Estimation of supercapacitor energy using a linear capacitance for applications in wireless sensor networks

Abstract

Supercapacitor-based energy storage systems have been developed for many applications. For wireless sensor network applications, supercapacitor energy is a critical parameter for system design and power management. However, estimation of supercapacitor energy using the rated capacitance may result in a significant error. To reduce the error, this paper proposes to use a linear capacitance for supercapacitor energy estimation based on the supercapacitor physics. A linear capacitance composed of a constant capacitance and a voltage-dependent capacitance is determined by conducting a constant power charge or discharge test. Evaluation results for a variety of supercapacitor samples with different rated capacitance and voltage demonstrate that using the proposed linear capacitance is more accurate than using the rated capacitance to estimate supercapacitor energy. Moreover, the results show that using the linear capacitance determined through a high power charge experiment can minimize the energy estimation error. This observation can serve as a guideline to design the experiment in a way that the determined linear capacitance results in the minimum supercapacitor energy estimation error.