Search

Large rotating machinery is widely applied in many industrial applications and shaft orbit classification plays an important role in fault diagnosis of large rotating machinery. Therefore, we propose a novel raw data feature-based automatic and deterministic shaft orbit classification framework for large rotating machinery. The main idea of the proposed framework is that it directly uses the most discriminative geometric features of each shaft orbit type for classification. Different from existing supervised learning shaft orbit classification methods, it does not need any labeled data and it is a deterministic, quite simple, light, easy to implement. Extensive experiments of synchronous integrated period sampling and fixed sampling at different rotating speeds conducted on real and simulated datasets have shown the superiority of proposed framework. This study provides a new deterministic interpretable and explainable approach to support predictive maintenance of large rotating machinery in various industrial fields even when no labeled data is available. [Display omitted] • ADSOC is a raw data feature-based classification framework so that it does not need any labeled data. • ADSOC is invariant to shaft orbit translation, scale and rotation. • ADSOC model and results are easy to interpret and explain. • ADSOC is independent from both shaft orbit starting point and rotating direction, and the shaft orbit does not need to be closed. • ADSOC is a deterministic framework and it is quite simple, light and easy to implement. [ABSTRACT FROM AUTHOR]

To achieve planetary gearbox fault classification, vibration signal analysis has been widely employed with rich information about the health status and easy measurement. It is critical to extract features with enough health status information for fault classification. The self-adaptation of ensemble empirical mode decomposition (EEMD) indicates the dependence between the raw vibration signal and EEMD-decomposed intrinsic mode functions (IMFs). In this study, we develop a novel fault feature vector based on the dependence. To develop the dependence-based feature vector, simulated vibration signals with different sun gear tooth crack levels are analyzed. The dependence between the raw signal and each IMF is investigated by Archimedean copulas. With the goodness-of-fit test, the copula estimation closest to the perfect fit is selected for dependence representation. The parameter of the selected copula is applied to develop the dependence-based feature vector. To test the ability of the dependence-based feature vector in fault classification for a real planetary gearbox, experimental vibration signals with different gear fault levels at different gears are classified by a multi-class support vector machine. The classification accuracy of the developed feature vector is compared with that of a reported indicator. Results show the dependence-based feature vector provides higher classification accuracy than the reported, indicating the developed feature vector contains more health status information. The developed feature vector can serve better for planetary gearbox fault classification. [ABSTRACT FROM AUTHOR]

The future of the wind energy industry passes through the use of larger and more flexible wind turbines in remote locations, which are increasingly offshore to benefit stronger and more uniform wind conditions. The cost of operation and maintenance of offshore wind turbines is approximately 15–35% of the total cost. Of this, 80% goes towards unplanned maintenance issues due to different faults in the wind turbine components. Thus, an auspicious way to contribute to the increasing demands and challenges is by applying low-cost advanced fault detection schemes. This work proposes a new method for detection and classification of wind turbine actuators and sensors faults in variable-speed wind turbines. For this purpose, time domain signals acquired from the operating wind turbine are represented as two-dimensional matrices to obtain grayscale digital images. Then, the image pattern recognition is processed getting texture features under a multichannel representation. In this work, four types of texture characteristics are used: statistical, wavelet, granulometric and Gabor features. Next, the most significant ones are selected using the conditional mutual criterion. Finally, the faults are detected and distinguished between them (classified) using an automatic classification tool. In particular, a 10-fold cross-validation is used to obtain a more generalized model and evaluates the classification performance. Coupled non-linear aero-hydro-servo-elastic simulations of a 5 MW offshore type wind turbine are carried out in several fault scenarios. The results show a promising methodology able to detect and classify the most common wind turbine faults. [ABSTRACT FROM AUTHOR]

Different health-monitoring techniques were considered in the literature to enhance the safety and stability of Connected Autonomous Vehicle (CAV) platoons. Mitigating these faults is faster and more reliable if the fault structure is known. In this paper, we propose using transmissibility operators, which are relationships that relate a set of velocities with another in the platoon, to classify the faults. Transmissibility operators were shown to be exceptional in signal estimation. However, it is also shown to be noncausal and thus can only be used offline. To this end, we propose using Data Aggregation (DAgger), an extension to imitation learning, to transfer the classification experience from transmissibility operators to a novice machine learning agent to be used online. The integration of transmissibility-DAgger gives the ability to adapt to new fault classes without the need to re-train the diagnosis model from the beginning. A heterogeneous CAV platoon was modeled with three different faults separately. These faults are actuator disturbances, false data injection attacks, and communication time delays. The classification scheme depends on estimating the faulty signal in the case of each fault class. Next, the measured faulty signal is compared with the three estimations, and the closer fault estimation to the measured one is considered the actual fault on the platoon. The proposed algorithm is then tested on the platoon model and then applied to an experimental setup that consists of three autonomous robots. The proposed results are compared with six different machine learning classification models. The overall classification accuracy achieved was 95.8% for the experiment using the proposed approach. [ABSTRACT FROM AUTHOR]

• We proposed Contrastive-weighted Self-supervised Model (CSM) for fault classification under long-tailed data. • We composited positive and negative samples for contrastive learning and transfer the encoder to downstream task for final classification. • The imbalanced learning strategy is adopted in the pretext task for improving the training efficiency. • The augmented vision transformer is adopted as the encoder to strengthen the representation learning ability of samples. In practical mechanical fault diagnosis, it's difficult to obtain fault data and the acquisition between normal and fault data is in great imbalance, usually presenting a long-tail distribution state. When training the long-tailed data directly, the imbalanced label may cause label bias, leading to better performance on dominant classes but poorer generalization on tail classes. To address this problem, we adopt two-stage training and propose contrastive-weighted self-supervised model (CSM) with augmented vision transformer which merges the imbalanced learning strategies during the pre-training. In the pretext task, we abandon the label information and adopt contrastive learning by constructing positive and negative sample pairs of each sample. Training it approach to positive sample and away from negatives via augmented vision transformer. The imbalanced strategies are implemented by adaptively weighting to the similarity loss with effective number of samples to learn a better original representation of long-tailed data. In the downstream task, long-tailed data with label are used to fine-tune the pre-trained encoder, which can effectively achieve the final classification task. The experiments under two datasets demonstrate that the pre-training stage can effectively learn a good initialized encoder and can be used in the downstream tasks for better long-tailed data classification. [ABSTRACT FROM AUTHOR]

Multi-stage planetary gearboxes are widely applied in aerospace, automotive and heavy industries. Their key components, such as gears and bearings, can easily suffer from damage due to tough working environment. Health condition identification of planetary gearboxes aims to prevent accidents and save costs. This paper proposes a method based on multiclass relevance vector machine (mRVM) to identify health condition of multi-stage planetary gearboxes. In this method, a mRVM algorithm is adopted as a classifier, and two features, i.e. accumulative amplitudes of carrier orders (AACO) and energy ratio based on difference spectra (ERDS), are used as the input of the classifier to classify different health conditions of multi-stage planetary gearboxes. To test the proposed method, seven health conditions of a two-stage planetary gearbox are considered and vibration data is acquired from the planetary gearbox under different motor speeds and loading conditions. The results of three tests based on different data show that the proposed method obtains an improved identification performance and robustness compared with the existing method. [ABSTRACT FROM AUTHOR]

Fault diagnosis of rotating machinery is receiving more and more attentions. Vibration signals of rotating machinery are commonly analyzed to extract features of faults, and the features are identified with classifiers, e.g. artificial neural networks (ANNs) and support vector machines (SVMs). Due to nonlinear behaviors and unknown noises in machinery, the extracted features are varying from sample to sample, which may result in false classifications. It is also difficult to analytically ensure the accuracy of fault diagnosis. In this paper, a feature extraction and evaluation method is proposed for fault diagnosis of rotating machinery. Based on the central limit theory, an extraction procedure is given to obtain the statistical features with the help of existing signal processing tools. The obtained statistical features approximately obey normal distributions. They can significantly improve the performance of fault classification, and it is verified by taking ANN and SVM classifiers as examples. Then the statistical features are evaluated with a decoupling technique and compared with thresholds to make the decision on fault classification. The proposed evaluation method only requires simple algebraic computation, and the accuracy of fault classification can be analytically guaranteed in terms of the so-called false classification rate (FCR). An experiment is carried out to verify the effectiveness of the proposed method, where the unbalanced fault of rotor, inner race fault, outer race fault and ball fault of bearings are considered. [ABSTRACT FROM AUTHOR]

Increasing advances in sensing technologies and analytics have led to the proliferation of sensors to monitor structural and infrastructural systems. Accurate sensor data can provide information about structural health, aid in prognosis, and help calculate forces for vibration control. However, sensors are susceptible to faults such as loss of data, random noise, bias, drift, etc., due to the aging of sensors, defects, or environmental factors. Although traditional signal processing techniques can detect and isolate faults and reconstruct corrupt or missing sensor data, they demand significant human intervention. The continuous rise in computational power and demonstrated efficacy in numerous domains motivates the use of deep learning to minimize human-in-the-loop techniques. In this work, we introduce a novel, deep learning framework for linear systems with time-invariant parameters that identifies the presence and type of fault in sensor data, location of the faulty sensor and subsequently reconstructs the correct sensor data for fault detection, fault classification, and reconstruction. In our framework, first, a Convolutional Neural Network (CNN) is used to detect the presence of a fault and identify its type. Next, a suite of individually trained Convolutional Autoencoder (CAE) networks corresponding to each type of fault are employed for reconstruction. We demonstrate the efficacy of our framework to address both single and multiple sensor faults in synthetically generated data of a simple shear-type structure and experimentally measured data from a simplified arch bridge. While the framework is agnostic of fault-type, we demonstrate its use for four types of fault namely, missing , spiky , random , and drift. For both simulated and experimental datasets with a single fault, our models performed well, achieving 100% accuracy in faulty sensor localization, more than 98.7% accuracy in fault type detection, and more than 99% accuracy in reconstruction. Our framework can also address multiple concurrent faults with similar accuracy. We empirically demonstrate that our proposed framework performs better than other state-of-the-art techniques in terms of computational efficiency with comparable accuracy. Adoption of our framework in online structural health monitoring applications can lead to minimal disruption to monitoring processes, reduced downtime for structures and infrastructure while simultaneously reducing uncertainty and improving the quality of sensor data for historical records. • Novel end-to-end deep learning framework for sensor fault diagnosis. • CNN to detect presence and type of fault, CAE to rectify fault. • Applied on small time-windows in continuous data, to process in real-time. • Validated numerically on shear-type structure, experimentally on scaled-bridge. • Computationally efficient when compared to state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

Abstract: This paper proposes singular value decomposition (SVD)-based feature extraction methods for fault classification of an induction motor: a short-time energy (STE) plus SVD technique in the time-domain analysis, and a discrete cosine transform (DCT) plus SVD technique in the frequency-domain analysis. To early identify induction motor faults, the extracted features are utilized as the inputs of multi-layer support vector machines (MLSVMs). Since SVMs perform well with the radial basis function (RBF) kernel for appropriately categorizing the faults of the induction motor, it is important to explore the impact of the values for the RBF kernel, which affects the classification accuracy. Likewise, this paper quantitatively evaluates the classification accuracy with different numbers of features, because the number of features affects the classification accuracy. According to the experimental results, although SVD-based features are effective for a noiseless environment, the STE plus SVD feature extraction approach is more effective with and without sensor noise in terms of the classification accuracy than the DCT plus SVD feature extraction approach. To demonstrate the improved classification of the proposed approach for identifying faults of the induction motor, the proposed SVD based feature extraction approach is compared with other state-of-the art methods and yields higher classification accuracies for both noiseless and noisy environments than conventional approaches. [Copyright &y& Elsevier]

Due to the importance of rolling bearings as one of the most widely used industrial machinery elements, development of proper monitoring and fault diagnosis procedure to prevent malfunctioning and failure of these elements during operation is necessary. For rolling bearing fault detection, it is expected that a desired time–frequency analysis method has good computational efficiency, and has good resolution in both, time and frequency domains. The point of interest of this investigation is the presence of an effective method for multi-fault diagnosis in such systems with optimizing signal decomposition levels by using wavelet analysis and support vector machine (SVM). The system that is under study is an electric motor which has two rolling bearings, one of them is next to the output shaft and the other one is next to the fan and for each of them there is one normal form and three false forms, which make 8 forms for study. The results that we achieved from wavelet analysis and SVM are fully in agreement with empirical result. [Copyright &y& Elsevier]

Abstract: Genetic programming (GP) is a stochastic process for automatically generating computer programs. In this paper, three GP-based approaches for solving multi-class classification problems in roller bearing fault detection are proposed. Single-GP maps all the classes onto the one-dimensional GP output. Independent-GPs singles out each class separately by evolving a binary GP for each class independently. Bundled-GPs also has one binary GP for each class, but these GPs are evolved together with the aim of selecting as few features as possible. The classification results and the features each algorithm has selected are compared with genetic algorithm (GA) based approaches GA/ANN and GA/SVM. Experiments show that bundled-GPs is strong in feature selection while retaining high performance, which equals or outperforms the two previous GA-based approaches. [Copyright &y& Elsevier]