Application Of Stochastic Resonance Algorithm In Magnetic Anomaly Signal Detection

Magnetic anomaly signal detection has been widely used in civil security,mineral mining,military defense and other fields.However,with the in-depth application and development of magnetic anomaly detection,many problems have arisen.For example,the targets to be detected are diverse,and the magnetic field noise is complex in the environment where the magnetic anomaly signal appears,which affects the accuracy of the magnetic anomaly signal detection.The strength of the magnetic field generated by the weak magnetic anomaly target is much lower than that of the ambient magnetic field,resulting in a very low signal-to-noise ratio of the magnetic anomaly signal,which makes the detection of the magnetic anomaly signal more difficult and so on.In order to solve the problem of detecting weak magnetic anomaly signals moving in the environmental magnetic field,thesis research on a magnetic anomaly detection method based on traditional stochastic resonance algorithm,and improves it on the basis of traditional stochastic resonance algorithm.At the same time,a large number of external field simulations are carried out.Experiments and driving tests have verified the effectiveness of the improved stochastic resonance algorithm in the application of magnetic abnormal signal detection.Thesis firstly introduces the research background and research significance of magnetic anomaly signal detection.Then,the research background of traditional stochastic resonance algorithm and the research significance in weak signal detection are introduced,and on this basis,the feasibility of applying stochastic resonance algorithm in magnetic anomaly signal detection is proposed,and an improved stochastic resonance magnetic anomaly signal detection is introduced.algorithm.The specific work and results of thesis are as follows:Aiming at the modeling of magnetic anomaly signals,the detection method of magnetic anomaly signals based on traditional stochastic resonance algorithm is studied.The traditional stochastic resonance theory can enhance the ability of weak signals to output signals in strong noise background,and the traditional stochastic resonance magnetic anomaly detection method is constructed.Model,when a magnetic abnormal signal appears,the bistable stochastic resonance system will switch in a steady state,so as to judge whether a magnetic abnormal signal is detected.And compared with the standard Orthonormal Basis Functions(OBF),a classic detection method in the field of magnetic anomaly detection,the driving test is carried out,which proves that the stochastic resonance algorithm is farther in detection distance than the OBF method.Aiming at the modeling of magnetic anomaly gradient signal,the application of the improved stochastic resonance algorithm based on gradient tensor signal in magnetic anomaly detection is studied.Through the gradient of the magnetic anomaly signal and the improvement of the potential well function of the traditional stochastic resonance algorithm,the output is smoother and less saturated.On this basis,the particle swarm algorithm is used to optimize the output of the improved stochastic resonance algorithm.The optimal system parameters are obtained.And a lot of experiments have been carried out to prove that the improved stochastic resonance algorithm is farther than the traditional stochastic resonance algorithm in detection distance.For the above algorithm,a large number of field experiments have been carried out to verify the effect.Through the designed magnetic anomaly signal detection platform,experiments on different magnetic anomaly targets such as magnets,cars,pickup trucks,and SUVs are carried out.