Scholarlink

<p>Wind Energy Conversion Systems (WECS) are vital for clean energy generation, but optimizing power extraction under fluctuating wind conditions remains a significant challenge. Traditional methods often lack efficiency and reliability in AC-DC conversion. This study proposes a novel approach combining the Portia Spider Optimization Algorithm (PSOA) with a Multi-Component Attention Graph Convolutional Neural Network (MCAGCNN) named PSOA-MCAGCNN to enhance the efficiency of WECS with three-phase AC-DC converters. The proposed method optimizes power production while supporting sustainable development by enhancing wind energy’s role in reducing carbon emissions. PSOA optimizes wind speed fluctuations, while MCAGCNN predicts parameters of the NB-IPCuk converter. The proposed strategy consistently outperforms existing techniques such as Long Short Term Memory (LSTM), Particle Swarm Optimization (PSO) and Recurrent Neural Network (RNN), achieving a power output of 2400&#xa0;kW with an efficiency of 98%. Additionally, operational losses are reduced, with semiconductor losses reduced by 12&#xa0;kW, conduction losses by 2&#xa0;kW, and switching losses by 9&#xa0;kW. The method also demonstrates a torque of 6.5&#xa0;N/m and achieves the lowest Total Harmonic Distortion (THD) of 1.76%, highlighting its superior performance and efficiency. These results demonstrate that the PSOA-MCAGCNN approach significantly enhances power output, operational efficiency, and grid stability, ensuring long-term sustainability under fluctuating wind conditions. The study suggests that the PSOA-MCAGCNN framework offers a promising approach for real-time optimization of WECS. Its performance indicates potential for application in grid-integrated renewable systems, with the capacity to improve energy reliability and contribute to reductions in carbon emissions.</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

Enhancing the efficiency of wind energy conversion systems with Three-Phase AC-DC converters using Multi-Component attention graph convolutional neural networks for dynamic power management

  • 期刊论文

  • 发表日期：

    2025年9月15日

查看全文
  • R. Kannan,
  • G. Rajendar,
  • P. Rajesh

摘要

Wind Energy Conversion Systems (WECS) are vital for clean energy generation, but optimizing power extraction under fluctuating wind conditions remains a significant challenge. Traditional methods often lack efficiency and reliability in AC-DC conversion. This study proposes a novel approach combining the Portia Spider Optimization Algorithm (PSOA) with a Multi-Component Attention Graph Convolutional Neural Network (MCAGCNN) named PSOA-MCAGCNN to enhance the efficiency of WECS with three-phase AC-DC converters. The proposed method optimizes power production while supporting sustainable development by enhancing wind energy’s role in reducing carbon emissions. PSOA optimizes wind speed fluctuations, while MCAGCNN predicts parameters of the NB-IPCuk converter. The proposed strategy consistently outperforms existing techniques such as Long Short Term Memory (LSTM), Particle Swarm Optimization (PSO) and Recurrent Neural Network (RNN), achieving a power output of 2400 kW with an efficiency of 98%. Additionally, operational losses are reduced, with semiconductor losses reduced by 12 kW, conduction losses by 2 kW, and switching losses by 9 kW. The method also demonstrates a torque of 6.5 N/m and achieves the lowest Total Harmonic Distortion (THD) of 1.76%, highlighting its superior performance and efficiency. These results demonstrate that the PSOA-MCAGCNN approach significantly enhances power output, operational efficiency, and grid stability, ensuring long-term sustainability under fluctuating wind conditions. The study suggests that the PSOA-MCAGCNN framework offers a promising approach for real-time optimization of WECS. Its performance indicates potential for application in grid-integrated renewable systems, with the capacity to improve energy reliability and contribute to reductions in carbon emissions.