Classifier For Fault Diagnosis Research Articles

Transfer learning based cross-process fault diagnosis of industrial robots

In the actual industrial application of robots, the characteristics of robot malfunctions change accordingly as the working environment becomes increasingly diverse and complex. Utilizing the original fault diagnosis models in new working environments correspondingly leads to a decline in the performance and the generalization capability of the model. Moreover, the monitoring data collected in new working processes often has limited or no labels, making the diagnosis models trained with this data unable to identify faults accurately. In this paper, we propose a Domain adaptive Cross-process Fault Diagnosis method (DCFD) to leverage knowledge from existing working processes for diagnosing faults in new working processes. DCFD uses Multi-Kernel Maximum Mean Discrepancy (MK-MMD) to measure the difference between the current working processes and the previous working processes, enhancing the fault diagnosis capability of the robotic system in cross-process scenarios. DCFD achieves an average fault classification accuracy of 98% on 12 types of migration tasks, which demonstrates the effectiveness of DCFD on cross-process fault diagnosis classification tasks in real-time industrial application scenarios.

Comparative performance analysis of different types of k-nearest neighbor (k-NN) classifiers for fault diagnosis of air compressor setup

This paper presents a method for identifying Reciprocating Air Compressor (RAC) faults using acoustic signals obtained from both healthy and unhealthy conditions. The entire procedure is carried out with Uni-directional microphones using a LABVIEW-based data collection interface and data acquisition (DAQ) hardware unit that has several ports. Accumulated one healthy and seven unhealthy signals of RAC setup processed using signal processing technique called Local Mean Decomposition (LMD). Further, six Statistical Properties (SPs) have been evaluated in order to extract fault features namely: mean (US), variance (σ s2), root square of mean (Mrms), root amplitude of mean (Mrma), absolute amplitude of mean (Mama), and Kurtosis index (Ki). Extracted fault features are classified using various types of k-NN classifiers namely: fine (f-kNN), medium (m-kNN), coarse (c-kNN) and weighted (w-kNN). It has been found that LMD along with 6 statistical properties and form different type of k-NN classifiers, the weighted k-NN classifier has a greater accuracy of 86.74%, which is quite accurate as compared to other k-NN classifiers.

Research on rolling bearing fault diagnosis method based on simulation and experiment fusion drive.

Aiming at the problem that the rolling bearing fault data are difficult to obtain and that the traditional fault diagnosis method does not consider the signal uncertainty characteristics and the low accuracy of models in the process of rolling bearing fault, a fault diagnosis method based on simulation and experiment fusion drive is proposed. First, the dynamics simulation model of rolling bearings under different fault conditions is established to obtain the bearing fault simulation signals. Second, a sequence generative adversarial network is used to fuse the simulation and experimental data. Bearing vibration signals are often very uncertain, so considering the probability characteristics of fault signals, the probability box model under different fault states is constructed by the direct probability box modeling method, and its characteristic vectors are extracted. Finally, an extreme gradient boosting Tree model for fault diagnosis classification is constructed to compare and evaluate the classification and diagnosis effects of bearing states before and after data fusion. The results show that the proposed method has a good diagnostic effect and is suitable for solving the fault diagnosis problem under the condition of insufficient data.

FedCAE: A New Federated Learning Framework for Edge-Cloud Collaboration Based Machine Fault Diagnosis

With the coming of the industrial big data era, data-driven fault diagnosis models emerge recently and show potential results in many studies. However, it is impractical to collect massive high-quality data in actual industrial applications, which strongly reduces the performance of fault diagnosis models. Therefore, multiple data owners are motivated to share their own data for training a brilliant model with the aggregated dataset. Notwithstanding, it is unrealistic to share data due to the potential conflict of interest. For addressing the issue of data island, a new federated framework namely FedCAE is proposed for fault diagnosis. In this framework, each client is equipped with a convolutional autoencoder (CAE) trained by its corresponding local data. Then, all CAEs are uploaded to a server and aggregated to a global CAE according to an adaptive weighted selection mechanism. Afterwards, the global CAE is downloaded to all clients for extracting low-level features from local datasets by its encoder and uploading these features to train a global fault diagnosis classifier in the server. At last, the classifier is downloaded to all clients for completing their own diagnosis tasks. Experiment results on two bearing datasets suggest FedCAE offers a promising solution on confidential decentralized learning.

Adaptive graph-guided joint soft clustering and distribution alignment for cross-load and cross-device rotating machinery fault transfer diagnosis

Domain adaptation (DA) is an effective solution for addressing the domain shift problem. However, existing DA techniques usually directly match the distributions of the data in the original feature space, where some of the features may be distorted by a large domain shift. Besides, geometric and clustering structures of the data, which play a significant role in revealing hidden failure patterns, are not considered in traditional DA methods. To tackle the above issues, a new joint soft clustering and distribution alignment with graph embedding (JSCDA-GE) method is proposed. Specifically, weighted subspace alignment is proposed to align bases of source and target subspaces by combining instance reweighting and subspace alignment strategies. Then, JSCDA-GE formulates an objective function by incorporating dynamic distribution alignment, soft large margin clustering, and GE in a unified structural risk minimization framework. Ultimately, JSCDA-GE aims to learn a generalization classifier for fault diagnosis. Its effectiveness and superiority have been confirmed through 36 tasks on 2 bearing databases.

Diagnosing faults in a photovoltaic system using the Extra Trees ensemble algorithm

The application of machine learning techniques for monitoring and diagnosing faults in photovoltaic (PV) systems has been shown to enhance the reliability of PV power generation. This research introduced a novel machine learning classifier for fault diagnosis in PV systems, utilizing an ensemble algorithm known as extra trees (ETC). The study initially proposed a system with two PV modules and developed a low-cost Arduino-based data logger to gather data from the PV system in free-fault and faulty conditions. Subsequently, the study evaluated six other advanced classifiers for fault diagnosis in PV systems, namely logistic regression (LR), k-nearest neighbor (kNN), support vector machine (SVM), decision tree (DT), AdaBoost, and random forest (RF) models using the collected data from the proposed PV system. The assessment of the various models' performance indicated that the extra trees model exhibits superior classification capabilities for partial shading (PS), open circuit (OCF), partial shading with bypass diode disconnected (PSBD), and combined partial shading with bypass diode disconnected plus open circuit (PSBDOC) faults. The results demonstrated that the new ETC classifier achieves an accuracy of 92%, surpassing the 91%, 87%, 7%, and 59% accuracy of the RF, DT, kNN, and LR classifiers, respectively. This highlights the effectiveness of the extra trees model in enhancing fault detection and classification by distinguishing between open circuits and twin faults. Consequently, these results can be utilized to develop advanced diagnostic tools for photovoltaic systems, thereby improving the reliability of solar technology and accelerating the rate of installation.

Adversarial Domain Adaptation with Dual Auxiliary Classifiers for Cross-Domain Open Set Intelligent Fault Diagnosis

Adversarial Domain Adaptation with Dual Auxiliary Classifiers for Cross-Domain Open Set Intelligent Fault Diagnosis

Chemical Process Fault Diagnosis Method Based on Deep Learning

To address the problem that the traditional fault diagnosis method for chemical processes under big data relies too much on expert experience and fault features are difficult to distinguish, a deep learning-based fault diagnosis method is proposed, which combines convolutional neural network (CNN), long and short-term memory (LSTM) and attention mechanism (AM). In this method, the spatial sequence features of the input signal are extracted by the CNN adaptively, while the LSTM extracts the time-series features of the signal. Finally, the model performance is enhanced by introducing the attention mechanism and using the SoftMax layer as a classifier for fault diagnosis, so that the model can notice the important features of the faults with the interference of noise. Simulation validation of the method in this paper is performed using the TE chemical process data set, and it is demonstrated that the method can be used for chemical process fault diagnosis studies. Finally, compared with other fault diagnosis methods, the method is more accurate and has certain superiority.

Multi-sensor data fusion-enabled semi-supervised optimal temperature-guided PCL framework for machinery fault diagnosis

Due to the extremely limited prior knowledge, machinery fault diagnosis under varying working conditions with limited annotation data is a very challenging task in practical industrial scenarios. To solve this issue, a multi-sensor data fusion-enabled semi-supervised optimal temperature-guided prototypical contrastive learning (OTPCL) framework is proposed for machinery fault diagnosis under varying working conditions with limited annotation data. First, a multi-sensor data adaptive weighting strategy (MSAWS) is built to reintegrate the multi-sensor data by adaptively learning the important information of different sensor data and assigning the corresponding weights. Then, an OTPCL method is proposed to reduce the domain shift between the source and target domains, where the vital parameter named the temperature coefficient are adaptively optimized for enhancing the ability of discriminant domain-invariant feature extraction. Finally, a semi-supervised learning (SSL) framework is constructed by combining the MSAWS, the OTPCL and a domain-shared classifier for machinery fault diagnosis. In two experimental analyses on machinery varying working conditions with limited annotation data, the superiority of the proposed MSAWS is validated by respectively using our fused multi-sensor data and different original single-sensor data for the SSL framework. The proposed method obviously outperforms the five advanced domain adaptation methods including CORAL, MMD, CMMD, JDA and PCL.

Fault Diagnosis of Rolling Bearing Based on HPSO Algorithm Optimized CNN-LSTM Neural Network.

The quality of rolling bearings is vital for the working state and rotation accuracy of the shaft. Timely and accurately acquiring bearing status and early fault diagnosis can effectively prevent losses, making it highly practical. To improve the accuracy of bearing fault diagnosis, this paper proposes a CNN-LSTM bearing fault diagnosis model optimized by hybrid particle swarm optimization (HPSO). The HPSO algorithm has a strong global optimization ability and can effectively solve nonlinear and multivariate optimization problems. It is used to optimize and match the parameters of the CNN-LSTM model and dynamically find the optimal value of the parameters. This model overcomes the problem that the parameters of the CNN-LSTM model depend on empirical settings and cannot be adjusted dynamically. This model is used for bearing fault diagnosis, and the accuracy rate of fault diagnosis classification reaches 99.2%. Compared with the traditional CNN, LSTM, and CNN-LSTM models, the accuracy rates are increased by 6.6%, 9.2%, and 5%, respectively. At the same time, comparing the models with different optimization parameters shows that the model proposed in this paper has the highest accuracy. The experimental results verified the superiority of the HPSO algorithm to optimize model parameters and the feasibility and accuracy of the HPSO-CNN-LSTM model for bearing fault diagnosis.

Block feature selection based on NSGA-II applied to fault diagnosis of gearboxes

A hybrid model with multiple sub-classifier is a powerful classifier for intelligent fault diagnosis of gearboxes. Each sub-classifier is used to deal with different gear or bearing faults, and thus each should have its own optimal feature group. However, these sub-classifiers in the hybrid model used the same feature group in most reported feature selection methods. To overcome this problem, this study brings some insights into optimization on sub-classifiers. This proposed method, known as block feature selection (BFS), optimizes the feature group for only one fault type in each sub-classifier. The selected groups are then amalgamated to form the optimal features for all types of gearbox failures, with corresponding labels in the hybrid model. Each sub-classifier is considered to be one block. In each block, the concept of NSGA-II is used to select features for its corresponding sub-classifier. Then, a new sorting algorithm is proposed to find the best feature group of sub-classifier. In general, BFS can optimize features of each sub-classifier to find the best feature group for hybrid model. The effectiveness of BFS is validated by experimental data from bearing and gear faults in a planetary gearbox rig. The robustness of BFS is verified by incorporating white noise at different levels.

A novel domain generalization network with multidomain specific auxiliary classifiers for machinery fault diagnosis under unseen working conditions

The domain adaptation-based intelligent diagnosis approaches have achieved promising performance on diagnosis tasks under different working conditions. However, these methods rely on a premise that the target data are available in the model training phase. In real industries, collecting interest data from target machines in advance may be infeasible, which greatly restricts the practicality of intelligent diagnosis approaches in reality. To solve this issue, this study proposes a novel domain generalization network for machinery fault diagnosis where interest data are completely unavailable during model training. In the proposed network, multiple domain-specific auxiliary classifiers are firstly designed to effectively learn domain-specific features from each source domain, and then, a convolutional auto-encoder module is further constructed to map raw signals into a new feature space where the learned domain-specific features are removed. Meanwhile, with the features outputted by the convolutional auto-encoder, a domain-invariant classifier with inter-domain alignment strategy is designed to learn generalization diagnostic knowledge among different source domains, thereby performing diagnosis tasks under unseen conditions. Experiments on three practical rotary machinery datasets validate the effectiveness of the proposed network, showing that the proposed network is promising for fault diagnosis tasks in practical scenarios.

Fault diagnosis of photovoltaic strings by using machine learning‐based stacking classifier

AbstractPhotovoltaic (PV) modules are prone to short circuits, open circuits, cracks, which can bring serious harmful effects. It is difficult to establish the corresponding PV fault models to diagnose the status of PV strings. The paper proposes a machine learning‐based stacking classifier (MLSC) for accurate fault diagnosis of PV strings. Specifically, for the operating state of PV modules, the parameter sensitivity algorithm is used to analyze the impact of characteristic factors on the characteristics of PV modules. Then based on the characteristic factors (irradiance, temperature, current, and power), MLSC is proposed to realize the accurate fault diagnosis of PV strings. This structure of MLSC is to integrate all kinds of classifiers by stacking to play the characteristics of each classifier. It combines the characteristics of various types of machine learning algorithms to improve the overall classification effect by using the advantages of each classifier. Finally, experiments reveal that MLSC improves the accuracy of PV fault diagnosis.

Bi-LSTM fault diagnosis method for rolling bearings based on segmented interception AR spectrum analysis and information fusion

The rolling bearing fault diagnosis is affected by industrial environmental noise and other factors, leading to the existence of some redundant components after signal decomposition. At the same time, the existence of the modal aliasing phenomenon in empirical mode decomposition (EMD) and the relevant improved algorithms also leads to the existence of many invalid features in the components. These phenomena have great influence on the bearing fault diagnosis. So a rolling bearing bidirectional-long short term memory (Bi-LSTM) fault diagnosis method was proposed based on segmented interception auto regressive (SIAR) spectrum analysis and information fusion. The ensemble empirical mode decomposition (EEMD), the complementary ensemble empirical mode decomposition (CEEMD) and the robust EMD (REMD) algorithms decompose the rolling bearing fault signals, and AR spectrum analysis is performed on the obtained components respectively. By comparing the AR spectra of the components corresponding to different fault locations, the effective AR spectral values are intercepted as the eigenvalues of the data, and finally all the eigenvalues are fused to achieve the purpose of screening effective features more efficiently so as to reduce the impact of feature redundancy caused by mode aliasing on neural network training. Then the Bi-LSTM neural network was used as a rolling bearing fault diagnosis classifier, and the simulation experiments were conducted based on the rolling bearing fault signal data from Case Western Reserve University to verify the effectiveness of the proposed feature extraction and fault diagnosis method.

Robust fault diagnosis of a high-voltage circuit breaker via an ensemble echo state network with evidence fusion

Reliable mechanical fault diagnosis of high-voltage circuit breakers is important to ensure the safety of electric power systems. Recent fault diagnosis approaches are mostly based on a single classifier whose performance relies heavily on expert prior knowledge. In this study, we propose an improved Dempster–Shafer evidence theory fused echo state neural network, an ensemble classifier for fault diagnosis. Evidence credibility is calculated through the evidence deviation matrix and the segmented circle function and employed as credibility weights to rectify the raw evidence. Then, an improved Dempster–Shafer evidence fusion algorithm is proposed to fuse evidence from different echo state network modules and sensors. Unlike conventional classifiers, the proposed methodology consists of multiple echo state neural network modules. It has better flexibility and stronger robustness, and its model performance is not sensitive to network parameters. Comparative analysis indicates that it can handle the paradox evidence fusion analysis and thus can achieve better diagnostic performance. The superiority of the reported fault diagnosis approaches is verified with the experimental data of a ZN12 high-voltage circuit breaker.

Exploring best-matched embedding model and classifier for charging-pile fault diagnosis

The continuous increase of electric vehicles is being facilitating the large-scale distributed charging-pile deployment. It is crucial to guarantee normal operation of charging piles, resulting in the importance of diagnosing charging-pile faults. The existing fault-diagnosis approaches were based on physical fault data like mechanical log data and sensor data streams. However, there are other types of fault data, which cannot be used for diagnosis by these existing approaches. This paper aims to fill this gap and consider 8 types of fault data for diagnosing, at least including physical installation error fault, charging-pile mechanical fault, charging-pile program fault, user personal fault, signal fault (offline), pile compatibility fault, charging platform fault, and other faults. We aim to find out how to combine existing feature-extraction and machine learning techniques to make the better diagnosis by conducting experiments on realistic dataset. 4 word embedding models are investigated for feature extraction of fault data, including N-gram, GloVe, Word2vec, and BERT. Moreover, we classify the word embedding results using 10 machine learning classifiers, including Random Forest (RF), Support Vector Machine, K-Nearest Neighbor, Multilayer Perceptron, Recurrent Neural Network, AdaBoost, Gradient Boosted Decision Tree, Decision Tree, Extra Tree, and VOTE. Compared with original fault record dataset, we utilize paraphrasing-based data augmentation method to improve the classification accuracy up to 10.40%. Our extensive experiment results reveal that RF classifier combining the GloVe embedding model achieves the best accuracy with acceptable training time. In addition, we discuss the interpretability of RF and GloVe.

Efficient fault diagnosis approach for solar photovoltaic array using a convolutional neural network in combination of generative adversarial network under small dataset

Accurate faults diagnosis for photovoltaic (PV) array is one of the vital factors that guarantee the reliable operation of PV power plant. Artificial intelligence (AI) based fault detection and diagnosis (FDD) models are promising techniques. In order to automatically extract the faults features from the raw electrical data of PV array and create efficient FDD model with small dataset, a FDD scheme using Wasserstein generative adversarial network (WGAN) and convolutional neural network (CNN) is designed. The proposed FDD model is consisting of three modules, a discriminator, a generator and a classifier for fault diagnosis. By analyzing sequential PV data in a 2-Dimension way, the proposed discriminator and generator learn the distribution of PV data under various PV system operations. Then they are utilized to generate more labeled samples to improve the performance of the CNN based classifier. Thus, the proposed FDD model can be trained only requiring minor labeled samples. A laboratory grid-connected PV system is established to experimentally investigate the performance of the developed method. The results demonstrate that the designed FDD model can accurately diagnose line-line and open circuit faults.

Fault diagnosis of PEMFC systems based on an auxiliary transfer network

Data-driven fault diagnosis methods require huge amounts of expensive experimental data. Due to the irreversible damage of severe fault embedding experiments to proton exchange membrane fuel cell (PEMFC) systems, rare available data can be obtained. In view of this issue, a fault diagnosis method based on an auxiliary transfer network (ATN) is proposed. This method uses two parallel neural networks (main and auxiliary neural network) and a prediction fusion module to realize fault diagnosis. The auxiliary neural network is a fault diagnosis classifier pretrained based on both slight and severe fault simulative data, and its weights are transmitted into the ATN structure and frozen. After that, the main neural network is trained based on a large number of slight fault experimental data and a small number of severe fault experimental data. Through ATN, the main neural network learns the abstract features of severe faults under the guidance of auxiliary neural network, and realizes the transfer learning from simulation-based fault diagnosis classifier to experiment-based fault diagnosis classifier. Through testing, the accuracy and precision of ATN-based fault diagnosis classifier with LSTM as both main and auxiliary neural network reaches 0.993 and 1.0 respectively, which is higher than the common data-driven methods.

Bearing fault diagnosis method based on variational mode decomposition optimized by CS-PSO

In order to solve the problem that bearing vibration signal fault feature is difficult to extract effectively under noise, a fault diagnosis method based on Variational Mode Decomposition (VMD) optimized by Cuckoo Search (CS) and Particle Swarm Optimization (PSO) is proposed. The effect of VMD is affected by the number of modes and the penalty parameter. The Levy flight strategy and elimination mechanism of the CS algorithm is added to PSO algorithm and the position updating process is optimized. According to the correlation coefficient and envelope entropy of the Intrinsic Mode Function (IMF), the objective function is constructed to search for the optimal combination of VMD mode number and penalty parameter. The bearing fault type is determined by analyzing the IMF envelope spectrum of the optimal objective function. The fault diagnosis classification task upon the bearing sample data from Case Western Reserve University demonstrates that the proposed CS-PSO algorithm improves the model’s classification performance by 5%, which is on average 7.3% higher than other state-of-the-art models. This model can provide an important reference for the accuracy of bearing fault diagnosis.

Convolutional autoencoder neural network for fault diagnosis with multi-sensor data

With machines in manufacturing industry being automated, complex and intelligent, its monitoring systems are equipped with more and more smart sensors. How to extract useful features from great volume of multi-sensor data become a great challenge to the field of fault diagnosis. To overcome such challenge, an improved convolutional autoencoder neural network (CANN) is proposed to fuse and extract effective features of the color images formed by multi-sensor data in this paper. Firstly, the vibration signals of different channels are jointly transformed into color images. Secondly, an improved CANN is constructed by introducing special convolution kernels and residual connection for multi-sensor data fusion and feature extraction. Finally, the encoder part of CANN is connected with the softmax classifier for fault diagnosis. Two datasets collected from Wind Power Test-Bed and Industrial Blower Fan System are used to fully validate the effectiveness of proposed CANN. The results show that it can effectively fuse multi-sensor data and mine the discriminative features. Furthermore, compared with some related state-of-art methods, the CANN obtains higher diagnostic accuracy, especially for less labeled data.