Scholarlink

<p>The rapid proliferation of electric vehicles (EVs) requires advanced control strategies to optimize energy efficiency, extend battery longevity, and ensure grid stability within vehicle-to-grid (V2G) frameworks. This systematic review provides a comprehensive evaluation of state-of-the-art intelligent control algorithms, specifically focusing on the mathematical formulation and performance benchmarking of Model Predictive Control (MPC), Reinforcement Learning (RL), and Fuzzy Logic Control (FLC). Following PRISMA guidelines, this study synthesizes research from 2015 to 2026, identifying critical transitions from individual vehicle management to IoT-enabled microgrid orchestration. Our analysis reveals that while MPC offers superior deterministic optimization reducing charging times by approximately 20% its reliance on high-fidelity models and significant computational overhead limits its deployment in cost-sensitive systems. Conversely, RL and deep learning architectures, such as Residual-Normalised Gated Recurrent Units (GRUs), demonstrate high adaptability in forecasting stochastic net-load fluctuations and managing battery state-of-health (SOH) across EV fleets. We further identify critical research gaps concerning algorithmic scalability and real-time execution latency in decentralized networks. The findings suggest that hybrid control frameworks, which integrate the robustness of fuzzy logic with the predictive precision of data-driven models, are essential for achieving long-term battery resilience and grid reliability. This paper contributes a novel taxonomy and a rigorous methodological roadmap to guide future developments in smart charging infrastructure, aligning algorithmic advancements with global 2030 sustainability and electrification targets.</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 Review Study On Advanced Intelligent Control Algorithms for Optimizing Charging Discharging, and Energy Efficiency in Electric Vehicles

  • 期刊论文

  • 发表日期：

    2026年6月15日

查看全文
  • Neethu Usha,
  • Deepu. S. R

摘要

The rapid proliferation of electric vehicles (EVs) requires advanced control strategies to optimize energy efficiency, extend battery longevity, and ensure grid stability within vehicle-to-grid (V2G) frameworks. This systematic review provides a comprehensive evaluation of state-of-the-art intelligent control algorithms, specifically focusing on the mathematical formulation and performance benchmarking of Model Predictive Control (MPC), Reinforcement Learning (RL), and Fuzzy Logic Control (FLC). Following PRISMA guidelines, this study synthesizes research from 2015 to 2026, identifying critical transitions from individual vehicle management to IoT-enabled microgrid orchestration. Our analysis reveals that while MPC offers superior deterministic optimization reducing charging times by approximately 20% its reliance on high-fidelity models and significant computational overhead limits its deployment in cost-sensitive systems. Conversely, RL and deep learning architectures, such as Residual-Normalised Gated Recurrent Units (GRUs), demonstrate high adaptability in forecasting stochastic net-load fluctuations and managing battery state-of-health (SOH) across EV fleets. We further identify critical research gaps concerning algorithmic scalability and real-time execution latency in decentralized networks. The findings suggest that hybrid control frameworks, which integrate the robustness of fuzzy logic with the predictive precision of data-driven models, are essential for achieving long-term battery resilience and grid reliability. This paper contributes a novel taxonomy and a rigorous methodological roadmap to guide future developments in smart charging infrastructure, aligning algorithmic advancements with global 2030 sustainability and electrification targets.