Neural Network Models of Process Equipment in a Monitoring and Predictive Analytics System
摘要
A neural network surrogate model of a gas turbine engine (GTE) has been developed, which approximates a more complex physico-mathematical model. The results generated by the model are demonstrated. A method for assessing the technical condition of an object is proposed, which is based on back-propagation in the artificial neural network. The main use cases are described and conclusions are made about the potential advantages of neural network surrogate models.