Search

Emotion causes constitute a pivotal component in the comprehension of emotional conversations. Recently, a new task named Causal Emotion Entailment (CEE) has been proposed to identify the causal utterances for the target emotional utterance in a conversation. Although researchers have achieved some progress in solving this problem, they failed to adequately incorporate speaker characteristics and overlooked the effects of temporal relations in conversation structures. To fill such a research gap to some extent, we propose a novel causal emotion entailment framework, namely MPEG (Multi-Perspective Enhanced Graph attention network). The training of MPEG consists of three stages. First, we utilize a speaker-aware pre-trained model and two attention mechanisms to obtain the utterance representations that incorporate local contexts as well as the speaker and emotional information. Then, these representations are fed into a graph attention network to model the conversation structures and emotional dynamics from both local and global perspectives. Finally, a fully-connected network is implemented to predict the relationships between emotional utterances and causal utterances. Experimental results show that MPEG achieves state-of-the-art performance. [ABSTRACT FROM AUTHOR]

Bridges are exposed daily to environmental and operational factors that may cause weariness, fatigue, and damage. Continuous structural health monitoring (SHM) has been crucial to ensuring public safety, preventing accidents, and avert costly damages. In this regard, advances in Machine Learning and Big Data technologies have enabled automated, real-time structural monitoring. However, challenges persist, notably the scarcity of labeled data, rendering supervised learning impractical. Additionally, state-of-the-art methods demand extensive training data to generalize and achieve satisfactory performance, which can be limited in real-world scenarios. This paper presents a novel three-step method supported by advanced Machine Learning and signal processing techniques aimed at detecting anomalous signals. This method is trained solely on structural acceleration signals, eliminating the need for labeled data. Among the contributions of this work, it can be mentioned that a remarkable accuracy in the detection of structural damage was demonstrated quantitatively. (F1 Score of 93%), while requiring significantly less training data volume than alternative methods (less than 25% of the total) and opening up different lines of research.

In recent times, digital transformations and Industry 4.0 have revolutionized real-time bridge monitoring and its inspection. The use of smart Structural Health Monitoring (SHM) techniques is becoming powerful with the competencies of Building Information Modeling (BIM) tools, Artificial Intelligence (AI), Internet of Things (IoT), and Virtual/Augmented (VR/AR) technologies. However, the lack of interconnectivity between these tools limits their functionality. This research has addressed this problem by developing an integrated framework to assess serviceability and implement a smart SHM for a newly constructed extradosed bridge. Using Finite Element Analysis (FEA), the study proposes an integrated SHM system that utilizes various IoT sensors, including Wired Strain Gauges (WSG), Liquid Levelling Sensors (LLS), MEMS accelerometers, and a Weather Monitoring Station (WMS) to monitor concrete deformations, vertical displacements, structural vibrations, and weather conditions. BIM tool is used to develop the virtual replica of the proposed SHM system which is then used in the 3D Game Engine (GE) to develop an AR application. This application is then successfully deployed and tested in the AR headset (HoloLens) where its capabilities for onsite bridge health monitoring are discovered. This approach overcomes the limitations of HoloLens devices by providing real-time access to SHM data through a web platform, enabling on-site or remote AR-based bridge health monitoring. Conclusively, this paper emphasizes the numerical modeling of bridges for the design of a health monitoring system, that highlights the importance of robust SHM techniques in assessing bridge conditions. Moreover, it introduces a novel approach for smart bridge inspection and onsite visualization of structural defects in an AR environment.

The emergence of FluidFM technology has presented a novel opportunity for selective aspiration and distribution of liquids at the sub-micron scale. However, precise control over droplet generation volume remains a challenging issue. This study investigates the impact of dynamic parameters between the tip and surface on droplet formation. Initially, we establish a finite element model to simulate the process from the formation of a liquid bridge between the hollow cantilever probe and the substrate to the rupture resulting in droplet formation. Additionally, we propose an image-based method for quantifying droplet volume. Subsequently, we construct a comprehensive experimental framework to delve into the influence of these parameters on microdroplet formation by monitoring the dynamic interaction parameters between the probe tip and the liquid surface. Finally, we optimize the parameters during the droplet generation process, enabling the formation of uniform arrays of microdroplets under controlled experimental conditions. Our study provides a reference for automation parameters of serial experiments and improves experimental throughput. [ABSTRACT FROM AUTHOR]

Aspect-based sentiment analysis is an important task in fine-grained sentiment analysis, which aims to infer the sentiment towards a given aspect. Previous studies have shown notable success when sufficient labeled training data is available. However, annotating adequate data is labor-intensive, which sets substantial barriers for generalizing the sentiment predictor to the new domain. Two main challenges exist in cross-domain aspect-based sentiment analysis. One challenge is acquiring the domain-invariant knowledge; the other challenge is mining the syntactic-related words towards the aspect-term. In this article, we propose a transformer-based semantic-primary knowledge transferring network (TSPKT) for cross-domain aspect-term sentiment analysis, which utilizes semantic-primary knowledge as a bridge to enable knowledge transfer across domains. Specifically, we first build an S-Graph from external semantic lexicons, and extract the semantic-primary knowledge from the S-Graph. Second, AoaGraphormer is proposed to learn the syntactically relevant words towards the aspect-term. Third, we extend the standard biLSTM classifier to fully integrate the semantic-primary knowledge by adding a novel knowledge-aware memory unit (KAMU) to the biLSTM cell. Extensive experiments on six cross-domain setups demonstrate the superiority of TSPKT against the state-of-the-art baseline methods. [ABSTRACT FROM AUTHOR]

In this article, a low computational burden model predictive control (MPC) strategy without weighting factor is proposed for the single-phase cascaded H-bridge CHB converters. To reduce the switching state candidates, a hierarchy control algorithm is proposed. The grid current is controlled by selecting a subregion from the designed 2-D control plane, instead of the entire area. Two vectors are chosen in one sampling period for more accurate tracking. Then, the voltage balancing is achieved by selecting the optimal switching state from the subregion candidates to form the above two vectors. The cost function can be constructed of one variable: load voltage. Therefore, the weighting factor can be eliminated. No tuning or retuning processes are required in the proposed method. To reduce the computational time further, the principle of eliminating the switching state candidates operating the same voltage balancing performance is proposed. Conventional and proposed MPC methods are verified by experimental tests via a laboratory setup of a three-cell connected CHB converter. Steady-state and transient operations demonstrate that the proposed method guarantees less distortion grid current and shorter execution time (reduced from 15 to 3 $\mu s$). Fast response speed to variations in voltage reference and load resistance can be achieved t. [ABSTRACT FROM AUTHOR]

In this article, a fixed on-time modulation (FTM) method for half-bridge series resonate converters (SRCs) is proposed to achieve the full range regulation for all loads. Additionally, the proposed method features low current stresses, zero voltage switching turn-on for transistors and zero current switching turn-off for diodes. A comprehensive analysis of the operation principle, voltage gain curves, component stresses, and comparisons with other methods is given. Besides, the extension of the FTM method to a family of half-bridge SRCs is discussed and the voltage gains of different topologies are provided. A 200 W prototype of the half-bridge SRC was built and tested to verify the theoretical analysis. [ABSTRACT FROM AUTHOR]

The goal of this article is to develop a new topology optimization framework that can secure the manufacturability of the synchronous reluctance motors (SynRMs) with multiple iron webs and bridges while satisfying the required electromagnetic and mechanical performances. In the SynRM, multiple iron webs and bridges are crucial to achieve a higher torque and a structural safety. A minimum thickness of the web and bridge should also be guaranteed to achieve manufacturability. To acquire the optimal SynRM that can satisfy the above characteristics, a two-stage topology optimization framework is proposed with an individual filtering-and-penalization scheme. This proposed scheme can individually control the number and thickness of multiple iron webs and bridges in the rotor to obtain the manufacturable and structurally safe SynRMs. For the abovementioned purpose, topology optimization is formulated to minimize a torque ripple while satisfying the desired average torque, structural compliance, and a rotational moment of inertia. Average torque and structural compliance are evaluated by performing the electromagnetic and structural finite element analyses, respectively, at each iteration of topology optimization. Here, design-dependent loads are considered in the radial and circumferential directions to reflect a high-speed rotation. Through investigating the optimized SynRM designs with various filter radii, the relationship among the torque, structural pattern, and stress distribution is examined. Finally, the experimental results with the manufactured prototypes validate the feasibility and potential of the proposed design method. [ABSTRACT FROM AUTHOR]

A major gap between few-shot and many-shot learning is the data distribution empirically oserved by the model during training. In few-shot learning, the learned model can easily become over-fitted based on the biased distribution formed by only a few training examples, while the ground-truth data distribution is more accurately uncovered in many-shot learning to learn a well-generalized model. In this paper, we propose to calibrate the distribution of these few-sample classes to be more unbiased to alleviate such an over-fitting problem. The distribution calibration is achieved by transferring statistics from the classes with sufficient examples to those few-sample classes. After calibration, an adequate number of examples can be sampled from the calibrated distribution to expand the inputs to the classifier. Specifically, we assume every dimension in the feature representation from the same class follows a Gaussian distribution so that the mean and the variance of the distribution can borrow from that of similar classes whose statistics are better estimated with an adequate number of samples. Extensive experiments on three datasets, miniImageNet, tieredImageNet, and CUB, show that a simple linear classifier trained using the features sampled from our calibrated distribution can outperform the state-of-the-art accuracy by a large margin. Besides the favorable performance, the proposed method also exhibits high flexibility by showing consistent accuracy improvement when it is built on top of any off-the-shelf pretrained feature extractors and classification models without extra learnable parameters. The visualization of these generated features demonstrates that our calibrated distribution is an accurate estimation thus the generalization ability gain is convincing. We also establish a generalization error bound for the proposed distribution-calibration-based few-shot learning, which consists of the distribution assumption error, the distribution approximation error, and the estimation error. This generalization error bound theoretically justifies the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Aging bridge infrastructure may jeopardize safety, and its economic impact has raised concerns worldwide. Reinforced concrete is a major component of bridge infrastructure, whose condition is directly related to reliability and safety. However, evaluating and maintaining reinforced concrete remains a complex issue, as it is often exposed to a dynamic and stochastic environment. In this article, we provide an integrated approach to capturing the characteristics of reinforced concrete deterioration by synthesizing knowledge from the physical model and a stochastic model. Then, a semi-Markov decision process is constructed for optimizing the maintenance of reinforced concrete. To improve the computation efficiency of the approach, we design the policy iteration algorithm with finite evaluations to reduce computation cost. The designed approach can further calibrate the temporal and complex dynamic information of the reinforced concrete deterioration and has a better performance than that of the Markov decision process. Finally, we explore the sensitivity of the maintenance policy under the changing of cost rates and durations of the maintenance activities. [ABSTRACT FROM AUTHOR]

The question of how institutional entrepreneurs working on megaprojects successfully enact change in regard to logic practices remains an unsolved problem in megaproject management literature. Building on an in-depth case study of a megaproject in China, we examine how institutional entrepreneurs manage in order to transform a conventional construction logic into a new logic of “mechanization, autoassembling, and factory construction” in steel bridge industry. We present a model of institutional work conducted by entrepreneurs for institutional entrepreneurship in the megaproject platform. Based on this case, we highlight the significance of institutional compatibility and institutional governmentality in implementing the new logic practices. This article contributes to the literature by examining the different forms of institutional work conducted by entrepreneurs with the aim of driving the institutional changes, and exploring the relationship between megaproject management and institutional entrepreneurship. [ABSTRACT FROM AUTHOR]

Interfacing low voltage dc to medium voltage three-phase ac grid is often based on series-stackable modular converter architectures. To minimize energy storage requirements, it is advantageous to employ a quadruple active bridge (QAB) stage operating as a “dc transformer” in each stackable module. The QAB stage offers three isolated dc link voltages, which then allow for flexible stacking of three single-phase dc-to-ac inverter stages. Each of the module phases processes a pulsating power having a component at twice the line frequency. This presents a challenge in maintaining zero voltage switching (ZVS) on the secondary sides of the QAB during low-power portions of the line cycle. This article is focused on the design of the QAB stage. A detailed analysis of ZVS switching waveforms is presented, including effects of nonlinear device capacitances. It is shown how ZVS can be achieved at all times using a relatively small circulating current provided by the magnetizing inductance of the high-frequency transformer. Analytical expressions are given for the optimal values of the magnetizing inductance and the dead times of the QAB primary and secondary bridges. The approach is verified by experimental results on a 1 kV, 10-kW SiC-based prototype, demonstrating a relatively flat efficiency curve with a peak efficiency of 97.1% at 75% load. [ABSTRACT FROM AUTHOR]

Offline reinforcement learning (RL) algorithms seek to learn an optimal policy from a fixed dataset without active data collection. Based on the composition of the offline dataset, two main methods are used: imitation learning which is suitable for expert datasets, and vanilla offline RL which often requires uniform coverage datasets. From a practical standpoint, datasets often deviate from these two extremes and the exact data composition is usually unknown. To bridge this gap, we present a new offline RL framework, called single-policy concentrability, that smoothly interpolates between the two extremes of data composition, hence unifying imitation learning and vanilla offline RL. Under this new framework, we ask: can one develop an algorithm that achieves a minimax optimal rate adaptive to unknown data composition? To address this question, we consider a lower confidence bound (LCB) algorithm developed based on pessimism in the face of uncertainty in offline RL. We study finite-sample properties of LCB as well as information-theoretic limits in multi-armed bandits, contextual bandits, and Markov decision processes (MDPs). Our analysis reveals surprising facts about optimality rates. In particular, in both contextual bandits and RL, LCB achieves a fast convergence rate for nearly-expert datasets, analogous to the one achieved by imitation learning, contrary to the slow rate achieved in offline RL. In contextual bandits, we prove that LCB is adaptively optimal for the entire data composition range, achieving a smooth transition from imitation learning to offline RL. We further show that LCB is almost adaptively optimal in MDPs. [ABSTRACT FROM AUTHOR]

Network theory has a big role in driving fields like communication, e-commerce, and internet. There are several graph theoretical structures to represent natural and man-made networks. Stability and connectivity are two major factors associated with networks. A fuzzy graph is an effective representation for a weight normalized network. Different connectivity parameters in fuzzy graphs can be used to find stability and optimum flow in a network. In this article, we discuss cyclic connectivity, one of the major parameter that is closely related with cyclic reachability within a normalized network. Cyclic connectivity status and cyclic status sequence associated with a fuzzy graph are discussed. A classification of vertices of a network based on the cyclic connectivity status is given. Algorithms related to the topic along with their illustration are presented. Scope and significance of this parameter are presented with the help of real-life applications. [ABSTRACT FROM AUTHOR]

A graph is an abstract model that represents relations among entities, for example, the interactions between characters in a novel. A background story endows entities and relations with real-world meanings and describes the semantics and context of the abstract model, for example, the actual story that the novel presents. Considering practical experience and prior research, human viewers who are familiar with the background story of a graph and those who do not know the background story may perceive the same graph differently. However, no previous research has adequately addressed this problem. This research article thus presents an evaluation that investigated the effects of background stories on graph perception. Three hypotheses that focused on the role of visual focus areas, graph structure identification, and mental model formation on graph perception were formulated and guided three controlled experiments that evaluated the hypotheses using real-world graphs with background stories. An analysis of the resulting experimental data, which compared the performance of participants who read and did not read the background stories, obtained a set of instructive findings. First, having knowledge about a graph's background story influences participants’ focus areas during interactive graph explorations. Second, such knowledge significantly affects one's ability to identify community structures but not high degree and bridge structures. Third, this knowledge influences graph recognition under blurred visual conditions. These findings can bring new considerations to the design of storytelling visualizations and interactive graph explorations. [ABSTRACT FROM AUTHOR]

Collaborative filtering (CF) is a classical model for recommender systems. Though neural network-based CF models have been shown to perform well in some cases, they still suffer from the sparsity and cold-start issues. To address these issues, we propose a novel neural network-based CF model, termed the cross-domain explicit–implicit-mixed CF neural network (CEICFNet). The proposed model utilizes deep neural networks to learn latent factors not only from the explicit ratings and the implicit interactions but also in a cross-domain manner. In particular, domain-shared multilayer perception (MLP) networks are designed to learn the user rating latent factors and the user interaction latent factors from the explicit ratings and the implicit interactions, respectively, which act as bridges for transferring knowledge across domains. On the other hand, domain-specific MLP networks are designed to learn the item rating latent factors and the item interaction latent factors from the explicit ratings and the implicit interactions, respectively. Then, in each domain, based on the user rating latent factors and the item rating latent factors, the rating predictive representation for each user–item pair can be learned by an MLP. Similarly, the interaction predictive representation for each user–item pair can be learned. For integrating the explicit ratings and the implicit interactions, in each domain, a fully connected layer is used to automatically assign different weights to the rating predictive representations and the interaction predictive representations, based on which the final interaction probability can be generated. Extensive experiments are conducted on five datasets and the results have confirmed the effectiveness of our model. [ABSTRACT FROM AUTHOR]

Bridges play an important role in transportation infrastructure systems. Intelligent and digital management of bridges group is an essential part of the future intelligent transportation infrastructure system. This paper proposes a digital twin system for bridges group in the regional transportation infrastructure network, which is interconnected by measured traffic loads. In physical space, a full-bridge traffic load monitoring system based on information fusion of weigh-in-motion (WIM) and multi-source heterogeneous machine vision is set up on the target bridge to measure traffic loads, also lightweight sensors are employed on the bridges group for structural response information. Furthermore, by establishing mechanical analysis models in the corresponding digital space and using the measured traffic loads as links, the working condition perception and safety warning of all bridges in the regional transportation network is achieved, forming an important support for further intelligent transportation infrastructure system. The proposed digital twin system has been preliminarily implemented in a bridges group around Shanghai, China, demonstrating the feasibility of the technical framework proposed in this paper and the bright prospects. [ABSTRACT FROM AUTHOR]

Conventional diagnosis methods are performed at the library cell level, which identify only defective cells. In reality, many defects affecting the inside of the cells cannot be identified, limiting yield improvement. To overcome the limit, cell-aware diagnosis has been introduced, but it is rarely used in scan chain domain due to the characteristics of a scan cell. Design of the scan cell is complex, but there are few I/O pins to control and observe. In this brief, the scan cell design is modified to observe more values inside the cell, distinguishing the many defect that could not be identified. Since the proposed method is not affected by logic and proceeds at the layout level of the scan cells, it is immune to type, size, or complexity of the circuit. Based on the layout data and some analog simulations, a new modified cell view for the intracell scan chain diagnosis is developed. [ABSTRACT FROM AUTHOR]

In recent years, owing to important improvements, the applicability of supervisor synthesis has significantly increased. We discuss notable developments that were pivotal in the application of supervisor synthesis to large infrastructural systems. [ABSTRACT FROM AUTHOR]

In this article, two configurations of the quadratic quasi-Z-source (Q-qZS) isolated dc–dc converters are presented, namely continuous input current and discontinuous input current configurations. The operating principles and steady state of the continuous input current topology analyzed in this article is characterized with the following features: 1) wider operating range due to buck and boost functions in a single stage; 2) draws continuous input current; 3) low input current ripple compared with other qZS-based isolated dc–dc converters; 4) better reliability and good immunity against EMI noise; 5) common ground between the input source and the inverting bridge; and 6) continuous control over the entire operating range. Compared with other qZS-based isolated dc–dc converters, lower current stresses on active and passive components used are realized. Hence, the total magnetic volume/energy required by the proposed converter is reduced with the potential of achieving high power density. A prototype of 40–140 V input voltage, 300 V output, and 550 W rated power is developed to validate the performance of the proposed converter presented. [ABSTRACT FROM AUTHOR]

Multiactive bridge (MAB) converters are attracting increasing attention due to their advantages of mutual electrical isolation and bidirectional power flow capability among different ports. However, the presence of transient dc bias in MAB can increase the current stress and lead to transformer saturation, thus affecting the system reliability. This letter establishes a quantitative and general description of the transient dc bias based on the instantaneous middle current of the high-frequency transformer and proposes a general dynamic modulation method to eliminate the bias, which is robust irrespective of the phase-shift modulation methods and the power flow directions. Also, it works at both the normal and startup modes. More importantly, the proposed method can ensure that the high-frequency transformer current reaches to new steady state without transient dc bias within one half cycle. Finally, the MAB experimental prototypes, including a triple active bridge and a modular MAB with a high-frequency ac bus, are built to verify the superior performance of the proposed method. [ABSTRACT FROM AUTHOR]

A dual active bridge converter is widely used in electric vehicle charging systems and other fields for its excellent features. Despite the diversity of relevant studies on the triple-phase-shift (TPS) modulation, the existing investigations fail in the zero-voltage-switching (ZVS) implementation over the whole voltage and load domains, especially in the transition process between different modes. To address the current state of research, the magnetic current is introduced to bridge the ZVS gap between different modes in this article. A simpler method of the acceptable maximal magnetic inductance design in view of the current stress and the transformer's turns ratio is presented. Afterward, the evaluation method of system efficiency over the full voltage range is raised. Most critically, the simplified optimal control trajectories considering the impact of the magnetic current are proposed for the first time for the TPS modulation, eliminating the nonmonotonicity and leapfrogging of the control variables. Compared with the existing works, the method proposed in this article accomplishes ZVS over the whole domain, thereby reducing potential electromagnetic interference and enhancing the efficiency. Eventually, the proposed method is verified by a 1.0-kW prototype. [ABSTRACT FROM AUTHOR]

Video captioning is a joint task of computer vision and natural language processing, which aims to describe the video content using several natural language sentences. Nowadays, most methods cast this task as a mapping problem, which learns a mapping from visual features to natural language and generates captions directly from videos. However, the underlying challenge of video captioning, i.e., sequence to sequence mapping across the different domains, is still not well handled. To address these problems, we introduce the polishing mechanism in an attempt to mimic human polishing process and propose a generate-and-polish framework for video captioning. In this paper, we propose a two-step transformer based polishing network (TSTPN) consisting of two sub-modules: the generation-module is to generate the caption candidate and the polishing-module is to gradually refine the generated candidate. Specifically, the candidate provides a global information of the visual contents in a semantically-meaningful order, where it is firstly considered as a semantic intersnubber to bridge the semantic gap between the text and video, with the cross-modal attention mechanism for better cross-modal modeling; and it secondly provides a global planning ability to maintain the semantic consistency and fluency of the whole sentence for better sequence mapping. In experiments, we present adequate evaluations to show that the proposed TSTPN achieves the comparable and even better performance than the state-of-the-art methods on the benchmark datasets. [ABSTRACT FROM AUTHOR]

Recently, domain-adaptation based transfer learning has been extensively studied and successfully achieved promising results in addressing the domain drift in closed-set scenarios. However, in the bridge damage diagnosis field, the target data-sets collected from bridges frequently present samples of new damages that were not observed in the source domain, which is known as the open-set domain adaptation problem. To address this problem, this paper proposes a new open-set deep transfer learning algorithm based on joint weighted sub-domain adaptation. First, a joint weighting mechanism is proposed based on adversarial learning and fuzzy theory to represent the similarity of target-domain samples with source-domain classes, and explore the method of separating the known and unknown classes in the target domain to solve the negative transfer problem. Then, to capture the fine-grained transferable information, a sub-domain adaptation algorithm based on minimizing the multi-channel multi-kernel weighted local maximum mean discrepancy (MCMK-WLMMD) is proposed to align the corresponding sub-domains in the two domains. Finally, membership is introduced to build an unsupervised fuzzy clustering model with evaluation indicator to recognize multiple unknown damages. Extensive experiments on open-set transfer tasks between three bridges verify the effectiveness of the algorithm. [ABSTRACT FROM AUTHOR]

Random traffic flow load (TFL) simulation is an important analysis method for bridge design and safety assessment, and accurate TFL modelling is a prerequisite for high-quality simulation. The existing TFL modelling methods almost all rely on the load data monitored by the weigh-in-motion system (WIM system). However, the WIM system has natural defects such as unsatisfactory measurement accuracy at low speed and the inability to measure vehicle lengths and transverse positions in the lane, limiting the improvement of TFL simulation accuracy. Regarding this, a TFL monitoring system that integrates the functions of machine vision and WIM system is developed in this paper. In this system, a deep learning method is applied, for the accurate detection of vehicles and wheels in the video, and the extraction of key parameters for TFL modelling based on detection results. According to the long-term monitoring value, statistical distributions of key parameters are determined, and then an intelligent TFL model is derived from the Intelligent Driver Model (IDM), considering the car-following behavior of vehicles. Correspondingly, this paper further suggests a TFL simulation method and achieves an accurate TFL simulation. A cable-stayed bridge is taken as an example to verify the feasibility of the method. The results show that, compared to the modelling and simulation methods that only rely on the WIM system, the proposed method not only reduces the measurement error of vehicle dimensions by nearly 4 times, but also performs higher resolution in time measurement. The proposed method effectively overcomes the shortcomings of existing schemes and has good application potential in engineering. [ABSTRACT FROM AUTHOR]

A radial nonlinear multi-spring-damper tire (RNMST) model is proposed in this paper for dynamics behavior analysis of the cooperative bridge-vehicle (CBV) system by considering different road surface roughness and separation. First, both the finite element (FE) and force analysis of the RNMST model were analyzed and studied under different velocity and different class of road surfaces. The three types of tire models, nonlinear single-spring tire (NSST) model, nonlinear multi-spring tire model and RNMST model, were introduced as the vehicle tire devices for a quarter vehicle models into nonlinear cooperative bridge-vehicle system. The cooperative dynamics problem, alongside separation and the responses of road surface in terms of its uneven nature, is a typical dynamics problem with the nonlinearity of material which can be solved using the Newmark- $\beta $ method. The comparison of the results demonstrated that the analysis of the dynamics of the cooperative bridge-vehicle system under the RNMST model and NMST model are generally similar. However, the radial nonlinear multi-spring-damper tire model is more accurate and closer to the stress state of the real tire model. Also, to further illustrate the comprehensive performance of the RNMST, the responses under different pavement excitations were analyzed by using the amplitude-frequency spectrum method. Finally, to research the influence dynamics of tire force on the CBV system, the horizontal component of tire force was analyzed. This is conducive to numerical research of the dynamic responses of both the bridge and the vehicle. [ABSTRACT FROM AUTHOR]

Multicore fibers (MCFs) have attracted intense attention to overcome the capacity limit of single-core optical fiber communication system. However, as indispensable connection devices between the MCF and the individual single mode fiber (SMF), fan-in and fan-out (FIFO) devices are challenging to achieve overall performance in terms of coupling efficiency, crosstalk, return loss, and device compactness. Here we develop an MCF reverse-tapering technique to increase the core-to-core spacing of the MCF and construct a spacing-tailored interface between the MCF and the normal SMF bundle. The fabricated FIFO device for the small-core MCF with narrow core-to-core spacing of 35 μm features a low average coupling loss (∼0.6 dB), extraordinarily low crosstalk (<−65 dB) and a return loss (<−60 dB) without using any bridge elements, where the remarkably low crosstalk is about 20 dB lower than using special bridge fibers for the same type of MCF. The flexibility of reverse tapering of MCFs can overcome its intrinsic integration challenge and provide a powerful approach to constructing all-fiber FIFO devices with compact construction and excellent overall performance. [ABSTRACT FROM AUTHOR]

Dynamic deformation measurement is an important approach to monitor structure stability. Due to its rapid imaging capabilities, ground-based multi-input–multi-output (GB-MIMO) radar has shown great application potential in bridge structure monitoring. This article proposes a dynamic deformation estimation method based on radar image series. First, an improved clutter suppression method is used to overcome the estimation error in complex working status. Second, a two-step pixel extraction method is adopted to ensure both the quantity and quality of sample pixels. Finally, frequency- and time-domain environmental stimulating methods are jointly considered for stable mode shape estimation. This article systematically measured the dynamic deformations of bridge structures, including two suspension bridges and a cable-stayed bridge. The time-series deformation, deflection, vibration frequency, and amplitude on the bridge structure are obtained by image-domain measurement. For the first time, the mode shapes of bridges are obtained through GB-MIMO radar. The correctness of structure parameter measurement is verified with the finite element model (FEM). Experimental results prove that with the proposed method, stable results could be acquired against weak and complex stimulation conditions. [ABSTRACT FROM AUTHOR]

In dual-active-bridge (DAB) converters, the driving-synchronization signal is supposed to be required to be transmitted across the primary bridge and the secondary bridge to realize a phase shift for power flow. In this letter, a novel bridge-to-bridge independent (B2I) control method is proposed. In this control method, cross-bridge transmission of sampling/driving signals is not required. Thus, full modularity of each side in the DAB converter was realized. B2I control is suitable for high-voltage series low-voltage parallel (HSLP) modular DAB interlinking converter in medium-voltage dc applications because: first, the bridge-level modularity of B2I control benefits electromagnetic compatibility and insulation design with the HSLP connection; second, B2I control features series-side voltage sharing; and third, B2I control can adapt for various operation modes of microgrids. [ABSTRACT FROM AUTHOR]

Deep learning based bridge damage diagnosis methods can successfully use labeled data to detect bridge damage. These successful applications usually need that the training samples (source domain) and test samples (target domain) obey the same probability distribution. However, it is difficult to acquire a large amount of labeled data with damage information from actual bridges. It is also difficult to apply a model trained with bridge A to diagnose bridge B because of the distribution discrepancy of data from different bridges or environments. Therefore, transferring a well-trained damage diagnosis model to another bridge with unlabeled data remains a major challenge. Motivated by transfer learning, this paper proposes a new intelligent damage diagnosis method for bridges, namely, sub-domain adaptive deep transfer learning network (SADTLN), to solve the feature generalization problem in different bridges. In our method, a multi-kernel local maximum mean discrepancy (MK-LMMD) based sub-domain adaptation module, including a domain classifier for aligning the global distribution and a sub-domain multi-layer adaptation for aligning local distribution, is proposed for transfer learning, so that the learned features are domain-invariant. Experiments prove the effectiveness and advancement of the proposed method. This exploration will promote the practical application of intelligent bridge structural damage diagnosis. [ABSTRACT FROM AUTHOR]

Since the high randomness feature of traffic flows, it is challenging to be aware of the spatial-temporal characteristics of dynamic vehicle loads on a long-span bridge. Compared with conventional measuring approaches, computer vision-based techniques have competitive advantages in the measurement of factual vehicle loads parameters, including the passing time, velocity, vehicle type, and corresponding locations on the bridge at each timestamp. In this paper, a vision-based framework for the identifications of dynamic vehicle loads on long-span bridges is proposed. To be specific, the framework consists of two major parts. The first part is the extraction of spatial-temporal factors of vehicles, including the detection of vehicles and the estimation of tracklets. Here, Single Shot Multi-Box Detection (SSD) architecture with MobileNets is utilized for detecting vehicles, while the data association technique with Kalman filter is applied on detection results for generating tracklets. Then, in the second part, based on the detection and tracking results, a Multi-layer Monte Carlo (MLMC) model is developed for axle-load intensities inference and estimation. The learned model is utilized to infer the dynamic load of vehicles from the detection results and generated tracklets. The proposed methodology is validated via an experiment of a case study on the Jiangyin bridge, Jiangsu Province, China, and several spatial-temporal distributions of vehicle loads are specifically analyzed and discussed. Experimental results indicate our proposed framework is a feasible and reliable solution to predict and identify dynamic vehicle loads on long-span bridges, without the assistance of the deck-embedded Weigh-in-Motion (WIM) system. [ABSTRACT FROM AUTHOR]

Current-fed dual active bridge converter offers multiport structure and can be used in applications, such as renewable energy systems, hybrid electric vehicles, etc. Dual pulsewidth modulation plus phase shift control provides three degrees of freedom for power control, which include duty ratios of primary and secondary side bridges (d1 and d2) and the phase difference between them (δ). Converter operation at the optimum point ($d_1^{\text{opt}}$ , $d_2^{\text{opt}},\ $ δopt) can help achieve multiple objectives, such as achieving zero voltage switching (ZVS) and minimizing current stress along with controlling the power flow. When a CF-DAB is operated as a multiport converter, the optimal control point ($d_1^{\text{opt}}$ , $d_2^{\text{opt}},\ $ δopt) depends on the voltages and currents at different ports, and determining it online is a challenge. In this article, ZVS behavior and current stress of CF-DAB converter are analyzed under various operating conditions pertaining to different source voltages/currents and load power. Based on the analysis, an algorithm has been proposed for determining ($d_1^{\text{opt}}$ , $d_2^{\text{opt}},\ $ δopt) for the given operating condition. Experimental results on a laboratory prototype have been presented, which validate the theoretical analysis and the proposed algorithm. [ABSTRACT FROM AUTHOR]

This work presents an radiation hardness assurance (RHA) methodology that combines both component and system level data to predict system-level reliability, based on system instrumentation for system-level testing. The methodology is illustrated by its application to two system-on-modules based on recent generations of system-on-chips irradiated with protons. [ABSTRACT FROM AUTHOR]

Visible-infrared person re-identification (VI-ReID) is challenging due to the large modality discrepancy between visible and infrared images. Existing methods mainly focus on learning modality-shared representations by embedding images from different modalities into a common feature space, in which some discriminative modality information is discarded. Different from these methods, in this paper, we propose a novel Modality-Specific Memory Network (MSMNet) to complete the missing modality information and aggregate visible and infrared modality features into a unified feature space for the VI-ReID task. The proposed model enjoys several merits. First, it can exploit the missing modality information to alleviate the modality discrepancy when only the single-modality input is provided. To the best of our knowledge, this is the first work to exploit the missing modality information completion and alleviate the modality discrepancy with the memory network. Second, to guide the learning process of the memory network, we design three effective learning strategies, including feature consistency, memory representativeness and structural alignment. By incorporating these learning strategies in a unified model, the memory network can be well learned to propagate identity-related information between modalities and boost the VI-ReID performance. Extensive experimental results on two standard benchmarks (SYSU-MM01 and RegDB) demonstrate that the proposed MSMNet performs favorably against state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Glass is very common in the real world. Influenced by the uncertainty about the glass region and the varying complex scenes behind the glass, the existence of glass poses severe challenges to many computer vision tasks, making glass segmentation as an important computer vision task. Glass does not have its own visual appearances but only transmit/reflect the appearances of its surroundings, making it fundamentally different from other common objects. To address such a challenging task, existing methods typically explore and combine useful cues from different levels of features in the deep network. As there exists a characteristic gap between level-different features, i.e., deep layer features embed more high-level semantics and are better at locating the target objects while shallow layer features have larger spatial sizes and keep richer and more detailed low-level information, fusing these features naively thus would lead to a sub-optimal solution. In this paper, we approach the effective features fusion towards accurate glass segmentation in two steps. First, we attempt to bridge the characteristic gap between different levels of features by developing a Discriminability Enhancement (DE) module which enables level-specific features to be a more discriminative representation, alleviating the features incompatibility for fusion. Second, we design a Focus-and-Exploration Based Fusion (FEBF) module to richly excavate useful information in the fusion process by highlighting the common and exploring the difference between level-different features. Combining these two steps, we construct a Progressive Glass Segmentation Network (PGSNet) which uses multiple DE and FEBF modules to progressively aggregate features from high-level to low-level, implementing a coarse-to-fine glass segmentation. In addition, we build the first home-scene-oriented glass segmentation dataset for advancing household robot applications and in-depth research on this topic. Extensive experiments demonstrate that our method outperforms 26 cutting-edge models on three challenging datasets under four standard metrics. The code and dataset will be made publicly available. [ABSTRACT FROM AUTHOR]

The article analyzes two concepts of conduction in a low-density plasma opening switch: based on the formation of a bipolar space charge layer subject to unlimited cathode emissivity (bipolar model) and of a unipolar ion layer due to electron motion off the cathode under an electric field induced by magnetic field penetration into the plasma (unipolar model). A comparison of unipolar-model expressions with widely known experimental data shows that the conduction current behaves strictly as $I_{c} \propto n^{1/2}$ at any plasma bridge length in the range of plasma densities from $\sim 10^{11}$ to $\sim 10^{14}$ cm−3 and current rise rates from ~0.3 to $\sim 4$ kA/ns. Holding the conduction current constant at any bridge length $l$ requires that $nl^{2}$ be constant. As the magnetic field rise rate is varied, the conduction current follows: $I_{c} \propto \dot {B}^{1/2}$. These results differ radically from what is predicted by the bipolar model. Also presented in the article are arguments of why the axial current channel width varies nonmonotonically during a pulse and estimates of the electron temperature at different plasma densities and field rise rates. [ABSTRACT FROM AUTHOR]

A defect-based bridge fault represents the faulty behavior of an interconnect short defect obtained by SPICE simulating the two shorted cells with the short defect injected. In this article, we have developed a framework to automatically extract defect-based bridge faults and utilize commercial automatic test pattern generation (ATPG) to generate corresponding test patterns for a given design. A defect-based bridge fault model can not only describe the faulty behavior of a short defect precisely but also result in collapsible faults at one shorted cell pair. As a result, using a defect-based bridge fault model for ATPG can lead to a significantly smaller bridge-fault test set when compared with a conventional four-way dominance bridge fault model, where four noncollapsible faults at one shorted cell pair are considered for ATPG. In addition, some short defects can only be detected by the test set for defect-based bridge faults but not by the test set for four-way dominance bridge faults with more test patterns. The runtime required for extracting 1-time-frame (1tf) defect-based bridge faults has been proven acceptable on industrial designs and some techniques were also proposed to speed up the runtime for extracting 2tf defect-based bridge faults. All experiments in this article are conducted based on industrial designs. [ABSTRACT FROM AUTHOR]

A contact force sensor is a critical clinical unit while performing cardiac ablation procedures. The success rate of these procedures is determined by the effective contact between the tip and the heart tissues. Using a miniaturized, susceptible tri-axial force sensor in the cardiac ablation catheter is quintessential to determine the force acting on the tip accurately. This paper discusses the design, fabrication, and comparison of two MEMS-based force sensors that can potentially be integrated with the catheter tip for better sensitivity. The working principle, shape (ring-type), and outer dimensions of both the sensors are similar, with a strategically positioned narrow rectangular cavity in the second sensor. Finite element analysis is performed to validate the performances of both sensors with the experimental results. The experiments show that sensor 2 (with the rectangular cavity on bridges) has $\sim 1.84\times $ higher sensitivity than sensor 1 (bridges with no rectangular cavity). The results demonstrate the possibility of integrating the sensor with the catheter tip for measuring catheter contact force. [ABSTRACT FROM AUTHOR]

Implementing test cases as programs to automate test execution is a popular testing practice. Current industrial practices usually use test functions to implement the test steps of a test case and then to compose the executable test case by choosing the test functions to call manually. It is time-consuming and could lead to invalid test results by selecting inappropriate test functions. In this article, we propose an automatic test function recommendation approach named Scenario-based Recommendation of TEst Function (SRTEF). Given a test step of a test case, SRTEF uses the weighted description similarity and the scenario similarity to recommend test functions. The description similarity utilizes the deep structured semantic model (DSSM) to measure the relatedness between a test step and a test function by their literal descriptions. The test scenario and the test function usage scenario are considered to calculate the scenario similarity. SRTEF has been successfully applied in Huawei. The systematic experiments have been conducted to evaluate SRTEF by using the dataset from Huawei and comparing with BiInformation source-based KnowledgE Recommendation (BIKER), reported as the best approach so far. The results show that SRTEF outperforms BIKER with significant positive ratios consistently in all the three selection strategies, i.e., Top-3, Top-5, and Top-10. The DSSM shows its advantage over word embedding by the double performance of capturing the semantic relatedness in SRTEF. [ABSTRACT FROM AUTHOR]

With the development of power system electronization, distributed new energy generation-based DC microgrid system has become a hot spot of the current research. Aiming to implement the collection, transformation and transmission of the new energy sources, the dual active bridge (DAB) converter has become a key energy conversion device. Aiming to promote the transmitted efficiency for such DAB converter, a deep reinforcement learning (DRL) optimization scheme with the triple-phase-shift (TPS) is proposed. Specifically, the deep deterministic policy gradient (DDPG) algorithm is used for solving optimal solutions with the minimum power losses. Then, the trained DDPG agent is used as a fast surrogate predictor, which giving the suitable control decisions under whole operation conditions. The main control objective is to minimize the power losses for getting the target maximum efficiency. Besides, the soft switching constrains are added to the DDPG algorithm. Based on this, the proposed optimized method can be used for different operation conditions with excellent performance. The operation principles, the loss analyses, and the training consideration and evaluation of the DDPG algorithm are given. Finally, simulation and experimental results using a 200 W hardware prototype are displayed and compared to prove to effectiveness of theoretical analysis. [ABSTRACT FROM AUTHOR]

Runaway Electrons (REs) generated during plasma disruptions in fusion grade tokamaks have the potential to severely damage the plasma-facing components. Designing optimal plasma discharge scenarios for RE suppression in future experiments requires interpretative modeling of current experiments. Multiple experiments have been carried out on ADITYA-U tokamak to design optimal plasma discharge scenarios for RE avoidance and suppression. In this article, we have numerically studied two representative pairs of ADITYA-U plasma discharges. In the first pair of discharges, plasma density was increased by gas puffs during the flap-top phase which is shown to suppress RE generation. In the second pair of the representative discharges, the effect of a lower ratio of peak electric field during breakdown to the pre-fill pressure is shown on RE generation. We simulate these plasma discharges using the PREDICT code to study the dynamics of Res. The results are consistent with the experimental Hard X-Ray diagnostic observations for both the pair of representative discharges. Additionally, the lowest RE-current is obtained in the discharges with a low ratio of the peak electric field during breakdown to the pre-fill pressure and high plasma density during the flap-top phase due to gas puffs. The suppression of REs is demonstrated by showing the effect of increasing plasma density on the separatrix in momentum space between thermal electrons and REs. [ABSTRACT FROM AUTHOR]

Overcoming the digital divide in rural and remote areas has always been a big challenge for Canada with its huge geographical area. In 2016, the Canadian Radio-television and Telecommunications Commission announced broadband Internet as a basic service available for all Canadians. However, approximately one million Canadians still did not have access to broadband services as of 2020. The COVID-19 pandemic has made the situation more challenging, as social, economic, and educational activities have increasingly been transferred online. The condition is more unfavorable for Indigenous communities. A key challenge in deploying rural and remote broadband Internet is to plan and implement high-capacity backbones, which are now available only in denser urban areas. For any Internet provider, it is almost impossible to make a viable business proposal in these areas. For example, the vast land of the Northwest Territories', Yukon's, and Nunavut's diverse geographical features present obstacles for broadband infrastructure. In this article, we investigate the digital divide in Canada with a focus on rural and remote areas. In so doing, we highlight two potential solutions using low Earth orbit (LEO) constellations to deliver broadband Internet in rural and remote areas to address the access inequality and the digital divide. The first solution involves integrating LEO constellations as a backbone for the existing 4G/5G telecommunications network. This solution uses satellites in a LEO constellation to provide a backhaul network connecting the 4G/5G access network to its core network. The 3rd Generation Partnership Project already specifies how to integrate LEO satellite networks into the 4G/5G network, and the Canadian satellite operator Telesat has already showcased this solution with one terrestrial operator, TIM Brasil, in their 4G network. In this way, users can seamlessly access broadband Internet via their mobile terminals. The second solution is based on the direct use of LEO constellations, such as Starlink, which are now operating in Canada, to deliver broadband Internet. As LEO satellites fly lower, their round-trip latency is lower, and the user terminals can receive Internet signals as long as they are pointing at the sky. An in-depth discussion of both solutions is presented in this work. [ABSTRACT FROM AUTHOR]

Key node identification is one of the most fundamental and significant problems in complex network research. This article presents a novel approach called information-theory-based centrality $C_{H}$ to distinguish influential nodes in complex networks. From an information-theoretic perspective, nodes involved in many information channels will improve the quality of communication. Thus, the greater the number of information channels a node participates, the more the importance of this node during information transmission. With this motivation, the local and global connectivity are measured according to the resistance distance, which captures the characteristic that multiple available routes reduce the resistance of information transmission on networks. We further model the probability of information diffusion according to the local and global connectivity and put forward an information-theory-based centrality to identify influential nodes. Finally, extensive evaluations are implemented on real-world networks to verify the feasibility and effectiveness of the proposed method. According to the results, our proposed method outperforms classical methods in identifying influential nodes and also indicates the potential for analyzing the influence evolution of networks, which shows a positive and effective impact on locating influencers and predicting potential key players in the future. [ABSTRACT FROM AUTHOR]

With multi-dimensional exploration toward oceans, enormous sensing data has been generated with significant volume, velocity, variety, and heterogeneity. The resulting big marine data (BMD) thus issue unprecedented architectural challenges on existing marine communication systems. Current dominant marine communication technologies (e.g., shore-based cellular stations, high frequency radio, and expensive satellites) suffer extremely from short coverage, low bandwidth, insecurity, and unavailable cross-domain transmission. In this article, a resilient vessel network (RVN) is proposed to fundamentally enhance BMD transmission. RVNs with widespread self-or- ganized vessels and opportunistic connections reveal advantages of ubiquity, resilience, low cost, and cross-domain transmission. To efficiently manage opportunistic vessel-to-vessel connections for optimal routing, social network analysis of historical vessel interactions is applied for vessel familiarity measurement and community detection. The performance of the proposed community-based routing (CBR) is comprehensively evaluated with real datasets of fishing vessel trajectories. It is demonstrated that CBR achieves much lower transmission cost with comparable delivery ratio compared to typical routing algorithms. [ABSTRACT FROM AUTHOR]

This article presents a novel control strategy for a dual active bridge series resonant converter. The strategy seeks to ensure the stability of the converter over its entire dynamic range while enhances the transient response. Both properties allow the use of the converter in new applications, where fast dynamics are required, surpassing the performance of existing feedback loops. Starting from the generalized averaged model of the converter, we propose a nonlinear control strategy by means of Lyapunov's stability theory. After that, we derive a series of modifications in order to implement the strategy in a microcontroller or a DSP, including a sensorless method to tackle the lack of measurements of certain variables and an adaptive control law to deal with uncertain parameters in the model. The strategy is evaluated in simulations and experiments, employing a commercial converter, and comparing the results with other control policies. [ABSTRACT FROM AUTHOR]

Despite being commonly used in big-data analytics; the outcome of dimensionality reduction remains a black-box to most of its users. Understanding the quality of a low-dimensional embedding is important as not only it enables trust in the transformed data, but it can also help to select the most appropriate dimensionality reduction algorithm in a given scenario. As existing research primarily focuses on the visual exploration of embeddings, there is still a need for enhancing interpretability of such algorithms. To bridge this gap, we propose two novel interactive explanation techniques for low-dimensional embeddings obtained from any dimensionality reduction algorithm. The first technique LAPS produces a local approximation of the neighborhood structure to generate interpretable explanations on the preserved locality for a single instance. The second method GAPS explains the retained global structure of a high-dimensional dataset in its embedding, by combining non-redundant local-approximations from a coarse discretization of the projection space. We demonstrate the applicability of the proposed techniques using 16 real-life tabular, text, image, and audio datasets. Our extensive experimental evaluation shows the utility of the proposed techniques in interpreting the quality of low-dimensional embeddings, as well as with selecting the most suitable dimensionality reduction algorithm for any given dataset. [ABSTRACT FROM AUTHOR]

The development of DNA sequencing technologies makes it possible to obtain reads originated from both copies of a chromosome (two parental chromosomes, or haplotypes) of a single individual. Reconstruction of both haplotypes (i.e., haplotype phasing) plays a crucial role in genetic analysis and provides relationship information between genetic variation and disease susceptibility. With the emerging third-generation sequencing technologies, most existing approaches for haplotype phasing suffer from performance issues to handle long and error-prone reads. We develop a divide-and-conquer algorithm, DCHap, to phase haplotypes using third-generation reads. We benchmark DCHap against three state-of-the-art phasing tools on both PacBio SMRT data and ONT Nanopore data. The experimental results show that DCHap generates more accurate or comparable results (measured by the switch errors) while being scalable for higher coverage and longer reads. DCHap is a fast and accurate algorithm for haplotype phasing using third-generation sequencing data. As the third-generation sequencing platforms continue improving on their throughput and read lengths, accurate and scalable tools like DCHap are important to improve haplotype phasing from the advances of sequencing technologies. The source code is freely available at https://github.com/yanboANU/Haplotype-phasing. [ABSTRACT FROM AUTHOR]

Spatiotemporal fusion is a topical framework for solving the mutual restricted problem between the spatial and temporal resolution of satellite images. We pioneer an approach to replace similarity measurement steps in spatiotemporal fusion algorithms with convolutional neural networks (CNNs), building a bridge between weight function-based models and the learning-based models. Specifically, we propose a nonlocal form that separates the relational computation part from the value representation part, and construct the CNN-based similarity weight learning block for learning normalized weights. The block can be inserted into spatial and temporal adaptive reflectance fusion model (STARFM) to replace the manually designed weight calculation rules common in weight function-based methods, or into the CNN model StfNet to better utilize neighboring high-resolution images. The trained model outputs a high-resolution prediction from each base date image pair. The final result is a combination of the two predictions. In this regard, we propose the standard deviation-based weights to combine two prediction results. Four experiments are performed on Landsat–Moderate-resolution Imaging Spectroradiometer (MODIS) image pairs to determine the following: 1) the performance of the model at the target training date; 2) the generalization of the model in the target training time period; and 3) the generalization of the model at different dates and different geographical locations, each considering the different cases of giving one and two pairs of known images. Experimental results demonstrate the superiority of the similarity weight learning block and standard deviation-based weights. Among them, STARFM with the similarity weight learning block exhibits strong generalization, which testifies to the practical value of our model. [ABSTRACT FROM AUTHOR]

This article proposes a new active power flow controller for dc–dc multiactive bridge (MAB) converters with minimum current point tracking (MCPT) to minimize conductive losses. Unlike recent efforts in this area, the proposed controller does not require look-up table for implementation nor complex nonlinear converter modeling and is not circuit parameter dependent. The proposed control scheme is also generalized for MAB converters with any number of ports and is derived in per unit form to enable standard implementation for buck/boost/unity-gain operating modes. The proposed MCPT algorithm is based on an innovative adaptive perturb and observe tracking of the minimum current point (MCP) on the MAB converter current characteristic plane. This MCP is tracked while active power regulation is simultaneously achieved based on a new phase shift decoupler to enable direct regulation of individual port active power. Detailed design procedure of the proposed controller is presented and extensive simulation is carried out to validate the effectiveness of the proposed MCPT closed loop controller. Experimental implementation is also provided to substantiate the simulation results. [ABSTRACT FROM AUTHOR]

In their quest toward Exascale, High Performance Computing (HPC) systems are rapidly becoming larger and more complex, together with the issues concerning their maintenance. Luckily, many current HPC systems are endowed with data monitoring infrastructures that characterize the system state, and whose data can be used to train Deep Learning (DL) anomaly detection models, a very popular research area. However, the lack of labels describing the state of the system is a wide-spread issue, as annotating data is a costly task, generally falling on human system administrators and thus does not scale toward exascale. In this article we investigate the possibility to extract labels from a service monitoring tool (Nagios) currently used by HPC system administrators to flag the nodes which undergo maintenance operations. This allows to automatically annotate data collected by a fine-grained monitoring infrastructure; this labelled data is then used to train and validate a DL model for anomaly detection. We conduct the experimental evaluation on a tier-0 production supercomputer hosted at CINECA, Bologna, Italy. The results reveal that the DL model can accurately detect the real failures, and, moreover, it can predict the insurgency of anomalies, by systematically anticipating the actual labels (i.e., the moment when system administrators realize when an anomalous event happened); the average advance time computed on historical traces is around 45 minutes. The proposed technology can be easily scaled toward exascale systems to easy their maintenance. [ABSTRACT FROM AUTHOR]