摘要
通过加入环境节点约束方程对多显著误差的同步识别并同步补偿法 (SEGE)进行改进研究。在多显著误差检测中的广义似然比法、同步识别并同步补偿法和顺序识别并同步补偿法等几种方法的原理对比基础上 ,考虑过程外部进料或出料流股中显著误差幅度与位于同一节点的其它显著误差幅度相近 ,容易被其它显著误差抵消的情况 ,利用这几种方法和改进的SEGE法对检测性能进行比较分析。仿真结果表明 ,改进后的SEGE法具有更好的显著误差检测性能。
The simultaneous estimation of gross errors(SEGE) is modified by adding environment node constraint to improve the detection performance.For the case of that the size of gross error of inlet stream or outlet stream are close to the other gross error in the same node,while the node effects are small,several detection methods,such as GLR、SEGE and SICC are introduced to compare the detection performance.Comparative assess of performance is performed on a recycle network and the simulation results show that the modified SEGE is more effective to identify this kind of gross error.
出处
《化工自动化及仪表》
CAS
北大核心
2004年第1期52-55,共4页
Control and Instruments in Chemical Industry
基金
国家杰出青年科学基金资助项目 (60 0 2 5 3 0 8)
国家"十五"科技攻关资助项目 (2 0 0 1BA2 0 4B0 1)
关键词
数据校正
显著误差检测
同步识别并同步补偿
data reconciliation
gross error detection
simultaneous estimation of gross errors
