模糊免疫网络算法在数字图像火焰监测中的应用 the application of fuzzy immune network algorithm to flame monitoring.pdfVIP

第27 卷 第2 期 中 国 电 机 工 程 学 报 Vol.27 No.2 Jan. 2007 2007 年1 月 Proceedings of the CSEE 2007 Chin.Soc.for Elec.Eng. 0258-8013 (2007) 02-0001-05 TK311 A 470 20 文章编号 中图分类号 文献标识码 学科分类号 模糊免疫网络算法在数字图像火焰监测中的应用 郭建民 刘 石 姜 凡 ( 中国科学院工程热物理研究所 北京市 海淀区 100080) The Application of Fuzzy Immune Network Algorithm to Flame Monitoring GUO Jian-min, LIU Shi, JIANG Fan (Institute of Engineering Thermophysics, Chinese Academy of Sciences, Haidian District, Beijing 100080,China) ABSTRACT: The method based on fussy immune network 0 引言 algorithm was presented, by which the flame digital images can 电站锅炉燃烧的基本要求是建立并维持稳定 be classified better. The algorithm involves two parts, immune 均匀的燃烧火焰 燃烧火焰是表征燃烧状态稳定与 network algorithm and fussy clustering algorithm. Fussy immune network algorithm is unsupervised algorithm and it need not to 否最直接的反映 燃烧不稳定不仅会降低锅炉热效 learn the transcendental knowledge of image classification. 率 产生噪声 增加污染物排放 而且在极端的情 Furthermore, dynamic clustering can be carried out along with 况下可能引起炉膛灭火 甚至引起炉膛爆炸 为提 the movement of combustion status by the algorithm. The 高燃烧效率及预防潜在的危险 必须对火焰进行切 method proposed overcomes the weakness of traditional fuzzy 实有效的监测 clustering that must determine the class number in advance. The 火焰具有丰富的信息 如几何参数(火焰位置 eigenvectors of flame image, which obtained from plant trial, has 尺寸) 亮度参数 温度场分布 闪烁频率等 当前 been extracted by digital image technology and classified, and