Scholarlink

<p>While the demand for energy is increasing and smart home environments are becoming more complex, finding the right balance between energy costs and the comfort of residents in smart homes remains a challenge. This paper introduces a scheduling framework based on the Improved Multi-Objective Fitness Dependent Optimizer (IMOFDO) and an innovative Comfort Violation Penalty Index (CVPI) to tackle this issue; unlike traditional scheduling metrics, the CVPI offers a normalized, occupancy-aware penalty that integrates temperature, humidity, and lighting into a single optimization target. To overcome the limitations of the standard Fitness Dependent Optimizer (FDO) evolutionary algorithm, the IMOFDO incorporates Lévy flight and Gaussian mutation, striking a balance between global exploration and local precision. We evaluated the framework using real-time pricing (RTP) schemes and 15-min sensor data from a smart home dataset. The proposed framework was tested through year-long simulations of ten homes utilizing RTP. The results indicated that IMOFDO was better than MOPSO, MOFDO, and MOANA at lowering costs, peak to average ratio (PAR), and keeping users comfortable; we observed significant cost savings of up to 16.4%, or $500–$950 annually, particularly in homes with higher energy consumption. Additionally, the framework reduced peak demand by up to 1.05 kW and improved the PAR load by 28.70%, which contributes to grid stability. Despite the decrease in cost and grid load, the IMOFDO framework maintained user comfort. These enhancements ensured stable indoor conditions during occupancy, highlighting the strength and reliability of the IMOFDO framework.</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

Optimal smart home energy management using the improved multi-objective fitness dependent optimizer with a comfort violation penalty index

  • 期刊论文

  • 发表日期：

    2026年7月20日

查看全文
  • Ardalan H. Awlla,
  • Tarik A. Rashid,
  • Ronak M. Abdullah

摘要

While the demand for energy is increasing and smart home environments are becoming more complex, finding the right balance between energy costs and the comfort of residents in smart homes remains a challenge. This paper introduces a scheduling framework based on the Improved Multi-Objective Fitness Dependent Optimizer (IMOFDO) and an innovative Comfort Violation Penalty Index (CVPI) to tackle this issue; unlike traditional scheduling metrics, the CVPI offers a normalized, occupancy-aware penalty that integrates temperature, humidity, and lighting into a single optimization target. To overcome the limitations of the standard Fitness Dependent Optimizer (FDO) evolutionary algorithm, the IMOFDO incorporates Lévy flight and Gaussian mutation, striking a balance between global exploration and local precision. We evaluated the framework using real-time pricing (RTP) schemes and 15-min sensor data from a smart home dataset. The proposed framework was tested through year-long simulations of ten homes utilizing RTP. The results indicated that IMOFDO was better than MOPSO, MOFDO, and MOANA at lowering costs, peak to average ratio (PAR), and keeping users comfortable; we observed significant cost savings of up to 16.4%, or $500–$950 annually, particularly in homes with higher energy consumption. Additionally, the framework reduced peak demand by up to 1.05 kW and improved the PAR load by 28.70%, which contributes to grid stability. Despite the decrease in cost and grid load, the IMOFDO framework maintained user comfort. These enhancements ensured stable indoor conditions during occupancy, highlighting the strength and reliability of the IMOFDO framework.