Critical Information Infrastructures Intelligent Protection: Digital Twins and Neural Network-Based Threat Detection Methods

The article discusses digital twins and neural network methods integration for critical information infrastructures integration for critical information infrastructure intelligent protection. An architecture is proposed in which digital twins are used to model attacks and test defense mechanisms, and neural network algorithms (autoencoders, LSTM, CNN) are used to analyze anomalies, predict threats, and optimize response strategies. The developed agent-based modeling allows for vulnerabilities efficient detection and defense measures adaptation in real time. Experimental results are presented demonstrating hybrid approach effectiveness combining digital twins and artificial intelligence.