STUDY ON RECOGNITION OF COMBUSTION CONDITION BASED ON ENERGY CHARACTERISTICS OF PRESSURE SIGNALS AND SVB CLASSIFIE
Wavelet packets analysis method is used to implement energy decomposing of boiler pressure signals in eight frequency bands. And the energies are used as the characteristics vector of combustion condition. Then SVM is designed to serve as a classifier for distinguishing abnormal combustion conditions of boiler. Experiments are carried out for choosing appropriate kernel functions and SVM parameters. The results from experiments indicate that using the SVM classifier to identify the abnormal combustion conditions will achieve higher correct rate.
Combustion condition Wavelet Packet Recognition Support Vector Machine
Liqun Han Lianyang He
School of Information Engineering,Beijing Technology and Business University,100037,Beijing
国际会议
2008高等智能国际会议(2008 International Conference on Advanced Intelligence)
北京
英文
2008-10-18（万方平台首次上网日期，不代表论文的发表时间）