This work constructed a machine learning(ML)model to predict the atmospheric corrosion rate of low-alloy steels(LAS).The material properties of LAS,environmental factors,and exposure time were used as the input,while ...This work constructed a machine learning(ML)model to predict the atmospheric corrosion rate of low-alloy steels(LAS).The material properties of LAS,environmental factors,and exposure time were used as the input,while the corrosion rate as the output.6 dif-ferent ML algorithms were used to construct the proposed model.Through optimization and filtering,the eXtreme gradient boosting(XG-Boost)model exhibited good corrosion rate prediction accuracy.The features of material properties were then transformed into atomic and physical features using the proposed property transformation approach,and the dominant descriptors that affected the corrosion rate were filtered using the recursive feature elimination(RFE)as well as XGBoost methods.The established ML models exhibited better predic-tion performance and generalization ability via property transformation descriptors.In addition,the SHapley additive exPlanations(SHAP)method was applied to analyze the relationship between the descriptors and corrosion rate.The results showed that the property transformation model could effectively help with analyzing the corrosion behavior,thereby significantly improving the generalization ability of corrosion rate prediction models.展开更多
Heavy components of low-alloy high-strength(LAHS) steels are generally formed by multi-pass forging. It is necessary to explore the flow characteristics and hot workability of LAHS steels during the multi-pass forging...Heavy components of low-alloy high-strength(LAHS) steels are generally formed by multi-pass forging. It is necessary to explore the flow characteristics and hot workability of LAHS steels during the multi-pass forging process, which is beneficial to the formulation of actual processing parameters. In the study, the multi-pass hot compression experiments of a typical LAHS steel are carried out at a wide range of deformation temperatures and strain rates. It is found that the work hardening rate of the experimental material depends on deformation parameters and deformation passes, which is ascribed to the impacts of static and dynamic softening behaviors. A new model is established to describe the flow characteristics at various deformation passes. Compared to the classical Arrhenius model and modified Zerilli and Armstrong model, the newly proposed model shows higher prediction accuracy with a confidence level of 0.98565. Furthermore, the connection between power dissipation efficiency(PDE) and deformation parameters is revealed by analyzing the microstructures. The PDE cannot be utilized to reflect the efficiency of energy dissipation for microstructure evolution during the entire deformation process, but only to assess the efficiency of energy dissipation for microstructure evolution in a specific deformation parameter state.As a result, an integrated processing map is proposed to better study the hot workability of the LAHS steel, which considers the effects of instability factor(IF), PDE, and distribution and size of grains. The optimized processing parameters for the multi-pass deformation process are the deformation parameters of 1223–1318 K and 0.01–0.08 s^(-1). Complete dynamic recrystallization occurs within the optimized processing parameters with an average grain size of 18.36–42.3 μm. This study will guide the optimization of the forging process of heavy components.展开更多
The mechanisms of oxide metallurgy include inducing the formation of intragranular acicular ferrite(IAF)using micron-sized inclusions and restricting the growth of prior austenite grains(PAGs)by nanosized particles du...The mechanisms of oxide metallurgy include inducing the formation of intragranular acicular ferrite(IAF)using micron-sized inclusions and restricting the growth of prior austenite grains(PAGs)by nanosized particles during welding.The chaotically oriented IAF and refined PAGs inhibit crack initiation and propagation in the steel,resulting in high impact toughness.This work summarizes the com-bined effect of deoxidizers and alloying elements,with the aim to provide a new perspective for the research and practice related to im-proving the impact toughness of the heat affected zone(HAZ)during the high heat input welding.Ti complex deoxidation with other strong deoxidants,such as Mg,Ca,Zr,and rare earth metals(REMs),can improve the toughness of the heat-affected zone(HAZ)by re-fining PAGs or increasing IAF contents.However,it is difficult to identify the specific phase responsible for IAF nucleation because ef-fective inclusions formed by complex deoxidation are usually multiphase.Increasing alloying elements,such as C,Si,Al,Nb,or Cr,con-tents can impair HAZ toughness.A high C content typically increases the number of coarse carbides and decreases the potency of IAF formation.Si,Cr,or Al addition leads to the formation of undesirable microstructures.Nb reduces the high-temperature stability of the precipitates.Mo,V,and B can enhance HAZ toughness.Mo-containing precipitates present good thermal stability.VN or V(C,N)is ef-fective in promoting IAF nucleation due to its good coherent crystallographic relationship with ferrite.The formation of the B-depleted zone around the inclusion promotes IAF formation.The interactions between alloying elements are complex,and the effect of adding dif-ferent alloying elements remains to be evaluated.In the future,the interactions between various alloying elements and their effects on ox-ide metallurgy,as well as the calculation of the nucleation effects of effective inclusions using first principles calculations will become the focus of oxide metallurgy.展开更多
基金the National Key R&D Program of China(No.2021YFB3701705).
文摘This work constructed a machine learning(ML)model to predict the atmospheric corrosion rate of low-alloy steels(LAS).The material properties of LAS,environmental factors,and exposure time were used as the input,while the corrosion rate as the output.6 dif-ferent ML algorithms were used to construct the proposed model.Through optimization and filtering,the eXtreme gradient boosting(XG-Boost)model exhibited good corrosion rate prediction accuracy.The features of material properties were then transformed into atomic and physical features using the proposed property transformation approach,and the dominant descriptors that affected the corrosion rate were filtered using the recursive feature elimination(RFE)as well as XGBoost methods.The established ML models exhibited better predic-tion performance and generalization ability via property transformation descriptors.In addition,the SHapley additive exPlanations(SHAP)method was applied to analyze the relationship between the descriptors and corrosion rate.The results showed that the property transformation model could effectively help with analyzing the corrosion behavior,thereby significantly improving the generalization ability of corrosion rate prediction models.
基金National Natural Science Foundation of China(No.52305373)Jiangxi Provincial Natural Science Foundation(No.20232BAB214053)+2 种基金Science and Technology Major Project of Jiangxi,China(No.20194ABC28001)Fund of Jiangxi Key Laboratory of Forming and Joining Technology for Aerospace Components,Nanchang Hangkong University(No.EL202303299)PhD Starting Foundation of Nanchang Hangkong University(No,EA202303235).
文摘Heavy components of low-alloy high-strength(LAHS) steels are generally formed by multi-pass forging. It is necessary to explore the flow characteristics and hot workability of LAHS steels during the multi-pass forging process, which is beneficial to the formulation of actual processing parameters. In the study, the multi-pass hot compression experiments of a typical LAHS steel are carried out at a wide range of deformation temperatures and strain rates. It is found that the work hardening rate of the experimental material depends on deformation parameters and deformation passes, which is ascribed to the impacts of static and dynamic softening behaviors. A new model is established to describe the flow characteristics at various deformation passes. Compared to the classical Arrhenius model and modified Zerilli and Armstrong model, the newly proposed model shows higher prediction accuracy with a confidence level of 0.98565. Furthermore, the connection between power dissipation efficiency(PDE) and deformation parameters is revealed by analyzing the microstructures. The PDE cannot be utilized to reflect the efficiency of energy dissipation for microstructure evolution during the entire deformation process, but only to assess the efficiency of energy dissipation for microstructure evolution in a specific deformation parameter state.As a result, an integrated processing map is proposed to better study the hot workability of the LAHS steel, which considers the effects of instability factor(IF), PDE, and distribution and size of grains. The optimized processing parameters for the multi-pass deformation process are the deformation parameters of 1223–1318 K and 0.01–0.08 s^(-1). Complete dynamic recrystallization occurs within the optimized processing parameters with an average grain size of 18.36–42.3 μm. This study will guide the optimization of the forging process of heavy components.
基金supported by the National Natural Science Foundation of China(No.U1960202).
文摘The mechanisms of oxide metallurgy include inducing the formation of intragranular acicular ferrite(IAF)using micron-sized inclusions and restricting the growth of prior austenite grains(PAGs)by nanosized particles during welding.The chaotically oriented IAF and refined PAGs inhibit crack initiation and propagation in the steel,resulting in high impact toughness.This work summarizes the com-bined effect of deoxidizers and alloying elements,with the aim to provide a new perspective for the research and practice related to im-proving the impact toughness of the heat affected zone(HAZ)during the high heat input welding.Ti complex deoxidation with other strong deoxidants,such as Mg,Ca,Zr,and rare earth metals(REMs),can improve the toughness of the heat-affected zone(HAZ)by re-fining PAGs or increasing IAF contents.However,it is difficult to identify the specific phase responsible for IAF nucleation because ef-fective inclusions formed by complex deoxidation are usually multiphase.Increasing alloying elements,such as C,Si,Al,Nb,or Cr,con-tents can impair HAZ toughness.A high C content typically increases the number of coarse carbides and decreases the potency of IAF formation.Si,Cr,or Al addition leads to the formation of undesirable microstructures.Nb reduces the high-temperature stability of the precipitates.Mo,V,and B can enhance HAZ toughness.Mo-containing precipitates present good thermal stability.VN or V(C,N)is ef-fective in promoting IAF nucleation due to its good coherent crystallographic relationship with ferrite.The formation of the B-depleted zone around the inclusion promotes IAF formation.The interactions between alloying elements are complex,and the effect of adding dif-ferent alloying elements remains to be evaluated.In the future,the interactions between various alloying elements and their effects on ox-ide metallurgy,as well as the calculation of the nucleation effects of effective inclusions using first principles calculations will become the focus of oxide metallurgy.
