化工过程工业控制系统的网络攻击路径识别与风险分析

打开文本图片集

中图分类号：TQ056 文献标志码：A DOI：10.3969/j.issn.1003-9015.2025.04.009

Abstract：To identify and quantifythe potential impacts ofcyberatack on industrial operational systems，aProcess Bayesian Network（PBN）model was proposed，integrating Process Hazard Impact Analysis （PHIA）and Bayesian Network（BN）theory.Themodel first identified key risk factorsand assessed the potential impacts of cyberattack on the process system，then incorporated the Leaky Noisy-OR（LNOR） logic to establish a probabilistic hybrid quantification framework. Combined with an optimized Expectation Maximum（EM）algorithm，the model addressed parameter uncertainty in traditional methods.Applied to a Continuous Stirred Tank Heater（CSTH）system，the PBN model updated fault probabilities under atack senarios and evaluated performance using Path Coverage Rate（PRC）.Results show that the highest risk propagation path occurs under combined attack，with PRC significantlyimproving compared to traditional attck tree models，enhancing network atack path identificationand risk assessment for optimal prevention strategies.

Key words： cyberattack；process hazard and impact analysis；bayesian network；risk analysis

1前言

随着石化企业数字化转型和工业4.0的到来，工业自动化控制系统（IACS）诸如生产调度与优化、自动控制系统、联锁保护系统等被广泛应用于化工厂。（剩余9856字）