Optimum power of a nonlinear piezomagnetoelastic energy harvester with using multidisciplinary optimization algorithms

Energy-harvesting devices have been widely used to generate electrical power. Through the use of energy harvesting techniques, ambient vibration energy can be captured and converted into usable electricity in order to create self-powering systems. In the present study, to further improve the efficiency of energy-harvesting devices, a nonlinear piezomagnetoelastic energy harvester is proposed in two different configurations that is parallel and series. In order to optimize the generated electrical power, the physical parameters of the harvester are chosen as the design variables. Classical and Metaheuristic algorithms, namely, random search, genetic algorithm, and simulated annealing are applied to optimize the output power regarding the stress and displacement constraints and feasible variable bounds. Finally, the results of the applied algorithms are compared together. The results demonstrate that most of the implemented algorithms converge to the similar objective function value. The constrained random search methods with SQP and active set algorithms converge faster with small iterations. However, the genetic algorithm and simulated annealing algorithm are more capable to find the global optimum. The obtained results revealed that, before the optimization, the average extracted power in specified time was 3.121 W in parallel configuration and 3.156 W in serial configuration. By using the optimization approaches, the power converged to 4.273 W in parallel configuration and 4.296 W in serial configuration that means the power is increased by 36.9% and 36.1% approximately.