Department of Electrical and Computer Engineering, Isfahan University of Technology, Isfahan, Iran.
Department of Electrical Engineering, Qatar University, Doha, Qatar.
ISA Trans. 2019 Apr;87:129-142. doi: 10.1016/j.isatra.2018.11.017. Epub 2018 Nov 27.
In this paper, the problem of distributed simultaneous fault detection and leader-following consensus control (SFDLCC) in a multi-agent network is investigated. In the proposed method, instead of designing two separate units for fault detection and control objectives, a single module is used that conducts both tasks. Based on the extended linear matrix inequality (LMI) technique, an SFDLCC module is designed for each agent which utilizes both the data received from the neighboring agents as well as the available local relative measurements. The SFDLCC unit in each agent generates both the control input and the residual signal such that the effect of the unknown inputs including faults, disturbances and noises on the tracking error and the effect of disturbances and noises on the residual signal are attenuated using finite frequency H performance index. On the other hand, the effect of fault inputs on the residual signals is enhanced using the finite frequency H performance index. Moreover, in the proposed algorithm, the group can not only isolate the faulty agent but also determine whether the fault is in the sensors or actuators. Finally, to illustrate the effectiveness of the proposed methodology, a simulation study is provided.
本文研究了多智能体网络中的分布式同时故障检测和领导者跟随一致性控制（SFDLCC）问题。在提出的方法中，不是为故障检测和控制目标设计两个单独的单元，而是使用一个单一的模块来执行这两个任务。基于扩展的线性矩阵不等式（LMI）技术，为每个智能体设计了一个 SFDLCC 模块，该模块利用从邻居智能体接收到的数据以及可用的本地相对测量值。每个智能体中的 SFDLCC 单元生成控制输入和残差信号，使得使用有限频率 H 性能指标来衰减未知输入（包括故障、干扰和噪声）对跟踪误差的影响，以及干扰和噪声对残差信号的影响。另一方面，使用有限频率 H 性能指标来增强故障输入对残差信号的影响。此外，在提出的算法中，组不仅可以隔离有故障的智能体，还可以确定故障是在传感器还是执行器中。最后，为了说明所提出方法的有效性，提供了一个仿真研究。
