Human motion energy harvesting: numerical analysis of electromagnetic swing-excited structures

Abstract

Energy harvesting from human motion has constantly attracted scientific interest over recent years. A location where a harvesting device can easily and unobtrusively be integrated is the shoe sole, which also protects the device from exterior influences. In this work a numerical system model is developed, which can be used to simulate different inductive harvester geometries and predict their power output. Real world acceleration data is used as a model input. The model is implemented in Matlab/Simulink and subdivided into a mechanical and an electromagnetic model. The key features including the motion model and the calculation of the electromagnetic coupling coefficient are explained in detail and the model is briefly evaluated experimentally. A total of six inductive architectures, i.e. different cylindrical and rectangular magnet-coil arrangements, are then investigated in detail. The geometrical parameters are optimized for each architecture to find the best geometry within the size of 71 mm × 37.5 mm × 12.5 mm, which can be integrated into the sole. With the best overall design an average power output of 42.7 mW is simulated across an ohmic load of 41 Ohms. In addition to the respective best designs, the (dis-)advantages of each architecture are explained.