Scholarlink

<p>Modern power systems demand innovative approaches for efficient monitoring, cost management, and power quality assurance. This paper presents a low-cost, open-source smart energy meter that integrates embedded technologies, IoT connectivity, and artificial intelligence. The system uses an ESP32 microcontroller to acquire real-time energy data and present key metrics via an interactive interface. A lightweight AI-based voltage anomaly classification model detects and classifies disturbances, such as sags and interruptions, in real time, requiring only 32.2 KB of memory and achieving 8&#xa0;s per classification and 4&#xa0;s for routine measurements. The meter also calculates energy tariffs and alerts users when consumption exceeds a customizable limit, promoting cost savings and responsible usage. Integration with Home Assistant enables remote monitoring, visualization, and control, making the system scalable for residential and broader smart grid applications. Testing demonstrates improved operational efficiency, user engagement, and practical applicability in modern energy networks.</p>

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

Design and Implementation of a Smart Energy Meter with Artificial Intelligence-Based Voltage Anomaly Classification

  • 期刊论文

  • 发表日期：

    2026年3月18日

查看全文
  • Bouchra Feriel Khaldi,
  • Fatma Zahra Dekhandji,
  • Abdelmadjid Recioui,
  • Anis Benfala,
  • Mohamed Bahri

摘要

Modern power systems demand innovative approaches for efficient monitoring, cost management, and power quality assurance. This paper presents a low-cost, open-source smart energy meter that integrates embedded technologies, IoT connectivity, and artificial intelligence. The system uses an ESP32 microcontroller to acquire real-time energy data and present key metrics via an interactive interface. A lightweight AI-based voltage anomaly classification model detects and classifies disturbances, such as sags and interruptions, in real time, requiring only 32.2 KB of memory and achieving 8 s per classification and 4 s for routine measurements. The meter also calculates energy tariffs and alerts users when consumption exceeds a customizable limit, promoting cost savings and responsible usage. Integration with Home Assistant enables remote monitoring, visualization, and control, making the system scalable for residential and broader smart grid applications. Testing demonstrates improved operational efficiency, user engagement, and practical applicability in modern energy networks.