Instruments and Systems: Monitoring, Control, and Diagnostics Annotation << Back Forecasting the reliability of technological equipment on the basis of identification of its conditions O.M. PROTALINSKY, I.A. SHCHERBATOV, I.O. PROTALINSKY In work using the technology of artificial neural networks implemented as a multi-layer perceptron of direct signal propagation, a model is constructed for predicting the reliability of technological equipment of power industry facilities. The features of a neural network structure choice in the form of a multilayer perceptron are shown. An experimental study of the proposed approach on a methodical example is carried out. The synthesized approach is implemented in the form of an appropriate structural scheme, which is used in the software of the EAM Optima information system. Keywords: reliability, identification of states, forecast, defect, failure, neural network, Optima, EAM. DOI: 10.25791/pribor.08.2018.136 Contacts: E-mail: protalinskiy@gmail.com Pp. 39-45.