Energy Storage Design for Energy Harvesting Sensors

Energy harvesting sensors scavenge energy from their surroundings to power themselves without a battery or utility-connected power supply. Sensors that avoid batteries and bespoke power wire connections offer flexibility for avoiding complications in safety and infrastructure. Energy from the sensor’s environment often arrives intermittently or stochastically, complicating the sensor design process. The size of on-board energy storage becomes a critical design decision. Energy storage allows the harvesting system to accumulate energy over time that can later be consumed for sensor tasks. This article presents a modeling and design guide for sizing sensor energy storage. These guidelines balance the tension between cold-start time and steady-state endurance. Coldstart time and steady-state endurance, as a function of energy storage design parameters, are quantified and analyzed with respect to both deterministic and stochastic energy harvest profiles. Results are demonstrated using experimentally measured power consumption data from an industrial machine on a microgrid. Two practical sensor storage design examples demonstrate the design guide. Simulation results highlight the very restrictive storage unit design space over which both fast boot-up and sufficient endurance are satisfied for a notional sensor application. The negative effect of oversized storage on overall sensor on-time over long time periods of thousands of hours is also demonstrated. These results emphasize the significant impact of storage unit startup and maximum voltage threshold design choices and their ability to reduce a required storage capacitance by over an order of magnitude to meet the same application requirements.