Machine Fault Detection Research Articles

Machine fault detection model based on MWOA-BiLSTM algorithm.

This paper proposes the Modulated Whale Optimization Algorithm(MWOA), an innovative metaheuristic algorithm derived from the classic WOA and tailored for bionics-inspired optimization. MWOA tackles common optimization problems like local optima and premature convergence using two key methods: shrinking encircling and spiral position updates. In essence, it prevents algorithms from settling for suboptimal solutions too soon, encouraging exploration of a broader solution space before converging, by incorporating cauchy variation and a perturbation term, MWOA achieve optimization over a wide search space. After that, comparisons were conducted between MWOA and seven recently proposed metaheuristics, utilizing the CEC2005 benchmark functions to assess MWOA's optimization performance. Moreover, the Wilcoxon rank sum test is used to verify the effectiveness of the proposed algorithm. Eventually, MWOA was juxtaposed with the BiLSTM classifier and six other meta-heuristics combined with the BiLSTM classifier. The aim was to affirm that MWOA-BiLSTM outperforms its counterparts, showcasing superior performance across crucial metrics such as accuracy, precision, recall, and F1-Score. The study results unequivocally demonstrate that MWOA showcases exceptional optimization capabilities, adeptly striking a harmonious balance between exploration and exploitation.

Deep learning-based anomaly detection using one-dimensional convolutional neural networks (1D CNN) in machine centers (MCT) and computer numerical control (CNC) machines

Computer numerical control (CNC) and machine center (MCT) machines are mechanical devices that manipulate different tools using computer programming as inputs. Predicting failures in CNC and MCT machines before their actual failure time is crucial to reduce maintenance costs and increase productivity. This study is centered around a novel deep learning-based model using a 1D convolutional neural network (CNN) for early fault detection in MCT machines. We collected sensor-based data from CNC/MCT machines and applied various preprocessing techniques to prepare the dataset. Our experimental results demonstrate that the 1D-CNN model achieves a higher accuracy of 91.57% compared to traditional machine learning classifiers and other deep learning models, including Random Forest (RF) at 89.71%, multi-layer perceptron (MLP) at 87.45%, XGBoost at 89.67%, logistic regression (LR) at 75.93%, support vector machine (SVM) at 75.96%, K-nearest neighbors (KNN) at 82.93%, decision tree at 88.36%, naïve Bayes at 68.31%, long short-term memory (LSTM) at 90.80%, and a hybrid 1D CNN + LSTM model at 88.51%. Moreover, our proposed 1D CNN model outperformed all other mentioned models in precision, recall, and F-1 scores, with 91.87%, 91.57%, and 91.63%, respectively. These findings highlight the efficacy of the 1D CNN model in providing optimal performance with an MCT machine’s dataset, making it particularly suitable for small manufacturing companies seeking to automate early fault detection and classification in CNC and MCT machines. This approach enhances productivity and aids in proactive maintenance and safety measures, demonstrating its potential to revolutionize the manufacturing industry.

A smart WSNs node with sensor computing and unsupervised One-Class SVM classifier for machine fault detection

This paper proposed a novel machine fault detection method based on a smart WSNs node with sensor computing and unsupervised learning. Firstly, sensor computing mode, which achieves machine fault detection on the WSNs node, has been employed to address the limitations of raw data transmission mode, such as huge payload transmission data and node energy consumption. Secondly, a fault classifier based on unsupervised one-class support vector machine (OC-SVM) is designed to overcome the drawbacks of supervised learning, like the requirement of myriad labeled samples for model training. Finally, a smart WSNs node with sensor computing and an unsupervised OC-SVM classifier is developed and fabricated as test beds. A set of experiments has been conducted to verify the proposed method and smart node. The results show that they can reduce 99% of payload transmission data and about 5% of node energy consumption while maintaining acceptable detection accuracy (above 99.2%).

The Use of eXplainable Artificial Intelligence and Machine Learning Operation Principles to Support the Continuous Development of Machine Learning-Based Solutions in Fault Detection and Identification

Machine learning (ML) revolutionized traditional machine fault detection and identification (FDI), as complex-structured models with well-designed unsupervised learning strategies can detect abnormal patterns from abundant data, which significantly reduces the total cost of ownership. However, their opaqueness raised human concern and intrigued the eXplainable artificial intelligence (XAI) concept. Furthermore, the development of ML-based FDI models can be improved fundamentally with machine learning operations (MLOps) guidelines, enhancing reproducibility and operational quality. This study proposes a framework for the continuous development of ML-based FDI solutions, which contains a general structure to simultaneously visualize and check the performance of the ML model while directing the resource-efficient development process. A use case is conducted on sensor data of a hydraulic system with a simple long short-term memory (LSTM) network. Proposed XAI principles and tools supported the model engineering and monitoring, while additional system optimization can be made regarding input data preparation, feature selection, and model usage. Suggested MLOps principles help developers create a minimum viable solution and involve it in a continuous improvement loop. The promising result motivates further adoption of XAI and MLOps while endorsing the generalization of modern ML-based FDI applications with the HITL concept.

Enhancing induction machine fault detection through machine learning: Time and frequency analysis of vibration signals

The integration of machine learning algorithms into fault diagnosis is regarded as an advanced and effective method for detecting electrical system faults. Vibration signals are identified as valuable and easily accessible data for fault identification in electrical machines. In this study, experiments were conducted to assemble a dataset consisting of 525 × 3 instances, each containing 200 three-axis vibration signal measurements. These measurements were obtained from a laboratory test bench equipped with a customized winding three-phase induction machine. Instead of directly analyzing the vibration signals, a feature extraction approach was employed, calculating a comprehensive set of statistical, harmonic, and impulsive features from the time domain (e.g., ShapeFactor, SINAD, ClearanceFactor, ImpulseFactor, CrestFactor, Kurtosis, Skewness, THD, Std, RMS, Mean, PeakValue, SNR), along with spectral features from the frequency domain (e.g., Peak Frequencies, Amplitudes, BandPower). To improve model efficiency, Analysis of Variance (ANOVA) was applied to select only the most significant features, with F-statistics used to rank feature importance. Features with higher F-statistics were prioritized for their ability to explain more variance in motor fault classification. This selective approach reduced model complexity, helping to minimize overfitting and focus on features that had a substantial impact on predictive accuracy. By concentrating on these high-impact features, a balance between simplicity and performance was achieved. These selected features were then used to train five machine learning classifiers: Ensemble (Bagged Trees), Quadratic Support Vector Machine, Weighted K-Nearest Neighbors, Efficient Logistic Regression, and Neural Networks. The performance of these classifiers was evaluated using various metrics, with validation accuracy rates ranging from 78.3% to 98.8%. A comparative study with similar works in the literature was conducted, demonstrating that high performances were obtained with the proposed approach while maintaining computational efficiency.

A Modified EMD Technique for Broken Rotor Bar Fault Detection in Induction Machines.

Induction machines (IMs) are commonly used in various industrial sectors. It is essential to recognize IM defects at their earliest stage so as to prevent machine performance degradation and improve production quality and safety. This work will focus on IM broken rotor bar (BRB) fault detection, as BRB fault could generate extra heating, vibration, acoustic noise, or even sparks in IMs. In this paper, a modified empirical mode decomposition (EMD) technique, or MEMD, is proposed for BRB fault detection using motor current signature analysis. A smart sensor-based data acquisition (DAQ) system is developed by our research team and is used to collect current signals wirelessly. The MEMD takes several processing steps. Firstly, correlation-based EMD analysis is undertaken to select the most representative intrinsic mode function (IMF). Secondly, an adaptive window function is suggested for spectral operation and analysis to detect the BRB fault. Thirdly, a new reference function is proposed to generate the fault index for fault severity diagnosis analytically. The effectiveness of the proposed MEMD technique is verified experimentally.

Machine Learning Approaches for Fault Detection and Diagnosis in Electrical Machines: A Comparative Study of Deep Learning and Classical Methods

Fault detection and diagnosis in electrical machines are crucial for ensuring their safe and reliable operation. In recent years, machine learning techniques have emerged as powerful tools for addressing this challenge, offering the potential for more accurate and efficient fault detection and diagnosis compared to traditional methods. Among these techniques, deep learning has gained significant attention due to its ability to automatically learn relevant features from raw data. However, the performance of deep learning models in this domain has not been extensively compared to classical methods. This paper presents a comparative study of deep learning and classical methods for fault detection and diagnosis in electrical machines. The study evaluates the performance of various machine learning algorithms, including deep neural networks, support vector machines, decision trees, and ensemble methods, in detecting and diagnosing faults such as stator winding faults, rotor faults, and bearing faults. The experimental evaluation is conducted using real-world datasets obtained from electrical machines in industrial settings. Performance metrics such as accuracy, precision, recall, and F1-score are used to assess the effectiveness of each approach in detecting and diagnosing faults accurately and efficiently. The results of the study indicate that deep learning approaches, particularly convolutional neural networks (CNNs) and recurrent neural networks (RNNs), outperform classical methods in terms of fault detection and diagnosis accuracy. These deep learning models demonstrate the ability to automatically extract informative features from raw sensor data, enabling them to effectively identify subtle patterns indicative of faults. The study investigates the interpretability of deep learning models compared to classical methods, examining the extent to which the models can provide insights into the underlying causes of faults. While deep learning models typically operate as black boxes, techniques such as layer-wise relevance propagation (LRP) are employed to enhance their interpretability and facilitate the identification of relevant features contributing to fault detection and diagnosis. This comparative study provides valuable insights into the strengths and limitations of deep learning and classical methods for fault detection and diagnosis in electrical machines, offering guidance for practitioners and researchers in selecting appropriate approaches for their specific applications.

A Comparative Study of Semi-Supervised Anomaly Detection Methods for Machine Fault Detection

A Comparative Study of Semi-Supervised Anomaly Detection Methods for Machine Fault Detection

DETECTION OF MACHINE FAILURES WITH MACHINE LEARNING METHODS

Abstract— This study aims to evaluate the effectiveness of classification algorithms such as Naive Bayes (NB), k-Nearest Neighbours (kNN) and Artificial Neural Networks (ANN) for machine fault detection and investigates the importance of feature selection. The dataset is analysed using cross-validation and the performance of the algorithms is evaluated in terms of AUC, accuracy, F1 Score, Precision and Sensitivity. Naive Bayes and ANN models have the highest AUC values and achieved 99.9% accuracy. kNN model has a lower AUC value than the others (76.0%), but has an accuracy of 97.2%. The feature selection analysis revealed that certain features such as HDF, OSF and PWF contribute significantly to the classification performance. These features have an important role to improve the effectiveness of classification algorithms in detecting faults. These results emphasize the effectiveness of algorithms and the importance of features in machine fault detection and contribute to the development of more reliable and efficient fault detection systems in industrial systems.

Open Fault Detection in Variable Phase-Pole Machines Based on Harmonic Plane Decomposition

Open Fault Detection in Variable Phase-Pole Machines Based on Harmonic Plane Decomposition

Performance analysis of genetically optimized 1D-convolutional neural network architecture for rotor system fault detection and diagnosis

Rotor mass imbalance is a critical problem in industries, which can lead to a breakdown or a catastrophic failure if unattended. Hence, Industry 4.0 uses deep learning approaches like convolutional neural network (CNN) for rotor fault detection and diagnosis of Industrial machines and turbomachinery where multifunctional structures are used. In machinery fault diagnosis, manual tuning of CNN hyperparameters is widely used, which is a tedious, time-consuming, and challenging task to achieve effective feature extraction. Additionally, overfitting of CNN during its training is an obstacle to achieving higher prediction accuracy. Hence, to address these issues, this research focuses on multi-class mass imbalance fault diagnosis using genetically optimized deep learning architecture of 1D-CNN to achieve higher diagnosis capability. The performance of the proposed methodology is evaluated through various case studies using experimental data from the in-house developed test rig. The test results are compared against various 1D-CNN architectures in which the hyperparameters and effective dropout layer positioning are genetically optimized. Also, the performance is benchmarked against standard machine learning algorithms. The results show that genetic algorithm (GA)-optimized 1D-CNN with dropout achieves highest fault prediction accuracy of 97.47% with reduced depth of CNN (with three convolution layers) compared to 84.16% of manually tunned CNN architecture (with seven convolutional layers) and outperformed standard machine learning algorithms. Best drop positioning (end of feature extraction part) reduced the learnable parameters to 95.9% with the highest prediction accuracy of 97.47% compared to adding dropout at all the CNN layers. The proposed GA-optimized 1D-CNN with effective dropout positioning eliminates human intervention and reduces the depth of CNN architecture; this, in turn, reduces computational load and time by reducing learnable parameters of CNN (network weights) with the highest prediction accuracy. Hence, the proposed approach shows promise in enhancing the performance and contributing to the advancement of 1D-CNN for rotor system fault detection and diagnosis.

Research on the dynamic characterization and detection of refrigerant leakage in multi-connected air-conditioning system

Microleakage, frequently arising from aging or inadequate installation, significantly affects the performance and longevity of refrigeration systems. The presence of microleakages, while not precipitating an immediate system failure, results in the gradual deterioration of components and escalates the energy consumption within buildings. Nonetheless, identifying these microleakages through conventional methodologies poses significant challenges. This study introduced a dynamic model specifically designed for detecting microleakages in multi-connected air conditioning systems, particularly in variable refrigerant volume (VRV) systems. Experiments performed under standard operating conditions were integral to enhancing the model's accuracy and furthering its validation. The model integrates the structural and thermophysical attributes of various components, thereby addressing the issues related to data acquisition in standard VRV during instances of microleakage. The analysis of dynamic parameter shifts during system microleakage led to the generation of a comprehensive dataset, which proves invaluable for machine learning and fault detection tasks. A detection and alerting protocol is written in Python presented, chiefly anchored on pressure deviations, augmented by temperature shifts and refrigeration and heating capacities, all represented as virtual volumetric alterations. The findings indicate a delayed transmission of the leakage signal from the leakage point to the system's detection points, with the maximal delay occurring at the compressor inlet under refrigeration conditions, amounting to 239 s. Evaluation under prescribed conditions indicated variances in pressure, temperature, and refrigeration and heating capacities by up to 13.2 %, thereby validating the model's precision and reliability. The program, tested with the generated dataset, accurately identifies leakage states within an average time of 20 min. This model and detection strategy provide a new perspective on microleakage in refrigeration systems, suggesting a robust strategy for their sustained and secure operation.

One-Dimensional LSTM-Regulated Deep Residual Network for Data-Driven Fault Detection in Electric Machines

Achieving a model which accurately diagnoses faults in electric machines is a vital step in data-driven fault detection approaches. To this aim, this paper proposes a Long Short-Term Memory (LSTM) regulated deep residual network for data-driven fault diagnosis purposes in electric machines. The advantages of the proposed network are that it is more general in terms of fault type and measurement, results in a more accurate model for fault classification, and has faster convergence compared with other networks, such as conventional deep residual networks. In order to prove these advantages of the proposed network, it is evaluated by two different types of datasets. One is the Inter-Turn Short Circuit (ITSC) fault in a Permanent Magnet Synchronous Motor (PMSM) with the data of measured three-phase current. The second one is the Case Western Reverse University (CWRU) bearing fault dataset with the vibration measurements. The performance of the network is also compared with other networks. Results reveal that the model can accurately detect both types of faults by two different measurements with a test accuracy of 100%. Furthermore, it converges faster than other networks in the training procedure.

Fuzzy clustering for feature extraction in wavelet-based fault gear identification of electrical machines

Fault detection and diagnosis in electrical machines are periodical for preventing operational interruptions and unexpected shutdowns. However, a Wavelet Feature-dependent Clustering Technique (WFCT) is introduced to address the cyclic fault detection between successive operation intervals. This technique identifies override features from the time-frequency operational wavelets throughout the machine running time. This grouping binds time and operational frequency for identifying override exceeding shutdown/ failure instances. Based on their revamping time, the identified instances are further grouped to prevent overrides in successive operational hours. The fuzzy clustering prevents variation features based on conventional to high-fuzzified extractions.

Interpretable temporal degradation state chain based fusion graph for intelligent bearing fault detection

Machine fault detection is a crucial task in prognostics health management (PHM), while most of the data-driven methods lack model interpretability and transparency. Aiming at promoting reliability, transparency, and anomaly detection performance, an interpretable temporal degradation state chain based fusion graph model is proposed to realize intelligent fault detection in this study. Based on fusion direction distance (FDD), a life-cycle fusion graph is first constructed with a temporal degradation state chain of vibration data. Along with preserving time information, the fusion graph can also comply with the irreversible degradation principle. Once the fusion graph is well developed, the degeneration track of the test sample can be located in the graph and the health status can be predicted, according to a mapping rule and a label determination rule developed in this paper. The anomaly detection accuracy achieves more than 97.8% on XJTU-SY bearing datasets and 92.8% on a NASA bearing dataset. Except for normal and abnormal states, the transition state can also be monitored by the proposed framework, in which incipient faults are initiated. Besides, a hyper-parameter selection method is generated based on Silhouette Coefficient. Moreover, experimental studies have shown the superiority, interpretability, and visibility of this proposed method.

Broken Rotor Bar Fault Detection of Induction Machine Based on Rectified Orthogonal Axes of Stator Current

Broken Rotor Bar Fault Detection of Induction Machine Based on Rectified Orthogonal Axes of Stator Current

General value functions for fault detection in multivariate time series data.

One of the greatest challenges to the automated production of goods is equipment malfunction. Ideally, machines should be able to automatically predict and detect operational faults in order to minimize downtime and plan for timely maintenance. While traditional condition-based maintenance (CBM) involves costly sensor additions and engineering, machine learning approaches offer the potential to learn from already existing sensors. Implementations of data-driven CBM typically use supervised and semi-supervised learning to classify faults. In addition to a large collection of operation data, records of faulty operation are also necessary, which are often costly to obtain. Instead of classifying faults, we use an approach to detect abnormal behaviour within the machine's operation. This approach is analogous to semi-supervised anomaly detection in machine learning (ML), with important distinctions in experimental design and evaluation specific to the problem of industrial fault detection. We present a novel method of machine fault detection using temporal-difference learning and General Value Functions (GVFs). Using GVFs, we form a predictive model of sensor data to detect faulty behaviour. As sensor data from machines is not i.i.d. but closer to Markovian sampling, temporal-difference learning methods should be well suited for this data. We compare our GVF outlier detection (GVFOD) algorithm to a broad selection of multivariate and temporal outlier detection methods, using datasets collected from a tabletop robot emulating the movement of an industrial actuator. We find that not only does GVFOD achieve the same recall score as other multivariate OD algorithms, it attains significantly higher precision. Furthermore, GVFOD has intuitive hyperparameters which can be selected based upon expert knowledge of the application. Together, these findings allow for a more reliable detection of abnormal machine behaviour to allow ideal timing of maintenance; saving resources, time and cost.

Multi-Rate Vibration Signal Analysis for Bearing Fault Detection in Induction Machines Using Supervised Learning Classifiers

Vibration signals carry important information about the health state of a ball bearing and have proven their efficiency in training machine learning models for fault diagnosis. However, the sampling rate and frequency resolution of these acquired signals play a key role in the detection analysis. Industrial organizations often seek cost-effective and qualitative measurements, while reducing sensor resolution to optimize their resource allocation. This paper compares the performance of supervised learning classifiers for the fault detection of bearing faults in induction machines using vibration signals sampled at various frequencies. Three classes of algorithms are tested: linear models, tree-based models, and neural networks. These algorithms are trained and evaluated on vibration data collected experimentally and then downsampled to various intermediate levels of sampling, from 48 kHz to 1 kHz, using a fractional downsampling method. The study highlights the trade-off between fault detection accuracy and sampling frequency. It shows that, depending on the machine learning algorithm used, better training accuracies are not systematically achieved when training with vibration signals sampled at a relatively high frequency.

The improved variational nonlinear chirplet mode decomposition via local maximum synchrosqueezing transform and recursive mode extracting scheme for robust estimation of nonlinear chirplet modes and application to fault detection of rotary machine

As an advanced time-frequency (TF) decomposition (TFD) method, variational nonlinear chirplet mode decomposition (VNCMD) decomposes the original signal into a series of nonlinear chirplet modes (NCMs), such that the inherent characteristic information contained in the signal can be revealed effectively. However, the decomposition ability of VNCMD is largely affected by the prior instantaneous frequency (IF) and the pre-set parameters. In practical engineering applications, the presence of noise and interference components often complicates the accurate determination of prior IFs and appropriate decomposition parameters. Considering the above issues, in order to precisely extract the NCMs and realize the effective analysis of mechanical vibration signals, this paper mainly focuses on the drawbacks of accurate prior IF and the decomposition parameters of VNCMD, and proposed an improved version via local maximum synchrosqueezing transform and a recursive mode extracting scheme. The performance of the proposed method is evaluated through simulation cases, and the results demonstrate its effectiveness. Finally, the proposed method is successfully applied to bearing data analysis and rub-impact fault detection.

Asymmetric air gap fault detection in linear permanent magnet Vernier machines

PurposeThe aim of this paper is to introduce an accurate asymmetric fault index for the diagnosis of the faulty linear permanent magnet Vernier machine (LPMVM).Design/methodology/approachThree-dimensional finite element method is applied to model the LPMVM. The geometrical and physical properties of the machine, the effect of stator and translator teeth, magnetic saturation of core and nonuniform air gap due to asymmetric fault are taken into account in the simulation. The air gap asymmetric fault is proposed. This analytical method estimates the air gap flux density of an LPMVM.FindingsThis paper presents an analytical method to predict the performance of a healthy and faulty LPMVM. The introduced index is based on the frequency patterns of the stator current. Besides, the robustness of the index in different loads and fault severity is addressed.Originality/valueIntroducing index for air gap asymmetry fault diagnosis of LPMVM.