Scholarlink

<p>This paper proposes a novel hybrid approach, called BWWPA-QPINN, which represents a new method for the optimal pricing of public Electric Vehicle Charging Stations (EVCS) embedded within Power Systems (PS). The Binary Water Wheel Plant Algorithm (BWWPA) optimizes, location wise, charging infrastructure to minimize power loss and voltage instability. The Quantum Physics-Informed Neural Network (QPINN) forecasts time varying prices for EVCS to optimize user costs, revenue for the operator and grid stability. Implemented on the MATLAB platform, the method is evaluated against existing optimization techniques including scenario-based stochastic optimization (SBSO), Soft Actor-Critic Reinforcement Learning Algorithm (SACRL), Particle Swarm Optimization (PSO), Deep Reinforcement Learning (DRL), and Arithmetic Optimization Algorithm (AOA). The results demonstrate that BWWPA-QPINN achieves a minimum power loss of 10.5&#xa0;kW, surpassing existing methods in computational efficiency, pricing accuracy 98%, and grid stability. The integration of BWWPA’s efficient spatial optimization with QPINN’s physics-informed learning constitutes a significant advancement in the design of scalable and grid-compliant EVCS pricing strategies.</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

A Hybrid Binary Water Wheel Plant Algorithm and Physics Informed Neural Network for Optimal Pricing of Public Electric Vehicle Charging Stations with Power Systems

  • 期刊论文

  • 发表日期：

    2025年10月18日

查看全文
  • M. Panneer Selvam,
  • H. Umesh Prabhu,
  • Babu Rajendra Prasad,
  • Koganti Srilakshmi

摘要

This paper proposes a novel hybrid approach, called BWWPA-QPINN, which represents a new method for the optimal pricing of public Electric Vehicle Charging Stations (EVCS) embedded within Power Systems (PS). The Binary Water Wheel Plant Algorithm (BWWPA) optimizes, location wise, charging infrastructure to minimize power loss and voltage instability. The Quantum Physics-Informed Neural Network (QPINN) forecasts time varying prices for EVCS to optimize user costs, revenue for the operator and grid stability. Implemented on the MATLAB platform, the method is evaluated against existing optimization techniques including scenario-based stochastic optimization (SBSO), Soft Actor-Critic Reinforcement Learning Algorithm (SACRL), Particle Swarm Optimization (PSO), Deep Reinforcement Learning (DRL), and Arithmetic Optimization Algorithm (AOA). The results demonstrate that BWWPA-QPINN achieves a minimum power loss of 10.5 kW, surpassing existing methods in computational efficiency, pricing accuracy 98%, and grid stability. The integration of BWWPA’s efficient spatial optimization with QPINN’s physics-informed learning constitutes a significant advancement in the design of scalable and grid-compliant EVCS pricing strategies.