A5100669772- Advanced Sensor and Energy Harvesting Materials
- Lignin and Wood Chemistry
- Advanced Materials and Mechanics
- Machine Learning in Materials Science
- Magnetic Properties and Applications
- High Temperature Alloys and Creep
- Machine Learning in Bioinformatics
- Microwave Engineering and Waveguides
- Enzyme-mediated dye degradation
- Antenna Design and Analysis
- MXene and MAX Phase Materials
- Aluminum Alloy Microstructure Properties
- Solidification and crystal growth phenomena
- Risk and Safety Analysis
- 3D Shape Modeling and Analysis
- Biochemical and biochemical processes
- Conducting polymers and applications
- Dielectric materials and actuators
- Microplastics and Plastic Pollution
- Iterative Learning Control Systems
- Chemistry and Chemical Engineering
- Evolutionary Algorithms and Applications
- Environmental remediation with nanomaterials
- Adaptive Control of Nonlinear Systems
- Domain Adaptation and Few-Shot Learning
Jilin Medical University
2024
Xidian University
2022-2023
Zhejiang University
2023
Dalian University of Technology
2022
Beijing Institute of Fashion Technology
2022
Beijing Forestry University
2016-2021
Northeast Forestry University
2021
Shanghai Tenth People's Hospital
2019
Tongji University
2019
Tianjin University of Technology
2018
Abstract Fibre‐based materials have received tremendous attention due to their flexibility and wearability. Although great efforts been devoted achieve high‐performance fibres over the past several years, it is still challenging for multifunctional macroscopic satisfy versatile applications. 2D transition metal carbides/nitrides (MXenes) with intriguing physical/chemical properties explored in broad application, may be able reinforce synthetic fibres. Inspired by natural materials, first...
Abstract Growing demand in intelligent wearable electronics raises an urgent requirement to develop deformable and durable power sources with high electrical performance. Here, a stretchable shape‐adaptive triboelectric nanogenerator (TENG) based on MXene liquid electrode is proposed. The open‐circuit voltage of MXene‐based TENG reaches up 300 V. excellent fluidity highly electronegativity the electrode, gives long‐term reliability stable output regardless diverse extreme deformations. With...
In this article, a compact differentially fed ±45° dual-polarized filtering magneto-electric (ME) dipole antenna with improved harmonic suppression is proposed and analyzed. First, the fusion of loop-slot half-wave vibrator calculated mentioned to furnish theoretical guidance. Then, by reasonably placing two pairs second-order step impedance feedlines patch, corresponding two-mode loop-vibrator combination an upper radiation null naturally constructed. Moreover, simply loading four...
Machine learning has huge potential to revolutionize the field of drug discovery and is attracting increasing attention in recent years. However, lacking domain knowledge (e.g., which tasks work on), standard benchmarks data preprocessing pipelines are main obstacles for machine researchers this domain. To facilitate progress discovery, we develop TorchDrug, a powerful flexible platform built on top PyTorch. TorchDrug variety important including molecular property prediction, pretrained...
We are now witnessing significant progress of deep learning methods in a variety tasks (or datasets) proteins. However, there is lack standard benchmark to evaluate the performance different methods, which hinders this field. In paper, we propose such called PEER, comprehensive and multi-task for Protein sEquence undERstanding. PEER provides set diverse protein understanding including function prediction, localization structure protein-protein interaction protein-ligand prediction. types...
Kraft lignin (KL) was valorized by introducing phenolic hydroxyl groups and adjusting the molecular weight oxidative depolymerization. Hydrogen peroxide with copper sulfate as a catalyst employed in this process. The depolymerized (DL) characterized differential ultraviolet spectrum (UV), Fourier transform infrared (FT-IR) spectrometry, gel permeation chromatography (GPC), heteronuclear single quantum correlation (HSQC) spectra. Both (Mw Mn) polydispersity of decreased compared default kraft...
Neural Code Intelligence -- leveraging deep learning to understand, generate, and optimize code holds immense potential for transformative impacts on the whole society. Bridging gap between Natural Language Programming Language, this domain has drawn significant attention from researchers in both research communities over past few years. This survey presents a systematic chronological review of advancements intelligence, encompassing 50 representative models their variants, more than 20...
In the speed control of permanent magnet synchronous motor (PMSM), traditional exponential reaching law has contradiction fast response and heavy chattering. The sliding mode controller based on a novel hybrid was proposed to solve these problem. This adds power law, it can problem that is slow when system far away from surface chattering in near surface. state variables good effect inhibiting caused by introduction index term. Based designed, Lyapunov stability theory used analyze system....
The widespread use of plastics in beverage packaging has led to the accumulation microplastics (MPs) and nanoplastics (NPs) beverages, which poses significant environmental health hazards. This review explores sources, entry pathways, risk factors for MPs NPs emphasizing their toxicological profiles harmful effects on human environment. Methods detecting beverages are discussed, highlighting need standardized testing protocols. Furthermore, future solution strategies, challenges, preventive...
Phase-field models have been widely used to investigate the phase transformation phenomena. However, it is difficult solve problems numerically due their strong nonlinearities and higher-order terms. This work devoted solving forward inverse of phase-field by a novel deep learning framework named Phase-Field Weak-form Neural Networks (PFWNN), which based on weak forms equations. In this framework, solutions are parameterized as neural networks with periodic layer, while test function space...
Unsupervised domain adaptation (UDA) aims to predict unlabeled data from target with access labeled the source domain. In this work, we propose a novel framework called SIDA (Surrogate Mutual Information Maximization Domain Adaptation) strong theoretical guarantees. To be specific, implements by maximizing mutual information (MI) between features. framework, surrogate joint distribution models underlying of Our analysis validates bounding expected risk on MI and bias. Experiments show that...
In this study, the electrochemical dechlorination of triclosan in aqueous solution on palladium/polypyrrolereduced graphene oxide/foamed nickel composite electrode (Pd/PPy-rGO/foam-Ni electrode) was investigated. The influences applied current, reaction temperature and initial concentration efficiency were studied. can be completely removed within 180 min under conditions current 7 mA 40 °C when 50 mg · L –1 .
With promising yet saturated results in high-resource settings, low-resource datasets have gradually become popular benchmarks for evaluating the learning ability of advanced neural networks (e.g., BigBench, superGLUE). Some models even surpass humans according to benchmark test results. However, we find that there exists a set hard examples settings challenge but are not well evaluated, which causes over-estimated performance. We first give theoretical analysis on factors bring difficulty...
Predicting the creep rupture time of Nickel-based alloys has been one central topics in materials science. This work is devoted to developing a comprehensive strategy based on machine learning for data generation and prediction. Our objective improve accuracy effectiveness To overcome limited availability experimental data, novel model designed Wasserstein Generative Adversarial Network with Dual Discriminators. The new capable generating batch high-quality synthetic datasets expanding scale...
Abstract In this paper, a broadband dipole antenna with filtering function is proposed. The proposed as whole comprises reflector, coaxial cables, main radiator, and cross‐center structure. First, two orthogonal dipoles are used the radiator form dual‐polarization. By adding Y‐shaped branches structure, base station realizes expands bandwidth. has on both sides of operating frequency band, out‐of‐band radiation suppression can be 16 dB at upper band. null controlled independently by...
The aromatic nature of lignin makes it a potential renewable source chemicals and other valuable products. Isolation from lignocellulosic biomass using organic solvents enables the production high-purity lignin. use formic acid in organosolv pulping fractionation process has been widely studied. Characterization is necessary to achieve valueadded applications To simplify isolation acid-treated lignin, herein, milled wheat straw (MWSL) was employed as an archetype for characterization...
In accordance with the problems of body measurement technology and difficulty in obtaining samples for common consumers large-scale customization clothing, simulation machine learning method positive images were proposed to be adopted predict data. Firstly, simulated shooting from different angles was conducted on basis 116 samples. addition, characteristic parameters obtained through extracting silhouette identifying key points multiple linear regression neural networks used conduct chest...
In order to research the protective performance of polyimide fabric as arc material, a synthetic circuit high current test platform was built simulate flashover between two copper rods with diameter 19 mm and length 450 mm. By controlling level 12.3 kA, 20 groups tests were done, energy measured by calorimeter. addition, keeping covered on surface calorimeter for tests, through measured. this paper, can play role when is under 110.7184 J/cm <sup xmlns:mml="http://www.w3.org/1998/Math/MathML"...