Substation Vulnerability Assessment Using Bayesian Network

Security of substations is important to ensure reliability and integrity of the power grid. In this paper, we provide a complete vulnerability assessment framework for a SAS substation. The framework uses CVE database as its foundation. We use an NLP-based model to identify and map the vulnerabilities present in each CVE entry. These mapped vulnerabilities are then used to determine various attack paths in a test network using Bayesian networks. We also compute the exploitation probabilities of the vulnerability nodes and the attack paths in the network. Our study provides an overall vulnerability landscape of the network which can be used to identify weak points and implement remedial security measures to fix them.