Scholarlink

<p>Extracting maximum power from photovoltaic systems (PV) is challenging due to the nonlinear relationship between voltage and current. This paper proposes a novel maximum power point tracking (MPPT) controller that leverages the combined strengths of artificial neural networks (ANNs) and robust terminal sliding mode control (RTSMC). An ANN trained using an adaptive particle swarm optimization (APSO) algorithm predicts the maximum power point (MPP) of the PV system. PSO optimizes the ANN’s structure and initial weights, leading to faster training times and improved prediction accuracy under varying environmental conditions. Subsequently, a TRSMC method ensures the system precisely and rapidly tracks the predicted MPP. TRSMC’s inherent robustness guarantees effective operation even under uncertain conditions like sudden changes in sunlight or temperature. This combined approach offers a high-accuracy, fast-response, and robust MPPT controller, maximizing power extraction from PV systems. The effectiveness of the proposed method is validated through simulations, demonstrating superior performance compared to other algorithms such as perturb and observe (P&amp;O) and a P&amp;O-based RTSMC controller (P&amp;O-RTSMC).</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

Efficiency optimization of a photovoltaic system using adaptive particle swarm optimization and neural network based on robust terminal sliding mode control for maximum power point tracking

  • 期刊论文

  • 发表日期：

    2025年5月20日

查看全文
  • Belgacem Mbarki,
  • Jaouher Chrouta,
  • Hechmi Khaterchi,
  • Fethi Farhani,
  • Abderrahmen Zaafouri

摘要

Extracting maximum power from photovoltaic systems (PV) is challenging due to the nonlinear relationship between voltage and current. This paper proposes a novel maximum power point tracking (MPPT) controller that leverages the combined strengths of artificial neural networks (ANNs) and robust terminal sliding mode control (RTSMC). An ANN trained using an adaptive particle swarm optimization (APSO) algorithm predicts the maximum power point (MPP) of the PV system. PSO optimizes the ANN’s structure and initial weights, leading to faster training times and improved prediction accuracy under varying environmental conditions. Subsequently, a TRSMC method ensures the system precisely and rapidly tracks the predicted MPP. TRSMC’s inherent robustness guarantees effective operation even under uncertain conditions like sudden changes in sunlight or temperature. This combined approach offers a high-accuracy, fast-response, and robust MPPT controller, maximizing power extraction from PV systems. The effectiveness of the proposed method is validated through simulations, demonstrating superior performance compared to other algorithms such as perturb and observe (P&O) and a P&O-based RTSMC controller (P&O-RTSMC).