Bearing Fault Diagnosis Using Multidomain Fusion-Based Vibration Imaging and Multitask Learning

Abstract

Statistical features extraction from bearing fault signals requires a substantial level of knowledge and domain expertise. Furthermore, existing feature extraction techniques are mostly confined to selective feature extraction methods namely, time-domain, frequency-domain, or time-frequency domain statistical parameters. Vibration signals of bearing fault are highly non-linear and non-stationary making it cumbersome to extract relevant information for existing methodologies. This process even became more complicated when the bearing operates at variable speeds and load conditions. To address these challenges, this study develops an autonomous diagnostic system that combines signal-to-image transformation techniques for multi-domain information with convolutional neural network (CNN)-aided multitask learning (MTL). To address variable operating conditions, a composite color image is created by fusing information from multi-domains, such as the raw time-domain signal, the spectrum of the time-domain signal, and the envelope spectrum of the time-frequency analysis. This 2-D composite image, named multi-domain fusion-based vibration imaging (MDFVI), is highly effective in generating a unique pattern even with variable speeds and loads. Following that, these MDFVI images are fed to the proposed MTL-based CNN architecture to identify faults in variable speed and health conditions concurrently. The proposed method is tested on two benchmark datasets from the bearing experiment. The experimental results suggested that the proposed method outperformed state-of-the-arts in both datasets.

Multi-Stream Convolutional Neural Networks for Rotating Machinery Fault Diagnosis under Noise and Trend Items

“…In order to test the improvement of information fusion on diagnostic accuracy and the robustness of the proposed model to interference, we first test the proposed model on five datasets with noise plus trend items. On the same datasets with the same noise and trend items, the proposed model is compared with the single-input fault diagnosis models in Section 3 and the methods from sources [ 25 , 41 ]. We denote the best performance, in Section 3 , with different input types and input sizes as the baseline.…”

“…The one-dimensional (1D) residual network proposed by [ 25 ] is chosen to be the comparison object, and its input is the 1D vibration signal without any process. Hasan et al [ 41 ] proposed a multidomain input type, where the three RGB channels of the input image are the time domain, the frequency domain, and the inclusive grayscale image, respectively. It is also compared with our proposed method using the same training set proportions and evaluation criteria as this experiment.…”

“…There are various reasons for trend items, which may be the influence of the signal acquisition system by temperature, humidity, electromagnetic field, etc., or the basic motion of the machine [ 33 , 39 , 40 , 41 ]. To simulate the diversity of trend items, we chose a series of basis functions.…”

A Hybrid Matching Network for Fault Diagnosis under Different Working Conditions with Limited Data

“…Qiao et al [22] built a dualinput model and achieved satisfactory antinoise and load adaptability based on a CNN and a long short-term memory neural network. e deep learning methods have discarded the traditional time-consuming and unreliable manual analysis, improving the e ciency of fault diagnosis [23][24][25][26][27][28] considerably.…”

Deep Unsupervised Domain Adaptation with Time Series Sensor Data: A Survey

“…Sensors capture physical parameters from the observed environment and convert them into observable electrical pulses [ 1 , 2 , 3 , 4 ]. A wide variety of sensors are used in a wide range of applications in manufacturing and machinery [ 5 , 6 , 7 ], transportation [ 8 , 9 , 10 , 11 , 12 , 13 ], healthcare [ 14 , 15 , 16 , 17 , 18 ] and many other aspects of our daily lives. In the field of mechanical engineering, for example, accelerometer sensors placed around gearboxes or bearings can capture the vibration signals of a machine and predict possible impending failures [ 7 ].…”