A5100323245- Metaheuristic Optimization Algorithms Research
- Computational Drug Discovery Methods
- Adaptive Dynamic Programming Control
- Advanced Multi-Objective Optimization Algorithms
- Protein Degradation and Inhibitors
- Advanced Algorithms and Applications
- Graph Labeling and Dimension Problems
- Adaptive Control of Nonlinear Systems
- Machine Learning in Materials Science
- Advanced Graph Theory Research
- Thin-Film Transistor Technologies
- Reinforcement Learning in Robotics
- Enzyme Catalysis and Immobilization
- vaccines and immunoinformatics approaches
- Model-Driven Software Engineering Techniques
- graph theory and CDMA systems
- Matrix Theory and Algorithms
- Silicon and Solar Cell Technologies
- Formal Methods in Verification
- Advanced Neural Network Applications
- Electron and X-Ray Spectroscopy Techniques
- Distributed Control Multi-Agent Systems
- Protein Structure and Dynamics
- Coding theory and cryptography
- Frequency Control in Power Systems
Jiangnan University
2018-2024
Northeast Normal University
2023
National Supercomputing Center in Wuxi
2022
Shangluo University
2021
National Computer Network Emergency Response Technical Team/Coordination Center of Chinar
2019
Xi’an University of Posts and Telecommunications
2017
Shandong Institute of Automation
2014-2016
Chinese Academy of Sciences
2014-2016
Institute of Automation
2016
Worcester Polytechnic Institute
2016
In order to effectively identify and classify weld defects of thin-walled metal canisters, a defect detection classification algorithm based on machine vision is proposed in this paper. With the categorized, modified background subtraction method Gaussian mixture models, extract feature areas defects. Then, we design an for according extracted features. Next, by using images sampled constructed system real-world production line, parameters classifiers are determined empirically. Experimental...
Protein-ligand docking has emerged as a particularly important tool in drug design and development, flexible ligand is widely used method for simulations. Many software packages can simulate docking, among them, Autodock used. Focusing on the search algorithm Autodock, many new optimization approaches have been proposed over last few decades. However, despite large number of alternatives, we are still lacking with high robustness performance. In this paper, conjunction popular software,...
The research on the hydrogen passivation process for silicon solar cells has been developed a long time. Recently, researchers have investigated use of lasers as light sources. It found that technology can improve minority carrier lifetime and passivate some impurities defects in cells. In this paper, hydrogenation platform based high intensity infrared LEDs 940 nm was built. After number comparative experiments, electrical characteristics, photo-luminescence images, beam induced current...
In general, flexible ligand docking is used for simulations under the premise that position of binding site already known, and meanwhile it can also be without prior knowledge site. However, most optimization search algorithms in popular software are far from being ideal first case, they hardly directly utilized latter case due to relatively large area. order design an algorithm flexibly adapt different sizes area, we propose effective swarm intelligence this paper, called...
Autodock and its various variants are widely utilized docking approaches, which adopt optimization methods as search algorithms for flexible ligand virtual screening. However, many of them have their limitations, such poor accuracy dockings with highly ligands low efficiency. In this paper, a multi-swarm algorithm integrated environment is proposed to design high-performance high-efficiency program, namely, MSLDOCK. The combination the random drift particle swarm novel strategy Solis Wets...
A high-quality docking method tends to yield multifold gains with half pains for the new drug development. Over past few decades, great efforts have been made development of novel programs efficiency and intriguing accuracy. AutoDock Vina (Vina) is one these achievements improved speed accuracy compared AutoDock4. Since it was proposed, some its variants, such as PSOVina GWOVina, also developed. However, all programs, there still large room performance improvement.In this work, we propose a...
Abstract Premature convergence is a thorny problem for particle swarm optimization (PSO) algorithms, especially on multimodal problems, where maintaining diversity crucial. However, most enhancement strategies PSO, including the existing diversity-guided strategies, have not fully addressed this issue. This paper proposes virtual position guided (VPG) strategy PSO algorithms. The VPG calculates values two different populations and establishes baseline. It then dynamically guides algorithm to...
In recent years, deep learning models have become predominant methods for computer vision tasks, but the large computation and storage requirements of many make them challenging to deploy on devices with limited resources. Knowledge distillation (KD) is a widely used approach model compression. However, when applied in object detection problems, existing KD either directly applies feature map or simply separate foreground from background by using binary mask, aligning attention between...
Metabolic engineering is a rapidly evolving field that involves optimizing microbial cell factories to overproduce various industrial products. To achieve this, several tools, leveraging constraint-based stoichiometric models and metaheuristic algorithms like particle swarm optimization (PSO), have been developed. However, PSO can potentially get trapped in local optima. Quantum-behaved (QPSO) overcomes this limitation, our study further enhances its binary version (BQPSO) with neighborhood...
Object-oriented development of real-time systems is becoming more and prevalent. Unified Modeling Language (UML) a standardized notation for describing object-oriented software design. While using UML to specify systems, the formal validation certain timing constraints becomes critical success systems. Current methods have not provided consistent support verifying models This paper presents specification & method, FORTS, system UML. The method extension UML; describes semantics extension;...
AutoDock is a widely used software for flexible ligand docking problems since it open source and easy to be implemented. In this paper, novel hybrid algorithm proposed applied in the environment of version 4.2.6 order enhance accuracy efficiency dockings with ligands. This search algorithm, called entropy-based Lamarckian quantum-behaved particle swarm optimization (ELQPSO), combination QPSO an update strategy Solis Wet local (SWLS) method. By using PDBbind core set v.2016, ELQPSO compared...
In this paper, we establish a decentralized control law to stabilize class of nonlinear interconnected large-scale systems using neural-network-based online model-free integral policy iteration algorithm. The approach can solve the problem for system which has unknown dynamics. stabilizing strategy is derived based on optimal policies isolated subsystems. algorithm developed problems subsystems with actor-critic technique and least squares implementation method are used obtain policies. A...