Dongguan is an important industrial city, located in the Pearl River Delta, South China. Recently, Dongguan city experienced a rapid urban growth with the locational advantage by transforming from traditional agricult...Dongguan is an important industrial city, located in the Pearl River Delta, South China. Recently, Dongguan city experienced a rapid urban growth with the locational advantage by transforming from traditional agricultural region to modern manufacturing metropolis. The urban transformation became the usual change in China under the background of urbanization which belongs to one trend of globalization in the 21st century. This paper tries to analyze urban growth simulation based on remotely sensed data of previous years and the related physical and socio-economic factors and predict future urban growth in 2024. The study examined and compared the land use/cover (LUC) changes over time based on produced maps of 2004, 2009, and 2014. The results showed that water and forest area decreased since the past years. In contrast, the urban land increased from 2004 to 2014, and this increasing trend will continue to the future years through the urbanization process. Having understood the spatiotemporal trends of urban growth, the study simulated the urban growth of Dongguan city for 2024 using neural network simulation technique. Further, the figure of merit (FoM) of simulated map of 2014 map was 8.86%, which can be accepted in the simulation and used in the prediction process. Based on the consideration of water body and forest, the newly growth area is located in the west, northeast, and southeast regions of Dongguan city. The finding can help us to understand which areas are going to be considered in the future urban planning and policy by the local government.展开更多
Instead of existing methods,a recurrent neural network is conceived to deal with three stages of portfolio management.Mainly,a deterministic annealing neural network is proposed for the approach to portfolio problem,w...Instead of existing methods,a recurrent neural network is conceived to deal with three stages of portfolio management.Mainly,a deterministic annealing neural network is proposed for the approach to portfolio problem,which is a kind of quadratic programming.Finally,through a real example,we verify that the neural network model proposed in this paper is a good tool to solve the portfolio problem.展开更多
LASP(large-scale atomistic simulation with neural network potential)software developed by our group since 2018 is a powerful platform(www.lasphub.com)for performing atomic simulation of complex materials.The software ...LASP(large-scale atomistic simulation with neural network potential)software developed by our group since 2018 is a powerful platform(www.lasphub.com)for performing atomic simulation of complex materials.The software integrates the neural network(NN)potential technique with the global potential energy surface exploration method,and thus can be utilized widely for structure prediction and reaction mechanism exploration.Here we introduce our recent update on the LASP program version 3.0,focusing on the new functionalities including the advanced neuralnetwork training based on the multi-network framework,the newly-introduced S^(7) and S^(8) power type structure descriptor(PTSD).These new functionalities are designed to further improve the accuracy of potentials and accelerate the neural network training for multipleelement systems.Taking Cu-C-H-O neural network potential and a heterogeneous catalytic model as the example,we show that these new functionalities can accelerate the training of multi-element neural network potential by using the existing single-network potential as the input.The obtained double-network potential Cu CHO is robust in simulation and the introduction of S^(7) and S^(8) PTSDs can reduce the root-mean-square errors of energy by a factor of two.展开更多
Memtransistors combine memristors and field-effect transistors, which can introduce multi-port control and have significant applications for enriching storage methods. In this paper, multilayer α-In2Se3and MoS2were t...Memtransistors combine memristors and field-effect transistors, which can introduce multi-port control and have significant applications for enriching storage methods. In this paper, multilayer α-In2Se3and MoS2were transferred to the substrate by the mechanical exfoliation method, then a heterojunction MoS_(2)/α-In_(2)Se_(3) memtransistor was prepared. Neural synaptic simulations were performed using electrical and optical pulses as input signals. Through measurements, such as excitatory/inhibitory post-synaptic current(EPSC/IPSC), long-term potentiation/depression(LTP/LTD), and paired-pulse facilitation/depression(PPF/PPD), it can be found that the fabricated device could simulate various functions of neural synapses well, and could work as an electronic synapse in artificial neural networks, proposing a possible solution for neuromorphic storage and computation.展开更多
Fast simulated annealing is implemented into the learning process of neural network to replace the traditional back-propagation algorithm. The new procedure exhibits performance fast in learning and accurate in predic...Fast simulated annealing is implemented into the learning process of neural network to replace the traditional back-propagation algorithm. The new procedure exhibits performance fast in learning and accurate in prediction compared to the traditional neural networks. Two numerical data sets were used to illustrate its use in chemistry.展开更多
Spiking Neural Network(SNN)simulation is very important for studying brain function and validating the hypotheses for neuroscience,and it can also be used in artificial intelligence.Recently,GPU-based simulators have ...Spiking Neural Network(SNN)simulation is very important for studying brain function and validating the hypotheses for neuroscience,and it can also be used in artificial intelligence.Recently,GPU-based simulators have been developed to support the real-time simulation of SNN.However,these simulators’simulating performance and scale are severely limited,due to the random memory access pattern and the global communication between devices.Therefore,we propose an efficient distributed heterogeneous SNN simulator based on the Sunway accelerators(including SW26010 and SW26010pro),named SWsnn,which supports accurate simulation with small time step(1/16 ms),randomly delay sizes for synapses,and larger scale network computing.Compared with existing GPUs,the Local Dynamic Memory(LDM)(similar to cache)in Sunway is much bigger(4 MB or 16 MB in each core group).To improve the simulation performance,we redesign the network data storage structure and the synaptic plasticity flow to make most random accesses occur in LDM.SWsnn hides Message Passing Interface(MPI)-related operations to reduce communication costs by separating SNN general workflow.Besides,SWsnn relies on parallel Compute Processing Elements(CPEs)rather than serial Manage Processing Element(MPE)to control the communicating buffers,using Register-Level Communication(RLC)and Direct Memory Access(DMA).In addition,SWsnn is further optimized using vectorization and DMA hiding techniques.Experimental results show that SWsnn runs 1.4−2.2 times faster than state-of-the-art GPU-based SNN simulator GPU-enhanced Neuronal Networks(GeNN),and supports much larger scale real-time simulation.展开更多
In this study, it is attempted to give an insight into the injection processability of three self-prepared polymers from A to Z. This work presents material analysis, injection molding simulation, design of ex- perime...In this study, it is attempted to give an insight into the injection processability of three self-prepared polymers from A to Z. This work presents material analysis, injection molding simulation, design of ex- periments alongside considering all interaction effects of controlling parameters carefully for green biodegradable polymeric systems, including polylactic acid （PLA）, polylactic acid-thermoplastic poly- urethane （PLA-TPU） and polylactic acid-thermoplastic starch （PLA-TPS）. The experiments were carried out using injection molding simulation software Autodesk Moldflov~~ in order to minimize warpage and volumetric shrinkage for each of the mentioned systems. The analysis was conducted by changing five significant processing parameters, including coolant temperature, packing time, packing pressure, mold temperature and melt temperature. Taguchi＇s [.27 （35） orthogonal array was selected as an efficient method for design of simulations in order to consider the interaction effects of the parameters and reduce spu- rious simulations. Meanwhile, artificial neural network （ANN） was also used for pattern recognition and optimization through modifying the processing conditions. The Taguchi coupled analysis of variance （ANOVA） and ANN analysis resulted in definition of optimum levels for each factor by two completely different methods. According to the results, melting temperature, coolant temperature and packing time had significant influence on the shrinkage and warpage. The ANN optimal level selection for minimiza- tion of shrinkage and/or warpage is in good agreement with ANOVA optimal level selection results. This investigation indicates that PLA-TPU compound exhibits the highest resistance to warpage and shrink- age defects compared to the other studied compounds.展开更多
Bioelectronics are powerful tools for monitoring and stimulating biological and biochemical processes,with applications ranging from neural interface simulation to biosensing.The increasing demand for bioelectronics h...Bioelectronics are powerful tools for monitoring and stimulating biological and biochemical processes,with applications ranging from neural interface simulation to biosensing.The increasing demand for bioelectronics has greatly promoted the development of new nanomaterials as detection platforms.Recently,owing to their ultrathin structures and excellent physicochemical properties,emerging two-dimensional(2D)materials have become one of the most researched areas in the fields of bioelectronics and biosensors.In this timely review,the physicochemical structures of the most representative emerging 2D materials and the design of their nanostructures for engineering highperformance bioelectronic and biosensing devices are presented.We focus on the structural optimization of emerging 2D material-based composites to achieve better regulation for enhancing the performance of bioelectronics.Subsequently,the recent developments of emerging 2D materials in bioelectronics,such as neural interface simulation,biomolecular/biomarker detection,and skin sensors are discussed thoroughly.Finally,we provide conclusive views on the current challenges and future perspectives on utilizing emerging 2D materials and their composites for bioelectronics and biosensors.This review will offer important guidance in designing and applying emerging 2D materials in bioelectronics,thus further promoting their prospects in a wide biomedical field.展开更多
文摘Dongguan is an important industrial city, located in the Pearl River Delta, South China. Recently, Dongguan city experienced a rapid urban growth with the locational advantage by transforming from traditional agricultural region to modern manufacturing metropolis. The urban transformation became the usual change in China under the background of urbanization which belongs to one trend of globalization in the 21st century. This paper tries to analyze urban growth simulation based on remotely sensed data of previous years and the related physical and socio-economic factors and predict future urban growth in 2024. The study examined and compared the land use/cover (LUC) changes over time based on produced maps of 2004, 2009, and 2014. The results showed that water and forest area decreased since the past years. In contrast, the urban land increased from 2004 to 2014, and this increasing trend will continue to the future years through the urbanization process. Having understood the spatiotemporal trends of urban growth, the study simulated the urban growth of Dongguan city for 2024 using neural network simulation technique. Further, the figure of merit (FoM) of simulated map of 2014 map was 8.86%, which can be accepted in the simulation and used in the prediction process. Based on the consideration of water body and forest, the newly growth area is located in the west, northeast, and southeast regions of Dongguan city. The finding can help us to understand which areas are going to be considered in the future urban planning and policy by the local government.
基金Supported by the National Science Foundatin of China (No.79670 0 64)
文摘Instead of existing methods,a recurrent neural network is conceived to deal with three stages of portfolio management.Mainly,a deterministic annealing neural network is proposed for the approach to portfolio problem,which is a kind of quadratic programming.Finally,through a real example,we verify that the neural network model proposed in this paper is a good tool to solve the portfolio problem.
基金supported by the National Key Research and Development Program of China (No.2018YFA0208600)the National Natural Science Foundation of China (No.91945301, No.22033003, No.92061112, No.22122301, and No.91745201)
文摘LASP(large-scale atomistic simulation with neural network potential)software developed by our group since 2018 is a powerful platform(www.lasphub.com)for performing atomic simulation of complex materials.The software integrates the neural network(NN)potential technique with the global potential energy surface exploration method,and thus can be utilized widely for structure prediction and reaction mechanism exploration.Here we introduce our recent update on the LASP program version 3.0,focusing on the new functionalities including the advanced neuralnetwork training based on the multi-network framework,the newly-introduced S^(7) and S^(8) power type structure descriptor(PTSD).These new functionalities are designed to further improve the accuracy of potentials and accelerate the neural network training for multipleelement systems.Taking Cu-C-H-O neural network potential and a heterogeneous catalytic model as the example,we show that these new functionalities can accelerate the training of multi-element neural network potential by using the existing single-network potential as the input.The obtained double-network potential Cu CHO is robust in simulation and the introduction of S^(7) and S^(8) PTSDs can reduce the root-mean-square errors of energy by a factor of two.
基金Project supported by the National Natural Science Foundation of China (Grant No. 51702245)。
文摘Memtransistors combine memristors and field-effect transistors, which can introduce multi-port control and have significant applications for enriching storage methods. In this paper, multilayer α-In2Se3and MoS2were transferred to the substrate by the mechanical exfoliation method, then a heterojunction MoS_(2)/α-In_(2)Se_(3) memtransistor was prepared. Neural synaptic simulations were performed using electrical and optical pulses as input signals. Through measurements, such as excitatory/inhibitory post-synaptic current(EPSC/IPSC), long-term potentiation/depression(LTP/LTD), and paired-pulse facilitation/depression(PPF/PPD), it can be found that the fabricated device could simulate various functions of neural synapses well, and could work as an electronic synapse in artificial neural networks, proposing a possible solution for neuromorphic storage and computation.
文摘Fast simulated annealing is implemented into the learning process of neural network to replace the traditional back-propagation algorithm. The new procedure exhibits performance fast in learning and accurate in prediction compared to the traditional neural networks. Two numerical data sets were used to illustrate its use in chemistry.
基金supported by the Key Research and Development Project of Guangdong Province(No.2021B0101310002)the National Key Research and Development Program of China(No.2021YFF1200104)+4 种基金the Strategic Priority CAS Project(No.XDB38050100)the National Natural Science Foundation of China(No.62272449)the Shenzhen Basic Research Fund(Nos.RCYX20200714114734194,JCYJ20210324102007021,and KQTD20200820113106007)the Open Fund of Guangdong Laboratory of Artificial Intelligence and Digital Economy(SZ)(No.GML-KF-22-13)the Shenzhen Key Laboratory of Intelligent Bioinformatics(No.ZDSYS20220422103800001).
文摘Spiking Neural Network(SNN)simulation is very important for studying brain function and validating the hypotheses for neuroscience,and it can also be used in artificial intelligence.Recently,GPU-based simulators have been developed to support the real-time simulation of SNN.However,these simulators’simulating performance and scale are severely limited,due to the random memory access pattern and the global communication between devices.Therefore,we propose an efficient distributed heterogeneous SNN simulator based on the Sunway accelerators(including SW26010 and SW26010pro),named SWsnn,which supports accurate simulation with small time step(1/16 ms),randomly delay sizes for synapses,and larger scale network computing.Compared with existing GPUs,the Local Dynamic Memory(LDM)(similar to cache)in Sunway is much bigger(4 MB or 16 MB in each core group).To improve the simulation performance,we redesign the network data storage structure and the synaptic plasticity flow to make most random accesses occur in LDM.SWsnn hides Message Passing Interface(MPI)-related operations to reduce communication costs by separating SNN general workflow.Besides,SWsnn relies on parallel Compute Processing Elements(CPEs)rather than serial Manage Processing Element(MPE)to control the communicating buffers,using Register-Level Communication(RLC)and Direct Memory Access(DMA).In addition,SWsnn is further optimized using vectorization and DMA hiding techniques.Experimental results show that SWsnn runs 1.4−2.2 times faster than state-of-the-art GPU-based SNN simulator GPU-enhanced Neuronal Networks(GeNN),and supports much larger scale real-time simulation.
文摘In this study, it is attempted to give an insight into the injection processability of three self-prepared polymers from A to Z. This work presents material analysis, injection molding simulation, design of ex- periments alongside considering all interaction effects of controlling parameters carefully for green biodegradable polymeric systems, including polylactic acid （PLA）, polylactic acid-thermoplastic poly- urethane （PLA-TPU） and polylactic acid-thermoplastic starch （PLA-TPS）. The experiments were carried out using injection molding simulation software Autodesk Moldflov~~ in order to minimize warpage and volumetric shrinkage for each of the mentioned systems. The analysis was conducted by changing five significant processing parameters, including coolant temperature, packing time, packing pressure, mold temperature and melt temperature. Taguchi＇s [.27 （35） orthogonal array was selected as an efficient method for design of simulations in order to consider the interaction effects of the parameters and reduce spu- rious simulations. Meanwhile, artificial neural network （ANN） was also used for pattern recognition and optimization through modifying the processing conditions. The Taguchi coupled analysis of variance （ANOVA） and ANN analysis resulted in definition of optimum levels for each factor by two completely different methods. According to the results, melting temperature, coolant temperature and packing time had significant influence on the shrinkage and warpage. The ANN optimal level selection for minimiza- tion of shrinkage and/or warpage is in good agreement with ANOVA optimal level selection results. This investigation indicates that PLA-TPU compound exhibits the highest resistance to warpage and shrink- age defects compared to the other studied compounds.
基金1.3.5 Project for Disciplines of Excellence,West China Hospital,Sichuan University,Grant/Award Number:ZYJC21047China Postdoctoral Science Foundation,Grant/Award Numbers:2021M692291,2021M692288,2021M702334+8 种基金Fundamental Research Funds for the Central Universities,Grant/Award Numbers:2021SCU12034,2021SCU12013Med-X Center for Materials,Sichuan University,Grant/Award Number:MCM202102National Natural Science Foundation of China,Grant/Award Numbers:82001824,82001829,51903178,81971622,52173133,82102064,82102065,82071938Post-Doctor Research Project,West China Hospital,Sichuan University,Grant/Award Numbers:2020HXBH071,2020HXBH126the National Key R D Program of China,Grant/Award Numbers:2021YFE0205000,2019YFA0110600,2019YFA0110601the Science and Technology Project of Sichuan Province,Grant/Award Numbers:2021YFH0087,2021YFH0135,2021YFS0050,2021YFH0180,2021YJ0434,2021YJ0554,21YYJC2714,21ZDYF376the Science and Technology Project of the Health Planning Committee of Sichuan,Grant/Award Number:20PJ049the State Key Laboratory of Polymer Materials Engineering,Grant/Award Number:sklpme2021-4-02Thousand Youth Talents Plan。
文摘Bioelectronics are powerful tools for monitoring and stimulating biological and biochemical processes,with applications ranging from neural interface simulation to biosensing.The increasing demand for bioelectronics has greatly promoted the development of new nanomaterials as detection platforms.Recently,owing to their ultrathin structures and excellent physicochemical properties,emerging two-dimensional(2D)materials have become one of the most researched areas in the fields of bioelectronics and biosensors.In this timely review,the physicochemical structures of the most representative emerging 2D materials and the design of their nanostructures for engineering highperformance bioelectronic and biosensing devices are presented.We focus on the structural optimization of emerging 2D material-based composites to achieve better regulation for enhancing the performance of bioelectronics.Subsequently,the recent developments of emerging 2D materials in bioelectronics,such as neural interface simulation,biomolecular/biomarker detection,and skin sensors are discussed thoroughly.Finally,we provide conclusive views on the current challenges and future perspectives on utilizing emerging 2D materials and their composites for bioelectronics and biosensors.This review will offer important guidance in designing and applying emerging 2D materials in bioelectronics,thus further promoting their prospects in a wide biomedical field.
