Search

Data augmentation (DA) has the potential to address the issue of imbalanced and insufficient datasets (I&ID) in pipeline fault diagnosis. However, the majority of existing DA methods for time series are inspired by computer vision techniques, ignoring the temporal dynamic properties and fine-grained fault features, which leads to limited performance of the augmentation. To tackle this problem, we introduce a novel DA approach called the subdomain-alignment adversarial self-attention network (SA-ASN), which takes into account both temporal association and semantic correlation. Our approach features a novel temporal association learning (TAL) mechanism, which transfers temporal information from the discriminator to the generator via a customized knowledge-sharing structure, improving the reliability of synthetic long-range associations. Additionally, we introduce a prototype-assisted subdomain alignment (PASA) strategy that forms a hierarchical structure in the synthetic dataset by incorporating local semantic correlation into the model training. With the support of TAL and PASA, our SA-ASN algorithm enhances the authenticity of temporal structure at the instance level and improves the discriminability of fault features at the category level. Our experimental results show that the SA-ASN algorithm provides a more diverse and accurate augmentation of pipeline data. The effectiveness of our SA-ASN algorithm encourages the use of data-driven diagnostic models in complex real-world oilfield pipeline networks.

This article is concerned with weak-shot learning, a practical yet challenging scenario in transfer learning where only a limited amount of weakly labeled data are available in the target domain. The key insights of weak-shot learning are focused on assigning fine-grained labels to target data and matching class-specific features across domains. In this article, a new prototype-assisted contrastive adversarial (PACA) network for weak-shot learning is proposed to make full use of the deterministic information from well-annotated data and the auxiliary information from weakly annotated data. Specifically, a prototypical pseudolabel learning mechanism is introduced to improve the credibility and robustness of pseudolabel estimation by fully exploiting prototype representations and weakly supervised information. Furthermore, a contrastive adversarial discrepancy strategy is developed to simultaneously reduce domain gaps at the global and local levels, providing compact intraclass features and distinguishable interclass features for weak-shot learning. The prototypical pseudolabel learning and contrastive adversarial discrepancy are designed to be updated alternately to eliminate pseudolabel noise in the target domain, which helps improve the transferability of domain-invariant features. Finally, extensive experiments are conducted on the cross-domain tasks of pipeline fault diagnosis, indicating that the proposed PACA network provides a promising tool for this practical industrial problem.

Healthy diet is essential for cardiovascular disease risk management, but its effects among Chinese patients, whose diets differ from Western diets, remain largely unknown.

The growing urbanization and the development of new energy automobiles have motivated a significant increase in the number of vehicles on the highway, which has put increasing pressure on existing transportation systems. Platooning control is an effective scheme to enhance road traffic safety and alleviate traffic congestion. In this paper, platooning control under multiple description encoding schemes (MDESs) is investigated for automated vehicles with nonlinear dynamics. A learning-based observer with MDESs is first utilized to acquire the vehicular state, where a neural network with a weight updating rule is created to approximate unknown nonlinearities. In light of estimated states as well as the features of encoding-decoding errors, some sufficient conditions are derived in the framework of the Lyapunov stability to attain the desired platoon with given constant inter-vehicle spacing. In addition, a set of matrix inequalities is used to solve the desired gain parameters based on the maximum and minimum eigenvalues of the communication topology. Finally, an illustrative example demonstrates the validity of the developed control scheme.

In this article, the recursive Bayesian estimation problem is investigated for a class of linear discrete-time systems subject to state-dependent packet dropouts. During the transmission to a remote estimator, the data packets carrying the local measurements might be dropped if the system state is located within certain occlusion region, and this gives rise to a nonstationary dropout process relying on real system states. In this scenario, due to the exponential growth of the computational cost, it is almost impossible to calculate the exact posterior distribution of the system state for the purpose of optimal state estimation. To address this issue, we propose a novel cross-coupled estimation framework consisting of two interactively working estimators, namely, a region-label estimator and a state estimator, where the former is utilized to obtain the optimal estimates of the region-label sequence in the maximum a posteriori sense, while the latter is adopted to achieve the optimal estimates of the system states in the minimum mean-square error sense. Moreover, a sufficient condition is obtained to ensure the mean-square boundedness of the resultant estimation error. The effectiveness of the proposed cross-coupled estimation framework is verified by a numerical simulation example.

This paper devotes itself to the issue of privacy-preserving distributed economic dispatch (ED) of microgrids. The desired algorithm in a distributed way not only ensures a balance between power supply and demand for the entire microgrid system at minimal cost but also effectively safeguards the confidentiality of sensitive information. To this end, both state decomposition and Laplace noises are embedded into a distributed optimization algorithm in the framework of consensus to improve the privacy-preserving capability. Then, both a sufficient condition about the strength of consensus coupling and the convergence accuracy induced by added noises are disclosed for the proposed algorithm via the Lyapunov stability theory and stochastic analysis. Furthermore, the privacy level is discussed in the framework of differential privacy and the lower bound of the convergence accuracy is found for the given privacy level and noise distribution. Finally, the effectiveness and correctness of our approach are demonstrated by applying an IEEE 39-bus system.

This article is concerned with the privacy-preserving distributed economic dispatch problem of microgrids. The main goal of this work is to develop a privacy-preserving distributed optimization algorithm over directed networks, aiming to achieve supply-demand balance at the lowest economic cost under practical constraints while preventing the leakage of power-sensitive information. For this purpose, a distributed optimization algorithm with a constant step size is proposed by combining the decentralized exact first-order algorithm with the push-sum protocol, which offers an advantage in terms of fast convergence. In addition, to ensure privacy preservation, a state-decomposition approach is employed by randomly dividing the state into two parts, where only partial state information is transmitted. Moreover, the effectiveness of the privacy-preserving scheme against honest-but-curious nodes and external eavesdroppers is demonstrated through rigorous analysis. Finally, simulation studies demonstrate the validity and superiority of the developed privacy-preserving distributed algorithm.

This paper investigates the anomaly-resistant decentralized state estimation (SE) problem for a class of wide-area power systems which are divided into several non-overlapping areas connected through transmission lines. Two classes of measurements (i.e., local measurements and edge measurements) are obtained, respectively, from the individual area and the transmission lines. A decentralized state estimator, whose performance is resistant against measurement with anomalies, is designed based on the minimum error entropy with fiducial points (MEEF) criterion. Specifically, 1) An augmented model, which incorporates the local prediction and local measurement, is developed by resorting to the unscented transformation approach and the statistical linearization approach; 2) Using the augmented model, an MEEF-based cost function is designed that reflects the local prediction errors of the state and the measurement; and 3) The local estimate is first obtained by minimizing the MEEF-based cost function through a fixed-point iteration and then updated by using the edge measuring information. Finally, simulation experiments with three scenarios are carried out on the IEEE 14-bus system to illustrate the validity of the proposed anomaly-resistant decentralized SE scheme.

The propagation velocity of seismic waves is a crucial parameter in seismic exploration, encompassing the entire process of seismic data acquisition, processing, and interpretation. Traditional model-driven full-waveform inversion (FWI) methods, which rely on an initial velocity, suffer from low computational efficiency. Conversely, data-driven deep-learning (DL) approaches heavily rely on extensive training data and lack interpretability due to overreliance on training data for generalization. To address these challenges, we present a seismic velocity inversion network model that incorporates prior knowledge and constraints based on physical laws. The proposed approach involves constructing a data-driven inversion network with dual encoders and single decoder structure, enabling the learning of nonlinear mappings from seismic data to velocity models. By incorporating prior well-logging data and attention mechanisms, the inversion process is improved. In addition, a seismic forward modeling network based on recurrent neural networks (RNNs) is developed to solve the acoustic wave equation. Leveraging the advantages of parallel computing, the forward modeling process achieves fast calculations. The automatic differentiation algorithm in DL facilitates gradient calculations, specifically back propagation of the residuals to incorporate the physical constraints. Ultimately, the proposed seismic velocity inversion network combines the two network structures while incorporating the constraint of wave field extrapolation law. Numerical experiments demonstrate that this network exhibits advantages in terms of result accuracy and model generalization.