Scholarlink

<p>This paper presents an intelligent Maximum Power Point Tracking (MPPT) control strategy for grid-connected photovoltaic (PV) systems, based on the integration of Artificial Neural Networks (ANN) and Model Predictive Control (MPC). The proposed ANN-MPC method is designed to overcome limitations of traditional MPPT algorithms, particularly their slow response and reduced efficiency under non-uniform and changing environmental conditions. The ANN component predicts optimal operating points using historical and real-time PV data, while MPC dynamically adjusts control actions to maximize power output and meet grid constraints. A detailed simulation study is conducted under constant irradiance, varying irradiance, and partial shading conditions. Results demonstrate that the ANN-MPC approach consistently achieves a high maximum power extraction ratio of 99.9%, low total harmonic distortion (THD) of grid current (as low as 1.48%), and system efficiency up to 98.3%. Comparisons with conventional methods including P&amp;O, Fuzzy, ANFIS, and ANN-PI further highlight the superior performance of the proposed control scheme in terms of precision, response time, and power quality. The system is implemented with a five-level NPC inverter and an LCL filter, ensuring grid compliance and high-quality power delivery. Although the environmental testing is based on stepwise irradiance changes and partial shading, the results validate the robustness of the proposed method. This research contributes a reliable, high-performance MPPT solution for PV systems, offering enhanced energy extraction and grid integration under changing climatic conditions.</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

ANN-MPC Based MPPT Control for Grid Connected PV Inverter System Under Dynamic Environmental Conditions

  • 期刊论文

  • 发表日期：

    2025年7月19日

查看全文
  • Ahmed Benfatah,
  • Noureddine Henini,
  • Abdelkader Morsli

摘要

This paper presents an intelligent Maximum Power Point Tracking (MPPT) control strategy for grid-connected photovoltaic (PV) systems, based on the integration of Artificial Neural Networks (ANN) and Model Predictive Control (MPC). The proposed ANN-MPC method is designed to overcome limitations of traditional MPPT algorithms, particularly their slow response and reduced efficiency under non-uniform and changing environmental conditions. The ANN component predicts optimal operating points using historical and real-time PV data, while MPC dynamically adjusts control actions to maximize power output and meet grid constraints. A detailed simulation study is conducted under constant irradiance, varying irradiance, and partial shading conditions. Results demonstrate that the ANN-MPC approach consistently achieves a high maximum power extraction ratio of 99.9%, low total harmonic distortion (THD) of grid current (as low as 1.48%), and system efficiency up to 98.3%. Comparisons with conventional methods including P&O, Fuzzy, ANFIS, and ANN-PI further highlight the superior performance of the proposed control scheme in terms of precision, response time, and power quality. The system is implemented with a five-level NPC inverter and an LCL filter, ensuring grid compliance and high-quality power delivery. Although the environmental testing is based on stepwise irradiance changes and partial shading, the results validate the robustness of the proposed method. This research contributes a reliable, high-performance MPPT solution for PV systems, offering enhanced energy extraction and grid integration under changing climatic conditions.