Implementação de framework em software livre com OpenDSS para simulação de ataques ciberfísicos em redes elétricas com recursos energéticos distribuídos

Abstract

The increasing digitalization of electrical distribution networks has expanded the integration of distributed energy resources and intelligent control devices, making these systems more efficient but vulnerable to cyber-physical attacks. Although previous studies have addressed isolated attacks or general security assessment methodologies, limitations remain regarding an in-depth analysis of cyberattacks applied specifically to distribution networks. In this context, this work aims to model, implement, and analyze different types of cyber-physical attacks in a distribution network, evaluating their effects on the operation of an energy storage element. The adopted methodology involves the computational modeling of the network in OpenDSS integrated with external routines in {Python} combined with the {py-dss-interface} library, enabling controlled attack insertion, data collection, and monitoring of the variables of interest under normal and adversarial conditions. The experiments consider scenarios with isolated and combined attacks, temporal variation of the injected signals, and multiple degrees of corruption applied to the sensors and actuators of the storage element. The results show significant distortions in control signals caused by the cyber-physical attacks, including deviations exceeding 10% in voltage levels and in the battery’s state of charge, highlighting the need for advanced detection and mitigation mechanisms to ensure secure network operation.