Scholarlink

<p>Photovoltaic (PV) systems face significant performance degradation under partial shading conditions (PSC), where conventional maximum power point tracking (MPPT) methods often converge to local maxima and lose efficiency. While gray wolf optimization (GWO)-based MPPT and adaptive neuro-fuzzy inference system (ANFIS) battery controllers have been studied separately, this work introduces a novel, fully integrated control framework that unifies both functions into a single, real-time capable system for hybrid PV–battery microgrids. The proposed strategy employs the GWO for global MPPT via a boost converter and an ANFIS with a PI regulator for intelligent battery management through a bidirectional sepic/zeta converter. A key innovation is the hybrid state of charge (SOC) estimation method, which integrates Coulomb counting (CC) with an ANFIS-based voltage model and a closed-loop PI corrector to significantly reduce estimation error. Simulation results under standard, dynamic, and partial shading conditions demonstrate the system’s robustness: The GWO–MPPT achieves 97% efficiency with zero steady-state oscillations, outperforming perturb and observe (P&amp;O) and incremental inductance (IC) methods. The ANFIS–PI controller ensures stable battery transitions without overshoot, and the hybrid SOC estimator maintains accuracy within ± 0.01%. Importantly, under variable irradiance and load steps, the system maintains continuous load supply and full grid independence, with smooth battery transitions and stable SOC operation within safe limits.&#xa0;These results confirm that the integrated intelligent framework enhances energy capture, storage reliability, and operational resilience in stand-alone microgrids.</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

Advanced control strategy for energy optimization in microgrids using intelligent algorithms: gray wolf optimizer and ANFIS

  • 期刊论文

  • 发表日期：

    2026年2月9日

查看全文
  • Aymen Lachheb,
  • Noureddine Akoubi,
  • Jamel Ben Salem

摘要

Photovoltaic (PV) systems face significant performance degradation under partial shading conditions (PSC), where conventional maximum power point tracking (MPPT) methods often converge to local maxima and lose efficiency. While gray wolf optimization (GWO)-based MPPT and adaptive neuro-fuzzy inference system (ANFIS) battery controllers have been studied separately, this work introduces a novel, fully integrated control framework that unifies both functions into a single, real-time capable system for hybrid PV–battery microgrids. The proposed strategy employs the GWO for global MPPT via a boost converter and an ANFIS with a PI regulator for intelligent battery management through a bidirectional sepic/zeta converter. A key innovation is the hybrid state of charge (SOC) estimation method, which integrates Coulomb counting (CC) with an ANFIS-based voltage model and a closed-loop PI corrector to significantly reduce estimation error. Simulation results under standard, dynamic, and partial shading conditions demonstrate the system’s robustness: The GWO–MPPT achieves 97% efficiency with zero steady-state oscillations, outperforming perturb and observe (P&O) and incremental inductance (IC) methods. The ANFIS–PI controller ensures stable battery transitions without overshoot, and the hybrid SOC estimator maintains accuracy within ± 0.01%. Importantly, under variable irradiance and load steps, the system maintains continuous load supply and full grid independence, with smooth battery transitions and stable SOC operation within safe limits. These results confirm that the integrated intelligent framework enhances energy capture, storage reliability, and operational resilience in stand-alone microgrids.