基于FPGA和神经网络逆控制的微型逆变器并网控制研究

发布时间：2019-03-26 13:31

【摘要】：化石能源的生产和消费所产生的大量的有毒污染物正严重加剧空气污染。作为一种可再生能源,清洁、无污染的太阳能引起了世界各国的广泛关注。利用太阳能发电是利用太阳能最为普遍的一种形式。微型逆变器作为光伏发电微型逆变器系统中重要的组成部分,对于微型逆变器的设计以及控制直接影响着整个光伏发电系统的效率与并网质量。本文将先进的神经网络逆控制方法应用到微型逆变器的控制中,通过仿真研究表明该控制方法能够使微型逆变器的并网电能质量得到提高,并对神经网络逆控制在实际微型逆变器上的应用进行实验。首先,对微型逆变器的分类,根据DC/DC变换电路和DC/AC变换电路的优缺点及微型逆变器的设计要点,确定了反激变换器和逆变全桥组成的电流型高频链微型逆变器的拓扑结构,并对常见的微型逆变器控制方法进行了描述,提出神经网络逆控制方法应用于微型逆变器的并网研究。其次,根据微型逆变器的拓扑结构,对其进行了建模,表明微型逆变器系统可逆;在此基础上,利用Matlab搭建了微型逆变器的仿真模型,建立了神经网络逆模型,并对神经网络逆控制下的微型逆变器系统进行了多种工况的仿真研究,表明该方法能改善微型逆变器的并网质量。最后,设计了基于FPGA和神经网络逆控制的微型逆变器实验平台,详细描述了实验硬件电路和软件的设计、FPGA与单片机之间的通讯协议。在此基础上,在实验平台开展了微型逆变器并网实验并记录了实验结果。
[Abstract]:Large quantities of toxic pollutants from fossil energy production and consumption are exacerbating air pollution. As a renewable energy, clean, pollution-free solar energy has attracted wide attention all over the world. The use of solar power is the most common form of solar energy. As an important part of the photovoltaic micro-inverter system, the design and control of the micro-inverter have a direct impact on the efficiency and grid-connected quality of the whole photovoltaic power generation system. In this paper, the advanced neural network inverse control method is applied to the control of micro-inverter. The simulation results show that the control method can improve the grid-connected power quality of micro-inverter. The application of neural network inverse control in practical micro inverter is also tested. First of all, the classification of micro-inverter, according to the advantages and disadvantages of DC/DC transform circuit and DC/AC converter circuit, and the key points of micro-inverter design, The topology of the current-source high-frequency chain micro-inverter composed of flyback converter and full-bridge inverter is determined, and the common control methods of micro-inverter are described. A neural network inverse control method is proposed and applied to the grid-connected research of micro-inverters. Secondly, according to the topology of the micro-inverter, the model is built, which shows that the micro-inverter system is reversible. On this basis, the simulation model of micro inverter is built by using Matlab, the inverse model of neural network is established, and the simulation research of micro inverter system under neural network inverse control is carried out under various working conditions. It is shown that this method can improve the grid-connected quality of micro-inverter. Finally, a micro inverter experiment platform based on FPGA and neural network inverse control is designed. The hardware circuit and software design of the experiment are described in detail, and the communication protocol between FPGA and single chip microcomputer is described in detail. On this basis, the micro inverter grid-connected experiment is carried out on the experimental platform and the experimental results are recorded.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM464

【参考文献】