Search

ObjectivePrior studies have shown that plaque inflammation on FDG-PET and the symptomatic carotid atheroma inflammation lumen-stenosis (SCAIL) score were associated with recurrent ischemic events, but the findings have thus far not been widely validated. Therefore, we aimed to validate the findings of prior studies.MethodsA single-center prospective cohort study that recruited patients with (1) recent TIA or ischemic stroke within the past 30 days, (2) ipsilateral carotid artery stenosis of ≥50%, and (3) were not considered for early carotid revascularization. The (1) maximum standardized uptake value (SUVmax) of the symptomatic carotid plaque, (2) the SCAIL score, and (3) stenosis severity of the symptomatic carotid artery were measured for all patients. The outcomes were (1) a 90-day ipsilateral ischemic stroke and (2) a 90-day ipsilateral symptomatic TIA or major adverse cardiovascular event (MACE).ResultsAmong the 131 patients included in the study, the commonest cardiovascular risk factor was hypertension (95 patients, 72.5%), followed by diabetes mellitus (77 patients, 58.8%) and being a current smoker (64 patients, 48.9%). The median (IQR) duration between the index cerebral ischemic event and recruitment to the study was 1 (0, 2.5) days. The median (IQR) duration between the index cerebral ischemic event and FDG-PET was 5 (4, 7) days. A total of 14 (10.7%) patients had a 90-day stroke, and 41 (31.3%) patients had a 90-day TIA or MACE. On comparison of the predictive performances of the SCAIL score and SUVmax, SUVmax was found to be superior to the SCAIL score for predicting both 90-day ipsilateral ischemic stroke (AUC: SCAIL = 0.79, SUVmax = 0.92; p < 0.001; 95% CI = 0.072, 0.229) and 90-day TIA or MACE (AUC: SCAIL = 0.76, SUVmax = 0.84; p = 0.009; 95% CI = 0.020, 0.143).ConclusionPlaque inflammation as quantified on FDG-PET may serve as a reliable biomarker for risk stratification among patients with ECAD and recent TIA or ischemic stroke. Future studies should evaluate whether patients with significant plaque inflammation as quantified on FDG-PET benefit from carotid revascularization and/or anti-inflammatory therapy.

Graph contrastive learning (GCL) has emerged as a dominant technique for graph representation learning which maximizes the mutual information between paired graph augmentations that share the same semantics. Unfortunately, it is difficult to preserve semantics well during augmentations in view of the diverse nature of graph data. Currently, data augmentations in GCL that are designed to preserve semantics broadly fall into three unsatisfactory ways. First, the augmentations can be manually picked per dataset by trial-and-errors. Second, the augmentations can be selected via cumbersome search. Third, the augmentations can be obtained by introducing expensive domain-specific knowledge as guidance. All of these limit the efficiency and more general applicability of existing GCL methods. To circumvent these crucial issues, we propose a \underline{Sim}ple framework for \underline{GRA}ph \underline{C}ontrastive l\underline{E}arning, \textbf{SimGRACE} for brevity, which does not require data augmentations. Specifically, we take original graph as input and GNN model with its perturbed version as two encoders to obtain two correlated views for contrast. SimGRACE is inspired by the observation that graph data can preserve their semantics well during encoder perturbations while not requiring manual trial-and-errors, cumbersome search or expensive domain knowledge for augmentations selection. Also, we explain why SimGRACE can succeed. Furthermore, we devise adversarial training scheme, dubbed \textbf{AT-SimGRACE}, to enhance the robustness of graph contrastive learning and theoretically explain the reasons. Albeit simple, we show that SimGRACE can yield competitive or better performance compared with state-of-the-art methods in terms of generalizability, transferability and robustness, while enjoying unprecedented degree of flexibility and efficiency., Comment: Accepted by The Web Conference 2022 (WWW 2022)

Contrastive Learning (CL) has emerged as a dominant technique for unsupervised representation learning which embeds augmented versions of the anchor close to each other (positive samples) and pushes the embeddings of other samples (negatives) apart. As revealed in recent studies, CL can benefit from hard negatives (negatives that are most similar to the anchor). However, we observe limited benefits when we adopt existing hard negative mining techniques of other domains in Graph Contrastive Learning (GCL). We perform both experimental and theoretical analysis on this phenomenon and find it can be attributed to the message passing of Graph Neural Networks (GNNs). Unlike CL in other domains, most hard negatives are potentially false negatives (negatives that share the same class with the anchor) if they are selected merely according to the similarities between anchor and themselves, which will undesirably push away the samples of the same class. To remedy this deficiency, we propose an effective method, dubbed \textbf{ProGCL}, to estimate the probability of a negative being true one, which constitutes a more suitable measure for negatives' hardness together with similarity. Additionally, we devise two schemes (i.e., \textbf{ProGCL-weight} and \textbf{ProGCL-mix}) to boost the performance of GCL. Extensive experiments demonstrate that ProGCL brings notable and consistent improvements over base GCL methods and yields multiple state-of-the-art results on several unsupervised benchmarks or even exceeds the performance of supervised ones. Also, ProGCL is readily pluggable into various negatives-based GCL methods for performance improvement. We release the code at \textcolor{magenta}{\url{https://github.com/junxia97/ProGCL}}., Comment: Accetpted at ICML 2022

A hybrid battery thermal management system (BTMS) with phase change material (PCM) coupled cooling plate arrayed in the manner of honeycomb is proposed for the cylindrical lithium battery pack in this paper. Paraffin wax and glycol are used as PCM and cooling liquid, respectively. 3D numerical simulations are carried out to analyze temperature characteristics of the system, coupling characteristics of PCM and cooling plate, mass flow rate of cooling liquid and channel number of cooling plate on the heat dissipation of battery pack. Results show that the temperature distribution of the battery pack has good uniformity, which is the most prominent advantage of the proposed system. The coupling of PCM and cooling plate greatly improve heat dissipation effect. Increasing mass flow rate of cooling liquid can significantly improve heat dissipation efficiency, but the maximum temperature difference (Δ T max ) of battery cell also increases. However, the influence of channel number of cooling plate on heat dissipation efficiency is lower than that of mass flow rate. Adding one channel can reduce the battery temperature by 2 K, but it has little effect on Δ T max . This study may provide effective guidance for the design of the PCM coupled liquid cooling BTMS.

Design a smart meter data acquisition and analysis system to collect and process electrical energy data, report remote data and collect electrical energy. The data collector of the smart meter supports multiple protocols and the collinear communication of the electric meters, and the collinear communication of the electric meters with different protocols, different models and different communication speeds is connected on the same communication line. The data is sent to the data collector through the RS-485 bus, and the data collector collects and processes the data and uploads it to the server through the WAN. The design of the smart meter data acquisition and analysis system effectively improves the scientific validity of electric energy billing.

This paper introduces a failure case of PhotoMOS relay applied to signal control. Its output is long-term or intermittent conduction when no input signal added as failure phenomenon. After eliminating the failure modes such as electronic over stress, ion pollution and structure aging etc., this paper reveals the failure mode and cause of PhotoMOS relay in the case, judges that the failure is a lot quality problem of products. Three prevention methods are studied on this kind of failure, and one method is suggested for the failure prevention of this PhotoMOS relay.

With the large-scale application of microgrids, the microgrids with three-phase/single-phase architecture, which are geographically relatively close can be interconnected to multi-microgrids systems with multi-phase architecture. For the regional autonomy multi-microgrids which are out of the power distribution network, the paper proposes a real-time control strategy based on Model Predictive Control (MPC). Based on multi-microgrids distributed and hierarchical control architecture, the proposed strategy considers the three-phase unbalance and the minimum deviation between the actual tie line power and the given reference value, which builds a dynamic prediction model with multi-input, multi-disturbance and multi-variable. Then transform it into a mixed integer nonlinear programming and solve it. With the introduction of feedback correction for prediction error correction, it finally realizes rolling finite time closed loop optimization. By the simulation results, the proposed strategy can adjust power of sources, loads and energy storage systems (ESSs) for the regional autonomy multi-microgrids as well as stabilize power fluctuation of sources and loads. Besides, it can keep track of tie line power so as to meet the unbalanced constraints, which is beneficial to the operation of the multi-microgrids that are out of the distribution network.

It is of great significance to study and analyze the measurement performance and sampling inspection methods of smart meters, and to conduct sampling inspection according to the quality status of smart meters, so as to improve the reliability of smart meters. On the basis of qualified rate test and error test of electric energy meters, radar chart and Wilcoxon rank sum test method are adopted to carry out quality inspection comparison of electric energy meters of the same specification from different manufacturers, and sampling statistical test method is studied accordingly. The method described in this paper is suitable for quality analysis and sampling inspection of electric energy meters.

With the development of the power market reform in China, the state chooses to carry out spot market pilot in eight provinces. PV(Photovoltaic) plant need to face real-time fluctuating prices in spot market, making the investment decisions become more complicated. Considering recent PV power plant investment don’t involve electricity market environment, this paper proposes three different trading patterns of PV plants: full-power day-ahead market settlement, signing long-term CFD or selling electricity to grid with fixed electricity. The PV plant revenue model uses dynamic investment payback period, internal rate of return and cost-benefit ratio to analyze the impact of different patterns on investment. Case study shows that in power pool, signing CFDs can effectively reduce the investment recovery period of PV plants and improve investment return.

In this paper, a case in which Bi migration leads to the failure of the chip resistor was analyzed, the source of Bi has been studied, and the analysis process of failure mode and failure mechanism were summarized. Migration of reduced isolation constituents (MRIC) model was described briefly. The preventive measures of the corresponding failure modes were also suggested in this paper.

Electrochemical migration (ECM) of metal has been studied for decades, including Ag, Ni, Sn, Pb, Cu and Au. These metal materials are commonly used in microelectronics devices, both in package and die. According to the failure mechanisms, some of the root causes are easy to find out except Au, since contaminant element is hard to be detected, and electrolyte is various besides water. In a real failure analysis case, we developed a method to analysis contaminant and alcohol induced ECM of Au bond by combining ion chromatography and TOF-SIMS characterization, and IVA analysis. Based on the coupled analysis, the origin of Au ECM is related to chlorine ions and dominant electrolyte formed in low temperature (-45°C) is alcohol. These results indicate alcohol control in microcircuit plays an important role in sealed package.

The chip solder layer of power semiconductor devices carries the mechanical, thermal and electrical loads in the working state of devices. With the increase of packaging density and power density, the reliability requirement of solder layer is increased. Because the solder layer area of power module is much larger than that of traditional solder, its failure mode and mechanism have certain uniqueness, and it is urgent to study deeply. Through the failure analysis case study, this paper expounds the mechanism and the influence factors for the formation of the voids. The detection and verification test methods of voids defect is also introduced.

Previous studies suggested that diets containing high levels of histamine influenced digestive system of aquatic animals. In addition, the exogenous histamine was first detoxified by diamine oxidase in the intestine, while the rest of histamine was further detoxified in the liver. Thus, based on the evidence from the previous studies, we hypothesized that high levels of histamine may lead to damage on liver of the aquatic animals. Here, in current attempt, we sought to investigate the toxic effect of histamine on yellow catfish (Pelteobagrus fulvidraco) liver physiology and pathogenesis. In the present study, yellow catfish were fed for 56 days on diets supplemented with 1000 mg kg-1 histamine (His) or a basal diet as the control group (Con). A significant change on the morphology of the intestine and liver was observed, followed with an induction of serum activity of aspartate aminotransferase (AST) and alanine aminotransferase (ALT). Furthermore, the transcriptomic analysis was performed to gain an overview of the gene expression profile in liver between control and histamine supplemented groups. Through the bioinformatics analysis, 431 differentially expressed genes were identified. Among these genes, Gene Ontology enrichment analysis (GO) suggests that immune-related genes are significantly dysregulated. In addition, TNF signaling pathway is enriched in Kyoto Encyclopedia of Genes and Genomes analysis (KEGG), and is also the dominant pathway in immune system, suggesting that the inflammatory response and apoptosis of hepatocytes are induced by exogenous histamine.

Graphene layers have been produced from multi-walled carbon nanotube (MWCNT) bulk materials by friction when polished on ground-glass, offering a novel and effective method to produce graphene layers, which, more importantly, could be transferred to other substrates by rubbing. Field emission scanning electron microscopy, Raman spectroscopy, atomic force microscopy, transmission electron microscopy and selected area electron diffraction confirmed the formation of graphene layers. They were thought to be peeled away from the MWCNT walls due to friction. The reflection spectra showed that absorption of as-produced graphene layers decreased with wavelength in the range of 250–400 nm, compared to the MWCNT bulk material having strong absorption at 350 nm. Nanoscratch test was used to determine the mechanical properties of graphene films, suggesting the tolerance of as-produced graphene film to flaws introduced by scratch.

Graphene/carbon composite films were prepared by electrospraying a graphene/polyacrylonitrile composite solution on SiO2-coated silicon substrates and subsequent heat treatment. The as-produced graphene/carbon composite films had a porous structure comprising graphene layers. With a magnetic field applied perpendicularly to the sample, an unexpectedly significant positive magnetoresistance attributed to e–e interaction and weak localization has been observed, which constantly increases with the magnetic field in the temperature range of 300–50 K from 0 to 80 kOe.

A general analytical model, formulated in 2-D polar coordinates, is developed to predict the unbalanced magnetic force, which results in permanent-magnet brushless ac and dc machines having a diametrically asymmetric disposition of slots and phase windings. It is shown that the unbalanced magnetic force can be significant in machines having a fractional ratio of slot number to pole number, particularly when the electric loading is high. The developed model is validated by finite-element calculations on 9-slot/8-pole and 3-slot/2-pole machines. In addition, the unbalanced magnetic force has been measured on a prototype 3-slot/2-pole machine and shown to be in excellent agreement with predicted results.

Two kinds of variational principles for numerical simulation of heat transfer and contact analysis are respectively presented. A finite element model for numerical simulation of the thermal contact problems is developed with a pressure dependent heat transfer constitutive model across the contact surface. The numerical algorithm for the finite element analysis of the thermomechanical contact problems is thus developed. Numerical examples are computed and the results demonstrate the validity of the model and algorithm developed.