Scholarlink

In this paper, we developed and implemented an innovative system that combines artificial intelligence (AI) and the Internet of Things (IoT) to effectively predict energy consumption and manage load in smart grids. The system aims to optimize the use of renewable energy sources, including solar panels, by adapting energy consumption to the needs of users. Using the Long-Short Term Memory (LSTM) algorithm and real-world monitoring data, we achieved an accuracy of 91% in forecasting solar power generation and daily energy consumption. The proposed system is based on a correlation function between the level of illumination and generation from solar panels. This basis allows the system to adjust generation forecasts and energy consumption plans in response to changes in illumination levels in real time. By continuously analyzing the relationship between illumination and energy generation, the model provides a dynamic approach to optimizing the use of solar energy resources. We proposed a mathematical model for solar energy management that integrates forecasting, correlation with illumination, and automatic switching on/off of household devices connected via IoT. This allows the system to effectively adapt and regulate energy consumption in households in the face of inaccurate forecasts. By integrating with the Smart Life platform, the system centralizes the management of household devices, ensuring their interaction via the IoT. The practical implementation of the proposed approach has confirmed its effectiveness in improving energy efficiency and optimizing the use of renewable energy sources. The use of the proposed system will enable users to also provide recommendations for effective planning and utilization of energy resources. By analyzing patterns and predicting power generation, users can optimize their energy consumption, reduce waste, and potentially lower costs. With accurate forecasting, it's possible to make informed decisions on when to store, use, or sell energy back to the grid, contributing to more efficient energy management overall.

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

AI-Based IoT System for Energy Forecasting and Load Management in Smart Grid

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Pavlo Beshley,
  • Natalia Kryvinska,
  • Mykola Beshley

摘要

In this paper, we developed and implemented an innovative system that combines artificial intelligence (AI) and the Internet of Things (IoT) to effectively predict energy consumption and manage load in smart grids. The system aims to optimize the use of renewable energy sources, including solar panels, by adapting energy consumption to the needs of users. Using the Long-Short Term Memory (LSTM) algorithm and real-world monitoring data, we achieved an accuracy of 91% in forecasting solar power generation and daily energy consumption. The proposed system is based on a correlation function between the level of illumination and generation from solar panels. This basis allows the system to adjust generation forecasts and energy consumption plans in response to changes in illumination levels in real time. By continuously analyzing the relationship between illumination and energy generation, the model provides a dynamic approach to optimizing the use of solar energy resources. We proposed a mathematical model for solar energy management that integrates forecasting, correlation with illumination, and automatic switching on/off of household devices connected via IoT. This allows the system to effectively adapt and regulate energy consumption in households in the face of inaccurate forecasts. By integrating with the Smart Life platform, the system centralizes the management of household devices, ensuring their interaction via the IoT. The practical implementation of the proposed approach has confirmed its effectiveness in improving energy efficiency and optimizing the use of renewable energy sources. The use of the proposed system will enable users to also provide recommendations for effective planning and utilization of energy resources. By analyzing patterns and predicting power generation, users can optimize their energy consumption, reduce waste, and potentially lower costs. With accurate forecasting, it's possible to make informed decisions on when to store, use, or sell energy back to the grid, contributing to more efficient energy management overall.