Adaptive Power Management in Solar Energy Harvesting Sensor Node Using Reinforcement Learning
- 27 September 2017
- journal article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Embedded Computing Systems
- Vol. 16 (5s), 1-21
- https://doi.org/10.1145/3126495
Abstract
In this paper, we present an adaptive power manager for solar energy harvesting sensor nodes. We use a simplified model consisting of a solar panel, an ideal battery and a general sensor node with variable duty cycle. Our power manager uses Reinforcement Learning (RL), specifically SARSA(λ) learning, to train itself from historical data. Once trained, we show that our power manager is capable of adapting to changes in weather, climate, device parameters and battery degradation while ensuring near-optimal performance without depleting or overcharging its battery. Our approach uses a simple but novel general reward function and leverages the use of weather forecast data to enhance performance. We show that our method achieves near perfect energy neutral operation (ENO) with less than 6% root mean square deviation from ENO as compared to more than 23% deviation that occur when using other approaches.Keywords
Funding Information
- JSPS KAKENHI (16K12405)
This publication has 17 references indexed in Scilit:
- Power management for Wireless Sensor Networks in underground miningPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2016
- Adaptive Duty Cycling in Sensor Networks With Energy Harvesting Using Continuous-Time Markov Chain and Fluid ModelsIEEE Journal on Selected Areas in Communications, 2015
- A Reinforcement Learning-Based ToD Provisioning Dynamic Power Management for Sustainable Operation of Energy Harvesting Wireless Sensor NodeIEEE Transactions on Emerging Topics in Computing, 2014
- A bird's eye view on reinforcement learning approaches for power management in WSNsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2013
- Analysis of adaptability of Reinforcement Learning approachPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- Dynamic power management utilizing reinforcement learning with fuzzy reward for energy harvesting wireless sensor nodesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- Event-driven adaptive duty-cycling in sensor networksInternational Journal of Sensor Networks, 2009
- Adaptive duty cycling for energy harvesting systemsPublished by Association for Computing Machinery (ACM) ,2006
- Performance aware tasking for environmentally powered sensor networksACM SIGMETRICS Performance Evaluation Review, 2004
- Battery-driven dynamic power managementIEEE Design & Test of Computers, 2001