Application of Multi-sensor Data Fusion in Hot Spot Detection for Air Preheater

Abstract

In power station, in order to prevent the air preheater on fire or control the fire damage in time, it is necessary to early detect air preheater ignition precursor, that is, the phenomenon of abnormal internal heat, and timely warning, prevent the fire from the fire. It is also the main target of the development of air preheater hot spot detection system. To improve environmental adaptability and anti-interference ability of the hot spot detection system, and reduce the false alarm rate, in this paper, multi sensor data fusion technology is used in the hot spot detection, that is to say, multi sensor data is analyzed and optimized, integrated to complete the decision. As one of the main methods of information fusion, D-S evidence theory is able to integrate the information provided by multiple sensors, so as to reduce the uncertainty of information. In D-S evidence theory, the synthetic method of conflict evidence is the key to solve the problem of evidence conflict. Therefore, this paper adopts the method of conflict evidence synthesis based on the similarity coefficient of evidence, so as to improve probability of the fire alarm.