RL-BAGS: A Tool for Smart Grid Risk Assessment

The security of critical infrastructure such as Smart Grid is of significant concern because cyber-physical attacks are becoming a frequent occurrence. Cybercriminals compromise cyberinfrastructure to control physical processes maliciously. It is the system administrator's goal to find vulnerabilities in the smart grid functions and patch them before they are compromised. Unfortunately, limited resources and a large attack surface make it difficult to decide which function to protect in a particular system state. In this research paper, we tackle the problem of resource allocation in the smart grid system by proposing a tool, Reinforcement Learning-Bayesian Attack Graph for Smart Grid System (RL-BAGS), which provides functionality to the system engineers to compute optimal policies on regular intervals about whether to SCAN or PATCH a particular function of the smart grid system. RL-BAGS considers functions and network architecture of the system to generate a Bayesian Network, which represents the state of the system. RL-BAGS implements two reinforcement learning algorithms, Q-Learning and SARSA learning, on the generated Bayesian Network to learn optimal policies. RL-BAGS assists system administrators performing in-depth studies of one of the functions of the smart grid system advising effective actions to scan or patch a system component.