摘要
准确识别流型是流化床气固二相流参数检测的一项重要内容,实验是在流化床气固二相流实验系统上进行的。首先,采集5种典型流型的压力波动信号,并以信号的统计参量作为流型特征。然后,将样本送入经过人工鱼群优化的BP神经网络进行训练。人工鱼群(AFSA)是一种新型的智能优化算法,具有全局收敛性好,鲁棒性强,对初值不敏感等特点,通过优化神经网络的权值使识别率得到明显提高,实现了气固流化床典型流型的快速、准确识别。实验结果表明,该方法对气固流化床5种典型流型的识别率达到97%,为在线识别气固流化床流型提供了一种新的有效方法。
The exact identification of flow regime is an important content to detect the parameters of gas-solid two-phase flow fluidized bed.The experiments were conducted on gas-solid fluidized bed system in the following steps： collect pressure fluctuation signals of 5 kinds of typical flow regimes,take statistical parameters of those signals as characteristics of flow regimes,and send those training samples to BP neural network optimized by artificial fish-swarm algorithm（AFSA）,which is a new kind of intelligence optimization algorithm,and has advantages of good global astringency,robustness and insensitive to initial value.The optimized weights of neural network can improve identification rate obviously.Thus,it can identify 5 typical flow regimes of gas-solid fluidized bed successfully,rapidly and accurately.The test results show that identification rate of this method can reach 97% and it provides a new effective way to identify the gas-solid flow regime of fluidized bed online.
出处
《化学工程》
CAS
CSCD
北大核心
2010年第7期39-42,50,共5页
Chemical Engineering(China)
关键词
BP神经网络
人工鱼群算法
气固流化床
统计参数
BP neural network
artificial fish-swarm algorithm
gas-solid fluidized bed
statistical parameters
作者简介
周云龙（1960-），男，教授，博士，从事多相流流动特性和多相流检测技术方面的研究，E-mail：zyl@mail．nedu．edu．cn。
