Supercapacitor modelling: a system identification approach

Recently a great deal of attention has been given to supercapacitors (SC) due to their outstanding power densities and long cycling lifetime. Their behavior has been extensively analyzed and tested through several types of modeling approaches. One common technique for modeling the dynamic operation of SCs is through an electrical circuit model (ECM). This article presents a new approach to identify ECM parameters by applying subspace system identification (SSID) algorithms. This novel application of SSID overcomes the disadvantages of previously presented parameter identification procedures for SCs with improved modeling accuracy. These disadvantages included manual tuning, local minima, and non-convergence. The novel approach was validated through experimental tests at several temperatures, reaching an improvement in model accuracy of almost 50% in some cases. The resulting models are ready to be used as building blocks in a wide range of applications such as energy management systems, renewable power generation, electric vehicles, and microgrids.