Investigation of a spherical pendulum electromagnetic generator for harvesting energy from environmental vibrations and optimization using response surface methodology

Abstract

A fundamental step towards the industrial application of an energy harvester device is the optimization of the output electrical power according to environmental conditions. In the present work, designing a pendulum spherical electromagnetic generator (S-EMG) is considered for harvesting energy from environmental vibrations with frequencies less than 12 Hz. Low frequencies pose significant challenges for vibration energy harvesters. The spherical structure creates a uniform and radial magnetic field, so the outputs will not depend on the direction of motion. Performance of S-EMG is evaluated considering different conditions, including the frequency, the load resistance, and the pendulum cord density. The output power density is optimized using the response surface methodology (RSM) based on the central composite design and the desirability function. In addition, the quadratic response surface regression model was used to predict the results. The optimized parameters were found as 5.966 Hz, 93.138 Ω, and 1154.070 kg/m3, for frequency, load, and density, respectively. The predicted power density was 4.357 µW/cm3, which is significant despite of the light weight and small size of the device. The results were also validated by re-testing inside and outside the design ranges. Furthermore, the importance of input parameters on system performance was evaluated using analysis of variance, which shows that frequency and density have the most effect (30.93 % and 30.66 %, respectively) on output power density. The RSM methodology is found to be helpful in determining the appropriate specifications for the energy harvesting device versus different environments.