← 返回
基于人工神经网络控制的单相离网型太阳能微逆变器
ANN Controlled Single-Phase Microinverter for Off-Grid Solar Application
| 作者 | Ajay Kumar Sahu · Ramnarayan Patel · Lalit Kumar Sahu |
| 期刊 | IEEE Journal of Emerging and Selected Topics in Power Electronics |
| 出版日期 | 2024年9月 |
| 技术分类 | 光伏发电技术 |
| 技术标签 | 储能系统 户用光伏 深度学习 |
| 相关度评分 | ★★★★★ 5.0 / 5.0 |
| 关键词 | 太阳能微逆变器 人工神经网络控制器 Z源变换器 稳定电压 光伏泄漏电流 |
语言:
中文摘要
太阳能板输出电压受光照、温度及遮蔽影响而波动,难以直接为家用电器提供稳定电压。本文提出一种采用人工神经网络(ANN)控制器的太阳能微逆变器,可在24–46 V输入范围内稳定输出230 V<sub>rms</sub>交流电压。通过带主动开关的改进型Z源变换器实现所需电压增益,ANN控制器动态调节占空比,以应对输入电压与负载变化,确保系统最优控制性能。该微逆变器实现光伏板与负载间的共地连接,有效抑制光伏泄漏电流。文章详述了电路工作原理,并通过实时实验验证了其在实际工况下的有效性,为太阳能离网供电提供稳定可靠的解决方案。
English Abstract
The voltage produced by solar panels fluctuates due to change in irradiance, temperature, and shading. This makes it challenging to obtain a stable voltage suitable for powering devices that use regular household electricity. This article proposes a solar microinverter with artificial neural network (ANN) controller that can handle input voltages ranging from 24 to 46 V and consistently produces an output voltage of 230V_ rms . A modified Z-source converter with an active switch helps achieve the requisite gain. The ANN controller enables dynamic adjustment of the duty ratio in response to fluctuations in input voltage or load, ensuring optimal control and performance. The proposed microinverter provides a common grounding feature between the solar panel and load; this suppresses the photovoltaic (PV) leakage current. This article provides a detailed circuit operation explanation, and real-time validation confirms the performance of microinverter under practical conditions. This solution addresses the challenge of obtaining stable voltages for household device power supply from solar panels.
S
SunView 深度解读
该ANN控制的微逆变器技术对阳光电源户用光伏产品线具有重要参考价值。其24-46V宽输入范围与改进型Z源拓扑可应用于SG系列户用逆变器的低压输入场景,特别是分布式MPPT优化。ANN动态占空比调节策略可融入阳光电源现有MPPT算法,提升复杂工况下的响应速度。共地连接抑制漏电流的设计与阳光电源储能系统ST系列的安全防护理念一致,可借鉴用于户用储能一体机。该深度学习控制思路为iSolarCloud智能诊断平台提供算法创新方向,可探索将ANN应用于逆变器自适应控制与预测性维护,提升系统智能化水平。