Scholarlink

<p>The widespread use of electronic appliances is driving the increasing electricity demand in residential sectors, putting immense pressure on local distribution grids. While smart grid technologies integrated with renewable energy systems are widely promoted for demand-side energy management, their effectiveness is often hindered by the intermittent nature of solar and wind power and the lack of user-independent control strategies. Most existing studies have not addressed the impact of human behavior on IoT-enabled smart grid performance in real-world residential settings, nor proposed adaptive control solutions that operate reliably across seasonal variations. This study aims to bridge this gap by experimentally analyzing a solar photovoltaic (PV)-powered household equipped with IoT-based energy monitoring systems and evaluating its seasonal energy performance. The test environment consisted of a family of four, and energy usage data were collected across summer, winter, and monsoon seasons. Initial assessments showed only marginal reductions in grid electricity demand, especially during the monsoon (1.5%), due to manual overrides interfering with IoT operations. To overcome this limitation, a novel hybrid machine learning algorithm, combining two adaptive models, was introduced to automate energy control and decision-making. The deployment led to grid load reductions of 69.0%, 41.0%, and 43.0% in the summer, winter, and monsoon seasons, respectively. The study demonstrates that integrating AI-driven automation with IoT systems significantly enhances the energy efficiency and autonomy of residential smart grids, offering a robust solution to overcome behavioral and seasonal variability.</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

Seasonal optimization of residential energy consumption using IoT and hybrid machine learning

  • 期刊论文

  • 发表日期：

    2026年1月22日

查看全文
  • P. Umaeswari,
  • M. P. Rajakumar,
  • R. Sonia,
  • Krishnamoorthy Ramalingam

摘要

The widespread use of electronic appliances is driving the increasing electricity demand in residential sectors, putting immense pressure on local distribution grids. While smart grid technologies integrated with renewable energy systems are widely promoted for demand-side energy management, their effectiveness is often hindered by the intermittent nature of solar and wind power and the lack of user-independent control strategies. Most existing studies have not addressed the impact of human behavior on IoT-enabled smart grid performance in real-world residential settings, nor proposed adaptive control solutions that operate reliably across seasonal variations. This study aims to bridge this gap by experimentally analyzing a solar photovoltaic (PV)-powered household equipped with IoT-based energy monitoring systems and evaluating its seasonal energy performance. The test environment consisted of a family of four, and energy usage data were collected across summer, winter, and monsoon seasons. Initial assessments showed only marginal reductions in grid electricity demand, especially during the monsoon (1.5%), due to manual overrides interfering with IoT operations. To overcome this limitation, a novel hybrid machine learning algorithm, combining two adaptive models, was introduced to automate energy control and decision-making. The deployment led to grid load reductions of 69.0%, 41.0%, and 43.0% in the summer, winter, and monsoon seasons, respectively. The study demonstrates that integrating AI-driven automation with IoT systems significantly enhances the energy efficiency and autonomy of residential smart grids, offering a robust solution to overcome behavioral and seasonal variability.