Metaheuristic optimization for stand-off current measurement employing vector magnetometer arrays

Current monitoring is essential in power grids and industrial applications. However, most existing current sensors require direct or near-contact placement on the wire, which introduces safety hazards and complicates installation and maintenance. To address these limitations, this study proposes a stand-off current measurement method that reconstructs the current by solving an electromagnetic inverse problem using metaheuristic optimization algorithms. The magnetic field generated by the target current is detected by a cross-shaped vector magnetometer array, and a differential objective function is used to inherently suppress common-mode interference, giving the method a clear robustness advantage over absolute field inversion approaches. Numerical simulations are conducted to evaluate the performance of differential evolution and particle swarm optimization enhanced with Lévy flight and to identify the optimal operating conditions of the sensor system. In the simulated scenario with a 3 m straight wire carrying a 100 A DC current and a Gaussian magnetic noise level of 0.1 nT, the inversion errors of both algorithms are within 0.1% at a measurement distance within 3 m. This approach provides valuable insights into the design of precision non-invasive current sensing systems and demonstrates strong potential for practical deployment in power system monitoring, particularly in scenarios requiring safe, flexible, and reliable sensing.