Research On Air Preheater Hot Spots Detection System Based On Image Of Temperature Field

This thesis is mainly concerned with mechanism of air preheater after-burning and air preheater hot spots detection technique based on image of temperature field.Air preheater is a kind of heat exchanger in power plant units. After-burning might be caused by inadequate combustion of fuel or badly heat-dispersed condition aroused by low air or gas velocity after boiler outage. At present, thermocouple or infrared thermometer are usually used as temperature measurement instrument in most air preheater hot spots detection systems. After-burning is alarmed or not is judged by the result of temperature measure value compared to the alarm threshold value set in advanced. However, when fire is brought out into air preheater, temperature degree covers a wide range .So, it is hard to set an appropriate value to detect all the hot spots at any time and much more experience and working conditions is depended on, therefore, sometimes warnings are lost or it makes a mistake to alarm. On the other hand, this kind of method need the emendation of specific air preheater which depresses the generality of the hot spots detection system.In our new system, the image of temperature field on preheater rotor is obtained by full-radiation infrared thermometry. After the image is pretreated, the wavelet energy feature is extracted from it and then being chosen from together with other features such as gray features and invariable squares according to the Fisher ratio and coefficient of cross-correlation. Then the image is classified by Sammon's best discriminant plane based on Fisher criterion so that the different operational states of air preheater can be identified. Experiments show that the method has some reference value in identifying abnormal operational state of air preheater.