A5100742675- Solar-Powered Water Purification Methods
- Membrane Separation Technologies
- Recommender Systems and Techniques
- Generative Adversarial Networks and Image Synthesis
- Pickering emulsions and particle stabilization
- Stochastic Gradient Optimization Techniques
- Traffic Prediction and Management Techniques
- Time Series Analysis and Forecasting
- Solar Thermal and Photovoltaic Systems
- Adsorption and Cooling Systems
- Surface Modification and Superhydrophobicity
- Advanced Malware Detection Techniques
- Privacy-Preserving Technologies in Data
- Software Engineering Research
- Advanced Neural Network Applications
- Advanced Bandit Algorithms Research
- Urban and Freight Transport Logistics
- Adversarial Robustness in Machine Learning
- Data Visualization and Analytics
- Artificial Intelligence in Healthcare and Education
- Intelligent Tutoring Systems and Adaptive Learning
- Aviation Industry Analysis and Trends
- BIM and Construction Integration
- Transportation Planning and Optimization
- Software Reliability and Analysis Research
Yanbian University of Science and Technology
2023-2024
Dalian Maritime University
2024
Zhejiang University
2023-2024
Beijing Jiaotong University
2021
Xi'an Technological University
2021
Abstract Solar‐driven evaporation process brings exciting opportunities to recover clean water and resources in a sustainable way from diverse sources like seawater wastewater. Separation membranes, as vital material many environmental energy applications, can contribute significantly this owing their structural features. However, the unique roles of membranes solar evaporator construction design are seldom recognized not summarized yet scientific principles application demands, which forms...
Abstract Salt scaling poses a significant obstacle to the practical implementation of solar‐driven evaporation for desalination. Attempts mitigate by enhancing mass transfer often lead compromise in efficiency due associated heat loss. In present work, novel seesaw evaporator with Janus structure harness periodic self‐descaling is reported. The evaporators are facilely fabricated delignifying balsa wood and subsequently single‐sided spray‐coating it soot polydimethylsiloxane (PDMS). This...
Recent years have witnessed an exponential growth of model scale in deep learning-based recommender systems---from Google's 2016 with 1 billion parameters to the latest Facebook's 12 trillion parameters. Significant quality boost has come each jump capacity, which makes us believe era 100 is around corner. However, training such models challenging even within industrial data centers. We resolve this challenge by careful co-design both optimization algorithm and distributed system...
Porous substrates act as open “interfacial reactors” during the synthesis of polyamide composite membranes via interfacial polymerization. However, achieving a thin and dense nanofilm with high permeance selectivity is challenging when using conventional substrate uniform wettability. To overcome this limitation, we propose use Janus porous confined reactors to decouple local monomer concentration from total amount By manipulating location hydrophilic/hydrophobic interface in substrate, can...
Janus membranes with unidirectional transport capacity have shown broad potential across various fields of mass–energy transmission and conversion. However, the underlying mechanisms driving this remain unclear. In study, a three‐stage model—encompassing liquid intrusion, capillary wetting, transport—is proposed to elucidate within membranes. Finite element analysis reveals two key mechanisms: direct contact‐driven intrusion in bilayer structures surface‐energy‐driven gradient structures. It...
Identifying the optimal cabin layout is an important way to improve efficiency of ship systems and ensure efficient circulation personnel materials. The task state refers maintained by joint actions machinery equipment, cargo, when facing different jobs tasks during operation. A that facilitates multitasking states can collaboration between operation ship. demand for human flow logistics in states. To express mathematical form, there a certain random uncertainty numerical quantification...
This paper proposes a collaborative optimization model of car-flow organization for freight trains based on adjacent technical stations to minimize the average dwell time train cars in yard. To solve problems, priority-based hump sequence, which depends available two stations, is adopted. Furthermore, meta-heuristic algorithm genetic and taboo search adopted model, introduction active scheduling method improves efficiency algorithm. Finally, applied problem results are compared with those...
Airport gate assignment performance indicator selection is a complicated decision-making problem with strong subjectivity and difficulty in measuring the importance of each indicator. A better indicators (PIs) can greatly increase airport overall benefit. We adopt multicriteria approach to quantify qualitative PIs conduct subsequent using fuzzy clustering method. First, we identify 21 commonly used through literature review survey. Subsequently, analytic hierarchy process technique was...
Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, procedure to optimizing mixing instruction datasets for LLM fine-tuning is still poorly understood. This study categorizes instructions into three primary types: NLP downstream tasks, coding, and general chat. We explore effects on different combinations performance, find that certain types are more advantageous specific applications but can negatively impact other areas....
Deep learning based models have dominated the current landscape of production recommender systems. Furthermore, recent years witnessed an exponential growth model scale--from Google's 2016 with 1 billion parameters to latest Facebook's 12 trillion parameters. Significant quality boost has come each jump capacity, which makes us believe era 100 is around corner. However, training such challenging even within industrial scale data centers. This difficulty inherited from staggering...