The excitation and vibration triggered by the long-term operation of railway vehicles inevitably result in defective states catenary support devices. With massive construction high-speed electrified railways, automatic defect detection diverse plentiful fasteners on device is great significance for safety cost reduction. Nowadays, devices are periodically captured cameras mounted inspection during night, but still mostly relies human visual interpretation. To reduce involvement, this paper...

Split pins (SPs) play an important role in fixing joint components on catenary support devices (CSDs) of high-speed railway. The occurrence loose and missing defects SPs could make the structure CSDs unstable. In this paper, we present a three-stage automatic defect inspection system for mainly based improved deep convolutional neural network (CNN), which is called PVANET++. First, are localized by PVANET++ Hough transform & Chan-Vese model, then, three proposed criteria applied to detect...

Previous reinforcement learning (RL) methods suffer significant performance degradation or collapse when deployed to the real world due huge sim-real gap. This article proposes a hybrid offline-and-online meta-RL (HOMRL) algorithm that leverages prior task experience learn and adapt new pantograph active control tasks in real-world applications. The policy process consists of three phases: offline meta-policy pretraining, online adaptation, fine-tuning. First, we construct an approach learns...

Multiwavelets possess better properties than traditional wavelets. Multiwavelet packet transformation has more high-frequency information. Spectral entropy can be applied as an analysis index to the complexity or uncertainty of a signal. This paper tries define four multiwavelet entropies extract features different transmission line faults, and uses radial basis function (RBF) neural network recognize classify 10 fault types power lines. First, preprocessing postprocessing problems...

The isoelectric line is an important component that connects the steady arm and drop bracket of catenary in high-speed railway. loose strands can be commonly observed real-life applications. In this paper, automatic fault detection system for proposed. This consists three stages. First, a convolutional neural network adopted to extract features. To accurately quickly learn these features, improved feature extraction network, called as presented. Using images captured from inspection...

The interaction between the catenary and pantograph is one of most crucial factors that determine train operation in high-speed railway. bad state able to directly influence power supply safety traction system. In this paper, four aspects on research railway are reviewed detail, namely solution methods for equilibrium state, dynamic modeling catenary, non-contact detection static evaluation catenary. addition, their recent advances described. For low accuracy initial structure finding method...

This paper aims to develop a high-precision detection approach for catenary geometry parameters of electrical railway based on vision technology. The novel idea the proposed method is that, image coordinates spot formed contact line by laser emitter are mapped world system obtain conductor height and stagger. In system, key factors that affect accuracy include two components, localization correction value. To solve these problems, genetic particle filter algorithm swarm optimization...

Brace sleeve screws are an important component of catenary support devices in high-speed railways. According to the statistical data, there is a much risk for failure brace during operation system. In order obtain better results on fault diagnosis screws, it vital importance develop high-precision detection approach screws. With aim solving problems that size screw too small and employed previous methods difficult detect accurately, novel object network based improved Faster RCNN proposed....

The goal of this article is to develop a universal anomaly detection approach for catenary support components (CSCs) based on the generative adversarial networks (GANs). As long-term operation railway system, wide range failures, which threaten safe vehicles, perhaps happen CSCs. Until now, it hard design generic system recognize all these kinds failures because each defect needs special detecting algorithm different fault signatures. lack samples also makes difficult supervised learning...

This article aims to present a defect detection approach for catenary rod-insulators based on unsupervised learning. The idea of the proposed method is that 1) regions insulator pieces are separated first; 2) then, real samples segmented reconstructed as foreground, and through difference between images, background; 3) levels evaluated according regions. separation effectiveness foreground background in second stage most vital key. Therefore, reconstruction classification convolutional...

The pantograph-catenary system (PCS) is the essential power supply in high-speed railway, but its coupling performance influenced significantly by rapidly increasing train speed. actively controlled pantograph one of promising technologies to suppress fluctuation contact force (PCCF). In this paper, we propose a novel control strategy based on deep reinforcement learning (DRL) overcome complex time-varying characteristic PCS, which distorts identification classical methods. First, non-linear...

As for the rapid expansion of high-speed electrified railway industry, current collection quality electric locomotives becomes one crucial technical issues stable operation trains. The key measurement index that can characterize is contact force between catenary and pantograph on locomotive roof. One major factors affect wire irregularity, which appears to be very difficult measured or detected at present. In recent years, frequency-domain indicators pantograph-catenary (PCCF) have been...

The catenary support components (CSCs) are the most important devices in high-speed railways to contact lines for powering trains. To estimate states of CSCs, it is very necessary locate their positions a monitoring system based on computer vision. Considering application scenarios and characteristics an automatic quick positioning designed this paper simultaneously position multiscale CSCs with 12 categories. In system, effective framework called network (CSCNET) presented, which cascades...

Active pantograph control is the most promising technique for reducing contact force (CF) fluctuation and improving train's current collection quality. Existing solutions, however, suffer from two significant limitations: 1) they are incapable of dealing with various types, catenary line operating conditions, changing speeds, contingencies well 2) it challenging to implement in practical systems due lack rapid adaptability a new pantograph-catenary system (PCS) conditions environmental...

ABSTRACT This study investigates the effect of venting ratio on responding intensity aluminized explosives using a fast cook‐off test vessel with adjustable areas. The is defined as vent area to surface area, and represented by variable R . Using polymer‐bonded explosive TB‐1 (RDX/Al/KP/Binder = 55/35/5/5) subject, we recorded overpressure, reaction temperatures, heating times, optical signals generated during tests. damage residue were also analyzed. results indicated that time exhibit...

Brace sleeve (BS) fasteners, i.e., nut and bolt, are small components but play essential roles in fixing BS cantilever railway catenary system. They commonly inspected by onboard cameras using computer vision to ensure the safety of operation. However, most fasteners cannot be directly localized because they too inspection images. Instead, is first for detecting fastener. This leads a new problem that boxes may not contain complete due low localization accuracy, making it infeasible further...

As a kind of high-efficiency explosive with compound destructive capability, the energy output law thermobaric explosives has been receiving great attention. In order to investigate effects main components on characteristics explosives, various high and oxidants were selected formulate five different types explosive. Then they tested in both open space closed respectively. Pressure measurement system, high-speed camera, infrared thermal imager multispectral temperature system used for...