Supercapacitor Parameter Identification Using Grey Wolf Optimization and Its Comparison to Conventional Trust Region Reflection Optimization

This work focuses on two parameter identification techniques using the traditional Trust Region Reflective (TRR) and the bio inspired Grey Wolf algorithm (GWO). The metaheuristic parameter identification technique is not widely used in the field of supercapacitor (SC) parameter identificaton since many researchers prefer to make use of classical search method. This work investigates the use of meta-heuristic based parameter identification for SC. Both the algorithms are applied on real data of SCs of different values and brands.This work also extends study on SC's impedance modeling from step input voltages and can estimate fractional impedance model parameters from time response data directly. Comparative results from classical and fractional impedance models are also shown which can extract the actual behavior of supercapacitor. The proposed identified fractional impedance parameters show less error in time domain. The SC model used in the article considers initial conditions (non-zero condition) on which identification is validated experimentally for 0.47F, 1F and 1.5F SCs and result is discussed to form a conclusion.