Fault Diagnosis of Rotary Machines Using Deep Convolutional Neural Network with Wide Three Axis Vibration Signal Input.

Davor Kolar,Danijel Pavković,Michał Pająk,Dragutin Lisjak

doi:10.3390/s20144017

Abstract

Fault diagnosis is considered as an essential task in rotary machinery as possibility of an early detection and diagnosis of the faulty condition can save both time and money. This work presents developed and novel technique for deep-learning-based data-driven fault diagnosis for rotary machinery. The proposed technique input raw three axes accelerometer signal as high definition 1D image into deep learning layers which automatically extract signal features, enabling high classification accuracy. Unlike the researches carried out by other researchers, accelerometer data matrix with dimensions 6400 × 1 × 3 is used as input for convolutional neural network training. Since convolutional neural networks can recognize patterns across input matrix, it is expected that wide input matrix containing vibration data should yield good classification performance. Using convolutional neural networks (CNN) trained model, classification in one of the four classes can be performed. Additionally, number of kernels of CNN is optimized using grid search, as preliminary studies show that alternating number of kernels impacts classification results. This study accomplished the effective classification of different rotary machinery states using convolutional artificial neural network for classification of raw three axis accelerometer signal input.

Highlights

Rotating machines in general consist of three major parts—a rotor, rolling or journal bearings, and a foundation
This study accomplished the effective classification of different rotary machinery states using convolutional artificial neural network for classification of raw three axis accelerometer signal input
It can be concluded that early fault detection is important, which emphasizes the necessity of maintenance in manufacturing operations

Summary

Introduction

Rotating machines in general consist of three major parts—a rotor, rolling or journal bearings (fluid or anti-friction bearings), and a foundation. A failure in rotating machinery results in the loss of productivity and in the delayed delivery of goods and services to customers and may even lead to safety, economic, and environmental problems. Both studies and experience show that faults develop and occur in rotating machines during normal operation. This results in a variety of failures, ending up in reduced availability of equipment and higher operating costs. It can be concluded that early fault detection is important, which emphasizes the necessity of maintenance in manufacturing operations. By looking at rotating machinery as a profit center that produces profit only when it is running, it can be concluded that by using modern condition-based maintenance strategy additional net revenue can be generated [1,2,3,4]

Methods

Results

Conclusion