摘要
提出一种基于人工神经网络的燃料电池混合动力系统。该系统采用两级循环控制,第一级采用神经网络控制,以实现能量跟踪的最大化,在不同的日光、温度和负载条件下提取最大的可利用能量;第二级采用实时/反应能量控制器,通过控制进入燃料电池堆的燃料,同时发送控制信号到能量条件子系统,以满足系统对反应能量的需求。通过时域仿真可知,该系统对混合动力系统稳定有效。
The purpose of this paper is to design a fuel cell hybrid power system based on artificial neural network.The system consists of two loops,the first loop is a neural network controller to maximize tracking power,which extracts maximum available power under vary. ing conditions of sunlight,temperature and system load.The second loop is a real/reactive power controller.The system＇s requirements for real and reactive powers are controlled by incoming fuel to fuel cell stacks,at the same time the switching control signals are sent to the power conditioning subsystem.Time-domain simulations prove this system is effective and stable to the hybrid power system.
出处
《电力电子技术》
CSCD
北大核心
2008年第10期50-51,57,共3页
Power Electronics
关键词
燃料电池
神经网络
最大能量
fuel cell. neural network ： maximum oower
