With the development of manufacturing,numerical control(NC)machining simulation has become a modern tool to obtain safe and reliable machining operations.Although some research and commercial software about NC machini...With the development of manufacturing,numerical control(NC)machining simulation has become a modern tool to obtain safe and reliable machining operations.Although some research and commercial software about NC machining simulations is available,most of them is oriented for G&M code.It is a low-level data model for computer numerical control(CNC),which has inherent drawbacks such as incomplete data and lack of accuracy.These limitations hinder the development of a real simulation system.Whereas,standard for the exchange of product data-compliant numerical control(STEP-NC)is a new and high-level data model for CNC.It provides rich information for CNC machine tools,which creates the condition for an informative and real simulation.Therefore,this paper proposes STEP-NC based high-level NC machining simulations solution integrated with computer-aided design/computeraided process planning/computer-aided manufacturing(CAD/CAPP/CAM).It turned out that the research provides a better informed simulation environment and promotes the development of modern manufacturing.展开更多
In the machining process of large-scale complex curved surface,workers will encounter problems such as empty stroke of tool,collision interference,and overcut or undercut of the workpieces.This paper presents a method...In the machining process of large-scale complex curved surface,workers will encounter problems such as empty stroke of tool,collision interference,and overcut or undercut of the workpieces.This paper presents a method for generating the optimized tool path,compiling and checking the numerical control(NC)program.Taking the bogie frame as an example,the tool paths of all machining surface are optimized by the dynamic programming algorithm,Creo software is utilized to compile the optimized computerized numerical control(CNC)machining program,and VERICUT software is employed to simulate the machining process,optimize the amount of cutting and inspect the machining quality.The method saves the machining time,guarantees the correctness of NC program,and the overall machining efficiency is improved.The method lays a good theoretical and practical foundation for integration of the similar platform.展开更多
This paper describes an algorithm of collision detection between moving objects in machin-ing process simulation. Graphical simulation of machining has been recognized to be useful for NCprogram verification , since t...This paper describes an algorithm of collision detection between moving objects in machin-ing process simulation. Graphical simulation of machining has been recognized to be useful for NCprogram verification , since the programmer of the machining operator can easily find some faults inthe NC program visually. But it is difficult to visually detect collisions arnong moving objects such ascutting tools , workpieces and fixtures, a data structure to represent moving objects and an algorithmof collision detection between moving objects are proposed. A moving object can be represented by ahierarchical sphere octree and its motion can be described by a quadratic function of time. A collisionoccurs in the case that the distance between any two sphere centers in the respective two moving ob-jects is equal to the sum of the radii of these two spheres, and the radii of these two spheres are lessthan a given precision. By solving the equations that satisfy the conditions of collision between thespheres recursively , we obtain the time and the position of the collision between these two moving ob-Jects.展开更多
Considering the deficiency in milling process parameters selection, based on the modelling of dynamic milling force and the deduction of chatter stability limits, the chatter stability lobes simulation program for mil...Considering the deficiency in milling process parameters selection, based on the modelling of dynamic milling force and the deduction of chatter stability limits, the chatter stability lobes simulation program for milling is developed with MAT- LAB. The simulation optimization application software of dynamics was designed using engineering simulation software Visio Basic. The chatter stability lobes for milling, which can be used as an instruction guide for the selection of process parameters, are simulated with frequency response functions （FRFs） gained by hammer test. The validation and accuracy of the simulation algorithm are verified by experiments. The simulation method has been used in a factory with an excellent application effect.展开更多
To makesystem-of-systems combat simulation models easy to be developed and reused, simulation model formal specification and representation are researched. According to the view of system-of-systems combat simulation,...To makesystem-of-systems combat simulation models easy to be developed and reused, simulation model formal specification and representation are researched. According to the view of system-of-systems combat simulation, and based on DEVS, the simulation model＇s fundamental formalisms are explored. It includes entity model, system-of-systems model and experiment model. It also presents rigorous formal specification. XML data exchange standard is combined to design the XML based language, SCSL, to support simulation model representation. The corresponding relationship between SCSL and simulation model formalism is discussed and the syntax and semantics of elements in SCSL are detailed. Based on simulation model formal specification, the abstract simulation algorithm is given and SCSL virtual machine, which is capable of automatically interpreting and executing simulation model represented by SCSL, is designed. Finally an application case is presented, which can show the validation of the theory and verification of SCSL.展开更多
Automated manufacturing system is characterized by flexibility. It aims at producing a variety of products with virtually no time loses to change over from one part to the next. In this paper, the Machining Process Si...Automated manufacturing system is characterized by flexibility. It aims at producing a variety of products with virtually no time loses to change over from one part to the next. In this paper, the Machining Process Simulator GMPS is introduced, which can be used as a supported environment for machining process. It can be executed off-line or on-line in manufacturing systems in order to predict the collisions of tool with machined workpieces, fixtures or pallets. First, the functional model of GMPS is described, then adopted critical techniques in the simulator are introduced. Finally, an application of GMPS in CIMS ERC of China is presented.展开更多
A CNC simulation system based on intemet for operation training of manufacturing facility and manufacturing process simulation is proposed.Firstly,the system framework and a rapid modeling method of CNC machine tool a...A CNC simulation system based on intemet for operation training of manufacturing facility and manufacturing process simulation is proposed.Firstly,the system framework and a rapid modeling method of CNC machine tool are studied under the virtual environment based on PolyTrans and CAD software.Then,a new method is proposed to enhance and expand the interactive ability of virtual reality modeling language(VRML)by attaining communication among VRML,JavaApplet,JavaScript and Html so as to realize the virtual operation for CNC machine tool.Moreover,the algorithm of material removed simulation based on VRML Z-map is presented.The advantages of this algorithm include less memory requirement and much higher computation.Lastly,the CNC milling machine is taken as an illustrative example for the prototype development in order to validate the feasibility of the proposed approach.展开更多
Molecular dynamics simulations are employed to study the nanometric machining process of single crystal nickel. Atoms from different machining zones had different atomic crystal structures owing to the differences in ...Molecular dynamics simulations are employed to study the nanometric machining process of single crystal nickel. Atoms from different machining zones had different atomic crystal structures owing to the differences in the actions of the cutting tool. The stacking fault tetrahedral was formed by a series of dislocation reactions, and it maintained the stable structure after the dislocation reactions. In addition, evidence of crystal transition and recovery was found by analyzing the number variations in different types of atoms in the primary shear zone, amorphous region, and crystalline region. The effects of machining speed on the cutting force, chip and subsurface defects, and temperature of the contact zone between the tool and workpiece were investigated. The results suggest that higher the machining speed, larger is the cutting force. The degree of amorphousness of chip atoms and the depth and extent of subsurface defects increase with the machining speed. The average friction coefficient first decreases and then increases with the machining speed because of the temperature difference between the chip and machining surface.展开更多
A method which adopts the combination of least squares support vector machine（LS-SVM） and Monte Carlo（MC） simulation is used to calculate the foundation settlement reliability.When using LS-SVM,choosing the traini...A method which adopts the combination of least squares support vector machine（LS-SVM） and Monte Carlo（MC） simulation is used to calculate the foundation settlement reliability.When using LS-SVM,choosing the training dataset and the values for LS-SVM parameters is the key.In a representative sense,the orthogonal experimental design with four factors and five levels is used to choose the inputs of the training dataset,and the outputs are calculated by using fast Lagrangian analysis continua（FLAC）.The decimal ant colony algorithm（DACA） is also used to determine the parameters.Calculation results show that the values of the two parameters,and δ2 have great effect on the performance of LS-SVM.After the training of LS-SVM,the inputs are sampled according to the probabilistic distribution,and the outputs are predicted with the trained LS-SVM,thus the reliability analysis can be performed by the MC method.A program compiled by Matlab is employed to calculate its reliability.Results show that the method of combining LS-SVM and MC simulation is applicable to the reliability analysis of soft foundation settlement.展开更多
Integration is a key component of CIMS and concurrent engineering is an important step in realizing CIMSprocess integration. Concurrent engineering is itself dependent on an effective machining process simulator that ...Integration is a key component of CIMS and concurrent engineering is an important step in realizing CIMSprocess integration. Concurrent engineering is itself dependent on an effective machining process simulator that verifies the machining process. A machining process simulator for concurrent engineering CMPS was developed at CIMSERC to meet the need for an effective simulator. This paper introduces the CMPS structure. key techniques. including geometry model construction and kinematics. NC code translation. collision and interference checks, material removal simulation. machining animation. etc. The model uses the solid Ray-representation method and the Voxelsplus B-representation algorithm. The Ray-representation method simplifies the Boolean operation process and improves the material removal simulation speed. The Voxels plus B-representation algorithm quickly detects collisionand interference problems. Finally. CMPS is applied to an actual NC milling process as an example.展开更多
The design and high-throughput screening of materials using machine-learning assisted quantummechanical simulations typically requires the existence of a very large data set,often generated from simulations at a high ...The design and high-throughput screening of materials using machine-learning assisted quantummechanical simulations typically requires the existence of a very large data set,often generated from simulations at a high level of theory or fidelity.Asingle simulation at high fidelity can take on the order of days for a complex molecule.Thus,although machine learning surrogate simulations seem promising at first glance,generation of the training data can defeat the original purpose.For this reason,the use of machine learning to screen or design materials remains elusive for many important applications.In this paper we introduce a new multi-fidelity approach based on a dual graph embedding to extract features that are placed inside a nonlinear multi-step autoregressive model.Experiments on five benchmark problems,with 14 different quantities and 27 different levels of theory,demonstrate the generalizability and high accuracy of the approach.It typically requires a few 10s to a few 1000’s of high-fidelity training points,which is several orders of magnitude lower than direct ML methods,and can be up to two orders of magnitude lower than other multi-fidelity methods.Furthermore,we develop a new benchmark data set for 860 benzoquinone molecules with up to 14 atoms,containing energy,HOMO,LUMO and dipole moment values at four levels of theory,up to coupled cluster with singles and doubles.展开更多
In this article,reference 43 was incorrectly given as‘Erhard,L.C.,Rohrer,V.,Albe,K.,and Deringer,V.L.,Modelling atomic and nanoscale structure in the silicon-oxygen system through active machine learning,https://arxi...In this article,reference 43 was incorrectly given as‘Erhard,L.C.,Rohrer,V.,Albe,K.,and Deringer,V.L.,Modelling atomic and nanoscale structure in the silicon-oxygen system through active machine learning,https://arxiv.org/abs/2309.03587(2023)’and should have read‘Erhard,L.C.,Rohrer,J.,Albe,K.et al.Modelling atomic and nanoscale structure in the silicon–oxygen system through active machine learning.Nat Commun 15,1927(2024).https://doi.org/10.1038/s41467-024-45840-9’.展开更多
Research and application of Concurrent Engineering have produced good results. A Chinese style concurrent engineering architecture and reference mode has been produced. A series of break thoughts in BPR (Business Proc...Research and application of Concurrent Engineering have produced good results. A Chinese style concurrent engineering architecture and reference mode has been produced. A series of break thoughts in BPR (Business Process Reengineering) have been made with organization of the IPT (Integrated Product Development Team) and engineering support technologies. Several prototype tools were developed, including product development process modeling and management, QFD based schema design and decision making, PDM based concurrent design, STEP based CAD/CAPP/CAM integration, design for assembly, design for manufacturing, computer aided fixture design, and machining process simulation. Finally, the research results were used in the development of two complex components in an aerospace application, and satisfactory results were obtained.展开更多
Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice.These interactions span large length and time sca...Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice.These interactions span large length and time scales,making them difficult to address with standard ab-initio techniques.This work addresses this challenge by employing accelerated machine learning(ML)molecular dynamics simulations through active learning.We conduct a comparative study of different ML-based interatomic potential schemes,including VASP,MACE,and CHGNet,utilizing various training strategies such as on-the-fly learning,pre-trained universal models,and fine-tuning.By considering different temperatures and concentration regimes,we obtain hydrogen diffusion coefficients and activation energy values which align remarkably well with experimental results,underlining the efficacy and accuracy of ML-assisted methodologies in the context of diffusive dynamics.Particularly,our procedure significantly reduces the computational effort associated with traditional transition state calculations or ad-hoc designed interatomic potentials.The results highlight the limitations of pre-trained universal solutions for defective materials and how they can be improved by fine-tuning.Specifically,fine-tuning the models on a database produced during on-the-fly training of VASP ML force-field allows the retrieval of DFT-level accuracy at a fraction of the computational cost.展开更多
基金supported by National Natural Science Foundation of China(No.51205054)National Key Technology Research and Development Program During the Twelfth Five-year Plan(Nos.2012BAF10B11,2012BAF12B08)
文摘With the development of manufacturing,numerical control(NC)machining simulation has become a modern tool to obtain safe and reliable machining operations.Although some research and commercial software about NC machining simulations is available,most of them is oriented for G&M code.It is a low-level data model for computer numerical control(CNC),which has inherent drawbacks such as incomplete data and lack of accuracy.These limitations hinder the development of a real simulation system.Whereas,standard for the exchange of product data-compliant numerical control(STEP-NC)is a new and high-level data model for CNC.It provides rich information for CNC machine tools,which creates the condition for an informative and real simulation.Therefore,this paper proposes STEP-NC based high-level NC machining simulations solution integrated with computer-aided design/computeraided process planning/computer-aided manufacturing(CAD/CAPP/CAM).It turned out that the research provides a better informed simulation environment and promotes the development of modern manufacturing.
基金supported by the Collaborative Innovation Center of Ma jor Machine Manufacturing in Liaoning
文摘In the machining process of large-scale complex curved surface,workers will encounter problems such as empty stroke of tool,collision interference,and overcut or undercut of the workpieces.This paper presents a method for generating the optimized tool path,compiling and checking the numerical control(NC)program.Taking the bogie frame as an example,the tool paths of all machining surface are optimized by the dynamic programming algorithm,Creo software is utilized to compile the optimized computerized numerical control(CNC)machining program,and VERICUT software is employed to simulate the machining process,optimize the amount of cutting and inspect the machining quality.The method saves the machining time,guarantees the correctness of NC program,and the overall machining efficiency is improved.The method lays a good theoretical and practical foundation for integration of the similar platform.
文摘This paper describes an algorithm of collision detection between moving objects in machin-ing process simulation. Graphical simulation of machining has been recognized to be useful for NCprogram verification , since the programmer of the machining operator can easily find some faults inthe NC program visually. But it is difficult to visually detect collisions arnong moving objects such ascutting tools , workpieces and fixtures, a data structure to represent moving objects and an algorithmof collision detection between moving objects are proposed. A moving object can be represented by ahierarchical sphere octree and its motion can be described by a quadratic function of time. A collisionoccurs in the case that the distance between any two sphere centers in the respective two moving ob-jects is equal to the sum of the radii of these two spheres, and the radii of these two spheres are lessthan a given precision. By solving the equations that satisfy the conditions of collision between thespheres recursively , we obtain the time and the position of the collision between these two moving ob-Jects.
基金Tianjin Municipal Association of Higher Vocational&Technical Education Projects(No.XIV412)
文摘Considering the deficiency in milling process parameters selection, based on the modelling of dynamic milling force and the deduction of chatter stability limits, the chatter stability lobes simulation program for milling is developed with MAT- LAB. The simulation optimization application software of dynamics was designed using engineering simulation software Visio Basic. The chatter stability lobes for milling, which can be used as an instruction guide for the selection of process parameters, are simulated with frequency response functions （FRFs） gained by hammer test. The validation and accuracy of the simulation algorithm are verified by experiments. The simulation method has been used in a factory with an excellent application effect.
文摘To makesystem-of-systems combat simulation models easy to be developed and reused, simulation model formal specification and representation are researched. According to the view of system-of-systems combat simulation, and based on DEVS, the simulation model＇s fundamental formalisms are explored. It includes entity model, system-of-systems model and experiment model. It also presents rigorous formal specification. XML data exchange standard is combined to design the XML based language, SCSL, to support simulation model representation. The corresponding relationship between SCSL and simulation model formalism is discussed and the syntax and semantics of elements in SCSL are detailed. Based on simulation model formal specification, the abstract simulation algorithm is given and SCSL virtual machine, which is capable of automatically interpreting and executing simulation model represented by SCSL, is designed. Finally an application case is presented, which can show the validation of the theory and verification of SCSL.
文摘Automated manufacturing system is characterized by flexibility. It aims at producing a variety of products with virtually no time loses to change over from one part to the next. In this paper, the Machining Process Simulator GMPS is introduced, which can be used as a supported environment for machining process. It can be executed off-line or on-line in manufacturing systems in order to predict the collisions of tool with machined workpieces, fixtures or pallets. First, the functional model of GMPS is described, then adopted critical techniques in the simulator are introduced. Finally, an application of GMPS in CIMS ERC of China is presented.
基金Selected from Proceedings of the 7th International Conference on Frontiers of Design and Manufacturing(ICFDM'2006)This project is supported by National Natural Science Foundation of China(No.50775047)Scientific and Technological Foundation of Guangdong Province,China(No.2004B10201032).
文摘A CNC simulation system based on intemet for operation training of manufacturing facility and manufacturing process simulation is proposed.Firstly,the system framework and a rapid modeling method of CNC machine tool are studied under the virtual environment based on PolyTrans and CAD software.Then,a new method is proposed to enhance and expand the interactive ability of virtual reality modeling language(VRML)by attaining communication among VRML,JavaApplet,JavaScript and Html so as to realize the virtual operation for CNC machine tool.Moreover,the algorithm of material removed simulation based on VRML Z-map is presented.The advantages of this algorithm include less memory requirement and much higher computation.Lastly,the CNC milling machine is taken as an illustrative example for the prototype development in order to validate the feasibility of the proposed approach.
基金supported by the National Natural Science Foundation of China(Grant Nos,51375082)
文摘Molecular dynamics simulations are employed to study the nanometric machining process of single crystal nickel. Atoms from different machining zones had different atomic crystal structures owing to the differences in the actions of the cutting tool. The stacking fault tetrahedral was formed by a series of dislocation reactions, and it maintained the stable structure after the dislocation reactions. In addition, evidence of crystal transition and recovery was found by analyzing the number variations in different types of atoms in the primary shear zone, amorphous region, and crystalline region. The effects of machining speed on the cutting force, chip and subsurface defects, and temperature of the contact zone between the tool and workpiece were investigated. The results suggest that higher the machining speed, larger is the cutting force. The degree of amorphousness of chip atoms and the depth and extent of subsurface defects increase with the machining speed. The average friction coefficient first decreases and then increases with the machining speed because of the temperature difference between the chip and machining surface.
文摘A method which adopts the combination of least squares support vector machine（LS-SVM） and Monte Carlo（MC） simulation is used to calculate the foundation settlement reliability.When using LS-SVM,choosing the training dataset and the values for LS-SVM parameters is the key.In a representative sense,the orthogonal experimental design with four factors and five levels is used to choose the inputs of the training dataset,and the outputs are calculated by using fast Lagrangian analysis continua（FLAC）.The decimal ant colony algorithm（DACA） is also used to determine the parameters.Calculation results show that the values of the two parameters,and δ2 have great effect on the performance of LS-SVM.After the training of LS-SVM,the inputs are sampled according to the probabilistic distribution,and the outputs are predicted with the trained LS-SVM,thus the reliability analysis can be performed by the MC method.A program compiled by Matlab is employed to calculate its reliability.Results show that the method of combining LS-SVM and MC simulation is applicable to the reliability analysis of soft foundation settlement.
文摘Integration is a key component of CIMS and concurrent engineering is an important step in realizing CIMSprocess integration. Concurrent engineering is itself dependent on an effective machining process simulator that verifies the machining process. A machining process simulator for concurrent engineering CMPS was developed at CIMSERC to meet the need for an effective simulator. This paper introduces the CMPS structure. key techniques. including geometry model construction and kinematics. NC code translation. collision and interference checks, material removal simulation. machining animation. etc. The model uses the solid Ray-representation method and the Voxelsplus B-representation algorithm. The Ray-representation method simplifies the Boolean operation process and improves the material removal simulation speed. The Voxels plus B-representation algorithm quickly detects collisionand interference problems. Finally. CMPS is applied to an actual NC milling process as an example.
文摘The design and high-throughput screening of materials using machine-learning assisted quantummechanical simulations typically requires the existence of a very large data set,often generated from simulations at a high level of theory or fidelity.Asingle simulation at high fidelity can take on the order of days for a complex molecule.Thus,although machine learning surrogate simulations seem promising at first glance,generation of the training data can defeat the original purpose.For this reason,the use of machine learning to screen or design materials remains elusive for many important applications.In this paper we introduce a new multi-fidelity approach based on a dual graph embedding to extract features that are placed inside a nonlinear multi-step autoregressive model.Experiments on five benchmark problems,with 14 different quantities and 27 different levels of theory,demonstrate the generalizability and high accuracy of the approach.It typically requires a few 10s to a few 1000’s of high-fidelity training points,which is several orders of magnitude lower than direct ML methods,and can be up to two orders of magnitude lower than other multi-fidelity methods.Furthermore,we develop a new benchmark data set for 860 benzoquinone molecules with up to 14 atoms,containing energy,HOMO,LUMO and dipole moment values at four levels of theory,up to coupled cluster with singles and doubles.
文摘In this article,reference 43 was incorrectly given as‘Erhard,L.C.,Rohrer,V.,Albe,K.,and Deringer,V.L.,Modelling atomic and nanoscale structure in the silicon-oxygen system through active machine learning,https://arxiv.org/abs/2309.03587(2023)’and should have read‘Erhard,L.C.,Rohrer,J.,Albe,K.et al.Modelling atomic and nanoscale structure in the silicon–oxygen system through active machine learning.Nat Commun 15,1927(2024).https://doi.org/10.1038/s41467-024-45840-9’.
文摘Research and application of Concurrent Engineering have produced good results. A Chinese style concurrent engineering architecture and reference mode has been produced. A series of break thoughts in BPR (Business Process Reengineering) have been made with organization of the IPT (Integrated Product Development Team) and engineering support technologies. Several prototype tools were developed, including product development process modeling and management, QFD based schema design and decision making, PDM based concurrent design, STEP based CAD/CAPP/CAM integration, design for assembly, design for manufacturing, computer aided fixture design, and machining process simulation. Finally, the research results were used in the development of two complex components in an aerospace application, and satisfactory results were obtained.
基金the“Doctoral College Advanced Functional Materials-Hierarchical Design of Hybrid Systems DOC 85 doc.funds”funded by the Austrian Science Fund(FWF)and by the Vienna Doctoral School in Physics(VDSP),For Open Access purposes,the author has applied a CC BY public copyright license to any author accepted manuscript version arising from this submission.D.M.and S.P.were supported by the European Union Horizon 2020 research and innovation program under Grant Agreement No.857470the European Regional Development Fund under the program of the Foundation for Polish Science International Research Agenda PLUS,grant No.MAB PLUS/2018/8+2 种基金the initiative of the Ministry of Science and Higher Education’Support for the activities of Centers of Excellence established in Poland under the Horizon 2020 program’under agreement No.MEiN/2023/DIR/3795L.P.and C.F.acknowledge the National Recovery and Resilience Plan(NRRP),Mission 4 Component 2 Investment 1.3-Project NEST(Network 4 Energy Sustainable Transition)of Ministero dell’Universitáe della Ricerca(MUR),funded by the European Union-NextGenerationEUL.L.and C.F.acknowledge the NRRP,CN-HPC grant no.(CUP)J33C22001170001,SPOKE 7,of MUR,funded by the European Union-NextGenerationEU.The computational results were obtained using the Vienna Scientific Cluster(VSC)and the LEONARDO cluster.We acknowledge access to LEONARDO at CINECA,Italy,via an AURELEO(Austrian Users at LEONARDO supercomputer)project.
文摘Understanding and accurately predicting hydrogen diffusion in materials is challenging due to the complex interactions between hydrogen defects and the crystal lattice.These interactions span large length and time scales,making them difficult to address with standard ab-initio techniques.This work addresses this challenge by employing accelerated machine learning(ML)molecular dynamics simulations through active learning.We conduct a comparative study of different ML-based interatomic potential schemes,including VASP,MACE,and CHGNet,utilizing various training strategies such as on-the-fly learning,pre-trained universal models,and fine-tuning.By considering different temperatures and concentration regimes,we obtain hydrogen diffusion coefficients and activation energy values which align remarkably well with experimental results,underlining the efficacy and accuracy of ML-assisted methodologies in the context of diffusive dynamics.Particularly,our procedure significantly reduces the computational effort associated with traditional transition state calculations or ad-hoc designed interatomic potentials.The results highlight the limitations of pre-trained universal solutions for defective materials and how they can be improved by fine-tuning.Specifically,fine-tuning the models on a database produced during on-the-fly training of VASP ML force-field allows the retrieval of DFT-level accuracy at a fraction of the computational cost.
