Scholarlink

<p>The growing adoption of electric vehicles (EVs) necessitates an extensive and efficient charging infrastructure, which must be reliably integrated into the smart grid. However, uncoordinated deployment of Electric Vehicle Charging Stations (EVCS) can compromise grid reliability, increase power losses, and contribute to environmental burdens. This research presents a novel hybrid methodology combining the Crayfish Optimization Algorithm (COA) with Finite Basis Physics-Informed Neural Networks (FBPINN) to strategically allocate EVCS for enhanced smart grid performance. The Proposed methodology is named COA-FBPINN. COA performs global optimization of EVCS locations, while FBPINN provides intelligent modeling of grid behaviour to ensure effective integration. The Simulation results obtained using MATLAB show that the COA-FBPINN approach significantly outperforms benchmark algorithms like Particle Swarm Optimization (PSO), Salp Swarm Algorithm (SSA), and Honey Badger Optimization (HBO). Specifically, the proposed technique attains a daily operational cost of $0.50, computation time of 5.1&#xa0;s, power loss of just 0.01&#xa0;MW, and emission levels of 2.5 × 10<sup>5</sup>&#xa0;g. Furthermore, it delivers high predictive performance with 95% accuracy and 98% precision in placement decisions. These results demonstrate that COA-FBPINN is highly effective in reducing operational costs, minimizing energy losses and emissions, and improving the entire reliability of the smart grid. The proposed approach is recommended as a robust solution for next-generation EVCS planning and smart grid resilience.</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 smart grid reliability through strategic allocation of electric vehicle charging stations using COA-FBPINNs approach

  • 期刊论文

  • 发表日期：

    2025年6月3日

查看全文
  • S. Sangeetha,
  • R. Saravanan,
  • G. Kannayeram

摘要

The growing adoption of electric vehicles (EVs) necessitates an extensive and efficient charging infrastructure, which must be reliably integrated into the smart grid. However, uncoordinated deployment of Electric Vehicle Charging Stations (EVCS) can compromise grid reliability, increase power losses, and contribute to environmental burdens. This research presents a novel hybrid methodology combining the Crayfish Optimization Algorithm (COA) with Finite Basis Physics-Informed Neural Networks (FBPINN) to strategically allocate EVCS for enhanced smart grid performance. The Proposed methodology is named COA-FBPINN. COA performs global optimization of EVCS locations, while FBPINN provides intelligent modeling of grid behaviour to ensure effective integration. The Simulation results obtained using MATLAB show that the COA-FBPINN approach significantly outperforms benchmark algorithms like Particle Swarm Optimization (PSO), Salp Swarm Algorithm (SSA), and Honey Badger Optimization (HBO). Specifically, the proposed technique attains a daily operational cost of $0.50, computation time of 5.1 s, power loss of just 0.01 MW, and emission levels of 2.5 × 105 g. Furthermore, it delivers high predictive performance with 95% accuracy and 98% precision in placement decisions. These results demonstrate that COA-FBPINN is highly effective in reducing operational costs, minimizing energy losses and emissions, and improving the entire reliability of the smart grid. The proposed approach is recommended as a robust solution for next-generation EVCS planning and smart grid resilience.