Scholarlink

<p>Electric vehicles (EVs) are increasingly recognized as a solution to transportation-related air pollution, yet their widespread adoption is limited by restricted driving range per charge. This manuscript proposes a novel hybrid technique to optimize energy use and extend EV range by integrating Sea-Horse Optimization (SHO) with Hamiltonian Deep Neural Networks (HDNNs), termed the SHO-HDNN technique. The proposed method enhances EV range and motor drive efficiency by using SHO to optimize energy consumption and HDNNs to predict EV range based on current driving conditions. The system's performance is evaluated through control error and speed analysis. The SHO-HDNN model is implemented in MATLAB and contrasted with other methods already in use, including Multi-Island Genetic Algorithm, Particle Swarm Optimization and Non dominated Sorting Genetic Algorithm (NSGA-II).Results demonstrate that the SHO-HDNN method improves prediction accuracy to 99% and reduces error value to 1.02%, outperforming other methods in optimizing energy consumption and motor drive efficiency. This approach not only extends the EV range but also enhances overall system performance and adaptability for various EV models, offering a significant contribution to the field of electric vehicle energy management.</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

Optimizing electric vehicle range through hybrid SHO-HDNN technique

  • 期刊论文

  • 发表日期：

    2025年2月3日

查看全文
  • N. Vengadachalam,
  • R. Ilango,
  • V. Subha Seethalakshmi

摘要

Electric vehicles (EVs) are increasingly recognized as a solution to transportation-related air pollution, yet their widespread adoption is limited by restricted driving range per charge. This manuscript proposes a novel hybrid technique to optimize energy use and extend EV range by integrating Sea-Horse Optimization (SHO) with Hamiltonian Deep Neural Networks (HDNNs), termed the SHO-HDNN technique. The proposed method enhances EV range and motor drive efficiency by using SHO to optimize energy consumption and HDNNs to predict EV range based on current driving conditions. The system's performance is evaluated through control error and speed analysis. The SHO-HDNN model is implemented in MATLAB and contrasted with other methods already in use, including Multi-Island Genetic Algorithm, Particle Swarm Optimization and Non dominated Sorting Genetic Algorithm (NSGA-II).Results demonstrate that the SHO-HDNN method improves prediction accuracy to 99% and reduces error value to 1.02%, outperforming other methods in optimizing energy consumption and motor drive efficiency. This approach not only extends the EV range but also enhances overall system performance and adaptability for various EV models, offering a significant contribution to the field of electric vehicle energy management.