Large-Scale Magneto-Stress Analysis of Complex Structures

The reduction of the underwater electromagnetic signature of a surface Navy vessel is critical to its survivability. A steel–hulled ship has a permanent magnetization and alters the Earth’s magnetic field, and as the ship moves through the water, it generates a magnetic signature possibly detectable by a hostile ordnance or device. Degaussing systems comprising current-carrying coils can significantly reduce a ship’s magnetic signature by partial cancellation. Since the Earth’s magnetic field changes relative to the ship’s axis and position, degaussing coil currents must be controlled to compensate for these predictable changes. Since steel is a hysteretic magnetic material, changes in the magnetic field change the magnetic properties of the steel and alter its permanent magnetization. Furthermore, dynamic mechanical stresses, e.g., varying sea states and hydrostatic pressure, encountered by the hull change the permanent magnetization of the ship due to reverse magnetostriction. Since such changes in the permanent magnetization cannot be well predicted currently, the ship’s degaussing system will slowly lose calibration. The objective of the proposed research is to further develop software simulation tools that can accurately and efficiently predict changes in induced and permanent magnetization of a Naval vessel undergoing dynamic changes in the applied magnetic field and mechanical stress due to hydrostatic pressure. The electromagnetics group at the University of Kentucky has been developing a software toolset (referred to as Magström/Stress3D) to model the magnetic signature of vessels as they move through the earth’s magnetic field and are subject to fluctuating stresses. Key goals are development of the Magström/Stress3D software to 1) improve the system conditioning (which impacts computational resources required by the fast MLSSM compression and LOGOS solution) through the use of alternative formulations, 2) improve accuracy by eliminating possible spurious modes in solutions, 3) reduce computational burden of eddy current simulations by applying a novel system of constraints to reduce the size of the system, 4) continue to address stability issues with the time-domain simulation of hysteretic materials in the presence of eddy currents, 5) improve efficiency and accuracy of the LOGOS fast-direct solver library, and 6) continue transition of software to NSWCCD.