In this article, we present a new method for the detection and localization of multiple dipole-like magnetic sources using magnetic gradient tensor data. We use the tilt angle that is the ratio of the vertical magnetic component to the horizontal field to determine the number of the magnetic sources and use the rotational-invariant normalized source strength (NSS) to estimate approximate horizontal coordinates of the magnetic sources. The differential evolution (DE) algorithm is used to estimate the locations and moments of the magnetic sources. Three synthetic examples, with and without random noise, are designed to validate the proposed method. The locations and magnetic moments of the magnetic sources can be well determined. The proposed method has also been applied to a field test with multiple magnetic sources and the result shows that the location of the magnetic source can be estimated well.
The triaxial magnetometer is a vector magnetic field sensor widely used in various fields. Due to limitations in manufacturing and installation processes, a non-orthogonal error, a scale factor error and a zero-bias error can occur. These sensor errors significantly affect the accuracy of magnetic field measurements, so calibration is essential to ensure reliable data. This paper proposes a novel error calibration method for triaxial magnetometers based on an improved Beluga Whale Optimization algorithm. In addition, three synthetic simulations, with and without noise, and with different initial search spaces and population sizes are designed to validate the proposed method. Ideally, the minimum absolute error between the calibrated and theoretical results is 10–13nT. Furthermore, when the calibration method is applied to the field experiment data, the total geomagnetic field fluctuation is reduced from 600nT to 9nT, which is about 66 times smaller. Both simulation and experimental results further confirm the effectiveness and practicality of the proposed algorithm.