Abstract Cyber-Physical Systems (CPSs) have become an inseparable part of the new industrial era referred to as Industry 4.0 the quintessential example of which are smart grids. With a cyber communication level integrated with the physical components, besides considerable merits came along some reliability and security concerns. To address these issues, there ought to be a supervisory level monitoring and control system to identify any faults within the system and balance out their potential damaging impacts. In this paper, we consider fault tolerant control (FTC) problem in one of the building blocks of smart grids i.e. microgrids. The FTC method is based on online recursive reduced-order model estimation and Klaman Filter (KF) residual generation. Then, two-sided cumulative sum (CUSUM) sequential change detection algorithm is employed for sensor fault or communication link failure detection. Lastly, a fault-tolerant logic is constructed to switch between actual data and the internal model for the control system.

中文翻译：

---

传感器和通信链路故障下孤岛交流微电网的容错控制使用在线递归降阶估计

摘要 信息物理系统 (CPS) 已成为被称为工业 4.0 的新工业时代不可分割的一部分，其中典型的例子是智能电网。通过与物理组件集成的网络通信级别，除了相当大的优点外，还带来了一些可靠性和安全性问题。为了解决这些问题，应该有一个监督级别的监控系统来识别系统内的任何故障并平衡其潜在的破坏性影响。在本文中，我们考虑了智能电网的构建块之一，即微电网中的容错控制 (FTC) 问题。FTC 方法基于在线递归降阶模型估计和克拉曼滤波器 (KF) 残差生成。然后，两侧累积和（CUSUM）顺序变化检测算法用于传感器故障或通信链路故障检测。最后，构建容错逻辑以在实际数据和控制系统的内部模型之间切换。