Vibration-based and computer vision-aided nondestructive health condition evaluation of rail track structures

Abstract

In railway engineering, monitoring the health condition of rail track structures is crucial to prevent abnormal vibration issues of the wheel–rail system. To address the problem of low efficiency of traditional nondestructive testing methods, this work investigates the feasibility of the computer vision-aided health condition monitoring approach for track structures based on vibration signals. The proposed method eliminates the tedious and complicated data pre-processing including signal mapping and noise reduction, which can achieve robust signal description using numerous redundant features. First, the method converts the raw wheel–rail vibration signals directly into two-dimensional grayscale images, followed by image feature extraction using the FAST-Unoriented-SIFT algorithm. Subsequently, Visual Bag-of-Words (VBoW) model is established based on the image features, where the optimal parameter selection analysis is implemented based on fourfold cross-validation by considering both recognition accuracy and stability. Finally, the Euclidean distance between word frequency vectors of testing set and the codebook vectors of training set is compared to recognize the health condition of track structures. For the three health conditions of track structures analyzed in this paper, the overall recognition rate could reach 96.7%. The results demonstrate that the proposed method performs higher recognition accuracy and lower bias with strong time-varying and random vibration signals, which has promising application prospect in early-stage structural defect detection.

Review of Data Processing Methods Used in Predictive Maintenance for Next Generation Heavy Machinery

“…Vibration signals have been found to transfer useful information about gearbox working states, such as speed and torque, and they were found to be useful for condition monitoring. Wang et al [13] present a new vibration signal-based method that uses computer vision to monitor rail track infrastructure's non-destructive health. This approach takes raw vibration signals and turns them into grayscale images directly, unlike tradi-tional approaches requiring noise elimination and feature reduction.…”

“…They evaluated how well three different approaches-neural network-based, support vector machine-based, and extreme gradient boosting-performed at detecting faults in the mill's recorded vibration data. Three different feature extraction techniques-entropy features, linear discriminant analysis, and artificial neural network, were combined by Wang et al [13]. Combining these features can improve the quality of feature representation by gathering additional information with respect to discriminative power, nonlinear relationships and signal complexity.…”

“…Machine learning and deep learning algorithms provide efficient methods for fault identification in vibration signals. In machine learning, features are usually extracted [13,28,31,36] from the signals and then machine learning classifiers like Random Forests [8,[37][38][39], Support Vector Machines (SVMs) [1,31,40,41] are applied. These classifiers are trained using extracted features, to differentiate between healthy and faulty signals.…”

An In-Depth Study of Vibration Sensors for Condition Monitoring

“…Several papers in our review collected vibration data from bearings or bearing-like components. Wang et al [89] acquired vibration data from wheel rails, which have qualities similar to bearings as rotating components. Although some papers did not specify the exact bearing type, some did.…”