Cyber-Resilient Controllers for Smart Distribution Grid Control Layers

This paper presents a novel cyber-resilient control strategy for enhancing the operational security of future smart distribution systems (SDSs) against compromised control setpoints originating from higher-level controllers. The proposed framework addresses the structure, control architecture, and cyber vulnerabilities of SDSs by embedding an anomaly detection and autonomous response mechanism within each control layer. An artificial neural network (ANN)-based detector is employed to identify non-implementable or malicious control commands based on local measurements and grid location data. Upon detecting a cyber anomaly, the controller avoids disconnection and enables droop-based autonomous operation, ensuring continued grid support. The proposed strategy was validated using MATLAB/Simulink R2022a under various dynamic test scenarios, demonstrating its ability to maintain system stability. Unlike prior studies that rely on offline anomaly detection, this study presents a real-time capable closed-loop control solution that detects anomalies during simulation runtime. The proposed method rejects erroneous commands arising from both cyber intrusions and human errors, thereby enhancing the cyber-resilience and reliability of SDS operations.