Using neural networks for detection of anomalous traffic in automation networks

Opening of local communication means of technological devices towards networks available to public, supervision of devices, and remote technological devices administration are the characteristics of modern automation. As a result of this process the intrusion of unwanted elements from the Internet to control networks is seen. Therefore, in communication and control networks we have to build in active means to ensure the access to individual technological process components.
The contribution is focused on the insurance of control systems data communication via neural networks technologies in connection with classical methods used in expert systems. The solution proposed defines the way of data elements identification in transfer network, solves the transformation of their parameters for neural network input and defines the type and architecture of a suitable neural network. This is supported by the experiments with various architecture types and neural networks activation functions and followed by subsequent real environment tests. A functional system proposal with possible practical application is the result.