Scholarlink

<p>This paper introduces an enhanced control scheme for an 11&#xa0;kW grid connected solar photovoltaic (PV) system, addressing two interrelated challenges: maximum power extraction under rapid irradiance changes and high-quality power injection into utility grid. A novel hybrid Fungal Growth Optimization-Artificial Neural Network (FGO-ANN) for Maximum Power tracking (MPT) is proposed and exploits the mycelial network- inspired exploration process for efficiently optimized weights of neural network. On the grid side control, an Improved Least Mean Square (ILMS) is implemented for inverter control. The ILMS controller ensures high-quality power injection into the utility grid, improves DC-link voltage regulation and dynamic performance. Additionaly, Lyapunov stability criteria is established for the proposed ILMS controller, providing a rigorous bound on the adaptation factor that ensures mean square convergence for all dynamic conditions. Rigorous comparative performance evaluation against hybrid Maximum Power Point Tracking (MPPT) algorithm, namely, Particle Swarm Optimization-Artificial Neural Network (PSO-ANN) and Hybrid Lion Optimization- Artificial Neural Network (HLOA-ANN) techniques has been carried out. The evaluation demonstrates the superiority of the proposed FGO-ANN in all evaluated parameters. The proposed FGO-ANN achieves faster tracking time and reduced steady-state error compared to other techniques. Further, as compared to conventional LMS technique, the ILMS controller achieves substantial harmonic reduction under voltage unbalancing, sag and swell condition. Results are evaluated and compared using the MATLAB/Simulink environment; and OPAL-RT for practical feasibility. The proposed control technique performs better under varying ambient conditions since the grid current is well within the IEEE-1547 standard limitations.</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 Novel Hybrid Fungal Growth Optimization -ANN Based MPPT with Improved LMS Adaptive Control for Grid Connected Solar PV Systems

  • 期刊论文

  • 发表日期：

    2026年8月4日

查看全文
  • Priya Rai,
  • Alka Singh,
  • Rachna Garg

摘要

This paper introduces an enhanced control scheme for an 11 kW grid connected solar photovoltaic (PV) system, addressing two interrelated challenges: maximum power extraction under rapid irradiance changes and high-quality power injection into utility grid. A novel hybrid Fungal Growth Optimization-Artificial Neural Network (FGO-ANN) for Maximum Power tracking (MPT) is proposed and exploits the mycelial network- inspired exploration process for efficiently optimized weights of neural network. On the grid side control, an Improved Least Mean Square (ILMS) is implemented for inverter control. The ILMS controller ensures high-quality power injection into the utility grid, improves DC-link voltage regulation and dynamic performance. Additionaly, Lyapunov stability criteria is established for the proposed ILMS controller, providing a rigorous bound on the adaptation factor that ensures mean square convergence for all dynamic conditions. Rigorous comparative performance evaluation against hybrid Maximum Power Point Tracking (MPPT) algorithm, namely, Particle Swarm Optimization-Artificial Neural Network (PSO-ANN) and Hybrid Lion Optimization- Artificial Neural Network (HLOA-ANN) techniques has been carried out. The evaluation demonstrates the superiority of the proposed FGO-ANN in all evaluated parameters. The proposed FGO-ANN achieves faster tracking time and reduced steady-state error compared to other techniques. Further, as compared to conventional LMS technique, the ILMS controller achieves substantial harmonic reduction under voltage unbalancing, sag and swell condition. Results are evaluated and compared using the MATLAB/Simulink environment; and OPAL-RT for practical feasibility. The proposed control technique performs better under varying ambient conditions since the grid current is well within the IEEE-1547 standard limitations.