Intelligent Fault Detection of Rotating Machinery Using Long-Short-Term Memory (LSTM) Network

Abstract

AbstractArtificial intelligence is one of the fastest-growing fields across the board. In every corner of the globe, researchers are attempting to harness its promise. Artificial intelligence’s capabilities began to be harnessed across all industries with the onset of the fourth industrial revolution. All smart industries follow the deployment of predictive maintenance with the assistance of AI. With the use of deep learning, a subset of artificial intelligence, this article describes a method for diagnosing defects in rotating machinery. The long-short-term memory framework, a class of recurrent neural network, is used to classify the faults of a rotating machine element. The experiment uses vibration data collected from rolling element bearings under various fault circumstances. The findings indicate that the LSTM network is a promising method for spotting faults in rotating machine parts such gears, rolling element bearings, shafts, rotors, and so on.KeywordsRotating machineryRolling element bearingsFault diagnosisLong-short-term memoryDeep learningArtificial intelligenceRecurrent neural network