Advanced programmable metamaterials with heterogeneous microstructures have become increasingly prevalent in scientific and engineering disciplines attributed to their tunable properties.However,exploring the structur...Advanced programmable metamaterials with heterogeneous microstructures have become increasingly prevalent in scientific and engineering disciplines attributed to their tunable properties.However,exploring the structure-property relationship in these materials,including forward prediction and inverse design,presents substantial challenges.The inhomogeneous microstructures significantly complicate traditional analytical or simulation-based approaches.Here,we establish a novel framework that integrates the machine learning(ML)-encoded multiscale computational method for forward prediction and Bayesian optimization for inverse design.Unlike prior end-to-end ML methods limited to specific problems,our framework is both load-independent and geometry-independent.This means that a single training session for a constitutive model suffices to tackle various problems directly,eliminating the need for repeated data collection or training.We demonstrate the efficacy and efficiency of this framework using metamaterials with designable elliptical holes or lattice honeycombs microstructures.Leveraging accelerated forward prediction,we can precisely customize the stiffness and shape of metamaterials under diverse loading scenarios,and extend this capability to multi-objective customization seamlessly.Moreover,we achieve topology optimization for stress alleviation at the crack tip,resulting in a significant reduction of Mises stress by up to 41.2%and yielding a theoretical interpretable pattern.This framework offers a general,efficient and precise tool for analyzing the structure-property relationships of novel metamaterials.展开更多
This paper mainly discusses the multiscale computation from a chemical engineering perspective.From the application designer's perspective,we propose a new approach to investigate and develop both flexible and eff...This paper mainly discusses the multiscale computation from a chemical engineering perspective.From the application designer's perspective,we propose a new approach to investigate and develop both flexible and efficient computer architectures. Based on the requirements of applications within one category,we first induce and extract some inherent computing patterns or core computing kernels from the applications.Some computing models and innovative computing architectures will then be developed for these patterns or kernels,as well as the software mapping techniques. Finally those applications which can share and utilize those computing patterns or kernels can be executed very efficiently on those novel computing architectures. We think that the proposed approach may not be achievable within the existing technology. However,we believe that it will be available in the near future. Hence,we will describe this approach from the following four aspects:multiscale environment in the world,mesoscale as a key scale,energy minimization multiscale(EMMS)paradigm and our perspective.展开更多
Based upon advances in theoretical algorithms, modeling and simulations, and computer technologies, the rational design of materials, cells, devices, and packs in the field of lithium-ion batteries is being realized i...Based upon advances in theoretical algorithms, modeling and simulations, and computer technologies, the rational design of materials, cells, devices, and packs in the field of lithium-ion batteries is being realized incrementally and will at some point trigger a paradigm revolution by combining calculations and experiments linked by a big shared database, enabling accelerated development of the whole industrial chain. Theory and multi-scale modeling and simulation, as supplements to experimental efforts, can help greatly to close some of the current experimental and technological gaps, as well as predict path-independent properties and help to fundamentally understand path-independent performance in multiple spatial and temporal scales.展开更多
In this paper,we formulate a two-way interfacial condition for simulating lattice dynamics in one space dimension.With a time history treatment,the incoming waves are incorporated into the motion of the boundary atoms...In this paper,we formulate a two-way interfacial condition for simulating lattice dynamics in one space dimension.With a time history treatment,the incoming waves are incorporated into the motion of the boundary atoms accurately.This condition reduces to the absorbing boundary condition when there is no incoming wave.Numerical tests validate the effectiveness of the proposed condition in treating simultaneously incoming waves and outgoing waves.展开更多
基金supported by the National Natural Science Foundation of China (Grant Nos.12102021,12372105,12172026,and 12225201)the Fundamental Research Funds for the Central Universities and the Academic Excellence Foundation of BUAA for PhD Students.
文摘Advanced programmable metamaterials with heterogeneous microstructures have become increasingly prevalent in scientific and engineering disciplines attributed to their tunable properties.However,exploring the structure-property relationship in these materials,including forward prediction and inverse design,presents substantial challenges.The inhomogeneous microstructures significantly complicate traditional analytical or simulation-based approaches.Here,we establish a novel framework that integrates the machine learning(ML)-encoded multiscale computational method for forward prediction and Bayesian optimization for inverse design.Unlike prior end-to-end ML methods limited to specific problems,our framework is both load-independent and geometry-independent.This means that a single training session for a constitutive model suffices to tackle various problems directly,eliminating the need for repeated data collection or training.We demonstrate the efficacy and efficiency of this framework using metamaterials with designable elliptical holes or lattice honeycombs microstructures.Leveraging accelerated forward prediction,we can precisely customize the stiffness and shape of metamaterials under diverse loading scenarios,and extend this capability to multi-objective customization seamlessly.Moreover,we achieve topology optimization for stress alleviation at the crack tip,resulting in a significant reduction of Mises stress by up to 41.2%and yielding a theoretical interpretable pattern.This framework offers a general,efficient and precise tool for analyzing the structure-property relationships of novel metamaterials.
文摘This paper mainly discusses the multiscale computation from a chemical engineering perspective.From the application designer's perspective,we propose a new approach to investigate and develop both flexible and efficient computer architectures. Based on the requirements of applications within one category,we first induce and extract some inherent computing patterns or core computing kernels from the applications.Some computing models and innovative computing architectures will then be developed for these patterns or kernels,as well as the software mapping techniques. Finally those applications which can share and utilize those computing patterns or kernels can be executed very efficiently on those novel computing architectures. We think that the proposed approach may not be achievable within the existing technology. However,we believe that it will be available in the near future. Hence,we will describe this approach from the following four aspects:multiscale environment in the world,mesoscale as a key scale,energy minimization multiscale(EMMS)paradigm and our perspective.
基金supported by the National Natural Science Foundation of China(Grant Nos.51372228 and 11234013)the National High Technology Research and Development Program of China(Grant No.2015AA034201)Shanghai Pujiang Program,China(Grant No.14PJ1403900)
文摘Based upon advances in theoretical algorithms, modeling and simulations, and computer technologies, the rational design of materials, cells, devices, and packs in the field of lithium-ion batteries is being realized incrementally and will at some point trigger a paradigm revolution by combining calculations and experiments linked by a big shared database, enabling accelerated development of the whole industrial chain. Theory and multi-scale modeling and simulation, as supplements to experimental efforts, can help greatly to close some of the current experimental and technological gaps, as well as predict path-independent properties and help to fundamentally understand path-independent performance in multiple spatial and temporal scales.
基金supported in part by NSFC under contract number 10872004,National Basic Research Program of China under contract number 2007CB814800 and 2010CB731500the Ministry of Education of China under contract numbers NCET-06-0011 and 200800010013.
文摘In this paper,we formulate a two-way interfacial condition for simulating lattice dynamics in one space dimension.With a time history treatment,the incoming waves are incorporated into the motion of the boundary atoms accurately.This condition reduces to the absorbing boundary condition when there is no incoming wave.Numerical tests validate the effectiveness of the proposed condition in treating simultaneously incoming waves and outgoing waves.
