David He

Deep Learning Based Approach for Bearing Fault Diagnosis

Bearing is one of the most critical components in most electrical and power drives. Effective bearing fault diagnosis is important for keeping the electrical and power drives safe and operating normally. In the age of Internet of Things and Industrial 4.0, massive real-time data are collected from bearing health monitoring systems. Mechanical big data have the characteristics of large volume, diversity, and high velocity. There are two major problems in using the existing methods for bearing fault diagnosis with big data. The features are manually extracted relying on much prior knowledge about signal processing techniques and diagnostic expertise, and the used models have shallow architectures, limiting their capability in fault diagnosis. Effectively mining features from big data and accurately identifying the bearing health conditions with new advanced methods have become new issues. This paper presents a deep learning-based approach for bearing fault diagnosis. The presented approach preprocesses sensor signals using short-time Fourier transform (STFT). Based on a simple spectrum matrix obtained by STFT, an optimized deep learning structure, large memory storage retrieval (LAMSTAR) neural network, is built to diagnose the bearing faults. Acoustic emission signals acquired from a bearing test rig are used to validate the presented method. The validation results show the accurate classification performance on various bearing faults under different working conditions. The performance of the presented method is also compared with other effective bearing fault diagnosis methods reported in the literature. The comparison results have shown that the presented method gives much better diagnostic performance, even at relatively low rotating speeds.

Using Deep Learning-Based Approach to Predict Remaining Useful Life of Rotating Components

In the age of Internet of Things and Industrial 4.0, prognostic and health management (PHM) systems are used to collect massive real-time data from mechanical equipment. PHM big data has the characteristics of large-volume, diversity, and high-velocity. Effectively mining features from such data and accurately predicting the remaining useful life (RUL) of the rotating components with new advanced methods become issues in PHM. Traditional data driven prognostics is based on shallow learning architectures, requires establishing explicit model equations and much prior knowledge about signal processing techniques and prognostic expertise, and therefore is limited in the age of big data. This paper presents a deep learning-based approach for RUL prediction of rotating components with big data. The presented approach is tested and validated using data collected from a gear test rig and bearing run-to-failure tests and compared with existing PHM methods. The test results show the promising RUL prediction performance of the deep learning-based approach.

Experimental study of dynamic strain for gear tooth using fiber Bragg gratings and piezoelectric strain sensors

It has always been a critical task to understand gear dynamics for gear design and condition monitoring. Many gear models have been proposed to simulate gear meshing dynamics. However, most of the theoretical models are based on simplified gear structure and may contain approximation errors. Direct measuring of gear strain is important for gear design validation, load analysis, reliability assessment, gear condition monitoring, etc. Most of the existing studies of tooth strain measurements are performed under static load condition. In this paper, we investigate new measuring techniques using fiber Bragg grating sensor and piezoelectric strain for gear dynamic strain measurement. We conduct gear dynamic strain measurement under both normal and pitted conditions to evaluate the strain transition process and pitting effect. The experiments are performed on an industrial gearbox with relatively small module gears. Multiple combinations of speed and load conditions are tested and the results are discussed and analyzed. We analyze multiple factors that affect the tooth root stress, including speed, load, extended tooth meshing, etc. It is found that under low operation speed range, the tooth root strain is mainly determined by the torque, while in the medium to high speed range, the tooth root strain is jointly affected by speed and torque. Extended tooth contact is shown in the measurement results with strong evidence. It conforms to earlier founding that the transmission error and dynamic load factor are overestimated for spur gear under heavy load. We also evaluate the change in dynamic strain caused by pitted tooth surface. It is shown that pitting faults lead to decreased bending strain on the tooth, especially in single-tooth contact zone, which represents a loss in torque and possibly reduced mesh stiffness. Numerical simulations are also provided to make comparisons and help to interpret the experimental results.