Scholarlink

<p>This paper presents an innovative hybrid electric vehicle (HEV) charging station design that adds a high-gain bidirectional DC-DC converter and a hybrid control system based on Greylag Goose Optimization (GGO) and a Quantum Convolutional Neural Network (QCNN). The system coordinates the flow of energy between solar photovoltaic (PV) panels, fuel cells (FC), batteries, and supercapacitors (SCs) on real-time energy management (EM). The GGO algorithm controls FC output based on solar irradiance and battery state of charge (SoC), whereas the QCNN predicts torque signals with high precision, thereby improving propulsion efficiency. The MATLAB simulations indicate that the suggested GGO-QCNN algorithm can reduce hydrogen consumption by 8.7% compared to the conventional Proportional Integral (PI) and State Machine Control (SMC) methods. The system maintains a stable PV voltage of 43 V and supports a variable PV current between 5-25 A, delivering a power range of 100-1000 W. It has a maximum EM accuracy of 99.97% across 1000 trials. The hybrid system provides optimal allocation of power and minimizes battery overcharging/discharging risks during high demand driving conditions (as is the case with the Worldwide Harmonized Light Vehicle Test Procedure Class 3 (WLTP-Class 3), which is used to test vehicles with aggressive speed profiles, and the Highway Fuel Economy Test (HWFET), which is used to test vehicles with high highway fuel efficiency). These findings validate the originality and high-quality performance of the GGO-QCNN-based system of augmenting energy consumption and fuel efficiency in the context of HEV charging systems.</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

High-Gain Bidirectional Converter with Hybrid Greylag Goose Optimization and Quantum Convolutional Neural Network for Intelligent Energy Management in Electric Vehicle Charging Systems

  • 期刊论文

  • 发表日期：

    2025年12月26日

查看全文
  • G. Rajendar,
  • B. Vijay Kumar

摘要

This paper presents an innovative hybrid electric vehicle (HEV) charging station design that adds a high-gain bidirectional DC-DC converter and a hybrid control system based on Greylag Goose Optimization (GGO) and a Quantum Convolutional Neural Network (QCNN). The system coordinates the flow of energy between solar photovoltaic (PV) panels, fuel cells (FC), batteries, and supercapacitors (SCs) on real-time energy management (EM). The GGO algorithm controls FC output based on solar irradiance and battery state of charge (SoC), whereas the QCNN predicts torque signals with high precision, thereby improving propulsion efficiency. The MATLAB simulations indicate that the suggested GGO-QCNN algorithm can reduce hydrogen consumption by 8.7% compared to the conventional Proportional Integral (PI) and State Machine Control (SMC) methods. The system maintains a stable PV voltage of 43 V and supports a variable PV current between 5-25 A, delivering a power range of 100-1000 W. It has a maximum EM accuracy of 99.97% across 1000 trials. The hybrid system provides optimal allocation of power and minimizes battery overcharging/discharging risks during high demand driving conditions (as is the case with the Worldwide Harmonized Light Vehicle Test Procedure Class 3 (WLTP-Class 3), which is used to test vehicles with aggressive speed profiles, and the Highway Fuel Economy Test (HWFET), which is used to test vehicles with high highway fuel efficiency). These findings validate the originality and high-quality performance of the GGO-QCNN-based system of augmenting energy consumption and fuel efficiency in the context of HEV charging systems.