Cloud Model for Security State Recognition Based on Factor Space

With the integration of informatization and industrialization, the oil-gas industry control system is currently confronting huge security risks. To more effectively assess and predict the security state of industry control systems, this study proposed a cloud model based on factor space for security state recognition. First, the malicious behavior factors of oil-gas industry control system were obtained and were described in the factor state space. Next, the expectation, entropy and hyper entropy of each factor were constructed to convert the fuzzy concepts in the cloud model into quantitative value. Finally, based on the forward and reverse generators, single-condition, multi-condition and multi-rule cloud generators were applied to rule-based reasoning model to realize the transformation from expected value to qualitative value. Experimental simulation demonstrates that cloud-reasoning model based on the factor space can predict the impact of unknown malicious program behavior on the security state and achieve a better evaluation effect. Moreover, this study can provide a new approach to the security state recognition of industrial control systems.