Toward resilient operation of large-scale systems

大型系统代表了社会中许多关键的工程基础设施系统，例如电网和供水网络。它们包括用于调节其操作的设备，例如控制器、传感器和致动器，其中设备之间的信息通常使用通信网络传输。如果设备随机无法正常运行，或者如果传输的信息被恶意攻击者操纵，则大型系统的运行将受到危害，可能导致社会上的负面后果。当这种异常发生时，需要在监测单元上有效显示警报消息和建议的动作，以供操作人员解决异常并防止进一步的后果。该研究旨在为大型系统的运行开发理论和计算设计方法：（i）通过研究检测和隔离故障和网络攻击的方法，提高对设备故障和通信网络攻击的适应能力，以及（ii）更有效地在监控单元上显示信息，以更好地处理此类异常情况，通过研究方法来建模和预测人类操作员对警报消息的响应方式。该研究还旨在使用模拟电力和供水系统操作的实验测试台系统来验证这些故障诊断，网络安全和警报管理方法。

Large-scale systems represent many critical engineering infrastructure systems in society, such as the power grid and water distribution networks. They include devices used in regulating their operation, such as controllers, sensors, and actuators, where information between the devices are commonly transferred using communication networks. If devices randomly fail to operator properly or if the transferred information is manipulated by a malicious attacker, the operation of large-scale systems is jeopardized, possibly leading to negative consequences in society. When such abnormalities occur, effective display of alarm messages and recommended actions on monitoring units is needed for human operators to resolve abnormalities and prevent further consequences. The research aims to develop theoretical and computational design approaches for the operation of large-scale systems to be: (i) more resilience to device faults and communication cyber attacks, by investigating methods for detection and isolation of faults and cyber attacks, as well as (ii) more effective in displaying information on monitoring units for better handling of such abnormalities, by investigating methods to model and predict the way human operators respond to alarm messages. The research also aims to validate these fault diagnosis, cyber security, and alarm management approaches using experimental testbed systems that mimic the operation of power and water systems.