大型复杂系统的动态SDG模型及传感器布置问题

符号有向图(SDG: signed directed graph)可以用于描述大型复杂系统及其变量之间的因果影响关系, 但是在描述故障传递关系时不能表示其动态传播规律. 本文在SDG支路上引入时间参数, 用于近似描述变量的变化在系统中的传递时间, 由此构建的SDG模型称为动态SDG模型. 另外, 要实现故障检测需要有传感器的信息, 传感器的布置直接影响着故障检测的性能. 本文在动态SDG框架下, 研究了故障的可检测性和可分辨性问题, 提出了一些一般性的结论, 并给出通过正向推理来求故障传播过程和传感器布置方案的方法. 最后, 通过实例验证该方法的有效性.

Dynamic SDG model and sensor location problem for large-scale complex systems

Signed directed graph (SDG) is used to describe large-scale complex systems and the cause-effect relations among variables. However, SDG cannot show the dynamic propagation patterns when expressing the fault propagation relation. In this paper, time parameters are introduced to the branches of SDG to approximately denote the propagation time of the variable changes in the systems. The SDG model constructed in this way is called the dynamic SDG model. Besides, fault detection is conducted by the use of sensors and is based on the information of sensors, thus the sensor location influences the performances of fault detection directly. The fault detectability and identifiability are then analyzed in the framework of dynamic SDG. A method is proposed to describe the fault propagation process by forward inference and to locate the sensors. Finally, an example is used to illustrate and validate the proposed method.