of errors generated in the localization process are analyzed.The reliability evaluation method of the localization results is constructed by using the information of the eigenvalues of the magnetic field gradient matrix,and the method is used to suppress the errors in the detection blind area,and the simulation and experimental results show that the method improves the localization accuracy to a certain extent.The magnetic multi-target localization method based on the clustering algorithm is studied.To address the problem of low localization accuracy of the traditional method under the conditions of fewer detection points or high percentage of invalid data,it is proposed to give different sample clustering localization weights to suppress the error by using the weight assignment error correction.The results of comparing the multi-target localization effect of density clustering algorithm and weight assignment optimized clustering algorithm in simulation and experimental tests show that the method improves the localization accuracy.An experimental platform for magnetic target detection and localization is built to verify the localization effect of the localization algorithm in actual detection.Preliminary verification of the feasibility of the application of the optimized algorithm proposed in this thesis is carried out.The experimental data results show that the optimized method can reduce the localization error to a certain extent.