Open AccessVibration Anomaly Detection using Deep Neural Network and Convolutional Neural Network
Author(s) -
Gicu Călin Deac,
Gicu Călin Deac,
Radu Constantin Parpală,
Cosmin Popa,
Costel Emil Cotet
Publication year - 2021
Publication title -
international journal of modeling and optimization
Language(s) - English
Resource type - Journals
ISSN - 2010-3697
DOI - 10.7763/ijmo.2021.v11.772
Subject(s) - computer science , convolutional neural network , accelerometer , artificial neural network , deep learning , artificial intelligence , raw data , anomaly detection , vibration , time domain , fault (geology) , pattern recognition (psychology) , series (stratigraphy) , domain (mathematical analysis) , time series , activation function , anomaly (physics) , machine learning , computer vision , seismology , mathematics , geology , acoustics , paleontology , mathematical analysis , physics , programming language , operating system , condensed matter physics
Identifying the “health state” of the equipment is the domain of condition monitoring. The paper proposes a study of two models: DNN (Deep Neural Network) and CNN (Convolutional Neural Network) over an existent dataset provided by Case Western Reserve University for analyzing vibrations in fault diagnosis. After the model is trained on the windowed dataset using an optimal learning rate, minimizing the cost function, and is tested by computing the loss, accuracy and precision across the results, the weights are saved, and the models can be tested on other real data. The trained model recognizes raw time series data collected by micro electro-mechanical accelerometer sensors and detects anomalies based on former times series entries.