A5012375264- Hydraulic flow and structures
- Complex Network Analysis Techniques
- Hydrology and Sediment Transport Processes
- Advanced Graph Neural Networks
- Hydrology and Watershed Management Studies
- Dam Engineering and Safety
- Advanced Text Analysis Techniques
- Text and Document Classification Technologies
- Perovskite Materials and Applications
- Opinion Dynamics and Social Influence
- Quantum Information and Cryptography
- Evaluation and Optimization Models
- Web Data Mining and Analysis
- Environmental and Sediment Control
- Plant biochemistry and biosynthesis
- Caching and Content Delivery
- Web visibility and informetrics
- Digital Transformation in Industry
- Ga2O3 and related materials
- High-Voltage Power Transmission Systems
- Transport and Logistics Innovations
- Color perception and design
- Collaboration in agile enterprises
- Genomics and Phylogenetic Studies
- Recommender Systems and Techniques
Tianjin Renai College
2022-2025
Taiyuan University of Technology
2021-2022
Northeast Normal University
2022
Tianjin University
2012-2020
Changjiang Institute of Survey, Planning, Design and Research
2016-2020
Tsinghua University
2010-2011
Beijing Jiaotong University
2010
Abstract 2D perovskites have attracted intensive attention by virtue of their excellent optical and electrical properties along with good stabilities. Herein, a highly sensitive self‐powered photodiode based on (PEA) 2 (MA) 4 Pb 5 I 16 (PEA=C 6 H (CH )NH 3 , MA=CH NH ) perovskite is demonstrated dual interface passivations. The Al O bottom passivation can reduce the pinhole defects in film suppress trap‐related recombination loss, bringing forward much reduced dark current increased...
Semi-supervised classification is a fundamental technology to process the structured and unstructured data in machine learning field. The traditional attribute-graph based semi-supervised methods propagate labels over graph which usually constructed from features, while convolutional neural networks smooth node attributes, i.e., real topology. In this paper, they are interpreted perspective of propagation, accordingly categorized into symmetric asymmetric propagation methods. From both...
The hydrodynamic properties of free surface vortices at hydraulic intakes were investigated. Based on the axisymmetric Navier–Stokes equations and empirical assumptions, two sets formulations for velocity distributions profiles are proposed validated against measurements available in literature. Compared with previous formulae, modifications based Mih's formula found to greatly improve agreement experimental data. Physical model tests also conducted study intake vortex Xiluodu hydroelectric...
Network contents including node and edge can be utilized for community detection in social networks. Thus, the topic of each extracted as its semantic information. A plethora models integrating model network topologies have been proposed. However, a key problem has not resolved that is division community. Since definition based on topology, might involve several topics. To ach
The study of community detection in networks has drawn great attention recent years. To find communities and to understand semantics, both network topology content are utilized. Unfortunately, none them can explain the driving factors generating structure with which is significant for understanding mechanisms generation. Our observations on a large number show that specific user social behaviors underlying generation structure. We exploit four types widely exist networks, i.e., reciprocity...
Community detection is an important research field in the understanding of networks. The definition network communities focuses on denser intracommunity links and sparpser intercommunity links. It cannot explain fundamental generation mechanisms two types links, which challenging to reveal. Unfortunately, none existing works can solve this challenge for accurately modeling community structures. This paper investigates a typical category networks possess contents Based analyses real networks,...
Graph convolutional networks (GCNs) have attracted increasing attention in various fields due to their significant capacity process graph-structured data. Typically, the GCN model and its variants heavily rely on transmission of node features across graph structure, which implicitly assumes that structure are consistent, i.e., they carry related information. However, many real-world networks, may unexpectedly mismatch with structural Existing GCNs fail generalize inconsistent scenarios even...
On the basis of minimum principle total cost, using structure logistics network and Mixed-integer Non-linear programming, we propose a mathematical model, taking uncertainty demand number recall return goods, dynamicity closed-loop network, environment E-commerce, time value currency into account. And then use LINGO to solve problem, in order determine constructing or expanding facilities, optimum flows transportation between various facilities. At end this paper, prove validity feasibility...
The surface vortex in a spillway tunnel intake was investigated physical model of the Xiluodu hydropower station. velocity fields were measured using particle tracking velocimetry technique. tangential formula flow field derived based on Navier-Stokes equations, and this greatly improved consistency between numerical experimental data. Also, has advantage describing more accurately than previous formulas. current research is established engineering practices, results provide valuable...
The high‐steep rock slope stability is one of the key technologies in construction water conservancy and hydropower projects, which affects restricts development hydraulic resources projects. In this paper, a three‐dimensional numerical model was built incorporating stratigraphy, geological structures, inverted mechanical parameters to perform displacement, stress, plastic zone analyses for an excavated China using FLAC3D software. simulation results after excavation show that deformation...
Learning effective graph representations in an unsupervised manner is a popular research topic data analysis. Recently, contrastive learning has shown its success representation learning. However, how to avoid collapsing solutions for methods remains critical challenge. In this paper, simple method proposed solve problem learning, which different from existing commonly used techniques (such as negative samples or predictor network). The model mainly relies on asymmetric design that consists...
Lots of high arch dam have the characteristics "High head, Large discharge and Narrow river valley", therefore, security researches energy dissipation were always focus in these hydro-projects. Statistically, trajectory type is most widely used built dams, water plunge poor set downstream body. However, protected need large investment with disadvantage complicated operation maintenance. Along construction concrete Southwest China, overburden cushion deep site, which becoming a new...
Lots of built or building dams, especially the high arch dam, located in Southwest China with abundant hydro-power resources.Most these dams have characteristics "High head, Large discharge and Narrow river valley", therefore, security researches energy dissipation were always focus hydro-projects.Statistically, trajectory type is most widely used at home abroad, which simple engineer structure, short construction period adaptability advantage on changing Tail water depth.And plunge poor set...
With the rapid development of Internet and Things, smart city home are gradually penetrating into people's lives. Through cognition existing furniture models in market, research is conducted on problem that wisdom only reflected electrical appliances, while resting sofas underdeveloped. The sofa model lies its ability to understand user's habits preferences through deep learning realize living needs terms movement or temporal characteristics. Using a combination qualitative quantitative...
With the deepening of development and application Internet Things technology in medical field, product design interaction have developed greatly. The market demand industry continues to expand, use environment equipment has expanded from hospitals families, users' requirements for rehabilitation also increased, forming higher emotional needs. Kano evaluation method was used obtain needs autistic patients transcranial electromagnetic therapy equipment, questionnaire survey screened sorted out...
It's important to eliminate noisy data for information extraction on the deep web. In this paper, we propose a new approach called ENDW(Eliminating Noisy Data in Web pages) based query keywords and DOM tools data. Query submitted backend databases always appear web pages. The boundary between useful region is concerned with position where appear. Once found boundary, could retain region. Our experiments show that effective stable.
For the third-party logistics company, services are base of its operation, and keeping competitive is fundamental to sustainable development enterprises. This paper establishes a game mechanism research on conditions service innovation provided by enterprise, then discusses management incentive which promotes innovation.