A two-dimensional(2-D)multi-component and multi-phase cellular automaton(CA)model coupled with the Calphad method and finite difference method(FDM)is proposed to simulate the gas pore formation and microstructures in ...A two-dimensional(2-D)multi-component and multi-phase cellular automaton(CA)model coupled with the Calphad method and finite difference method(FDM)is proposed to simulate the gas pore formation and microstructures in solidification process of hypoeutectic Al-Si-Mg alloys.In this model,the pore growth,and dendritic and eutectic solidification are simulated using a CA technique.To achieve the equilibrium among multiple phases during ternary Al-based alloy solidification,the phase transition thermodynamics and kinetics are evaluated by adopting the Calphad method.The diffusion equations of hydrogen and two solutes are solved by FDM.The developed CA-FDM coupled model can be used for simulating the evolution of gas microporosity and microstructures,involving dendrites and irregular binary and ternary eutectics,of ternary hypoeutectic Al-Si-Mg alloys.It has the capability of reproducing the interactions between the hydrogen microporosity formation and the growth of dendrites and eutectics,the competitive growth among the growing gas pores of different sizes,together with the time-evolving concentration fields of hydrogen and solutes.The simulated morphology of gas pore and microstructure has a good agreement with the experimental observation.The influences of the initial hydrogen concentration and cooling rate on the microporosity formation are investigated.It is found that the main portion of porosity formation occurs in the eutectic solidification stage through analyzing the profiles of porosity percentage and solid fraction varying with solidification time.The varying features of simulated porosity percentage,the maximum and average pores radii indicate that increasing initial hydrogen concentration promotes the formation of higher final porosity percentage and larger pores,while the size of gas pores will significantly reduce with increasing cooling rate,leading to a lower final porosity percentage.展开更多
Wind speed prediction is of great importance because it affects the efficiency and stability of power systems with a high proportion of wind power.Temporal-spatial wind speed features contain rich information;however,...Wind speed prediction is of great importance because it affects the efficiency and stability of power systems with a high proportion of wind power.Temporal-spatial wind speed features contain rich information;however,their use to predict wind speed remains one of the most challenging and less studied areas.This paper investigates the problem of predicting wind speeds for multiple sites using temporal and spatial features and proposes a novel two-layer attentionbased long short-term memory(LSTM),termed 2Attn-LSTM,a unified framework of encoder and decoder mechanisms to handle temporal-spatial wind speed data.To eliminate the unevenness of the original wind speed,we initially decompose the preprocessing data into IMF components by variational mode decomposition(VMD).Then,it encodes the spatial features of IMF components at the bottom of the model and decodes the temporal features to obtain each component's predicted value on the second layer.Finally,we obtain the ultimate prediction value after denormalization and superposition.We have performed extensive experiments for short-term predictions on real-world data,demonstrating that 2Attn-LSTM outperforms the four baseline methods.It is worth pointing out that the presented 2Atts-LSTM is a general model suitable for other spatial-temporal features.展开更多
We simulate the evolution of hydrogen concentration and gas pore formation as equiaxed dendrites grow during solidification of a hypoeutectic aluminum-silicon(Al-Si)alloy.The applied lattice Boltzmann-cellular automat...We simulate the evolution of hydrogen concentration and gas pore formation as equiaxed dendrites grow during solidification of a hypoeutectic aluminum-silicon(Al-Si)alloy.The applied lattice Boltzmann-cellular automaton-finite difference model incorporates the physical mechanisms of solute and hydrogen partitioning on the solid/liquid interface,as well as the transports of solute and hydrogen.After the quantitative validation by the simulation of capillary intrusion,the model is utilized to investigate the growth of the equiaxed dendrites and hydrogen porosity formation for an Al-(5 wt.%)Si alloy under different solidification conditions.The simulation data reveal that the gas pores favorably nucleate in the corners surrounded by the nearby dendrite arms.Then,the gas pores grow in a competitive mode.With the cooling rate increasing,the competition among different growing gas pores is found to be hindered,which accordingly increases the pore number density in the final solidification microstructure.In the late solidification stage,even though the solid fraction is increasing,the mean concentration of hydrogen in the residue melt tends to be constant,corresponding to a dynamic equilibrium state of hydrogen concentration in liquid.As the cooling rate increases or the initial hydrogen concentration decreases,the temperature of gas pore nucleation,the porosity fraction,and the mean porosity size decrease,whilst the mean hydrogen concentration in liquid increases in the late solidification stage.The simulated data present identical trends with the experimental results reported in literature.展开更多
Numerical estimates of the components of yield strength of a high strength Fe-C-Mn-P-N-Si enameling steel were determined using empirical relationships between microstructure and yield strength. Results are reported f...Numerical estimates of the components of yield strength of a high strength Fe-C-Mn-P-N-Si enameling steel were determined using empirical relationships between microstructure and yield strength. Results are reported for both the hot rolled （HR） and cold rolled （CR） forms before and after simulating an enamel-fire anneal （EFA）. To determine the solubilities of alloying elements, thermodynamic calculations were performed in combination with the considerations of process conditions and the element diffusivities. The results show that the main solid solution strengtheners were the elements Mn, Si, and P, while the elements C and N were nearly completely tied up as precipitates. The yield strength reduction, due to the EFA, resulted primarily from an increase in grain size and a decrease in dislocation density, and the EFA appeared to have a negligible effect on the element solubilities.展开更多
基金the National Natural Science Foundation of China(No.51371051)Jiangsu Key Laboratory of Advanced Metallic Materials(No.BM2007204)。
文摘A two-dimensional(2-D)multi-component and multi-phase cellular automaton(CA)model coupled with the Calphad method and finite difference method(FDM)is proposed to simulate the gas pore formation and microstructures in solidification process of hypoeutectic Al-Si-Mg alloys.In this model,the pore growth,and dendritic and eutectic solidification are simulated using a CA technique.To achieve the equilibrium among multiple phases during ternary Al-based alloy solidification,the phase transition thermodynamics and kinetics are evaluated by adopting the Calphad method.The diffusion equations of hydrogen and two solutes are solved by FDM.The developed CA-FDM coupled model can be used for simulating the evolution of gas microporosity and microstructures,involving dendrites and irregular binary and ternary eutectics,of ternary hypoeutectic Al-Si-Mg alloys.It has the capability of reproducing the interactions between the hydrogen microporosity formation and the growth of dendrites and eutectics,the competitive growth among the growing gas pores of different sizes,together with the time-evolving concentration fields of hydrogen and solutes.The simulated morphology of gas pore and microstructure has a good agreement with the experimental observation.The influences of the initial hydrogen concentration and cooling rate on the microporosity formation are investigated.It is found that the main portion of porosity formation occurs in the eutectic solidification stage through analyzing the profiles of porosity percentage and solid fraction varying with solidification time.The varying features of simulated porosity percentage,the maximum and average pores radii indicate that increasing initial hydrogen concentration promotes the formation of higher final porosity percentage and larger pores,while the size of gas pores will significantly reduce with increasing cooling rate,leading to a lower final porosity percentage.
基金This work is supported in part by the Priority Academic Program Development of Jiangsu Higher Education Institutions,Natural Science Foundation of China(No.61103141,No.61105007 and No.51405241)NARI Nanjing Control System Ltd.(No.524608190024).
文摘Wind speed prediction is of great importance because it affects the efficiency and stability of power systems with a high proportion of wind power.Temporal-spatial wind speed features contain rich information;however,their use to predict wind speed remains one of the most challenging and less studied areas.This paper investigates the problem of predicting wind speeds for multiple sites using temporal and spatial features and proposes a novel two-layer attentionbased long short-term memory(LSTM),termed 2Attn-LSTM,a unified framework of encoder and decoder mechanisms to handle temporal-spatial wind speed data.To eliminate the unevenness of the original wind speed,we initially decompose the preprocessing data into IMF components by variational mode decomposition(VMD).Then,it encodes the spatial features of IMF components at the bottom of the model and decodes the temporal features to obtain each component's predicted value on the second layer.Finally,we obtain the ultimate prediction value after denormalization and superposition.We have performed extensive experiments for short-term predictions on real-world data,demonstrating that 2Attn-LSTM outperforms the four baseline methods.It is worth pointing out that the presented 2Atts-LSTM is a general model suitable for other spatial-temporal features.
基金Project supported by the National Natural Science Foundation of China(Grant No.51901148)the Fund of the State Key Laboratory of Solidification Processing(Northwestern Polytechnical University),China(Grant No.SKLSP202006)the State Key Lab of Advanced Metals and Materials(University of Science and Technology Beijing),China(Grant No.2019-Z15).
文摘We simulate the evolution of hydrogen concentration and gas pore formation as equiaxed dendrites grow during solidification of a hypoeutectic aluminum-silicon(Al-Si)alloy.The applied lattice Boltzmann-cellular automaton-finite difference model incorporates the physical mechanisms of solute and hydrogen partitioning on the solid/liquid interface,as well as the transports of solute and hydrogen.After the quantitative validation by the simulation of capillary intrusion,the model is utilized to investigate the growth of the equiaxed dendrites and hydrogen porosity formation for an Al-(5 wt.%)Si alloy under different solidification conditions.The simulation data reveal that the gas pores favorably nucleate in the corners surrounded by the nearby dendrite arms.Then,the gas pores grow in a competitive mode.With the cooling rate increasing,the competition among different growing gas pores is found to be hindered,which accordingly increases the pore number density in the final solidification microstructure.In the late solidification stage,even though the solid fraction is increasing,the mean concentration of hydrogen in the residue melt tends to be constant,corresponding to a dynamic equilibrium state of hydrogen concentration in liquid.As the cooling rate increases or the initial hydrogen concentration decreases,the temperature of gas pore nucleation,the porosity fraction,and the mean porosity size decrease,whilst the mean hydrogen concentration in liquid increases in the late solidification stage.The simulated data present identical trends with the experimental results reported in literature.
基金financially supported by AO Smith Corporate Technology Center,USA
文摘Numerical estimates of the components of yield strength of a high strength Fe-C-Mn-P-N-Si enameling steel were determined using empirical relationships between microstructure and yield strength. Results are reported for both the hot rolled （HR） and cold rolled （CR） forms before and after simulating an enamel-fire anneal （EFA）. To determine the solubilities of alloying elements, thermodynamic calculations were performed in combination with the considerations of process conditions and the element diffusivities. The results show that the main solid solution strengtheners were the elements Mn, Si, and P, while the elements C and N were nearly completely tied up as precipitates. The yield strength reduction, due to the EFA, resulted primarily from an increase in grain size and a decrease in dislocation density, and the EFA appeared to have a negligible effect on the element solubilities.
