Electric Power Prediction of 2D FE Magnetic Spring Energy Harvester Based on Coil using Support Vector Machines

Magnetic spring energy harvesters are considered to be a promising solution for powering small electronic devices because they employ the magnetic interaction between two fixed permanent magnets and one that is levitated vertically. This paper presents the two dimensional (2D) Finite Element Method (FEM) model of magnetic spring energy harvester based on coil layout. The simulation provides an optimized arrangement of number of turn and geometrical parameters of coil in the magnetic spring. The method based on Support Vector Machine (SVM) is proposed to predict electrical power of magnetic spring and to establish its dependence on the coil parameters. Furthermore an experimental analysis was performed to evaluate the electrical performance of the proposed magnetic spring system.