Rational design of Pt single-atom catalysts provides a promising strategy to significantly improve the electrocatalytic activity for hydrogen evolution reaction. In this work, we presented novel and efficient utilizing low electron-density region substrate effectively trap confine high metal atoms. The catalyst supported by nickel selenide with rich vacancies was prepared via hydrothermal-impregnation stepwise approach. Through experimental testation DFT theoretical calculation, confirm that...

Anomaly detection in the video is an important research area and a challenging task real applications. Due to unavailability of large-scale annotated anomaly events, most existing (VAD) methods focus on learning distribution normal samples detect substantially deviated as anomalies. To well learn motion appearance, many auxiliary networks are employed extract foreground object or action information. These high-level semantic features effectively filter noise from background decrease its...

In this paper, based on the traditional voltage-lift-technique, a modified voltage-lift cell is proposed. A single-switch high gain quadratic boost converter obtained by applying to converter, voltage of improved and switch stress also reduced. The operation principle with analyzed. Compared other highest, lowest. Experimental results are presented verify theoretical analysis results.

Abstract Rational design of Pt single‐atom catalysts provides a promising strategy to significantly improve the electrocatalytic activity for hydrogen evolution reaction. In this work, we presented novel and efficient utilizing low electron‐density region substrate effectively trap confine high metal atoms. The catalyst supported by nickel selenide with rich vacancies was prepared via hydrothermal‐impregnation stepwise approach. Through experimental testation DFT theoretical calculation,...

With the continuous development of economy and society, power demand forecasting has become an important task industry. Accurate can promote operation supply However, since consumption is affected by a number factors, it difficult to accurately predict data. accumulation data in industry, machine learning technology shown great potential forecasting. In this study, gradient boosting decision tree (GBDT), extreme (XGBoost) light (LightGBM) are integrated stacking build XLG-LR fusion model...

Notice of Violation IEEE Publication Principles <br><br> "User Power Interaction Behavior Clustering Analysis That is Based on the Self- Organizing-Center K-Means Algorithm" <br> by Shiyu Zhang and Shaoyun Ge in Access, Volume 7, December 2019 After careful considered review content authorship this paper a duly constituted expert committee, has been found to be violation IEEE's Principles. This translated copied from original cited below. The was without attribution (including appropriate...

This paper investigates the beamforming for three- dimensional multiple input output (3D-MIMO) systems with inaccurate channel state information (CSI). From view of angle-domain, 3D-MIMO is sparse on high 3D resolution provided by planar antenna array large number elements at base station (BS) in systems. Prior knowledge sparsity not only beneficial to estimation literature, but also implies more efficient CSI transmitters as discussed this paper. We prove that optimal vector correspondingly...

Chinese cooking produces large numbers of particles that can cause both indoor and outdoor air quality problems. To reduce the extraction to air, this investigation studied capture efficiency a rotating disk in an exhaust hood. The studies were performed experimentally wind tunnel numerically by computational fluid dynamics (CFD) models with Lagrangian method for tracking particle trajectories. experimental data used identify best turbulence model among three tested CFD simulations. results...

Snapshot compressive imaging (SCI) can record the 3D information by a 2D measurement and from this to reconstruct original reconstruction algorithm. As we see, algorithm plays vital role in SCI. Recently, deep learning show its outstanding ability, outperforming traditional Therefore, improve accuracy is an inevitable topic for Besides, algorithms are usually limited scalability, well trained model general not be applied new systems if lacking training process. To address these problems,...

A Project risk forecast model was investigated using least square support vector machine(LS-SVM) method. Risk estimation data of experts acted as eigenvector learning samples to train the constructed LS-SVM regression for realizing mapping relationship between and characteristic. The test were used compare BP neural network. result showed that has high prediction accuracy strong generalization ability. So it is suitable large scale project risk.

In a deregulated market, with the differences of location, generation power form each plant have different network loss levels and injecting make distinct "quality". It is desirable to take into account while applying market-clearing mechanism. this paper novel economic dispatch proposed, which involves utilization rate at time day-ahead trade planning. The IEEE 30-bus test system used demonstrate proposed method simulation results show that it could offer more equitable market clearing...

This paper stands on the perspective of smart grid and electricity needs, presenting concept distributed power, making depth study two levels "Plug Play" device information exchange management technology after that. Firstly, presents status research studies at home abroad. On this basis, for different forms power access, proposes types "plug play" solutions, makes system development, finally shows system.

Aiming at the problem of energy efficiency public institution to be promoted, this paper propose a method analysis use basing multi-level transfer model. Firstly, analyze characteristics institution, so as preliminary ascertain consumption systems institution. Then, construct model including input and use, procedure Finally, carry out instance analysis, verify effectiveness research result.

In the light of characteristics electricity measurement automation system in Power Company, safe protection is researched from three perspectives:master station,communication channel and terminal equipment.As for master station,the formulated accordance with four levels safety protective measures-boundary, network, host application.As communication channel, made scheduling data networks wireless public network.As equipment,a set hardware encryption security mechanism established.

This paper investigates the beamforming for three- dimensional multiple input output (3D-MIMO) systems with inaccurate channel state information (CSI). From view of angle-domain, 3D-MIMO is sparse on high 3D resolution provided by planar antenna array large number elements at base station (BS) in systems. Prior knowledge sparsity not only beneficial to estimation literature, but also implies more efficient CSI transmitters as discussed this paper. We prove that optimal vector correspondingly...

Data analysis reflecting the social politics, economy, public opinion and management of sentiment, especially those in such areas based on questionnaire survey, often need to study main factors affecting a problem. The traditional statistical method is count factor which selected most frequently among influencing question, dee me d as principal factors. But are not necessarily sensitive factors, that is, highest An entropy proposed determine multiple choices survey for empirical study,...