All

What are you looking for?

All
Projects
Results
Organizations

Quick search

  • Projects supported by TA ČR
  • Excellent projects
  • Projects with the highest public support
  • Current projects

Smart search

  • That is how I find a specific +word
  • That is how I leave the -word out of the results
  • “That is how I can find the whole phrase”

Q-learning Energy Management Strategy for TEG-powered Environmental Monitoring IoT Devices: A Pilot Study

The result's identifiers

  • Result code in IS VaVaI

    <a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27240%2F22%3A10251746" target="_blank" >RIV/61989100:27240/22:10251746 - isvavai.cz</a>

  • Result on the web

    <a href="https://ieeexplore.ieee.org/abstract/document/10022025" target="_blank" >https://ieeexplore.ieee.org/abstract/document/10022025</a>

  • DOI - Digital Object Identifier

    <a href="http://dx.doi.org/10.1109/SSCI51031.2022.10022025" target="_blank" >10.1109/SSCI51031.2022.10022025</a>

Alternative languages

  • Result language

    angličtina

  • Original language name

    Q-learning Energy Management Strategy for TEG-powered Environmental Monitoring IoT Devices: A Pilot Study

  • Original language description

    In this pilot study, we describe self-learning energy management principles for energy harvesting environmental monitoring nodes using Internet of Things (IoT) communications technology. The solution is powered with ambient energy harvested by a thermoelectric generator (TEG) and stored in an internal supercapacitor. We present a hardware-based model derived from a DC/DC converter, microcontroller and LoRaWAN IoT interface, which is detailed in the paper. The simulation applied historical temperature data obtained at several soil depths. The study&apos;s contribution is a reinforcement learning (Q-learning) method to achieve an optimal energy management strategy to maximize data collection and minimize failure. The results demonstrate that the designed approach was capable of operating more effectively (up to approx. 96 % ratio between complete and missed cycles) than reference solutions with a fixed duty-cycle configuration. We support our conclusions with results from 10 candidate Q-learning controllers which apply various learning and discount factor configurations and demonstrate superior complete/missed cycles ratios than the reference solutions. (C) 2022 IEEE.

  • Czech name

  • Czech description

Classification

  • Type

    D - Article in proceedings

  • CEP classification

  • OECD FORD branch

    20202 - Communication engineering and systems

Result continuities

  • Project

    <a href="/en/project/EF16_019%2F0000867" target="_blank" >EF16_019/0000867: Research Centre of Advanced Mechatronic Systems</a><br>

  • Continuities

    P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Others

  • Publication year

    2022

  • Confidentiality

    S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Data specific for result type

  • Article name in the collection

    2022 IEEE Symposium Series on Computational Intelligence, SSCI 2022 : proceedings : 4-7 december 2022, Singapore

  • ISBN

    978-1-66548-769-6

  • ISSN

  • e-ISSN

  • Number of pages

    6

  • Pages from-to

    211-216

  • Publisher name

    IEEE - Institute of Electrical and Electronics Engineers

  • Place of publication

    Piscataway

  • Event location

    Singapur

  • Event date

    Dec 4, 2022

  • Type of event by nationality

    WRD - Celosvětová akce

  • UT code for WoS article