N.V. Sukhanova, S.A. Sheptunov
The automation control systems software tools in mechanical engineering are distinguished by the large amount of data and complex algorithms for their processing. The purpose of the study is to automate testing of software tools. Research methods - modeling, artificial neural networks. The new neural network models for testing automation control systems are
proposed. Models are realized on the basis of the switching neural networks. The switching structure of neural networks allows you to modify, supplement, complicate tests, set and change the sequence of their execution.
Keywords: testing, software, switching structure, artificial neural networks, neural network model.