Distributed Fusion Algorithms in Embedded Network On-line Fusion System

In order to solve effectively the issue of remote online detection and fault diagnosis of hydroelectric generating units in FengMan hydropower plant, a kind integrated project of combining embedded network fusion with multi-sensor consistency information fusion algorithm and fuzzy comprehensive assessment algorithm is proposed based on distributed and embedded architecture. Multi-sensor consistency information fusion algorithm is adopted in local fusion center. Fuzzy comprehensive assessment algorithm is introduced into traditional decision fusion technology under the “soft” decision architecture. The process of fusion is composed of the comprehensive operation and the global decision through fusing the local decision of multiple sensors for obtaining the global decision of the concerned object in global fusion center. These algorithms have been successfully applied in fault detection and diagnosis of hydroelectricity digital simulation system and hydropower plant of Jilin Fengman. Tests on experimental device show that the proposed algorithms are effective and can be applied widely.