In this paper, an edge computing system for IoT-based (Internet of Things) smart grids is proposed to overcome the drawbacks in current cloud paradigm power systems, where many problems have yet be addressed such as fully realizing requirements high bandwidth with low latency. The new mainly introduces traditional cloud-based and establishes a hardware software architecture. Therefore, considerable amount data generated electrical grid will analyzed, processed, stored at network. Aided...

In this paper, a deep learning (DL)-based physical (PHY) layer authentication framework is proposed to enhance the security of industrial wireless sensor networks (IWSNs). Three algorithms, neural network (DNN)-based nodes' method, convolutional (CNN)-based and convolution preprocessing (CPNN)-based have been adopted implement PHY-layer in IWSNs. Among them, improved CPNN-based algorithm requires few computing resources has extremely low latency, which enable lightweight multi-node...

Industrial robots are one of the most typical machines in smart manufacturing systems. Their joint bearing faults account for a significant portion failures. Data-driven fault diagnosis methods, especially deep learning have become research hotspot due to development industrial Internet Things and big data. However, varying working conditions robots, such as continuous changing load speed, challenge existing data-driven methods. Although adversarial-based domain adaptive methods promising...

In this study, a method based on Differential Privacy Federated Learning (DPFL) is adopted, which suitable for safe and efficient processing of power grid data. Combined with new algorithm called DP-FedSAM, combines the sharpness-aware minimization (SAM) optimizer differential privacy technology, it aims to solve problems protection model performance degradation in data analysis. DP-FedSAM avoids need centralized storage by performing training locally each operating node, thereby reducing...

In this paper, we investigate the security threats in mobile edge computing (MEC) of Internet things, and propose a deep-learning (DL)-based physical (PHY) layer authentication scheme which exploits channel state information (CSI) to enhance MEC system via detecting spoofing attacks wireless networks. Moreover, three gradient descent algorithms are adopted accelerate training deep neural networks, enables smaller computation overheads lower energy consumptions. addition, maximum likelihood...

Network intrusion detection system (NIDS) plays an important role in network security. It can detect the malicious traffic and prevent intrusion. Traditional methods used machine learning techniques such as support vector machine, Bayesian classification, decision tree k-means. The traditional first need to manually select features has obvious limitations. In this paper, we propose a novel NIDS based on convolutional neural network. We train deep-learning models using both extracted original...

Pressure transmitters are widely used in the process industry for pressure measurement. The sensing line, a core component of sensor transmitter, significantly impacts accuracy transmitter’s output. reliability is critical nuclear power industry. Blockage recognized as common failure lines; therefore, novel detection method based on Trend Features Time–Frequency domain characteristics (TFTF) proposed this paper. dataset comprises both fault and normal data. This innovatively integrates...

Polymerase chain reactions (PCR), thermally activated chemical which are widely used for nucleic acid amplification, have recently received much attention in microelectromechanical systems and micro total analysis because a wide variety of DNA/RNA molecules can be enriched by PCR further analyses. In the present work, droplet-based oscillating-flow chip was designed fabricated silicon microfabrication technique. Three different temperature zones, were stable at denaturation, extension...

Industrial and automation control systems require that data be delivered in a highly predictable manner terms of time. Time-sensitive Networking (TSN), an extension the Ethernet, is set protocols developed maintained by IEEE 802.1 Task Group; deal with time synchronization, traffic scheduling, network configuration, etc. TSN yields promising solutions for real-time deterministic networks. Here, we develop simulation model based on OMNET++; TSN-enabled switch schedules using gate lists...

Most of the physical-layer authentication methods are based on pre-knowledge channel information and threshold is fixed before communication process. In this paper, we propose a physical layer method machine learning with dynamic in wireless environment. method, differences determined by latest system state according to ε-greedy strategy. The simulation results illustrate efficiency method.

This paper deals with the fault detection of a pneumatic control valve using canonical variate analysis (CVA). CVA can find optimal linear combinations p-window and f-window data, so that correlation between these be maximized. Based on CVA, data is considered by traditional hotelling T <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> statistic squared prediction error (SPE) indicators, corresponding rates (FDR) are low. In order to...

With the wide application of PMU in power system, real time prediction and emergency control system transient instability using wide-area measurements has brought attention. There are two sub-problems: postfault rotor-angle trajectory detection. three categories for former sub-problem. They network reduction based super-real simulation, curve-fitting extrapolation angle speed integration methods. The includes polynomial function, auto regression model trigonometric function model. types...

The prognosis of time‐to‐failure for a battery can avoid the failure caused by performance loss. In this paper, novel and effective algorithm is proposed to predict remaining useful life lithium‐ion batteries. extended Kalman particle filter used improve degradation problem existing in standard algorithm. order fit capacity degradation, transformed model based on double exponential empirical model. It reduce number parameters training difficulty parameters; it also matches form state...

With the increasing demand of fuel and emission hybrid electric vehicles (HEV) with dual power sources, engine motor, has been one development ways clean automobiles. In research on vehicles, control powertrain is key issue. According to parallel this paper used fuzzy realize assist strategy membership function was optimized for economy decreasing emissions

With the development of Internet technology, software vulnerabilities have become a major threat to current computer security. In this work, we propose vulnerability detection for source code using Contextual LSTM. Compared with CNN and LSTM, evaluated CLSTM on 23185 programs, which are collected from SARD. We extracted features through program slicing. Based features, used natural language processing analysis programs code. The experimental results demonstrate that has best performance...

This paper investigates the physical layer (PHY-layer) authentication that exploits channel state information (CSI) to enhance multiple-input multiple-output orthogonal frequency division multiplexing (MIMO-OFDM) system security by detecting spoofing attacks in wireless networks. A multi-user is proposed using convolutional neural networks (CNNs) which also can distinguish spoofers effectively. In addition, mini batch scheme used train and accelerate training speed. Meanwhile, L1...

In this paper, we propose a clustering based physical-layer authentication scheme (CPAS) to overcome the drawback of traditional cipher-based schemes that suffer from heavy costs and are limited by energy-constrained intelligent devices. CPAS is novel cross-layer secure approach for edge computing system with asymmetric resources. The combines lightweight symmetric cipher channel state information provide two-way between terminals By taking advantage temporal spatial uniqueness in physical...

Data-driven bearing-fault diagnosis methods have become a research hotspot recently. These to meet two premises: (1) the distributions of data be tested and training are same; (2) there large number high-quality labeled data. However, machines usually work under different working conditions in practice, which challenges these prerequisites due fact that different. In this paper, one-dimensional Multi-Scale Domain Adaptive Network (1D-MSDAN) is proposed address issue. The 1D-MSDAN kind deep...

Industrial communication requires highly reliable networks with hard temporal constraints. Thus, the IEEE 802.1 Task Group proposed a promising technology, namely, time-sensitive networking (TSN), to complement determinism and real-time (RT) capabilities of Ethernet via set enhanced standards. To explore feasibility applicability TSN, we developed simulation model for TSN using module-based design method. Our TSN-compliant modules implement time-based traffic scheduling functionality...

A small coil was made and tested for the purpose of magnet technology development YBa2Cu3O7−x (YBCO) high field coils. The has a winding inner diameter 24 mm, outer 36 length 46 mm in pancake wound construction 5 double pancakes. YBCO conductor is SuperPower SCS4050 with 2×20 μm copper stabilizer. user facility NHMFL 31 T background magnet. At this field, operated to current 325 produced increment 2.8 record total 33.8 T. average density windings at full very 446 A/mm2. test complicated by...