Key Technology Research On Airborne Magnetic Anomaly Detection

The earth is a huge magnetic ellipsoid generating stationary magnetic fieldsaround20000nT to50000nT, called geomagnetic filed. Magnetic objects in thegeomagnetic fields will be magnetized leading to change the geomagnetic aroundthem. These magnetic anomalies can release the information of magnetic target.Airborne magnetic anomaly detection (MAD), with this principle, is to find andlocate the magnetic target by processing data from the magnetometer loaded onthe plane and detecting its magnetic anomaly. MAD technology has been widelyused in geophysical detection, geomagnetic navigation, and many other fields. Ithas attracted more and more concerns and attentions.Since the target magnetic signal is often submerged in noises andinterferences caused by various irrelevant magnetic fields, magnetic anomalydetection algorithm cannot be used to detect target directly due to the bad result.According to what is mentioned above, MAD has two key technologies includingmagnetic compensation and magnetic anomaly detection. Among them, the basisof magnetic compensation is the TOLLES-LAWSON equation which has anunreasonable hypothesis about the constant geomagnetic. With the advances ofthe detection mode of magnetometers and especially their improving accuracy,it’s necessary to be re-examined and improved. Otherwise, only under thecircumstance of Gaussian white noise, the classic OBF algorithm can obtainoptimal results. But actually, the compensation results have colored noise, thusit’s important to offer an improved detection algorithm.The following are innovative researches around magnetic compensation andmagnetic anomaly detection included in this thesis:Firstly, the probable factors causing the error of magnetic compensation hadbeen numerical simulated respectively, and two key factors came into our sight.One is the error from instruments and aircraft’s maneuver, another is theunreasonable assumption of T-L equation, especially earth magnetic field’sconstancy, is one of main error sources. They’ll inspire us to improve thehypothesis towards real various geomagnetic fields.Secondly, taking the advantage of Kalman filter, we proposed a method to predict various earth magnetic fields. By analyzing real earth magnetic fields, wefound the approximate linear change law and the result that second orderdifferences of geomagnetic fields are smallest. With these analysis, the stateequations were respectively established. T-L equation helped to establish themeasurement equation. These two predicted results could effectively improve thecompensation accuracy without changing the existing magnetic compensationalgorithm.Finally, we simply improve the classic OBF detection algorithm. As to ahigh false alarm rate of OBF, the improved detection algorithm based onnarrowed-band filter was proposed. With the help of narrowed-band filter, thedetective results reduced the false alarm rate and improve the accuracy ofdetection.