Enhancing Network Security: Machine Learning Evaluation for Intrusion Detection in Power Load Management System
Bin Li, Jiyuan Zhong, Shuyang Wang, Haoyang Yu, Qiang Guo, Mingyuan Ren, Ke Chen · 2024
In the era of increasingly sophisticated cyber threats, ensuring the security of critical infrastructure such as power load management system is paramount. This paper presents a comprehensive evaluation of machine learning techniques for network intrusion detection within a simulated power load management environment. The dataset used in this study was created to mimic a real-world power load management system, subjected to various attacks. Each TCP/IP connection within the dataset is characterized by 41 features, both quantitative and qualitative, and is labeled as either normal or anomalous.We employed three machine learning models including Logistic Regression, Random Forest, and K-Nearest Neighbors— to analyze the dataset. By comparing the performance of these models in terms of accuracy, precision, recall, and F1 score, we identified the most effective approach for detecting anomalies in this context. Our findings provide a robust framework for enhancing network security in power load management system. This study contributes to the development of more resilient and secure power management infrastructures by leveraging advanced machine learning techniques.