High Efficiency Micro-electromagnetic Energy Harvesting System for Self-powered Smart Environment--EMPOWER

The evolution of wireless sensor networks (WSNs) underpins the Internet of Things (IoT). The potential of such a wirelessly connected world spans a wide range of applications, including transportation, home automation, healthcare, smart environments, and many more. Although the performance of WSNs is both pervasive and self-organizing, the widespread deployment of sensor nodes for wireless communication is restricted by the lack of alternative energy sources to power the continuous operation of the sensor nodes. The need to replace and recharge batteries is the greatest weakness of WSNs and requires extensive system maintenance. Vibration-driven energy harvesting, including piezoelectric, electrostatic, and electromagnetic mechanisms, is a smart method of generating electrical energy from the mechanical energy of ambient sources. Macro/mesoscale techniques have emerged over the years and offer attractive solutions to the hazards of batteries. However, major challenges that restrict these harvester systems' implementation at the micro-scale include low transduction, limited operational bandwidth, and inefficient power conversion. The primary objective of this project is to develop a new class of highly efficient, miniaturized, integrated, wideband micro-power energy harvesters compatible with CMOS processing technology. Novel device concepts have been explored to overcome the fundamental limitations of harvester performance under realistic operating conditions. The design methodology is expected to enable self-powered, sensor-driven health and environment monitoring applications that will lead to an "Ambient Assisted Living" scenario. An advanced two-stage tuning approach has been proposed to enhance the reliability and efficiency of microgenerators in scavenging the mechanical energy from low-frequency vibrations. Based on the thoroughly analyzed performance, the proposed approach shows potential advantages over the conventional methods to overcome the fundamental challenges of MEMS-based energy harvesting technology. Employing the two-stage tuning transducers in the MEMS-based energy harvesters is capable of maintaining an extensive tuning range with high reliability and then making the harvesters more efficient when driven by low-frequency vibrations. A MEMS prototype was designed and fabricated by SOI-MUMPS processing with a device layer thickness of 25 µm and an active dimension of 4 mm x 5 mm to validate the proof-of-concept. This standard microfabrication process is compatible with CMOS technology in integrating the ASIC power management unit into the energy harvesting system. Experimental measurements have verified the device's performance in comparison with the theoretical analysis. The performance of a hybrid energy harvesting system consisting of electromagnetic and piezoelectric/electrostatic transductions is addressed in the project. Compared to a single-mehanism system, a hybrid energy harvester is expected to boost space utilization efficiency and increase total power harvested from the vibrational surroundings. Analytical solutions to the maximum power delivered to the load have been derived using circuit theory and impedance matching techniques by considering the parasitic losses of realistic energy harvesting systems. The conditions between effective figures of merit of the electromagnetic and piezoelectric transductions have been identified, under which one system yields more benefits than the other in terms of power. When both transductions have the same effective figure of merit, optimizing efficiency and maximizing power can only be achieved as a unique solution to the hybrid mechanism system. A potential design principle has been introduced to make the microgenerator substantially adaptive to various types of vibrational excitations, including both narrow- and wideband. The effective stiffness and electromechanical coupling are steadily controlled by the mechanical interplay between the curved beam and electrostatic force in the form of a nontrivial electrode. The comprehensive analysis shows that the harvester performance can be driven from linear to bi-stable regimes to maximize the harvested power, depending on the nature of vibration sources. In particular, the zero-stiffness operation can be achieved by precisely balancing the spring and transduction forces for the off-resonance regime. The unique design feature allows the energy harvester's performance to be more versatile under different operating conditions.

The primary outcome achieved is the development of innovative device concepts that enable CMOS-based energy harvesting technology to overcome fundamental limitations for health/environmental monitoring applications, thereby leading to the "Ambient Assisted Living" scenario. The implementation has proven the proposed concepts as an alternative solution to the performance challenges of miniaturized harvesters. The research provides an adaptive design methodology to maintain the highly efficient operation of energy harvesters even under broadband or low-frequency vibrations. In addition, the new findings on the performance of a hybrid energy harvester consisting of electromagnetic and piezoelectric/electrostatic transducers further improve the scavenged output power and the space utilization efficiency in the miniaturized system. The obtained achievements have been presented and published at international conferences and peer-reviewed scientific journals, giving scope to continue that trend. The outcomes are, therefore, expected to contribute to advancing the state-of-the-art in the field, which underpin the future growth of the overall ICT sectors.

Internet of Things (IoT) is regarded among the fastest growing technological platforms in the coming decades. These advanced systems enable intelligent decision-making for a smart environment around human beings. However, the technological limitations of batteries for powering IoT have led to research into harvesting alternative ambient energy sources, such as mechanical vibrations due to its widespread abundance in nature. Lately, Mechanical Energy Harvesting (MEH) devices have stimulated a lot of research efforts, however, without any major breakthrough in the outcome due to unfavorable scaling effects in the power-density. In this project, we aim to challenge the fundamental limitations of highly miniaturized MEH devices. We propose a solution that involves significant innovation in device physics, engineering of proficient device designs, next-generation material development for ultra-sophisticated microfabrication, and circuit design. Towards achieving the goal of delivering a realistic solution for powering IoT-based applications, we focus on addressing four critical challenges as follows. (i) Developing highly anisotropic, nano-structured, micro-patterns of exchange-coupled permanent magnets with high energy product. (ii) Significant increase in the electromagnetic (EM) coupling in a novel MEMS topology-critical to high-performance MEMS EM-MEH devices. (iii) Optimized design of MEMS device incorporating nonlinear oscillation to simultaneously increase the overall power-density and the operational bandwidth, for efficient operation under low-amplitude external vibrations. (iv) Development of high-efficiency power management IC, including an integrated energy storage unit, which is designed for low-voltage sources to overcome the essential impediments in EM transduction at the micro-scale.