Online model parameter identification for supercapacitor based on weighting bat algorithm

To support real-time power management of the supercapacitor-powered embedded systems, an online model parameter identification method is proposed for predicting the supercapacitor behavior. In the proposed method, an optimization problem is formulated based on our previously developed supercapacitor model, and a weighting bat algorithm (WBA) with the weighting solution update method is proposed for solving this problem in each model parameter updating time window. Simulation and experimental results show that the proposed online model parameter identification method can accurately capture the terminal behavior of a supercapacitor, and the proposed WBA-based optimization method has better performance for the supercapacitor model parameter identification compared with other benchmark algorithms.