High-fidelity analysis and experiments of a wireless sensor node with a built-in supercapacitor powered by piezoelectric vibration energy harvesting
Takaharu Yamada, Haruhiko ASANUMA, Yushin Hara, Alper Ertürk
Abstract
• A supercapacitor charged by a vibration energy harvester (VEH) was investigated. • A wireless sensor node (WSN) powered by a piezoelectric VEH was modelled. • Nonlinear piezoelectricity, three-branch model, and diode model were considered. • Analysis technique was established based on differential algebraic equation (DAE). • Simulation predicted charging/discharging characteristics for the WSN accurately. A supercapacitor or an electric double layer capacitor (EDLC) is an essential component of a high-performance wireless sensor node (WSN) that can transmit the three-axis acceleration waveform data of structural vibrations and is powered by a piezoelectric vibration energy harvester (VEH). However, the intrinsically slow charging rate of the piezoelectric VEH triggers charging of extra capacitors formed inside the complex-shaped electrodes of the supercapacitor, which complicates the charging/discharging characteristics and thus estimation of the power generation and energy storage in the supercapacitor. Therefore, in this work, we establish an analytical methodology that allows accurate prediction of the charging/discharging characteristics of the supercapacitor built into the WSN powered by the piezoelectric VEH, considering (1) the nonlinear piezoelectricity, (2) a three-branch circuit model of the supercapacitor, and (3) the current–voltage relation equation for the diode. We develop a parameter identification technique for both the nonlinear piezoelectric VEH and the supercapacitor, and then establish a coupled analysis technique based on the differential algebraic equation (DAE). Using the DAE, we investigate the charging/discharging characteristics of the supercapacitor by considering the actual working steps of the WSN. We revealed that the second branch in the three-branch model contributes not only to an increase in the effective capacitance of the supercapacitor, but also to faster recovery of the voltage across the supercapacitor during WSN operation. We also revealed that a longer charging time led to more energy being stored in the capacitor in the third branch, from which it is difficult to extract energy within a short time period such as the WSN’s transmission process due to its long time constant. The contribution of the second and third branches must be considered to enable accurate prediction of the charging/discharging characteristics of the supercapacitor when built into the WSN powered by the piezoelectric VEH.