Micro Scale Energy Harvesting For Ultra-Low Power Systems

K T, Hafeez and Dutta, Asudeb (2011) Micro Scale Energy Harvesting For Ultra-Low Power Systems. Masters thesis, Indian Institute of Technology Hyderabad.

Preview

Text
EE09G004.pdf - Submitted Version
Download (5MB)

Abstract

Ultra-low power systems such as Wireless sensor network (WSN) nodes have emerged as an active research topic due to their vast application areas. Such WSNs would be able to perform their sensing functions and wireless communication without any supervision, configuration, or maintenance. These systems have to cope with severe power supply constraints. The need shared by most WSNs for long lifetimes and small form factors does not match up well with the power density of available battery technology. This could limit the use of WSNs due to the need for large batteries. It is not expected that better batteries for small devices will become available in the near future. Energy harvesting could therefore be a solution to making WSNs autonomous and could thus enable widespread use of these systems in many applications. Energy harvesting is becoming more and more popular for micro-power applications where the environmental energy is used to power up the systems. As sensors have become smaller, cheaper, and increasingly abundant, there have been commensurate reductions in the size and cost of computation and wireless communication. In context of micro scale solar energy harvesting systems, the design of ecient energy conversion unit and accurate maximum power point tracking(MPPT) unit becomes a tremendous challenge due to area constraint and very low (W) output power. This thesis presents a novel MPP tracking method including a charge pump based DC-DC converter for extracting energy from a tiny single PV cell (open circuit voltage 0.4V). We have used a feed-forward (FF) unit to track maximum power point. The design of FF MPP is derived from the operating point of solar cell under dierent solar intensity. This scheme consumes very little power and is faster when compared to other methods. This method eliminates the use of current sensor and other power hungry elements in the MPPT unit. The proposed method tracks the MPP with less than 2 % error and gives eciency of 63.50% through FF MPPT. The complete circuit has been simulated using 0.18 m CMOS process.

[error in script]

IITH Creators: