Anomaly detection for resilient control systems using fuzzy-neural data fusion engine

摘要: Resilient control systems in critical infrastructures require increased cyber-security and state-awareness. One of the necessary conditions for achieving desired high level resiliency is timely reporting understanding status behavioral trends system. This paper describes design development a fuzzy-neural data fusion system state-awareness resilient systems. The proposed consists dedicated engine each component Each implements three-layered alarm consisting of: 1) conventional threshold-based alarms, 2) anomalous behavior detector using self-organizing maps, 3) prediction error based alarms neural network signal forecasting. was integrated with model Idaho National Laboratory Hytest facility, which testing facility hybrid energy Experimental results demonstrate that implemented provides plant performance monitoring cyber-state reporting.