Industrial control system network virus detection system and method

The invention relates to an industrial control system network virus detection system, the system comprises a collector and a processor, the collector extracts information from a real virus feature library, and the processor carries out detection according to the information extracted from the real virus feature library according to the following steps: using a generator model of a generative adversarial neural network, generating a large number of virtual virus feature libraries based on a very small number of real virus library features; inputting the real virus library features and the virtual virus library features into an adversarial neural network discriminator model for adversarial training, and performing iterative updating on the virtual virus feature library; when the discriminator model cannot recognize the virtual data, inputting the discriminated feature data into the AI recognition model for recognition training; and deploying the finally trained AI identification model to an industrial control s