Scholarlink

In response to the growing demand for sustainable and low-maintenance power solutions in IoT networks, this paper explores the integration of thermoelectric generators (TEGs) as solid-state energy harvesters. With IoT anticipated to encompass 42 billion devices by 2025, TEGs emerge as a reliable means to convert thermal energy into electricity, addressing challenges of battery consumption and environmental impact. The authors conduct a thorough review, leveraging machine learning (ML) approaches to manage and predict energy availability in TEG-powered IoT devices. The study outlines diverse application areas for TEG-driven devices, utilizing temperature differentials in environments, biological structures, and machinery. Despite TEGs exhibiting scalability and suitability for ubiquitous temperature variations, challenges such as low energy efficiency (approximately 10%) and the need for a consistent heat source are acknowledged. The research aims to enhance energy prediction and management through ML methods, fostering adaptive and dynamic operational capabilities for TEG-powered IoT applications.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A Comprehensive Review of Machine Learning Paradigms to Forecast Energy in Thermoelectric Powered IoT Devices

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Rohit Kumar Singh Gautam,
  • R. Kavitha,
  • Kalyan Acharjya,
  • Rahul Mishra,
  • Mohammad Amir

摘要

In response to the growing demand for sustainable and low-maintenance power solutions in IoT networks, this paper explores the integration of thermoelectric generators (TEGs) as solid-state energy harvesters. With IoT anticipated to encompass 42 billion devices by 2025, TEGs emerge as a reliable means to convert thermal energy into electricity, addressing challenges of battery consumption and environmental impact. The authors conduct a thorough review, leveraging machine learning (ML) approaches to manage and predict energy availability in TEG-powered IoT devices. The study outlines diverse application areas for TEG-driven devices, utilizing temperature differentials in environments, biological structures, and machinery. Despite TEGs exhibiting scalability and suitability for ubiquitous temperature variations, challenges such as low energy efficiency (approximately 10%) and the need for a consistent heat source are acknowledged. The research aims to enhance energy prediction and management through ML methods, fostering adaptive and dynamic operational capabilities for TEG-powered IoT applications.