CONVOLUTIONAL AND LONG SHORT-TERM MEMORY NEURAL NETWORKS BASED MODELS FOR REMAINING USEFUL LIFE PREDICTION

The development of industry leads to the growth of the complexity of equipment used in enterprises. Corrective and preventive maintenance are replaced by predictive maintenance. Predicting the remaining useful life of equipment with high accuracy allows carry out repairs or replacement of equipment in terms maximally near to its failures. It will allow increase the reliability and safety of systems, and reduce maintenance costs. In Industry 4.0 conception the most preferable are approaches based on processing of large amounts of data using machine learning methods. In this study it is proposed deep learning models based on convolutional and long short-term memory neural network which allow to improve prediction accuracy. The high efficiency of proposed models is demonstrated by comparison with other models used in predicting of remaining useful life of aero engines.