The multi scale modeling method was utilized to study the bending characteristics of a carbon nanotube （CNT） and CNT reinforced composites. Through combining molecular dynamics and continuum mechanics, the tensional...The multi scale modeling method was utilized to study the bending characteristics of a carbon nanotube （CNT） and CNT reinforced composites. Through combining molecular dynamics and continuum mechanics, the tensional and flexural modulus of a CNT were calculated by a finite element model constructed by reticulate beams with solid cylinder shape and energy equal to C-C bonds. Then, another beam element with hollow cylinder shape and equivalent stiffness was utilized in place of a CNT in a matrix, thus, a multi scale representative volume element （RVE） model of CNT reinforced composite was established. Using this RVE model, the bending behavior of CNT based composites was analyzed. The influence of diameter D, length L, aspect ratio L/D, volume fraction, chiral of CNTs and shape of RVE as well as the arrangement of CNTs in matrix on the rein forcement effect of flexural modulus of resultant nanocomposites were further discussed. The ob tained data provide useful information for the design of CNT reinforced composites.展开更多
Protein science is an interdisciplinary research field of understanding the structure,function,and interactions of proteins,as well as their functions in biological processes.Protein science provides particularly impo...Protein science is an interdisciplinary research field of understanding the structure,function,and interactions of proteins,as well as their functions in biological processes.Protein science provides particularly important tools for drug discovery,1 a process characterized by long cycling,high risk,and high investment.To accelerate drug discovery,several challenges remain in advancing protein science:(1)Multi-scale modeling:elements related to proteins have a variety of scales,from atoms to cells;(2)Property estimation:the property of proteins is determined by many factors,which challenges accurate measurement;(3)Structure understanding:the complex and dynamic biological structures of proteins are hard to estimate.While traditional methods have been developed based on biological experiments,they are expensive and time-consuming.With the availability of big data and high computing power,emerging data-driven technologies powered by high-performance computing have been revolutionizing scientific discovery due to their significant advantages in pattern recognition and predictive modeling.Fig.1 shows an overview of various protein science-related artificial intelligence(AI)models.A landmark model,AlphaFold2,scored 92.4 points on the CASP14 standard dataset in the protein folding prediction task,indicating that its predicted structures closely match the real structure.2 The high performance of AI models on complex conformation prediction shows the great potential of AI in biological modeling,demonstrating the possibility of AI technology enhancing the efficiency of biological research to accelerate the traditional pathway of drug discovery.展开更多
基金Supported by the National Natural Science Foundation of China(50975011)
文摘The multi scale modeling method was utilized to study the bending characteristics of a carbon nanotube （CNT） and CNT reinforced composites. Through combining molecular dynamics and continuum mechanics, the tensional and flexural modulus of a CNT were calculated by a finite element model constructed by reticulate beams with solid cylinder shape and energy equal to C-C bonds. Then, another beam element with hollow cylinder shape and equivalent stiffness was utilized in place of a CNT in a matrix, thus, a multi scale representative volume element （RVE） model of CNT reinforced composite was established. Using this RVE model, the bending behavior of CNT based composites was analyzed. The influence of diameter D, length L, aspect ratio L/D, volume fraction, chiral of CNTs and shape of RVE as well as the arrangement of CNTs in matrix on the rein forcement effect of flexural modulus of resultant nanocomposites were further discussed. The ob tained data provide useful information for the design of CNT reinforced composites.
基金supported by the National Science and Technology Major Project(2023ZD0121401).
文摘Protein science is an interdisciplinary research field of understanding the structure,function,and interactions of proteins,as well as their functions in biological processes.Protein science provides particularly important tools for drug discovery,1 a process characterized by long cycling,high risk,and high investment.To accelerate drug discovery,several challenges remain in advancing protein science:(1)Multi-scale modeling:elements related to proteins have a variety of scales,from atoms to cells;(2)Property estimation:the property of proteins is determined by many factors,which challenges accurate measurement;(3)Structure understanding:the complex and dynamic biological structures of proteins are hard to estimate.While traditional methods have been developed based on biological experiments,they are expensive and time-consuming.With the availability of big data and high computing power,emerging data-driven technologies powered by high-performance computing have been revolutionizing scientific discovery due to their significant advantages in pattern recognition and predictive modeling.Fig.1 shows an overview of various protein science-related artificial intelligence(AI)models.A landmark model,AlphaFold2,scored 92.4 points on the CASP14 standard dataset in the protein folding prediction task,indicating that its predicted structures closely match the real structure.2 The high performance of AI models on complex conformation prediction shows the great potential of AI in biological modeling,demonstrating the possibility of AI technology enhancing the efficiency of biological research to accelerate the traditional pathway of drug discovery.
