Impact of the Convolutional Neural Network Structure and Training Parameters on the Effectiveness of the Diagnostic Systems of Modern AC Motor Drives

Currently, AC motors are a key element of industrial and commercial drive systems. During normal operation, the machines may become damaged, which pose threat to users. Therefore, it is important develop fault detection method that allows for at an early stage. Among currently used diagnostic systems, applications based on deep neural structures dynamically developed. Despite many examples learning methods, there no formal rules selecting network structure parameters training process. Such methods would make possible shorten implementation process networks in systems machines. The article presents detailed analysis influence convolutional hyperparameters procedures precision interturn short-circuits system. studies take into account direct phase currents through induction permanent magnet synchronous motors. research results presented extension authors’ previous research.