Flame stability recognizing of HTAC based on texture and structure analysis

Author

Dong, Zengshou ; Wang, Yun ; Zhuo, Dongfeng ; Li, Meiling

Author_Institution

Eng. Inst. of Electron. Inf., Taiyuan Univ. of Sci. & Technol., Taiyuan

fYear

2008

fDate

25-27 June 2008

Firstpage

5040

Lastpage

5044

Abstract

The relation between the following 6 features and combustion Index was discussed in this paper based on the texture and structure analysis of the flame image of the coal combusting in high temperature air combustion .The figures are: the autocorrelation function average, the information entropy of the co-occurrence matrix in horizontal direction and of the histogram, the pq moment, the feature and the complexity of color.While ldquocombustion indexrdquo was presented based on the discussion and was verified by the improved PSO-AWNN algorithm. It is concluded that the recognition results obtained by the combustion index agree basically with that by PSO-AWNN, which shows that the combustion index can be regarded as a detective model of the flame combustion stability in HTAC.

Keywords

combustion; entropy; flames; high-temperature techniques; neural nets; particle swarm optimisation; wavelet transforms; HTAC; PSO-AWNN algorithm; adaptive wavelet neural network; autocorrelation function average; co-occurrence matrix; combustion index; flame stability; high temperature air combustion; information entropy; particle swarm optimization; structure analysis; texture analysis; Algorithm design and analysis; Autocorrelation; Combustion; Fires; Image analysis; Image color analysis; Image texture analysis; Information analysis; Stability analysis; Temperature; an improved particle swarm optimization based on self-adaptive escape velocity (AEPSO); high temperature air combustion (HTAC); the adaptive wavelet neural network (AWNN); the combustion index(CI); the texture and structural character of image;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Control and Automation, 2008. WCICA 2008. 7th World Congress on

Conference_Location

Chongqing

Print_ISBN

978-1-4244-2113-8

Electronic_ISBN

978-1-4244-2114-5

Type

conf

DOI

10.1109/WCICA.2008.4593747

Filename

4593747