Degaussing currents optimal calibration method based on linear neutral network

The magnetic field created by ferromagnetic ships may endanger their invisibility. Nowadays, a new technique called closed-loop degaussing system can reduce the magnetic anomaly especially permanent one in real-time. The key technique of this system is to predict degaussing currents from onboard measurements. To achieve it, researchers often solve the problem by Offboard Magnetic Field Method (OMFM) or Coils´ Current Database Method (CCDM). In this paper, a linear neural network has been proposed to get the currents directly from onboard magnetic field. The neural network can prevent many troubles from other methods. Its high precision and good robustness have been tested by a mockup test.