基于分布式结构的判决反馈数据融合算法

New Decision Feedback Data Fusion Algorithm in Distributed Fault Detection System

提出了一种新的分布式结构下的故障检测算法·在该算法中 ,不仅用于故障检测的多传感器数据被传送到数据融合中心 ,而且上一时刻的判决输出也反馈到融合中心·从理论上分析了应用此算法时全局判决概率的动态收敛性能 ,并证明了在平稳环境中 ,融合中心DFC(datafusioncenter)的判决以概率 1收敛·该算法处理动态问题优于传统的无反馈的融合算法·

A new data fusion algorithm applied to distributed fault detecting system was deduced. Along with the local decisions, the DFC(data fusion center) remembers and uses its most recent decision in synthesizing each new decision through the new suggested algorithm decision feedback data fusion algorithm. In a stationary environment, the suggested algorithm converges to a steady state decision in finite time with probability one, and its detection performance during convergence and in steady state is strictly determined. The given method is better in dynamic problem than traditional method.