Research Trend and Technical Challenges on Small Scale Energy Harvesting

Dong Ha (ha@vt.edu)

Harvesting small scale energy from otherwise wasted ambient energy sources has attracted immense research efforts for applications such as wireless sensor networks for environmental and habitat monitoring, implantable devices and biosensors, nano-robotics, and micro-electromechanical systems. Replacing or recharging batteries for those applications is inconvenient, expensive (eg. implant devices), or impractical (eg. wireless sensor nodes for wildlife tracking). Energy harvesting (EH) from ambient sources offers a promising solution to the problem. Ambient energy sources include solar, indoor light, vibrations from industry machines, human body heat and motion, TV broadcasting, and so on. In fact, there are already numerous EH products and applications ranging from integrated circuits to wireless sensor nodes to floor lighting at dance clubs.

Four distinctive forms of energy subject to EH are photovoltaic, kinetic, thermal gradient, and radio frequency (RF). Table I shows energy density of the four forms of energy sources in a approximate scale [‎1]. The outdoor solar energy has the highest energy density followed by temperature gradient, vibration/motion, and RF the least, which explains the reason that outdoor solar EH is most pervasive these days, while RF EH is still in the research stage.

Key Applications

Over one billion wireless sensor nodes (WSNs) are deployed every year. WSNs find a variety of applications from smart buildings to structural health monitoring to healthcare. Most existing WSNs are powered by batteries, which are costly to replace or recharge. EH is a promising solution to remove or replenish batteries. Figure 1 shows a perpetually powered wireless sensor node envisioned by Texas Instruments [2] in which ambient energy is harvested, conditioned, and stored in a storage device such as a rechargeable battery or supercapacitor. Another key application of EH is medical implant devices such as pacemakers, defibrillators, infusion pumps, and neuro-stimulators. Energy harvested from body motion and/or heat, along with wireless power transmission, is a promising solution to power implant devices.

Technical Challenges on Power Management Circuits

A typical EH system consists of three components: a transducer, a power management circuit, and a storage element. A power management circuit (PMC) rectifies the input voltage, if necessary (such as for electromagnetic and piezoelectric generators), and regulates the output voltage to the desired level to charge the battery or supercapacitor. The core part of a PMC is a DC/DC converter, and the key requirement for a PMC is to maximize the net energy harvested by (i) transferring maximum power from the source to the storage device and (ii) dissipating minimal power. Meeting the two requirements needs a trade-off in PMC design for small scale EH. To transfer the maximum power, the load impedance should match the source impedance, and the source impedance changes as operating conditions changes. Matching between the source and load are accomplished using maximum power point tracking (MPPT), which dynamically adjusts the load impedance (or the load voltage for photovoltaic cells), specifically the input impedance of the DC/DC converter. To minimize power dissipation of a PMC, one needs to select an MPPT algorithm judiciously by considering the type of transducer, the amount of energy available, and its application. For example, employment of a sophisticated MPPT algorithm harvests more energy from the energy source, but the PMC dissipates more power to execute the MPPT algorithm. Another rather unique problem for PMCs for EH applications is cold start. A PMC is typically powered by the storage device of an EH system, as the storage voltage is regulated. When the storage is drained completely, the PMC fails to start. Two solutions are feasible; one is to provide a dedicated backup battery for cold start, and the other one use the source power to start. Both approaches have pros and cons. A PMC for EH should be able to harvest energy from a low input voltage and should have low standby power dissipation, and desirably a battery controller circuit.

Some design challenges of PMCs are specific to energy sources. A rectifier circuit for a piezoelectric and electromagnetic generator causes a substantial loss for small scale vibration EH. Some vibration energy such as human motion and traffic on a bridge is intermittent and random. Implementation of a power efficient MPPT is difficult for such vibration energy, and incorporation of a sleep mode may be necessary. Some vibration signals in nature such as automobiles and railcars are wideband, and existing impedance matching schemes are designed to tune for a single frequency. Low irradiance for indoor applications possibly combined with a wide dynamic range of the irradiance level (such as near a window) poses a challenge for ultra low-power design of PMCs. A PMC for a wearable solar jacket should be able to adjust to rapidly changing operating conditions as well as partial shading. Thermoelectric generators for human body heat are constrained in size and generate low output voltage due to a low temperature gradient and small size. Hence, a PMC for such applications should operate at very low input voltage. A PMC for RF EH needs an efficient voltage multiplier operating at a low voltage.

Research Trend

Energy harvesting has key applications in wireless sensor nodes and implant devices. Current energy harvesting systems except photovoltaic systems suffer from low energy density, low level of energy harvested, high cost, and a large form factor. To address the problems, extensive research is being conducted on power management circuits as well as transducers and storage devices. We expect energy harvesting will make continuous advancements in all the three areas, transducers, power management circuits, and storage, and the current trend in research will continue for the next five to ten years owing the key applications of energy harvesting.

Reference:

1. C. O. Mathuna, T. O'Donnell, R. V. Martinez-Catala, J. Rohan, and B. O'Flynn, "Energy scavenging for long-term deployable wireless sensor networks," Talanta, vol. 75, pp. 613-623, 2008.
2. Murugavel Raju, and Mark Grazier, “Energy Harvesting,” Texas Instruments, 2010.