Search

In order to address the challenge of the small-scale, small-target, and complex scenes often encountered in offshore remote sensing image datasets, this paper employs an interpolation method to achieve super-resolution-assisted target detection. This approach aligns with the logic of popular GANs and generative diffusion networks in terms of super-resolution but is more lightweight. Additionally, the image count is expanded fivefold by supplementing the dataset with DOTA and data augmentation techniques. Framework-wise, based on the Faster R-CNN model, the combination of a residual backbone network and pyramid balancing structure enables our model to adapt to the characteristics of small-target scenarios. Moreover, the attention mechanism, random anchor re-selection strategy, and the strategy of replacing quantization operations with bilinear interpolation further enhance the model's detection capability at a low cost. Ablation experiments and comparative experiments show that, with a simple backbone, the algorithm in this paper achieves a mAP of 71.2% on the dataset, an improvement in accuracy of about 10% compared to the Faster R-CNN algorithm. [ABSTRACT FROM AUTHOR]

Accurate state-of-health (SOH) estimation is critical for reliable and safe operation of lithium-ion batteries. However, reliable and stable battery SOH estimation remains challenging due to diverse battery types and operating conditions. In this paper, we propose a physics-informed neural network (PINN) for accurate and stable estimation of battery SOH. Specifically, we model the attributes that affect the battery degradation from the perspective of empirical degradation and state space equations, and utilize neural networks to capture battery degradation dynamics. A general feature extraction method is designed to extract statistical features from a short period of data before the battery is fully charged, enabling our method applicable to different battery types and charge/discharge protocols. Additionally, we generate a comprehensive dataset consisting of 55 lithium-nickel-cobalt-manganese-oxide (NCM) batteries. Combined with three other datasets from different manufacturers, we use a total of 387 batteries with 310,705 samples to validate our method. The mean absolute percentage error (MAPE) is 0.87%. Our proposed PINN has demonstrated remarkable performance in regular experiments, small sample experiments, and transfer experiments when compared to alternative neural networks. This study highlights the promise of physics-informed machine learning for battery degradation modeling and SOH estimation. Reliable lithium-ion battery health assessment is vital for safety. Here, authors present a physics-informed neural network for accurate and stable state-of-health estimation, overcoming challenges of varied battery types and usage conditions. [ABSTRACT FROM AUTHOR]

Convolutional neural network (CNN), with the ability of feature learning and nonlinear mapping, has demonstrated its effectiveness in prognostics and health management (PHM). However, an explanation on the physical meaning of a CNN architecture has rarely been studied. In this article, a novel wavelet-driven deep neural network, termed as WaveletKernelNet (WKN), is presented, where a continuous wavelet convolutional (CWConv) layer is designed to replace the first convolutional layer of the standard CNN. This enables the first CWConv layer to discover more meaningful kernels. Furthermore, only the scale parameter and translation parameter are directly learned from raw data at this CWConv layer. This provides a very effective way to obtain a customized kernel bank, specifically tuned for extracting defect-related impact component embedded in the vibration signal. In addition, three experimental studies using data from laboratory environment are carried out to verify the effectiveness of the proposed method for mechanical fault diagnosis. The experimental results show that the accuracy of the WKNs is higher than CNN by more than 10%, which indicate the importance of the designed CWConv layer. Besides, through theoretical analysis and feature map visualization, it is found that the WKNs are interpretable, have fewer parameters, and have the ability to converge faster within the same training epochs. [ABSTRACT FROM AUTHOR]

Existing data-driven fault diagnosis methods assume that the label sets of the training data and test data are consistent, which is usually not applicable for real applications since the fault modes that occur in the test phase are unpredictable. To address this problem, open set fault diagnosis (OSFD), where the test label set consists of a portion of the training label set and some unknown classes, is studied in this article. Considering the changeable operating conditions of machinery, OSFD tasks are further divided into shared-domain open set fault diagnosis (SOSFD) and cross-domain open set fault diagnosis (COSFD) in this article. For SOSFD, 1-D convolutional neural networks are trained for learning discriminative features and recognizing fault modes. For COSFD, due to the distribution discrepancy between the source and target domains, the deep model needs to learn domain-invariant features of shared classes and separate features of outlier classes. Thus, by utilizing the output of an additional domain classifier, a model named bilateral weighted adversarial networks is proposed to assign large weights to shared classes and small weights to outlier classes during the feature alignment. In the test phase, samples are classified according to the outputs of the deep model and unknown-class samples are rejected by the extreme value theory model. Experimental results on two bearing datasets demonstrate the effectiveness and superiority of the proposed method. [ABSTRACT FROM AUTHOR]

Unsupervised domain adaptation (UDA)-based methods have made great progress in mechanical fault diagnosis under variable working conditions. In UDA, three types of information, including class label, domain label, and data structure, are essential to bridging the labeled source domain and unlabeled target domain. However, most existing UDA-based methods use only the former two information and ignore the modeling of data structure, which make the information contained in the features extracted by the deep network incomplete. To tackle this issue, a domain adversarial graph convolutional network (DAGCN) is proposed to model the three types of information in a unified deep network and achieving UDA. The first two types of information are modeled by the classifier and the domain discriminator, respectively. In data structure modeling, a convolutional neural network (CNN) is first employed to exact features from input signals. After that, the CNN features are input to the proposed graph generation layer to construct instance graphs by mining the relationship of structural characteristics of samples. Then, the instance graphs are modeled by a graph convolutional network, and the maximum mean discrepancy metric is leveraged to estimate the structure discrepancy of instance graphs from different domains. Experimental results conducted on two case studies demonstrate that the proposed DAGCN can not only obtain the best performance among the comparison methods, but also can extract transferable features for domain adaptation. The code library is available at: https://github.com/HazeDT/DAGCN. [ABSTRACT FROM AUTHOR]

Traditional deep learning assumes that the probability distributions of the test set and the training set are exactly the same. When the test samples and the training samples are from different working conditions, the distribution gap between different conditions might dramatically lower the accuracy. To address this problem, domain adaptation (DA) via joint using training and test sets is adopted to increase the final accuracy. The disadvantage of DA is that it requires test samples to participate in the training phase. However, sometimes it is impossible to collect the test samples beforehand. To solve this problem, this article proposes a conditional adversarial domain generalization with a single discriminator, which aims to extract domain-invariant features from different working conditions and generalize the features to unseen working conditions. To realize conditional adversarial training, a new conditional adversarial strategy that the feature extractor can allow the discriminator to distinguish the fault classes, but cannot distinguish the domains, is designed to better confuse the discriminator and generalize the features. Meanwhile, three loss functions, including mean square error (MSE), cross-entropy coupled with entropy, and cross-entropy combined with traditional adversarial methods, are proposed to achieve the new conditional adversarial strategy. The effectiveness of the conditional adversarial strategy is verified by training conditional generative adversarial networks (CGANs) on two image data sets, and the performance of the proposed domain generalization method for bearing fault diagnosis is also verified by bearing data sets. Compared with traditional conditional adversarial domain generalization, the proposed method can save computing resource consumption and improve accuracy that does not appear in the training phase. [ABSTRACT FROM AUTHOR]

The tool wear condition monitoring is key to ensuring product quality. This article develops a direct technique dealing with cutting tool images to automate the tool wear detection and identification. The constructed U-Net-based network can realize an effective and reliable extraction of the tool wear area. The introduction of deep supervision with a Matthews correlation coefficient (MCC)-based surrogate loss function helps to address the few-shot and data imbalance issues. Experiments on the images with wear on the flank face of cutting tools from a computer numerical control (CNC) turning machine show the effectiveness, competitiveness, and reliability of the proposed method under different types of loss functions. [ABSTRACT FROM AUTHOR]

Anomaly detection is one of the most fundamental and indispensable components in predictive maintenance. In this article, anomaly detection is modeled as a one-class classification problem. Based on the scenario that the training data only include healthy state data, a fault-attention generative probabilistic adversarial autoencoder (FGPAA) is proposed to automatically find low-dimensional manifold embedded in high-dimensional space of the signal. Benefited from the characteristics of autoencoder, the signal information loss in feature extraction is reduced. Then, the fault-attention abnormal state indictor can be constructed with the distribution probability of low-dimensional feature and reconstruction error. Effectiveness of the model is verified with fault classification datasets and run-to-failure experimental datasets. The results show that FGPAA outperforms both GPAA and other traditional methods and can be processed in real time. It not only can obtain high accuracy for both classification data and run-to-failure data, but also achieve a certain trend index for run-to-failure data. [ABSTRACT FROM AUTHOR]

Deep learning-based multi-sensor fusion approach has been widely used for machine condition monitoring. Complementary information from different physical sensors or the same sensors on multiple locations of the monitored target can effectively improve the accuracy of condition monitoring. While, how to distinguish importance of every single sensor for condition monitoring is rarely researched. To address this gap, an adaptive channel weighted convolutional neural network (ACW-CNN) is proposed in this paper to investigate importance of different sensors in fusion approach. The designed ACW layer in a deep neural network can calibrate sensors weight according to sensors importance. The recalibrated channel weights can be used as the guidance for sensor position optimization. Furthermore, a new loss function, that is, Focal Loss, is introduced into ACW-CNN to deal with class imbalance generated in real helicopter transmission system (HTS) monitoring. To validate the effectiveness of the proposed ACW-CNN for condition monitoring, two case studies including condition monitoring to gearbox transmission system and helicopter transmission system are carried out. The results of comparative experiments show the proposed method, that is ACW-CNN with Focal Loss, is superior to other methods in classification accuracy, G-mean, and F-measure for condition monitoring of HTS. [ABSTRACT FROM AUTHOR]

Bearing fault diagnosis is becoming more and more important for current rotating machinery. How to extract bearing fault signals submerged by heavy background noise is still a challenging problem, especially in the case of multiple faults coupled with each other. In this paper, a novel multifault model called sparse multiperiod group lasso (SMPGL) is proposed to extract the fault feature of every single fault from multifault signals based on the sparsity within and across groups (SWAG) property and the separably periodic prior. Moreover, a fast algorithm is deduced based on the majorization–minimization (MM) algorithm for solving the proposed multifault model and its convergence condition is also analyzed. We investigate the parameter selection thoroughly and provide a deterministic rule for the parameter selection of SMPGL. The main advantages of the proposed method are that users can set the number of compound faults freely, the algorithm is very fast, and parameters are set adaptively. The effectiveness and robustness of SMPGL are verified by simulation studies and two experiment cases. Furthermore, the comparison study shows that the proposed SMPGL method gives more satisfying results than other state-of-the-art methods, including the L1-based method and spectral kurtosis (SK). [ABSTRACT FROM AUTHOR]

Deep learning with ability to feature learning and nonlinear function approximation has shown its effectiveness for machine fault prediction. While, how to transfer a deep network trained by historical failure data for prediction of a new object is rarely researched. In this paper, a deep transfer learning (DTL) network based on sparse autoencoder (SAE) is presented. In the DTL method, three transfer strategies, that is, weight transfer, transfer learning of hidden feature, and weight update, are used to transfer an SAE trained by historical failure data to a new object. By these strategies, prediction of the new object without supervised information for training is achieved. Moreover, the learned features by deep transfer network for the new object share joint and similar characteristic to that of historical failure data, which is beneficial to accurate prediction. Case study on remaining useful life (RUL) prediction of cutting tool is performed to validate effectiveness of the DTL method. An SAE network is first trained by run-to-failure data with RUL information of a cutting tool in an off-line process. The trained network is then transferred to a new tool under operation for on-line RUL prediction. The prediction result with high accuracy shows advantage of the DTL method for RUL prediction. [ABSTRACT FROM AUTHOR]

A sparse deep learning method is proposed to overcome overfitting risk of deep networks with a large number of nodes and layers. Deep stacking network (DSN) is a classic and effective deep learning method, and its sparse form is presented to generate the sparse deep learning method. In DSN, output labels are encoded as a series consisted of 1 and 0. This coding strategy makes output labels to be sparse. However, sparsity of output labels is not considered in DSN model. Considering this limitation, sparse DSN (SDSN) is developed in this paper. The SDSN extends tradition DSN in sparsity characterization using a sparse regularization term. By this term, predicted output label is constrained to be similar with ideal output label that is binary and consisted of continuous 1 and 0 with a sidestep shape. The sparse regularization term is used as a soft threshold strategy to set irrelevant element to be zero, by which effectiveness of SDSN is enhanced. Case studies about fault diagnosis of motor are used to validate performance of SDSN. Comparison between SDSN and commonly used deep networks is further conducted. The results show advance of SDSN for fault classification. [ABSTRACT FROM AUTHOR]

• Time-series augmentation and balanced loss to address minority overfitting. • Data-rebalanced sampler and boundary balancing to address majority domination. • Integrate minority overfitting-based and majority domination-based strategies. • A comprehensive model-agnostic framework for class-imbalanced diagnosis. Considering the difficulty of data acquisition in industry, especially for failure data of large-scale equipment, classification with these class-imbalanced datasets can lead to the problems of minority categories overfitting and majority categories domination. A model-agnostic framework towards class-imbalanced fault diagnosis requirement is proposed to systematically alleviate these problems. Four sub-modules, including Time-series Data Augmentation, Data-Rebalanced sampler, Balanced Margin Loss, and classifier with Dynamic Decision Boundary Balancing are performed to improve recognition accuracy of minority categories without performance degradation on majority categories. Meanwhile, the framework is compatible with general neural networks and provides flexible model candidates to meet the need of feature extraction for different data types. Three case studies on public datasets demonstrate that proposed framework outperformed various state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Powder bed quality is critical in Laser Powder Bed Fusion (LPBF) as defects in the powder bed will affect the quality of the forming parts. Recently, numerous powder bed defect detection methods based on semantic segmentation algorithms have been developed. However, due to the computational resource constraints and real-time requirements in industry scenarios, these heavy models face challenges in deploying into the LPBF process monitoring system. Meanwhile, existing lightweight models struggle to achieve sufficient performance on industrial data due to their limited feature extraction capabilities. To address the challenges in industrial deployment, this paper proposes a lightweight and efficient LPBF layer-wise defect detection paradigm to achieve a balance between performance and efficiency. Aiming at improving the performance of the lightweight model, this paper proposes a structured Knowledge Distillation framework, in which a pre-trained high-performance segmentation model (UNet++) is used to guide the training process of a lightweight model. To address the impact of model discrepancies on the effectiveness of Knowledge Distillation and further improve the inference speed, this paper constructs a specific lightweight model and applies Structural Re-parameterization to deeply compress the model. Moreover, we created a dataset comprising 406 images of powder-bed defects, with each image annotated at the pixel level. Massive experiments based on the dataset demonstrate the effectiveness of the proposed paradigm, with an excellent trade-off in performance and efficiency. The speed of the proposed model is approximately 5 times faster than that of high-performance models while maintaining comparable performance. • A lightweight and efficient paradigm for LPBF in-situ defect detection. • Introduced Structure Re-parameterization to enhance the Knowledge Distillation. • An excellent trade-off in performance and efficiency for model deployment. [ABSTRACT FROM AUTHOR]

As a multi-resolution analysis method rooted rigorously in mathematics, wavelet transform (WT) has shown its great potential in rotary machine fault diagnosis, characterized by continued development and innovative new applications. In traditional fault diagnosis, WT has been widely used for fault feature extraction and extensively studied for performance improvement. With the emergence of data-driven intelligent fault diagnosis, especially deep learning techniques, WT has attracted renewed attention for its ability of adding interpretability into the intelligent diagnosis models. This paper aims to highlight the advancement of WT-based fault diagnosis research over the last decade. Toward this end, a comprehensive overview of WT method is given, followed by a summary of WT for fault diagnosis from two perspectives: traditional fault diagnosis and intelligent fault diagnosis. Finally, future research trends are discussed, including benchmarking, wavelet base design, integration with other methods, and enhancement through deep learning. [ABSTRACT FROM AUTHOR]

• A reassignment-enable reweighted sparse time–frequency method is proposed for the feature extraction of signals with fast-varying IF. • An acceleration strategy is proposed for FISTA to speed up the convergence for solving the Re2STF model. • The simulation study proves that Re2STF has the best TF performance among the considered methods. • The application in aeroengine vibration analysis verifies that Re2STF is effective for aeroengine rub-impact fault diagnosis. Time-frequency analysis (TFA) is an effective tool for the feature extraction of nonstationary signals. However, the energy concentration of the TF representation (TFR) obtained by conventional TFA is weakened when the instantaneous frequency (IF) of signals varies fast, which adversely affects the feature extraction and reconstruction. In this paper, a reassignment-enable reweighted sparse TF method (Re2STF) is proposed to achieve high energy concentration as well as highly accurate reconstruction when analyzing signals with fast-varying IF, and thus is effective to the vibration analysis of aeroengine rub-impact fault. A reweighted sparse TF model is constructed, which iteratively employs the TF reassignment representation into the reweighted strategy to suppress noise and match the fast-varying TF structure better. In particular, the reassignment-enable weight can highlight the TF ridge and suppress the TF coefficients elsewhere. Furthermore, an effective acceleration strategy for the fast iterative shrinkage threshold algorithm is proposed to speed up the model solving. The simulation study and the application in the vibration analysis of a dual-rotor turbofan engine show the improvement of Re2STF in denoising, TF energy concentration and reconstruction, thus proving the effectiveness in aeroengine rub-impact fault diagnosis. [ABSTRACT FROM AUTHOR]