基于图像纹理特征和Elman神经网络的气液两相流流型识别.pdfVIP

第 27 卷 第 29 期 中 国 电 机 工 程 学 报 Vol.27 No.29 Oct. 2007 2007 年 10 月 Proceedings of the CSEE ©2007 Chin.Soc.for Elec.Eng. 0258-8013 (2007) 29-0108-05 O359 A 470 10 文章编号： 中图分类号： 文献标识码： 学科分类号： ⋅ 基于图像处理和 Elman 神经网络的 气液两相流流型识别 周云龙，陈 飞，刘 川 (东北电力大学，吉林省 吉林市132012) Identification Method of Gas-liquid Two-phase Flow Regime Based on Images Processing and Elman Neural Network ZHOU Yun-long, CHEN Fei, LIU Chuan (Northeast Dianli University, Jilin 132012, Jilin Province, China) ABSTRACT: Gas-liquid two-phase flow widely exists in 关键词：两相流；流型识别；图像处理；Elman 神经网络 modern industry process. Two-phase flow and heat transfer 0 引言 character are extremely influenced by the flow regimes. Therefore, a flow regime identification method based on 两相流动现象广泛存在于自然界和许多工业 images statistical features of gray histogram and Elman neural 过程中。两相流动介质的分布状况，即流型，极大 network was proposed. Gas-liquid two-phase flow images were 影响两相流动压力损失和传热传质等特性，同时影 captured by high speed video system in horizontal pipe. The [1] 响流动参数的准确测量以及流动系统的运行特性 。 images statistical features of the gray histogram were extracted 气液两相流流型识别主要有两种方法：一类是 using image processing techniques. Then ，images statistical 根据两相流流动图像的形式直接确定流型，如目测 eigenvectors of flow regime were established. Elman neural 法、高速摄影法、过程层析成像法等；另一类是间 network was trained using those eigenvect