Scholarlink

<p>Increasing reliance on renewable energy sources (RES) within smart grid systems, ensuring power balance amid fluctuations in energy production and load demand presents a significant challenge. This study proposes a novel hybrid approach, termed the GJO-THDCNN technique, which integrates Golden Jackal Optimization (GJO) with a Tree Hierarchical Deep Convolutional Neural Network (THDCNN) to address this issue effectively. The proposed approach uses advanced controllers and power electronic converters to improve overall performance while integrating battery storage with solar and wind energy conversion systems. GJO generates optimized control signals, while the THDCNN enhances prediction accuracy by considering power demand, state-of-charge (SoC), and RES availability. Implemented in MATLAB, the model showcases superior performance compared to existing methods, achieving a remarkable 20% improvement in power output stability and a 30% reduction in response time to load variations. These findings underscore the GJO-THDCNN technique's potential for advancing energy management strategies in smart grids.</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

Hybrid optimization for energy management in smart grids using Golden Jackal algorithm and deep convolutional neural networks

  • 期刊论文

  • 发表日期：

    2025年2月2日

查看全文
  • S. Chithra,
  • C. Arunachalaperumal,
  • R. Rajagopal,
  • P. Meenalochini

摘要

Increasing reliance on renewable energy sources (RES) within smart grid systems, ensuring power balance amid fluctuations in energy production and load demand presents a significant challenge. This study proposes a novel hybrid approach, termed the GJO-THDCNN technique, which integrates Golden Jackal Optimization (GJO) with a Tree Hierarchical Deep Convolutional Neural Network (THDCNN) to address this issue effectively. The proposed approach uses advanced controllers and power electronic converters to improve overall performance while integrating battery storage with solar and wind energy conversion systems. GJO generates optimized control signals, while the THDCNN enhances prediction accuracy by considering power demand, state-of-charge (SoC), and RES availability. Implemented in MATLAB, the model showcases superior performance compared to existing methods, achieving a remarkable 20% improvement in power output stability and a 30% reduction in response time to load variations. These findings underscore the GJO-THDCNN technique's potential for advancing energy management strategies in smart grids.