Uncertainty quantification of a corrosion-enabled energy harvester for low-power sensing applications

New developments in novel energy harvesting schemes for structural health monitoring sensor networks have progressed in parallel with advancements in low-power electronic devices and components. Energy harvesting from galvanic corrosion is one such scheme that has shown to be a viable solution for powering sensing platforms for marine infrastructure. However, with this particular energy harvesting scheme, the power output is current limited as a result of a high terminal resistance that increases with time. In addition, the output voltage is non-stationary, and is a function of several environmental parameters and the applied resistive load. Variability in the power source requires a robust conditioning circuit design to produce a regulated power supply to the sensing and computing electronics. This paper experimentally investigates the non-stationary power characteristics of a galvanic corrosion energy harvester; and uncertainty quantification (UQ) is performed on the measured power characteristics for two experimental specimens subject to resistive load sweeps. The effects on designing a low-power sensor node are considered, and the uncertainty characteristics are applied to a low-power boost converter by means of a Monte Carlo simulation. Lastly, the total energy harvester capacity (measured in mA-Hr) is approximated from the data and is compared to a conventional battery.