부산대학교 도서관

The triaxial magnetometer is an essential device for measuring the weak magnetic field in space. Although the three axes of the ideal magnetometer are entirely coincident with the marked axis, they are not strictly coincident due to the sensor manufacturing limitation. Moreover, the effects of zero drift and long-term magnetic offset make it not ideal for measuring magnetic field. Thus, an error correction model is required. In this article, the three-axis noncoincident error, zero deviation error, and sensitivity coefficient error of a single sensor are remodeled, and the triaxial noncoincident correction model is proposed for the first time. Moreover, trigonometric function fitting is utilized to determine the triaxial sensitivity coefficients $k_{x}$ , $k_{y}$ , and $k_{z}$ , zero deviation coefficients $B_{x0}$ , $B_{y0}$ , and $B_{z0}$ , and triaxial noncoincident angles $\alpha _{1}$ , $\alpha _{2}$ , $\alpha _{3}$ , $\beta _{1}$ , $\beta _{2}$ , and $\beta _{3}$ and verify the three-axis integrated error correction. In addition, the feasibility of the calibration model is evaluated through simulations. Ideally, the corrected absolute error of the triaxial geomagnetic field reduces from 103 to $10^{-11}$ nT, which is almost negligible. The maximum absolute errors under the measurement noise levels of 5, 15, 30, and 50 nT are obtained as 3, 6, 18, and 35 nT, respectively. This means that the general experimental correction requirements are fulfilled. Finally, the feasibility of the calibration model is verified by experiment, and the fluctuation of the total magnetic field intensity is 60 nT, which is reduced by about 40 times.