Domain adaptation (DA) enhances the scalability of remaining useful life (RUL) prediction technologies, providing a reliable foundation for maintenance decisions across diverse equipment within Industrial Internet of Things (IIoT). Traditional DA approaches typically necessitate simultaneous access to both source and target domain data, which often conflicts with data privacy concerns prevalent in IIoT. Furthermore, the substantial storage resources required for source domain data hinder the implementation of efficient DA on resource-constrained edge devices. To address these challenges, we propose a source-free DA (SFDA) framework leverages uncertainty estimation pseudo-labels to conduct cross-domain RUL prediction in the absence of source domain data. Initially, we develop a self-supervised knowledge distillation framework that adapts efficiently to domain shift based on pseudo-labeling. Building on this, we propose an uncertainty estimation pseudo-label guided loss reweighting strategy to mitigate the impact of pseudo-label noise. This strategy prioritizes highly reliable pseudo-labels by assessing discrepancies in feature distributions from different augmented views of the input samples. Additionally, a reconstruction-based training method is designed to align the feature distributions. Extensive experimental evaluations demonstrate that our proposed method outperforms current state-of-the-art techniques, ensuring reliable RUL predictions in scenarios characterized by domain shift and the absence of source domain data.

Harnessingthe data processing and communication capabilities of the Internet of Things (IoT), smart grids can seamlessly share power information across wired and wireless networks, enhancing the grid's reliability, stability, sustainability, and energy efficiency. Since smart meters' communication in the IoT network is carried out on the public channel, there is significant uncertainty about the authenticity of the transmitted electricity data and the security of the smart meter keys. Existing solutions cannot well-handle the above security issues. Therefore, in this article, an efficient authentication scheme with key leakage-resistance, signing key leakage-resistant authentication (SKLRA), for the IoT-enabled smart grid is proposed for IoT-integrated smart grids. The SKLRA scheme is designed to overcome the performance shortcomings and security limitations found in traditional certificateless key-insulated signature mechanisms and certificateless signature approaches that do not utilize random oracles, particularly for smart grid applications. We also provide a rigorous security analysis alongside experimental evaluations to demonstrate the robustness and practical viability of the SKLRA scheme in real-world applications.

Generative Adversarial Networks (GANs) for generating realistic data, have substantially improved fault diagnosis algorithms in various Internet of Things (IoT) systems. However, challenges such as training instability and dynamical inaccuracy limit their utility in high-speed rail (HSR) bogie fault diagnosis. To address these challenges, we introduce the Pulse Voltage-Guided Conditional Diffusion Model (VGCDM). Unlike traditional implicit GANs, VGCDM adopts a sequential U-Net architecture, facilitating multi-steps denoising diffusion for generation, which bolsters training stability and mitigate convergence issues. VGCDM also incorporates control pulse voltage by cross-attention mechanism to ensure the alignment of vibration with voltage signals, enhancing the Conditional Diffusion Model’s progressive controlablity. Consequently, solely straightforward sampling of control voltages, ensuring the efficient transformation from Gaussian Noise to vibration signals. This adaptability remains robust even in scenarios with time-varying speeds. To validate the effectiveness, we conducted two case studies using SQ dataset and high-simulation HSR bogie dataset. The results of our experiments unequivocally confirm that VGCDM outperforms other generative models, achieving the best RSME, PSNR, and FSCS, showing its superiority in conditional HSR bogie vibration signal generation. For access, our code is available athttps://github.com/xuanliu2000/VGCDM.

Liquid Rocket Engine, as the key power device of the space transportation system, the anomaly detection of operation status is the key to its reliable operation. However, in the face of multi-sensor high-frequency monitoring signals under extreme operating conditions, limited by the ability of model data modeling, the existing methods, based on classification and reconstruction strategies, are difficult to further improve the anomaly localization precision. To address the challenges and overcome the limitations of existing methods, this paper proposes a Dual-control Inference Diffusion Model (DIDM), which reconstructs and inferences on specified sensor samples to achieve accurate anomaly detection. The reverse diffusion inference process is controlled by the channel condition and mask prior, combined with two loss functions for alternating training, which enables inference for samples from specified sensors at specific moments. We evaluate the model based on the static ignition test data of a certain type of LRE. The results show that DIDM outperforms the state-of-the-art methods in terms of detection accuracy, which demonstrates the effectiveness and superiority of DIDM. Furthermore, by combining the error distributions of the inference results, we can achieve a more accurate location of anomaly in the time and frequency domains, which could increase the efficiency of rocket launches and air and space transportation, and enhance the potential of the academic results for industrial applications.

With the improvement of intelligence and integration, automatic supervision of large-scale systems is a current challenge in guaranteeing the high-reliability of edge devices. Hence, fast & accurate anomaly detection (AD) has become an urgent need via the edge computing of the industrial Internet of Things (IIoT). For this purpose, this paper creatively proposes a dual agents based on two-phase adversarial training strategy (2P-DAs) to perform rapid, stable and unsupervised AD for large-scale IIoT-edge devices. It integrates the superiorities of deep autoencoder (AEs) and generative adversarial network (GANs), utilizing normal multivariate time-series as inputs, 1-Encoder vs. 2-Decoders architecture as backbone, and two-phase unsupervised adversarial learning to make it isolate anomalies while providing efficient training. On the one hand, this allows the inherent limitations of AEs to be overcome by training a model capable for recognizing non-anomalies and thus performing a good reconstruction. On the other hand, dual structures allow for stability in adversarial training, thereby solving the issues of collapse and non-convergence encountered in GANs. Two practical industrial data, as cloud & edge data, are used to verify the robustness, inference speed and high detection performance of 2P-DAs in IIoT-edge AD, which demonstrates an impressive performance under multiple evaluation indexes.

Developing a trustworthy framework for intelligent fault diagnosis (IFD) of machines has two major challenges: confidently recognizing known faults and precisely detecting novel faults. However, current IFD frameworks are typically based on the closed-world assumption and tackle the two issues independently, making it hard to meet the expectations of reliable diagnosis in complicated working situations. In this paper, aProbabilistic framework forMechanical fault diagnosis (ProMo) is proposed to integrate misidentification and out-of-distribution (OOD) detection for high-speed train bogie. ProMo uses variational parameters to capture uncertainties at both the feature and prediction levels, in this way the misclassified and OOD samples with high predictive uncertainty are differentiated. To begin, a Bayesian deep neural network with a hierarchical classifier is built, offering diverse predictions and enabling ensemble uncertainty estimate. Then, a probabilistic null space analysis technique for post hoc OOD detection is presented, in which the magnitude of feature projections indicates the OOD-ness. Additionally, a novel metric assessing classification correctness and prediction reliability simultaneously is proposed. Extensive experiments revealed that ProMo outperforms state-of-the-art methods and achieves reliable fault diagnosis, even in the presence of covariate shift in monitoring data.

Deep graph neural networks (GNNs) have demonstrated their exceptional expressive capability in detecting fault features from multisensor signals for fault diagnosis. However, the excessive depth of these models often hinders their deployment on Internet of Things (IoT) systems. In order to facilitate the implementation of graph-based models on IoT devices for fault diagnosis of HSR bogie, this paper proposes a novel compression technique called GraKD. GraKD distills the latent knowledge from the teacher model to a lightweight student model in an adversarial manner. The discriminator of GraKD comprises a representation identifier and a logit identifier. The former effectively distinguishes between the node representations of the student and the teacher by discerning the local affinity of connected node patch-patch pairs and the global affinity represented by the patch-global pairs. The latter employs a residual multi-layer perceptron block to differentiate between the logits of the student and the teacher. The effectiveness of GraKD is validated using data collected from test rigs of bogies. A series of experiments substantiate the broad applicability of GraKD in compressing various GNN-based models.

Fault detection, also known as anomaly detection (AD), is at the heart of prediction and health management (PHM), which plays a vital role in ensuring the safe operation of mechanical equipment. Nonetheless, the lack of anomaly data creates a significant obstacle to the AD of the mechanical system. In particular, the complex modulation effects induced by time-varying speeds make AD much more challenging. For rapid and accurate AD, a stable knowledge distillation decoupling net (DecouplingNet) is provided to overcome these difficulties. First, an adversarial network consisting of an encoder, a decoder, and an encoder-discriminator is developed to model normal samples well by imposing constraints on the latent space. Then, a causal decoupling framework is suggested to disentangle equipment state-related information from operating conditions-related features, enabling stable condition monitoring at varying speeds. Finally, feature-based knowledge distillation is employed to boost the efficiency of AD while maintaining the detection accuracy. The proposed method is tested on two experimental scenarios and compared with some typical AD methods. The finding demonstrates that the net outperforms others in terms of accuracy and efficiency when it comes to detecting anomalies in the mechanical equipment that runs under varying speeds.

Fault diagnosis is vital to ensuring the security of rotating machinery operations. While fault data obtained from mechanical equipment for this issue are often insufficient and of no labels. In this case, supervised algorithms cannot come into play. Hence, this article proposes a self-supervised simple Siamese framework (SSF) for bearing fault diagnosis based on the contrastive learning algorithm SimSiam which uses a simplified Siamese network to find the distinguishable features of different fault categories. SSF consists of a weight-sharing encoder applied on two inputs, a nonlinear predictor and a linear classifier. SSF learns invariant characteristics of fault samples via maximizing the similarity between two views of each inputted sample. Several data augmentation (DA) methods for vibration signals, which provide different sample views for the model, are also studied, for it is crucial for contrastive learning. After fine-tuning the learned encoder and a linear layer classifier with a small subset of labeled data (1%-5% of the total samples), the network achieves satisfactory performance for bearing fault diagnosis. A series of experiments based on the data from three different scenarios are used to verify the proposed methods, getting 100%, 99.38%, and 98.87% accuracy separately.

Recent advances in data-driven methods have significantly promoted intelligent fault diagnostics for varied industrial applications. However, due to the limitations of machine fault data and the varied scenarios in the context of industrial working conditions, existing diagnostic models can hardly achieve satisfactory results. In this article, we propose a domain discrepancy-guided contrastive feature learning framework for few-shot fault diagnosis under varied working conditions. Unlike the conventional contrastive learning paradigm using manually augmented data, a sample pairs construction is implemented based on the differences between domain distributions for data acquired under different working conditions. The similarity contrast learns the domain-invariant features from a small number of sample pairs. The learned fault features can then be used for fault identification without parameter fine-tuning. In two case studies, we validated the performance of the proposed framework with small training samples under varying speeds, loads, and significant noises. Compared with the state-of-the-art methods, the proposed solution achieved higher diagnostic accuracy for the targeted applications.

Deep learning has demonstrated splendid performance in mechanical fault diagnosis on condition that source and target data are identically distributed. In engineering practice, however, the domain shift between source and target domains significantly limits the further application of intelligent algorithms. Despite various transfer techniques proposed, either they focus on single-source domain adaptation (SDA) or they utilize multisource domain globally or locally, which both cannot address the cross-domain diagnosis effectively, especially with category shift. To this end, we propose globally localized multisource DA for cross-domain fault diagnosis with category shift. Specifically, we construct a GlocalNet to fuse multisource information comprehensively, which consists of a feature generator and three classifiers. By optimizing the Wasserstein discrepancy of classifiers locally and accumulative higher order multisource moment globally, multisource DA is achieved from domain and class levels thus to reduce the shift on domain and category. To refine the classifier at sample level, a distilling strategy is presented. Finally, an adaptive weighting policy is employed for reliable result. To evaluate the effectiveness, the proposed method is compared with multiple methods on four bearing vibration datasets. Experimental results indicate the superiority and practicability of the proposed method for cross-domain fault diagnosis.

To achieve reliable and automatic anomaly detection (AD) for large equipment such as liquid rocket engine (LRE), multisource data are commonly manipulated in deep learning pipelines. However, current AD methods mainly aim at single source or single modality, whereas existing multimodal methods cannot effectively cope with a common issue, modality incompleteness. To this end, we propose an unsupervised multimodal method for AD with missing sources in LRE system. The proposed method handles intramodality fusion, intermodality fusion, and decision fusion in a unified framework composed of multiple deep autoencoders (AEs) and a skip-connected AE. Specifically, the first module restores missing sources to construct a complete modality, thus advancing the secondary reconstruction. Different from vanilla reconstruction-based methods, the proposed method minimizes reconstruction loss and meanwhile maximizes the dissimilarity of representations in two latent spaces. Utilizing reconstruction errors and latent representation discrepancy, the anomaly score is acquired. At decision level, the model performance can be further enhanced via anomaly score fusion. To demonstrate the effectiveness, extensive experiments are carried out on multivariate time-series data from static ignition of several LREs. The results indicate the superiority and potential of the proposed method for AD with missing sources for LRE.

Industrial Internet of Things (IIoT) is mainly a data-oriented network, so intelligent processing of massive data is desiderated to realize the interconnection between machines. Currently, deep-learning-based methods are widely applied for intelligent construction of the IIoT, so as to maximize the self-monitoring and self-management capabilities of various machines. However, the quantity and quality of data and the optimization of parameters greatly limit the properties of such methods. As a breakthrough of artificial intelligence (AI), deep reinforcement learning (DRL) provides inspiration and direction, which combines the advantages of deep learning and reinforcement learning to construct an end-to-end fault identification system. Therefore, a novel deep dual reinforcement learning model was proposed, which consisted of an actor model and a critic model. The dual structures avoid the over-self-optimization of the network. The action model continually learns the knowledge of identifying unknown samples by the$\varepsilon $-$greedy$algorithm, while the critic model dynamically adjusts the policy to guide the action model in right training direction. The effectiveness of the proposed method was verified by three bearing data sets. The results indicate that the proposed method enables agents to independently realize precise fault quantitative identification. The establishment of an experience storage unit overcomes the problem of insufficient samples, which avoids blind trial and error of the proposed mode.

With the increasing automation and integration of equipment, it is urgent to carry out anomaly detection (AD) for the large-scale system to ensure security, in virtue of Industrial Internet of Things (IIoT). Recently developed intelligent methods focus on component-level diagnosis or detection, resulting in difficulty in the health assessment of system with multisource data coupling. In addition, data-driven methods rarely emphasize the use of knowledge from the real physical system. In this article, we propose a full graph autoencoder to perform one-class group AD for the large-scale IIoT system. The proposed model takes as input data of normal status at training and only comprises several normalized graph convolutional layers, thus it is simple and fast. Different from Euclidean-based methods, the proposed model can handle various irregular structures together. For graph learning, multivariate time series are converted into graph data fused with prior knowledge. To achieve AD, we propose to reconstruct the full graph for the first time to obtain a reliable anomaly score. Besides, we extend a variational model to fully learn the graph representation. Moreover, a graph augmentation operation is employed to improve the accuracy and robustness. The proposed models are evaluated on two multisensor data sets from liquid rocket engine (LRE) systems, and the experimental results demonstrate the effectiveness and generalization of the IIoT system.

Anomaly detection (AD) for liquid rocket engine (LRE) is essential to improve the reliability and safety of space launch missions. However, it is difficult for existing methods to implement fast and precise detection with single-source information and absent anomalous samples. The Internet of Things (IoT) enables intelligent LRE AD via edge computing using multisource data. This article proposed an edge-based detection algorithm named stepwise generative adversarial network (StepGAN) for AD of LRE with multisource fusion at the Edge of IoT. StepGAN incorporates deep autoencoder and relativistic generative adversarial network. Through multistage stepwise training with unlabeled normal data, an encoder-generator-discriminator network is obtained to identify LRE anomalies at the edge. In addition, by fusing multisource information at feature level and aggregating neighboring information at decision level during real-time detection, the performance of StepGAN is further improved. To demonstrate its ability for IoT edge-based AD, the proposed method is compared with typical methods, using real static ignition data of LRE as cloud and edge data, and achieves the best result under multiple evaluation indexes. To evaluate the proposed fusing approach, a channel-varying case is carried out and extensive experimental results indicate its effectiveness.

In spite of widely discussed, drawing order recovery (DOR) from static images is still a great challenge task. Based on the idea that drawing trajectories are able to be recovered by connecting their trajectory components in correct orders, this work proposes a novel DOR method from static images. The method contains two steps: firstly, we adopt a convolution neural network (CNN) to predict the next possible drawing components, which is able to covert the components in images to their reasonable sequences. We denote this architecture as Im2Seq-CNN; secondly, considering possible errors exist in the reasonable sequences generated by the first step, we construct a sequence to order structure (Seq2Order) to adjust the sequences to the correct orders. The main contributions include: (1) the Img2Seq-CNN step considers DOR from components instead of traditional pixels one by one along trajectories, which contributes to static images to component sequences; (2) the Seq2Order step adopts image position codes instead of traditional points’ coordinates in its encoder-decoder gated recurrent neural network (GRU-RNN). The proposed method is experienced on two well-known open handwriting databases, and yields robust and competitive results on handwriting DOR tasks compared to the state-of-arts.

Class imbalance issue has been a major problem in mechanical fault detection, which exists when the number of instances presenting in a class is significantly fewer than that in another class. This article focuses on the problem of zero-shot fault detection of rolling bearing, which is the extreme case of class imbalance. Aiming at this problem, a two-stage zero-shot fault recognition method is proposed. First, inspired by the conditional generative adversarial network, a novel feature generating network which is composed of a feature extractor, a discriminator, and a generator is designed to capture the potential distribution of normal samples. Then, the generator will generate abundant pseudofault features by adding an additional sequence to the condition. Second, an improved deep neural network is trained with these synthetic pseudofault features as the classifier. Specially, a condition index is designed to represent different fault classes so that it can recognize the unseen fault samples. Finally, the effectiveness of the proposed method is verified by three datasets and a comparison method is also given to show the superiority. Results show that the feature generation network can effectively detect the typical faults even though the fault data are unavailable during training, which is practical for industrial application.

Data-driven intelligent method has shown superior performance in remaining useful life (RUL) prediction. However, the model training is difficult due to the limited degradation data. To address the challenges of small data set, a Sequence Adaptation Adversarial Network (SAAN) is proposed in this paper. SAAN could expand training data with auxiliary set by sequence domain adaption. We verify the proposed method with C-MAPSS dataset. By comparing with the literature methods, results show SAAN could significantly improve the accuracy of RUL prediction under small data set, and also keeps a competitive performance on sequence life prediction.

Recently, intelligent fault diagnosis has achieved fruitful research results. However, the small sample is still the major problem in fault diagnosis owing to lacking fault data of machines. In view of this, a signals augmented semi-supervised learning scheme is proposed for intelligent fault diagnosis in the case of small sample. In the proposed method, fault signal samples are generated by generative adversarial networks (GAN). The fault classifier is trained in a semi-supervised way using the generated samples and a small number of real samples. Besides, attention mechanism is applied in the fault classifier for sensitive feature extraction. The trained fault classifier is capable of accurate fault classification. Results indicate that the proposed method is effective in mechanical fault diagnosis under the small sample condition.

Data imbalance is a major problem in intelligent fault diagnosis. Aiming at this problem, the paper proposed a novel adversarial learning framework for zero-shot fault recognition of mechanical systems. The proposed network consists of three parts which are the feature extractor, the generator and the discriminator. Trained with normal samples, the proposed method is capable of generating unseen fault samples by changing the condition of the generator. After, these synthetic samples are used to train an improved deep neural network for fault recognition. Results show that the proposed method can recognize the unseen faults even though none of fault samples are available during training, which is meaningful for industry application.

Intelligent fault detection is an important application of artificial intelligence and has been widely used in many mechanical systems. The shipborne antenna that is a typical and an important mechanical system plays an irreplaceable role in ships. Considering the tough working environment and heavy background noise, fault detection is difficult for the shipborne antenna. Therefore, the paper presents an intelligent fault detection method via multiscale inner product with locally connected feature extraction for shipborne antenna fault detection. Inspired by inner product principle, this paper takes advantage of inner product to capture fault information in the vibration signals and detect the faults in rolling bearing of the shipborne antenna. Meanwhile, multiscale analysis is employed in two layers of the network to improve the feature extraction ability. The local features under different scales are collected and used for fault classification. Finally, the proposed method is verified by three datasets and comparison methods are also developed to show its superiority. Results show that the proposed method can learn sensitive features directly from raw vibration signals and detect the faults in rolling bearing of shipborne antenna effectively.

Confusion of frequency spectrum under varying speed condition leads to the efficacy degradation of fault recognition for rolling element bearings. Meanwhile, a widely-used method is transforming non-stationary vibration signals into angular domain based on angular re-sampling to reduce influence of speed fluctuation. However, single domain usually cannot describe health information comprehensively. Thus a multi-domain indexes description method is proposed in this paper to improve fault recognition by multi-domain fusion. First the input indexes of multiple domains including time domain, frequency domain, angular domain and order domain are organized to train SOM neural network. Then an optimized fault cognition model is established based on the SOM input planes. Finally, faults are recognized based on multi-domain indexes model and extra complexity indexes are added to enhance recognition of compound faults. Experimental results show that the proposed method can distinguish different faults effectively and have good practical significance.