Characterization of energy storage devices for constant power applications

In this paper, a few known characterization techniques for supercapacitors have been evaluated with the purpose of being able to predict the roundtrip efficiency of an energy storage system under constant power application. It is found that the user cannot rely on a single characterization method since the methods based on small signal disturbance are not very precise when the model is subjected to large signal variations. Also, methods based on large signal disturbance will not be very precise when subjected to small signal tests. By comparing the analytically estimated constant power round trip efficiency with experimental data, it is concluded that both models can give acceptable accuracy if a relevant DC bias is used.