Non-Intrusive Magneto-Optic Detecting System for Investigations of Air Switching Arcs

In current investigations of electric arc plasmas, experiments based on modern test- ing technology play an important role. To enrich the testing methods and contribute to the understanding and grasping of the inherent mechanism of air switching arcs, in this paper, a non- intrusive detecting system is described that combines the magneto-optic imaging (MOI) technique with the solution to inverse electromagnetic problems. The detecting system works in a sequence of main steps as follows: MOI of the variation of the arc flux density over a plane, magnetic field information extracted from the magneto-optic (MO) images, arc current density distribution and spatial pattern reconstruction by inverting the resulting field data. Correspondingly, in the system, an MOI set-up is designed based on the Faraday effect and the polarization properties of light, and an intelligent inversion algorithm is proposed that involves simulated annealing (SA). Experiments were carried out for high current (2 kA RMS) discharge cases in a typical low-voltage switchgear. The results show that the MO detection system possesses the advantages of visual- ization, high resolution and response, and electrical insulation, which provides a novel diagnostics tool for further studies of the arc.