Abstract Non-resonant energetic particle modes (EPMs) are extremely common in tokamak experimental phenomenon, which can disrupt the plasma balance, subsequently reducing device confinement performance. Nevertheless, it should be stressed that hybrid simulation of EPMs requires considerable time and computational resources study such phenomena. To solve this issue, research proposes Machine Learning (ML) approaches to predict linear instability non-resonant EPMs. Here, current compares four...

Extracellular vesicles (EVs) are important paracrine mediators derived from various cells and biological fluids, including plasma, that capable of inducing regenerative effects by transferring bioactive molecules such as microRNAs (miRNAs). This study investigated the effect mesenchymal stem cell-derived extracellular (MSC-EVs) isolated umbilical cord blood human plasma-derived (UCB-EVs) on wound healing scar formation reduction. Spatial transcriptomics (ST) was used to MSC-EVs UCB-EVs...

Abstract Improving the prediction accuracy and stability of non-resonant high-order harmonics energetic particle modes (EPMs) is crucial for achieving tokamak plasmas steady state operation magnetohydrodynamics (MHD) activities control. This study proposed a novel stacking ensemble learning model EPMs prediction. adopts 2-layer structure with base learner meta-learner: first layer, learners include K-Nearest Neighbor regression (KNN), Extreme Gradient Boosting (XGBoost), Regression (GBR),...

Industrial sensor signals are essentially non-Euclidean graph structures due to the interplay between process variables; thus, convolutional networks (GCNs) have been widely studied and applied. However, most of existing GCN-based methods may suffer from two drawbacks: 1) it is difficult characterize multiple interactions among nodes 2) input constructed original data contain errors missing edges, which will degenerate fault diagnosis performance. To address abovementioned issues, this...

Existing methods regarding the influential nodes identification in complex networks usually assume that structures of are fully known. However, many cases, knowing full structure one network is hard or impossible, and each participant can only obtain partial networks. Therefore, be viewed as a private (PN) original network, they form multiple PNs. Then, question arises: how to collaboratively identify PNs while protecting privacy PN. To this end, secure multiparty computation ranking...

In general, the safety and efficiency of thermal power plants require collaboration multiple coal mills. However, running data different mills will introduce significant inconsistent distribution, resulting in suboptimal performance or even unavailability conventional diagnosis methods. For this end, paper presents an advantageous discriminant analysis aided collaborative alignment network (DA-CAN) for cross-device fault diagnosis. Firstly, contribution each feature distinguishing source...

Recently, the problem of user identification across multiple social networks (UIAMSNs) has attracted considerable attention since it is a prerequisite for many downstream tasks and applications. Although substantial network feature-based approaches have been proposed to solve UIAMSNs' problem, matching degree in most current works given by experience, which lacks solid theoretical basis. To alleviate above predicament, we propose algorithm based on naive Bayes model (UI-NBM) within...

Remaining useful life (RUL) prediction of bearings has extraordinary significance for prognostics and health management (PHM) rotating machinery. RUL approaches based on deep learning have been dedicated to finding a nonlinear mapping relationship between non-stationary monitoring data RUL. However, most existing pay little attention the degradation trend diverse stages bearing lack discriminative power crucial features, resulting in loss some important information associated with To address...

Federated learning (FL)-based fault diagnosis is being widely developed. However, most of the existing FL methods may suffer from two drawbacks: 1) they are limited to a single task, and this be insufficient when comprehensive health status information needed 2) them work offline, thus neglecting useful contained in newly collected operation data. For end, article proposes multitask federated incremental (multitask-FIL) framework. First all, feature sharing network established by assigning...

Due to the continuous technological innovation in industrial processes, many deep learning based methods have shown powerful capability handing equipment status monitoring, but most of them ignore temporal features and dynamic changes diverse spatial structure raw data. Meanwhile, these usually focus on handling a single downstream task, rarely consider different tasks simultaneously. To solve issues, this paper proposes more flexible monitoring framework dynamic-multilayer graph convolution...

In many cases, a network may be dispersedly recorded by different participants and each participant is one part of the original network, no willing to share its data due commercial competition. Therefore, forms an independent private they form " multiple networks" regarding network. Existing methods only use structure itself predict missing links, leading underutilized information deteriorated prediction accuracy. One natural question arises: how integrate networks formulating security...

Abstract User alignment across online social network platforms (OSNPs) is a growing concern with the rapid development of internet technology. In reality, users tend to register different accounts on multiple OSNPs, and are reluctant share structure user’s information due business interest privacy protection, which brings great obstacles cross-platform user alignment. view this, we propose homomorphic encryption-based (HE-SNA) algorithm from perspective leakage. Specifically, first consider...

Local Fisher discriminant analysis (LFDA) has been widely applied to dimensionality reduction and fault classification fields. However, it often suffers from small sample size (SSS) problem incorporates all process variables without emphasizing the key faulty ones, thus leading degraded diagnosis performance poor model interpretability. To this end, paper develops sparse selection based exponential local (SELFDA) model, which can overcome two limitations of basic LFDA concurrently. First,...

Abstract Real-time prediction of the growth rate internal kink modes and timely implementation control measures are crucial for stable operation tokamak devices. In this study, machine learning techniques combined with three-dimensional toroidal nonlinear magnetohydrodynamic code CLT (Ci-Liu-Ti) to enable real-time modes. The is first utilized generate a dataset mode rates 15 input features. Based on dataset, four algorithms applied predict within timescale less than 25 microseconds....

Fault diagnosis of industrial equipments is extremely important for the safety requirements modern production processes. Lately, deep learning (DL) has been mainstream fault tool due to its powerful representational ability in and flexibility. However, most existing DL-based methods may suffer from two drawbacks: Firstly, only one metric used construct networks, thus multiple kinds potential relationships between nodes are not explored. Secondly, there few studies on how obtain better node...

The traditional kernel principal components analysis (KPCA) and linear discriminant (LDA) have been verified to be two effective approaches for fault detection diagnosis in recent years. Nevertheless, the conventional method corresponding improved ones still exposed their deficiencies some ways. Facing this dilemma, paper presents a combination of optimized KPCA modified LDA (OKPCA-MLDA), which OKPCA avoids loss original features after centralizing data eigenspace by adjusting covariance...

Existing methods cannot depict the intrinsic correlations of time series accurately. Aiming at above problem, this paper proposes a novel causal analysis scheme called GWO-KGC, which incorporating grey wolf optimization algorithm into kernel-ganger analysis. Specifically, model removes redundant and unrelated variables, also recognizes complex nonlinear relationships between thus providing an optimal subset input features. Meanwhile, in order to verify correctness results, GWO is introduced...

Predicting the remaining useful life (RUL) of bearings is important for secure running and reducing maintenance costs rotating machinery. In this article, a novel feature extraction RUL prediction framework bearing proposed by utilizing time-frequency representations (TFRs) deep neural network with double attention (DA-DNN). Firstly, TFRs are employed to extract degradation features associated prediction, which can effectively cope non-stationary properties vibration data. The reduced...