摘要
提出一种基于观测器的在线动态系统故障检测快速生成方案,利用NetCon系统实现控制器与故障检测系统运行在同一个智能节点上,使得残差的生成与系统的性能联系起来,实现不仅能监控系统的运行情况,还能完成故障检测的功能,同时能够避免闭环控制对故障检测的影响。针对一个实际的伺服电机,是进行离线辨识获得电机的模型;设计一个输出估计器——动态观测器,生成残差信号;通过离线仿真和在线实验证实该方案的有效性,完成从理论到实验仿真,再到实际运用的系统开发过程。
A rapid formation scheme on observer-based on-line fault detection for dynamic system was proposed. With the NetCon system, the controller and fault detection system operated at the same intelligent node. A connection between the generation of residual and the performance of the system was established. Not only the operation could be monitored, but also the fault detection function could be achieved. The scheme could avoid the effect of the closed-loop control. As to an actual servo motor, firstly, the model was obtained by the identification; secondly, an output estimator （an dynamic observer） was designed; finally, the feasibility was tested by both off-line and on-line real-time simulation. Thus, the developing press of a fault detection system was achieved from the theoretic research to simulation in laboratory, and then to practicail applications.
出处
《系统仿真学报》
EI
CAS
CSCD
北大核心
2007年第1期226-229,共4页
Journal of System Simulation
基金
中科院海外杰出学者基金项目(2003-1-10)
