Font Size: a A A

Research On Technologies Of GMI Magnetic Sensor And Magnetic Anomaly Detection

Posted on:2015-11-23Degree:DoctorType:Dissertation
Country:ChinaCandidate:X H NieFull Text:PDF
GTID:1108330509460949Subject:Instrument Science and Technology
Abstract/Summary:PDF Full Text Request
Compared with the usual traditional magnetic sensors based on fluxgate, Hall effect, magneto-resistance(MR) effect, etc, the magnetic sensors based on giant magneto-impedance(GMI) effect has many advantages, such as higher sensitiveness, small size, fast response, small hysteresis and low power. Then the GMI sensors can be widely utilized in the field of weak magnetic measurement. On the other hand, the magnetic anomaly detection(MAD) in the geomagnetic field, has the characteristics of passive detection, good concealment and strong anti-jamming, etc., which is a potential technology for traffic, resource reconnoitering, military targets detecting and tracking, etc. Therefore, the research on GMI magnetic sensor and MAD is important for academic significance and engineering practial. This dissertation is going to centrically lucubrate the design of GMI magnetic sensor and the problem of the detection of magnetic anomaly signal caused by ferromagnetic targets. The main contents include:(1) Both the principle of GMI effect and the relation with the frequency of driving current are briefly introduced. Then based on the research results of the frequency characteristics of our used amorphous wire, the frequency of driving current and the linear working range are selected. By using a DC current to generate the bias magnetic field, a differential-type sensing head is designed and a possible conditioning circuit utilized orthogonal vector lock-in differential amplifier instead of peak-value detector is also presented in this paper. On the same time, the principle of conditioning circuit and the performance of our GMI magnetic sensor are analyzed in detail. On the other hand, based on the principle of second generation current conveyor(CCII) and operational trans-conductance amplifier(OTA), a quadrature oscillator with the characteristic that the oscillation condition and oscillation frequency can be adjusted independently by different passive elements, is designed by employing available IC AD844 s in this paper, while the principle of this oscillator and its non-ideal effects are also presented.(2) The MAD model and the expressions of magnetic anomaly signal based on three orthonormal basis functions(OBFs) are presented in this paper. Aiming at the case that the characteristic time of magnetic anomaly signal is unknown, and based on the theory of the orthonormal matching bank filter, a MAD method using the OBFs matching bank filters is proposed and analyzed for the case of white noise background. Then based on the above MAD method, a farther MAD method is also proposed for the case of color noise background, which is based on OBF matching bank filters and a whitening filter using LMS adaptive AR model algorithm. Finally, some simulations in the 1/f noise background, which is one of major noise sources, are performed to analysis the effects of adaptive whitening filter and the performances of t our proposed MAD system. In addition, a further real-world experiment is also done to demonstrate the advantages of this MAD method.(3) Based on the theory of ARMA model, both the principle and process of ARIMA model which can be utilized in the case of non-stationary time serials the are briefly introduced. For the case that the characteristic time of magnetic anomaly target signal is unknown and the background noise is non-stationary, based on the L-steps forecasting formulas of ARMA model, a novel MAD method based on the forecasting formulas of ARIMA mode is proposed in this paper. Some experiment data and simulations are done to prove the performance of this method, while some disadvantages are also given out. On the other hand, based on the principle of grey incidence, another farther MAD method based on ARIMA residual and grey incidence is also presented, while many simulations are done to analyses its performance.(4) Based on the foundation of discrete wavelet transform(DWT) and its inverse transform, the concept of undecimated discrete wavelet transform(UDWT) is derived form the problem that the length of input data would halve when the decomposition level increases during DWT. Considering the time domain-frequency domain localizing characteristic and the disadvantages existing in the traditional hard-thresholding and soft-thresholding noise reduction, a novel noise reduction method based on UDWT and weighed fuzzy logical is proposed in this paper, while its performance and advantages are also presented out. On the other hand, based on the characteristic that the orthonormal wavelet decomposition can play a role of whitening filter to the 1/f noise, an energy MAD detection based on UDWT is also proposed in this paper. At the same time, the probabilities of detection and false alarm in theory are expressed, and many simulations are performed to analyze the performance of our proposed detector.
Keywords/Search Tags:Giant Magneto-Impedance(GMI), Magnetic Anomaly Detection(MAD), Magnetic Sensor, 1/f Noise, Orthonormal Basis Functions, ARIMA model, Undecimated Discrete Wavelet Transform(UDWT)
PDF Full Text Request
Related items