Traditional fault diagnosis models often overlook the interconnections between segments of vibration data, resulting in the loss of critical feature information. Additionally, the vibration signals of rolling bearings exhibit non-linear behaviors during operation. Therefore, an efficient fault diagnosis model tailored for rolling bearings is proposed in this paper. In the proposed model, the 1D vibration signals are first preprocessed using ensemble empirical mode decomposition (EEMD). This technique generates multiple intrinsic mode functions (IMF) as individual nodes. The percentage distance between each node is calculated using the Wasserstein distance (WD) to capture the relationships between nodes and use it as the edge weights to construct a node graph. This unique approach enhances the transformation of 1D vibration signals into a node graph representation, preserving important information. An improved multi-head graph attention network (MGAT) model is established to extract features and perform classification on the node graph. This MGAT model effectively utilizes the relationships between nodes and enhances the accuracy of fault diagnosis. The experimental results demonstrate that the proposed method achieves higher accuracy compared to similar modelswhile requiring less processing time. The proposed approach contributes significantly to the field of fault diagnosis for rolling bearings and provides a valuable tool for practical applications.
The reconstruction-based (RB) approach can effectively suppress the misdiagnosis problem due to the smearing effect in fault isolation. However, the current exploration of the RB approach for large-scale nonlinear systems is still limited. Therefore, this paper proposes a reliable and effective fault diagnosis method based on a reconstruction-based stacked sparse autoencoder (RBSSAE) for high-dimensional industrial systems. In RBSSAE, a reconstruction-based index achieved by the Steffensen iterative method is developed to check whether the given variable(s) are responsible for the faults efficiently. However, the number of possible faulty variable combinations grows exponentially with the system dimensionor actual abnormal variables, causing an unbearable computational burden for variable combination optimization. Hence, the proposed RBSSAE utilizes a sequential floating forward selection approach to rapidly isolate the most decisive combination of fault variables, meeting a requirement of online fault diagnosis. Finally, the effectiveness of the RBSSAE is verified on a numerical example and a real industrial case. Comparisons with other state-of-the-art methods are also presented.
The suspension system in an automobile is essential for comfort and control. Implementing a monitoring system is crucial to ensure proper function, prevent accidents, maintain performance, and reduce both downtime and costs. Traditionally, diagnosing faults in suspension systems has relied on specialized setups and vibration analysis. The conventional approach typically involves either wavelet analysis or a machine learning approach. While these methods are effective, they often demand specialized expertise and time consumable. Alternatively, using deep learning for suspension system fault diagnosis enables faster and more precise real-time fault detection. This study explores the use of vision transformers as an innovative approach to fault diagnosis in suspension systems, utilizing spectrogram images. The process involves extracting spectrogram images from vibration signals, which serve as inputs for the vision transformer model. The test results demonstrate that the proposed fault diagnosis system achieves an impressive accuracy rate of 98.12% in identifying faults.
This paper tries to introduce a new intelligent method for the early fault diagnosis of diesel engines. Firstly, infrared thermography (IRT) is introduced into diesel engine condition monitoring, then infrared images of diesel engines in four health states, such as normal condition, single-cylinder misfire, multi-cylinder misfire and air filter blockage, are collected and the region of interest (ROI) of infrared images are extracted. Next, conditional generative adversarial network (CGAN) is deployed to perform data augmentation on infrared image datasets. Then, deep convolutional neural network (DCNN) and Softmax regression (SR) classifier are used for automatically extracting infrared image fault features and pattern recognition, respectively. Finally, a comparison with three deep learning (DL) models is performed. The validation results show that the data augmentation method proposed in the paper can significantly improve the early fault diagnosis accuracy, and DCNN has the best fault diagnosis effect andresistance to temperature fluctuation interference among the four DL models.
This research aims to provide a fault diagnosis approach for Hybrid Dynamic (SDHs), Systems and FaultTolerant Control synthesis, while also ensuring the smooth operation of industrial settings. This study is a part of the larger topic of Hybrid Dynamic System control and fault diagnosis. The primary focus is on modelling strategies designed expressly for Hybrid Dynamic Systems, with a concentration on combining continuous and event-driven components. Much work is devoted to developing a model that can incorporate both kinds of elements. A system model that can track several modes without explicit identification can be created by utilizing Neuro-Fuzzy Networks, providing a thorough overview. On the basis of this synthesized model, an AI-based fault diagnosis method is subsequently developed.
6
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Condition monitoring and problem diagnostics have drawn more attention recently in the industrial sector. One of the most crucial parts of rotating machinery are rolling-element bearings. Bearing faults are a common cause of machinery failures. To identify failing bearings early, vibration condition monitoring of rotating machinery has emerged as the preferred technique. Several signal analysis techniques can extract useful information from vibration data. The non-stationary analysis signals that are typically associated with machine defects cannot be handled by frequency-based approaches. Non-stationary signals are analyzed effectively by applying time-frequency techniques. The use of wavelet transform has increased in bearing monitoring research for the last 20 years to obtain correlated time-frequency information. This paper presents a discrete wavelet transform (DWT) and energy distribution-based bearing defect diagnostic technique. The "db3" wavelet form of DWT is used to decompose vibration signals under both normal and faulty (inner race-fault and outer race-fault) bearing conditions at various frequency ranges. Due to the default, the energy distribution for every decomposition level is calculated to detect which frequency band contains the harmonics. The results obtained from healthy and defective bearings are compared. The wavelet coefficient with the highest value of the energy distribution is employed in the Fourier analysis to pinpoint the site of the fault. The monitoring results demonstrate that the suggested approach is effective in finding and analyzing faults.
PL
Monitorowanie stanu i diagnostyka problemów przyciągnęły ostatnio więcej uwagi w sektorze przemysłowym. Jedną z najbardziej kluczowych części maszyn wirujących są łożyska toczne. Usterki łożysk są częstą przyczyną awarii maszyn. W celu wczesnej identyfikacji uszkodzonych łożysk, monitorowanie stanu wibracji maszyn wirujących stało się preferowaną techniką. Kilka technik analizy sygnału może wydobyć użyteczne informacje z danych o drganiach. Niestacjonarne sygnały analizy, które są zwykle związane z uszkodzeniami maszyn, nie mogą być obsługiwane przez podejścia oparte na częstotliwości. Sygnały niestacjonarne są skutecznie analizowane poprzez zastosowanie technik czasowoczęstotliwościowych. Zastosowanie transformaty falkowej wzrosło w badaniach nad monitorowaniem łożysk przez ostatnie 20 lat w celu uzyskania skorelowanej informacji czasowo-częstotliwościowej. W niniejszej pracy przedstawiono dyskretną transformatę falkową (DWT) oraz technikę diagnostyczną opartą na rozkładzie energii. Forma falkowa "db3" DWT jest używana do dekomponowania sygnałów drganiowych w warunkach łożyska zarówno normalnego, jak i wadliwego (wewnętrznego i zewnętrznego) w różnych zakresach częstotliwości. Ze względu na domyślność, rozkład energii dla każdego poziomu dekompozycji jest obliczany w celu wykrycia, które pasmo częstotliwości zawiera harmoniczne. Wyniki uzyskane z łożysk zdrowych i uszkodzonych są porównywane. Współczynnik falkowy o największej wartości rozkładu energii jest wykorzystywany w analizie Fouriera w celu określenia miejsca uszkodzenia. Wyniki monitorowania pokazują, że proponowane podejście jest skuteczne w wyszukiwaniu i analizie uszkodzeń.
This paper proposes a method for the diagnosis of stator inter-turn short-circuit fault for permanent magnet synchronous generators (PMSG). Inter-turn short-circuit currents are among the most critical in PMSG. For safety considerations, a fast detection is required when a fault occurs. This approach uses the parameter estimation of the per-phase stator resistance in closed-loop control of variable speed of wind energy conversion system (WECS). In the presence of an incipient short-circuit fault, the estimation of the resistance of the stator in the d-q reference frame does not make it possible to give the exact information. To solve this problem, a novel fault diagnosis scheme is proposed using parameter estimation of the per-phase stator resistance. The per-phase stator resistance of PMSG is estimated using the MRAS algorithm technique in real time. Based on a faulty PMSG model expressed in Park’s reference frame, the number of short-circuited turns is estimated using MRAS. Fault diagnosis is on line detected by analysing the estimated stator resistance of each phase according to the fault condition. The proposed fault diagnosis scheme is implemented without any extra devices. Moreover, the information on the estimated parameters can be used to improve the control performance. The simulation results demonstrate that the proposed method can estimate the faulty phase.
Marine electronically controlled (ME) two-stroke diesel engines occupy the highest market share in newly-built ships and its fuel injection system is quite different and important. Fault diagnosis in the fuel injection system is crucial to ensure the power, economy and emission of ME diesel engines, so we introduce hierarchical multiscale fluctuation dispersion entropy (HMFDE) and a support matrix machine (SMM) to realise it. We also discuss the influence of parameter changes on the entropy calculation’s accuracy and efficiency. The system simulation model is established and verified by Amesim software, and then HMFDE is used to extract a matrix from the features of a high pressure signal in a common rail pipe, under four working conditions. Compared with vectorised HMFDE, the accuracy of fault diagnosis using SMM is nearly 3% higher than that using a support vector machine (SVM). Experiments also show that the proposed method is more accurate and stable when compared with hierarchical multiscale dispersion entropy (HMDE), hierarchical dispersion entropy (HDE), multiscale fluctuation dispersion entropy (MFDE), multiscale dispersion entropy (MDE) and multiscale sample entropy (MSE). Therefore, the proposed method is more suitable for the modelling data. This research provides a new direction for matrix learning applications in fault diagnosis in marine two-stroke diesel engines.
As an important component of the fuel injection system, the fuel injector is crucial for ensuring the power, economy, and emissions for a whole ME (machine electronically-controlled) marine diesel engine. However, injectors are most prone to failures such as reduced pressure at the opening valve, clogged spray holes and worn needle valves, because of the harsh working conditions. The failure characteristics are non-stationary and non-linear. Therefore, to efficiently extract fault features, an improved refined composite multi-scale dispersion entropy (IRCMDE) is proposed, which uses the energy distribution of sampling points as weights for coarse-grained calculation, then fast correlation-based filter (FCBF) and support vector machine (SVM) are used for feature selection and fault classification, respectively. The experimental results from a MAN B&W 6S35ME-B9 marine diesel engine show that the proposed algorithm can achieve 92.12% fault accuracy for injector faults, which is higher than multiscale dispersion entropy (MDE), refined composite multiscale dispersion entropy (RCMDE) and multiscale permutation entropy (MPE). Moreover, the experiment has also proved that, due to the double-walled structure of the high-pressure fuel pipe, the fuel injection pressure signal is more accurate than the vibration signal in reflecting the injector operating conditions.
Bearings are important components of rotating machinery and transmission systems, and are often damaged by wear, overload and shocks. Due to the low resolution of traditional time-frequency analysis for the diagnosis of bearing faults, a synchrosqueezed wavelet transform (SSWT) is proposed to improve the resolution. An improved convolutional neural network fault diagnosis model is proposed in this paper, and a Bayesian optimisation method is applied to automatically adjust the structure and hyperparameters of the model to improve the accuracy of bearing fault diagnosis. Experimental results from the accelerated life testing of bearings show that the proposed method is able to accurately identify various types of bearing fault and the different status of these faults under complex running conditions, while achieving very good generalisation ability.
Marine centrifugal fans usually work in harsh environments. Their vibration signals are non-linear. The traditional fault diagnosis methods of fans require much calculation and have low operating efficiency. Only shallow fault features can be extracted. As a result, the diagnosis accuracy is not high. It is difficult to realize the end-to-end fault diagnosis. Combining the Complete Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN) and lightweight neural network, a fault classification method is proposed. First, the CEEMDAN can decompose the vibration signal into several intrinsic modal functions (IMF). Then, the original signals can be transformed into 2-D images through pseudocolour coding of the IMFs. Finally, they are fed into the lightweight neural network for fault diagnosis. By embedding a convolutional block attention module (CBAM), the ability of the network to extract critical feature information is improved. The results show that the proposed method can adaptively extract the fault characteristics of a marine centrifugal fan. While the model is lightweight, the overall diagnostic accuracy can reach 99.3%. As exploratory basic research, this method can provide a reference for intelligent fault diagnosis systems on ships.
Induction motors (IMs) have a crucial and significant role in various industrial sectors. With the prolonged operation of IMs, faults tend to develop that can be classified into five major categories, i.e., broken rotor bars, stator winding faults, air-gap eccentricity, bearing faults, and load torque fluctuations. If the faults go undetected, it may lead to catastrophic failure. Hence, the predictive-based condition monitoring technique has evolved as a real-time fault diagnosis that exploits the revolutionary idea of cyber-physical system (CPS). Furthermore, motor current signature analysis (MCSA) is a non-invasive fault diagnosis technique of a motor that can be used to investigate the presence of five fault types. However, the major constraint that industries face today is the on-field implementation of MCSA-based fault diagnosis involving CPS-based architecture, executed in an automated manner. Hence, the present article depicts algorithms that aim at real-time monitoring of IMs through a CPS framework. The proposed methodology is automated, does not involve any human intervention, and has been validated with real-time experiments, depicting its effectiveness and practicality.
Equipment condition monitoring is essential to maintain the reliability of the electromechanical systems. Recently topics related to fault diagnosis have attracted significant interest, rapidly evolving this research area. This study presents a non-invasive method for online state classification of a squirrel-cage induction motor. The solution utilizes thermal imaging for non-contact analysis of thermal changes in machinery. Moreover, used convolutional neural networks (CNNs) streamline extracting relevant features from data and malfunction distinction without defining strict rules. A wide range of neural networks was evaluated to explore the possibilities of the proposed approach and their outputs were verified using model interpretability methods. Besides, the top-performing architectures were optimized and deployed on resource-constrained hardware to examine the system's performance in operating conditions. Overall, the completed tests have confirmed that the proposed approach is feasible, provides accurate results, and successfully operates even when deployed on edge devices.
Deep groove ball bearings are widely used in rotary machinery. Accurate for bearing faults diagnosis is essential for equipment maintenance. For common depth learning methods, the feature extraction of inverse time domain signal direction and the attention to key features are usually ignored. Based on the long short term memory(LSTM) network, this study proposes an attention-based highway bidirectional long short term memory (AHBi-LSTM) network for fault diagnosis based on the raw vibration signal. By increasing the Attention mechanism and Highway, the ability of the network to extract features is increased. The bidirectional LSTM network simultaneously extracts the raw vibration signal in positive and inverse time-domains to better extract the fault features. Six deep groove ball bearings with different health conditions were used to validate the AHBi-LSTM method in an experiment. The results showed that the accuracy of the proposed method for bearing fault diagnosis was over 98%, which was 8.66% higher than that of the LSTM model. The AHBi-LSTM model is also better than other relevant models for bearing fault diagnosis.
The fault diagnosis for maintenance of machines operating in variable conditions requires special dedicated methods. Variable load or temperature conditions affect the vibration signal values. The article presents a new approach to diagnosing rotating machines using an artificial neural network, the training of which does not require data from the damaged machine. This is a new approach not previously found in the literature. Until now, neural networks have been used for machine diagnosis in the form of classifiers, where data from individual faults were required. A new diagnostic parameter rDPNS (Relative Differences Product of Network Statistics) as a function of the machine's shaft order was proposed as a kind of new order spectrum independent of the machine's operating conditions. The presented work analyses the use of the proposed method to diagnose misalignment and unbalance. The results of an experiment carried out in the laboratory demonstrated the effectiveness of the proposed method.
Industrial high-speed rotating machines entail constant and consistent monitoring to prevent downtime, affecting quantity and quality. Complex machines need advanced intelligent fault diagnosis showing minimal errors. This work offers a MATLAB-based fault diagnosis for sugar industry machines. The vibration behavior of physical industrial machines is obtained, and the signals are provided to a MATLAB program to identify the fault. The information helps to suggest remedies to include in the maintenance schedule. The ease and comprehensible nature of the method reduce time and enhance the reliability of condition monitoring for industrial machines.
The three-phase induction motor is well suited for a wide range of mobile drives, specifically for electric vehicle powertrain. During the entire life cycle of the electric motor, some types of failures can occur, with stator winding failure being the most common. The impact of this failure must be considered from the incipient as it can affect the performance of the motor, especially for electrically powered vehicle application. In this paper, the intern turn short circuit of the stator winding was studied using Fast Fourier transform (FFT) and Shor-Time Fourier transform (STFT) approaches. The residuals current between the estimated currents provided by the extended Kalman filter (EKF) and the actual ones are used for fault diagnosis and identification. Through FFT, the residual spectrum is sensitive to faults and gives the extraction of inter-turn short circuit (ITSC) related frequencies in the phase winding. In addition, the FFT is used to obtain information about when and where the ITSC appears in the phase winding. Indeed, the results allow to know the faulty phase, to estimate the fault rate and the fault occurrence frequency as well as their appearance time.
The research was focused on analyzing the causes of high-pressure die-casting defects, more specifically on casting leakage, which is considered perhaps the most important and common defect. The real data used for modelling was obtained from a high-pressure die-casting foundry that manufactures aluminum cylinder blocks for the world’s leading automotive brands. This paper compares and summarizes the results of applying advanced modelling using artificial neural networks, regression trees, and support vector machines methods to select artificial neural networks as the most effective method to perform a multidimensional optimization of process parameters to diagnose the causes of die-casting defects and to indicate the future research scope in this area. The developed system enables the prediction of the level of defects in castings with satisfactory accuracy and is therefore a highly relevant reference for process engineers of high-pressure foundries. This article indicates exactly which process parameters significantly influence the formation of a defect in a casting.
This paper presents a novel fault detection algorithm for a three-phase interleaved DC–DC boost converter integrated in a photovoltaic system. Interleaved DC–DC converters have been used widely due to their advantages in terms of efficiency, ripple reductions, modularity and small filter components. The fault detection algorithm depends on the input current waveform as a fault indicator and does not require any additional sensors in the system. To guarantee service continuity, a fault tolerant topology is achieved by connecting a redundant switch to the interleaved converter. The proposed fault detection algorithm is validated under different scenarios by the obtained results.
In modern drive systems, the high-efficient permanent magnet synchronous motors (PMSMs) have become one of the most substantial components. Nevertheless, such machines are exposed to various types of faults. Hence, on-line condition monitoring and fault diagnosis of PMSMs have become necessary. One of the most common PMSM faults is the stator winding fault. Due to the destructive character of this failure, it is necessary to use fault diagnostic methods that allow fault detection at its early stage. The article presents the results of experimental studies obtained from fast Fourier transform (FFT) and short-time Fourier transform (STFT) analyses of the stator phase current, stator phase current envelope and stator phase current space vector module. The superiority of the proposed method over the classical approach based on the stator current analysis using FFT is highlighted. The proposed solution is experimentally verified under various motor operating conditions. The application of STFT analysis discussed so far in the literature has been limited to the fault diagnosis of induction motors and the narrow range of the analysed motor operating conditions. Moreover, there are no works in the field of motor diagnostics dealing with STFT analysis for stator windings based on the stator current envelope and the stator current space vector module.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.