曹 流
博士，讲师，硕士生导师
系统流变学研究所、智能制造工程系
邮箱：caoliu@gzhu.edu.cn
主要研究方向[ResearchGate]：材料成形过程数值模拟、金属增材制造(3D打印)、基于物理知识的机器学习、金属流变学
学术兼职及荣誉：
最新消息：
《材料加工理论与数值计算方法》讲义 v1.0
前言：起初并非奔着整理出本讲义的目标而来，而是想着将材料加工数值模拟所涉及的基础数学与物理知识做个简略的归纳，以便于下一步深入开展自己的科研工作，同时也为指导学生提供一个理论讲解大纲。然而，随着整理工作的推进，逐渐回忆起自己近十年针对材料加工数值模拟的学习历程，发现可供学习的中、英文教材很多，网上的各类学习资料更是数不胜数，但总体上存在两个明显的困境。第一个困境是：对于从事材料加工数值模拟技术开发的学生，其学习的起点往往是所在课题组积累的资料，但由于绝大多数课题组已形成了成熟的研究方向，所积累的学习资料针对性很强，同时也意味着存在一定的片面性，导致学生所掌握的基础数学与物理知识存在一定的局限，进而影响其深入开展数值模拟技术开发。学习各类相关的中、英文教材是克服此困境的通用方法，但各个教材的关注点不一，而且单本教材往往就单个理论或技术的讲解较深，致使学生需要反复穿插于多本中、英文教材之间，以期建立起自己的完整知识架构，此过程需耗费极大的精力，而且往往难以取得较好的效果，我自己就对此深有感触。第二个困境是：对于只是想利用数值模拟技术来辅助实验研究的学生，其往往会选择一款商用化软件，再找一本名为“某某软件从入门到精通”的软件操作书，依葫芦画瓢地开展针对自己所研究工艺的模拟仿真。由于这类学生几乎从未接触过材料加工数值模拟所涉及的基础数学与物理知识，导致其在操作软件过程中对很多设置似懂非懂，甚至无法判断计算结果的物理合理性，从而难以最大程度地发挥商用化软件作用，以致于很多从事实验研究的学生形成了数值模拟只是用来装装样子的认识。本讲义以期为破解上述两个困境贡献出绵薄之力，故讲义的内容涉及数值模拟的三要素“数学基础—物理控制方程—数值计算方法”，本讲义想回答的最根本的问题是“各类描述物理规律的数学公式是如何被推导出来的？以及如何处理这些物理公式以预测各类现象”。
引用：曹流. 材料加工理论与数值计算方法[R]. 广州: 广州大学, 2022. https://pan.baidu.com/s/1onDnY6F70EBPlgeFCqtFXg?pwd=2022.
成果动画：
教育及科研经历：
教育及工作经历：
科研经历：
期刊论文：
Eulerian-Lagrangian Numerical Simulation of Powder Bed Denudation and Spatter Behavior During Powder Bed Fusion Process
Liu Cao*, Rui-fan Meng, Qin-dan Zhang, Zhen-zhen Gui
引用：Liu Cao, Rui-fan Meng, Qin-dan Zhang, Zhen-zhen Gui. Eulerian-Lagrangian Numerical Simulation of Powder Bed Denudation and Spatter Behavior During Powder Bed Fusion Process[J]. Metallurgical and Materials Transactions A, 2023, 54: 2771-2790
摘要：Powder bed fusion using a laser beam (PBF-LB) is rapidly developing toward high laser power, high scanning speed, and multiple lasers, however, spatter has become one of the major bottlenecks limiting PBF-LB to form large metal parts. In this paper, the Eulerian-Lagrangian numerical simulation study was carried out for the spatter formation and spatter-protective gas flow coupling behavior. Among them, the Eulerian model considered the effect of particles on the gas-phase flow, and the Lagrangian model took into account the fluid drag force, gravity, buoyancy, non-uniform pressure distribution, additional mass force, and particle-particle collisions. For the effect of metal vapor on spatter and powder bed denudation, it was found that the larger the metal vapor jet speed or jet area, the faster the spatter speed and the larger the width of the powder bed denudation zone; the larger the lateral dimension of the metal vapor action zone (the scanning direction is longitudinal), the larger the number of spatters. For the effect of protective gas flow velocity on spatter movement, it was found that the longitudinal movement distance of spatter was significantly larger for larger protective gas flow velocity, and almost all spatters fell on the powder bed or substrate for smaller protective gas flow velocity. This paper is expected to provide scientific guidance for regulating PBF-LB spatter.
金属增材制造条件下二元合金三维微观组织的相场法-有限元法预测
曹流*，孟瑞繁，张沁丹，张珞
期刊：铸造（中文核心）
引用：曹流，孟瑞繁，张沁丹，张珞. 金属增材制造条件下二元合金三维微观组织的相场法-有限元法预测[J]. 铸造, 2023, 72(4): 365-374
摘要：基于开源相场法代码PRISMS-PF所提供的相场法-有限元法计算架构，开展了A356铝合金MAM过程的模拟仿真研究。计算分析了工艺参数（温度梯度与扫描速率）以及初始固相形貌对溶质浓度、一次枝晶臂间距和晶粒形貌的影响，再现了微观偏析、晶粒竞争生长以及二次枝晶臂等现象，并与实验结果进行了对比验证。对比不同工艺参数(温度梯度与扫描速率)情形下的预测结果，发现一次枝晶臂间距随温度梯度与扫描速率乘积的减小而增大，而随着温度梯度与扫描速率的比值逐渐减小，晶粒形貌由平面晶逐渐向胞状晶、柱状晶以及树枝晶转变。对比不同初始固相形貌情形下的预测结果，发现当温度梯度与扫描速率的比值较小，即液-固界面前沿的成分过冷区越大时，非平整的初始固相形貌对晶粒生长过程造成的影响越明显。
CFD-DPM Simulation Study of the Effect of Powder Layer Thickness on the SLM Spatter Behavior
Liu Cao*, Qin-dan Zhang, Rui-fan Meng
期刊：Metals（JCR2区）
引用：Liu Cao, Qin-dan Zhang, Rui-fan Meng. CFD-DPM Simulation Study of the Effect of Powder Layer Thickness on the SLM Spatter Behavior[J]. Metals, 2022, 12: 1897
摘要：Selective Laser Melting (SLM) has significant advantages in manufacturing complex structural components and refining the alloy microstructure, however, spatter, as a phenomenon that accompanies the entire SLM forming process, is prone to problems such as inclusions, porosity and low powder recovery quality. In this paper, a Computational Fluid Dynamics-Discrete Particle Method (CFD-DPM) simulation flow for predicting the SLM spatter behavior is established based on the open-source code OpenFOAM. Among them, the single-phase flow Navier-Stokes equation is used in Eulerian framework to equivalently describe the effect of metal vapor and protective gas on the flow field of the forming cavity, and the DPM method is used in Lagrangian framework to describe the metal particle motion, and the factors affecting the particle motion include particle-particle collision, particle-wall collision, fluid drag force, gravity, buoyancy force, and additional mass force. In addition, the equivalent volume force and fluid drag force are used to characterize the fluid-particle coupling interaction. For the spatter behavior and powder bed denudation phenomenon, the calculation results show that the spatter height and the drop location show a clear correlation, and the powder bed denudation phenomenon is caused by the high-speed gas flow causing the surrounding gas to gather in the forming area, which in turn drives the motion of the powder bed particles. For the effect of powder layer thickness on spatter and powder bed denudation, the calculation results show that the effect of powder layer thickness on the number of spatters is large (when the thickness was increased from 50 μm to 100 μm, the number of spatters increased by 157%), but the effect on spatter height and drop location distribution is small. When the powder layer thickness is small, the width of the denudation zone is significantly larger, but when the powder layer reaches a certain thickness, the width of the denudation zone does not show significant changes. It should be noted that the presented model is NOT directly validated by experiments so far due to the difficulty of tracking the large-scale motion of SLM spatter in real time by current experimental means.
Analyzing effects of temperature gradient and scan rate on metal additive manufacturing microstructure by using phase field-finite element method
Liu Cao*, Luo Zhang, Rui-fan Meng, Qin-dan Zhang
引用：Liu Cao, Luo Zhang, Rui-fan Meng, Qin-dan Zhang. Analyzing effects of temperature gradient and scan rate on metal additive manufacturing microstructure by using phase field-finite element method[J]. Modelling and Simulation in Materials Science and Engineering, 2022, 30: 034003
摘要：Predicting the evolutionary behavior of microstructures with the help of numerical simulation techniques has become an essential tool for studying the solidification process of metal additive manufacturing. As a mesoscopic model based on the diffusion interface theory, phase field method (PFM) can be used to predict the evolution of solidification microstructure. The open-source PFM framework PRISMS-PF can not only efficiently solve systems of equations with billions of degrees of freedom, but also provide a simple adaptive mesh control module. In this paper, based on the open-source PFM framework PRISMS-PF, a phase field-finite element method (PFM-FEM) simulation flow for the solidification process of A356 aluminum alloy additive manufacturing in the two-dimensional case was established. The effects of temperature gradient, scan rate and initial solid-phase morphology on solute concentration, dendrite spacing and dendrite morphology were analyzed and compared with experimental results for verification. Analyzing the results for different temperature gradients and scan rates cases, it was found that the increase of temperature gradient or scan rate made the primary dendrite arm space decrease; as the ratio of temperature gradient to scan rate decreased, the solidification morphology gradually changed from flat crystal to cellular crystal, columnar crystal, and even dendritic structure. Analyzing the results for different initial solid-phase morphology cases, it was found that the influence of initial solid-phase morphology on dendrite growth increased as the ratio of temperature gradient to scan rate decreased. The above influence rules were mainly related to the composition overcooling zone under different conditions. This paper is expected to provide a theoretical support for the effective regulation of solidification microstructure in metal additive manufacturing.
Simulation and analysis of LPBF multi-layer single-track forming process under different particle size distributions
Liu Cao*, Wei Guan
引用：Liu Cao, Wei Guan. Simulation and analysis of LPBF multi-layer single-track forming process under different particle size distributions[J]. International Journal of Advanced Manufacturing Technology, 2021, 114: 2141-2157
摘要：Laser powder bed fusion (LPBF) has the advantages of rapid and customized production, and has gradually been used in aerospace, biomedicine and other fields. However, the current LPBF technology still faces many obstacles, such as the failure to fully explore the influence of the powder layer parameters (particle size distribution, thickness of spreading layer, etc.) on the LPBF multi-layer forming process. Based on the open-source discrete element method framework Yade and the open-source finite volume method framework OpenFOAM, the simulation flow for the LPBF multi-layer single-track process was established. Among them, the considered force influence factors included gasification recoil, surface tension, Marangoni effect, viscous force, mushy zone drag force and gravity, and the considered thermal influence factors included gasification, convection and radiation heat dissipation. In order to analyze the influence of particle size distribution and spreading layer thickness on the LPBF multi-layer process, the forming processes of the first eight layers of powder bed with three different particle size distributions and two different spreading layer thicknesses were calculated. Regarding the surface roughness of the formed layer, it was found that when the proportion of large-size particles increased, the surface roughness of the formed layer would increase. Regarding the actual thickness of the powder layer and the depth of the solidified track, it was found that they were mainly related to the thickness of the spreading layer, but were not affected by the particle size distribution. Regarding the porosity of the formed zone, it was found that when the spreading layer thickness was smaller, the higher the proportion of small-size particles, the lower the porosity of the formed zone; when the spreading layer thickness was larger, the smaller the proportion of large-size particles, the lower the porosity of the formed zone. This paper is expected to provide support for in-depth understanding of the effects of particle size distribution and spreading layer thickness on the LPBF multi-layer forming process.
Workpiece-scale numerical simulations of SLM molten pool dynamic behavior of 316L stainless steel
Liu Cao*
引用：Liu Cao. Workpiece-scale numerical simulations of SLM molten pool dynamic behavior of 316L stainless steel[J]. Computers and Mathematics with Applications, 2021, 96: 209-228
摘要：316L stainless steel is currently one of the most critical stainless-steel materials due to its excellent corrosion resistance and comprehensive mechanical properties. Selective laser melting (SLM), as an additive manufacturing technology for directly forming complex metal parts, has been applied in the production of 316L stainless steel components. By introducing reasonable and comprehensive equivalent processing models (e.g., gasification pressure, gasification heat dissipation, and equivalent physical parameters), a predictive model of the dynamic behavior of the molten pool on the workpiece scale (two-phase flow model) was established for the SLM process of 316L stainless steel. The related equivalent processing models were customized by secondary development means based on the commercial software Fluent. By comparing and analyzing the different calculation schemes, it was found that surface tension stabilizes the liquid metal surface, while the Marangoni effect and the gasification recoil force cause the liquid metal surface to appear concave. The tangential movement of the liquid metal surface, caused by the Marangoni effect, causes the molten metal to accumulate around the central region, forming a liquid surface morphology resembling a crater. The influence of different processing parameters (scanning speed and laser power) on the SLM process of 316L stainless steel was analyzed. The simulated and experimentally obtained solidified track sizes were in good agreement.
Numerical investigation on molten pool dynamics during multi-laser array powder bed fusion process
Liu Cao*
引用：Liu Cao. Numerical investigation on molten pool dynamics during multi-laser array powder bed fusion process[J]. Metallurgical and Materials Transactions A, 2021, 52: 211-227
摘要：Multi-laser powder bed fusion (MLPBF) has become the most promising technology for rapid manufacturing of large metal parts. As a branch of MLPBF, multi-laser array powder bed fusion (MLA-PBF) has gradually attracted the attention of the industry, because of its advantages such as significantly speeding up production efficiency and low technical implementation difficulty. However, there is currently a lack of simulation studies based on the mesoscopic scale to describe the dynamic behavior of the MLA-PBF molten pool. The MLA-PBF spreading powder process was calculated herein based on the open source DEM framework Yade, the MLA-PBF molten pool dynamics was described based on the open source CFD framework OpenFOAM, and a multi-laser heat source model for real-time tracking of changes in the metal-phase and gas-phase interface was proposed. Aiming at the single-line mode of MLA-PBF, it was found that the dual-laser forming with low-front and high-rear could be used to preheat and pre-sinter the metal particles that were about to enter the molten pool, which was beneficial to reduce the pore defect in the solidified track, and a moderate laser beam space should be used. Aiming at the multi-line mode of MLA-PBF, it could form a molten pool with a significantly larger width and length than in the case of a single-laser beam, which was beneficial to eliminate pore defect in the formed zone, obtain a flat solidified track surface, and improve forming efficiency. When the laser power was low or the laser beam space was large, a large number of pores were prone to appear in the formed zone. As the laser power increased or the laser beam space decreased, when the laser energy was sufficient to melt the metal particles located in the lower part of the powder bed, a smooth surface of the solidified track and fewer pore defect would be obtained. This paper is expected to provide theoretical support for deepening the application of MLA-PBF in metal additive manufacturing.
Mesoscopic-scale numerical investigation including the influence of scanning strategy on selective laser melting process
Liu Cao*
期刊：Computational Materials Science（JCR2区）
引用：Liu Cao. Mesoscopic-scale numerical investigation including the influence of scanning strategy on selective laser melting process[J]. Computational Materials Science, 2021, 189: 110263
摘要：Selective laser melting (SLM) has become one of the most widely used laser additive manufacturing technologies today. However, the proper selection of scanning strategies is still a key concern for SLM production. In order to quantify the influences of different scanning strategies (One-direction, Two-directions, and Pre-sinter) on the SLM process, this paper calculated the spreading powder process based on the open-source discrete element method framework Yade, predicted the SLM mesoscopic molten pool dynamics behavior based on the open-source finite volume method framework OpenFOAM, and established the simulation flow of the SLM multi-layer multi-path forming process. The influences of different scanning strategies on grain orientation, pore defect, and surface roughness were analyzed and verified in comparison with experimental results. It is found that the grain orientation of the current formed layer under scanning strategies One-direction and Pre-sinter was almost the same as that of the previous formed layer, while the grain orientation of the current formed layer under scanning strategy Two-directions was significantly different from that of the previous formed layer. In addition, the porosity and surface roughness under scanning strategy Pre-sinter were lower than those under scanning strategies One-direction and Two-directions. This paper is expected to provide a basis for the selection of scanning strategies in practical SLM production.
Mesoscopic-Scale Numerical Investigation Including the Influence of Process Parameters on LPBF Multi-Layer Multi-Path Formation
Liu Cao*
引用：Liu Cao. Mesoscopic-Scale Numerical Investigation Including the Influence of Process Parameters on LPBF Multi-Layer Multi-Path Formation[J]. Computer Modeling in Engineering & Sciences, 2021, 126(1): 5-23
摘要：As a typical laser additive manufacturing technology, laser powder bed fusion (LPBF) has achieved demonstration applications in aerospace, biomedical and other fields. However, how to select process parameters quickly and reasonably is still the main concern of LPBF production. In order to quantitatively analyze the influence of different process parameters (laser power, scanning speed, hatch space and layer thickness) on the LPBF process, the multi-layer and multi-path forming process of LPBF was predicted based on the open-source discrete element method framework Yade and the open-source finite volume method framework OpenFOAM. Based on the design of experiments method, a four-factor three-level orthogonal test scheme was designed, and the porosity and surface roughness data of each calculation scheme were extracted. By analyzing the orthogonal test data, it was found that as the laser power increased, the porosity decreased, and as the scanning speed, hatch space, and layer thickness increased, the porosity increased. In addition, the influence of laser power and scanning speed on surface roughness showed a trend of decreasing first and then increasing, while the influence of scanning distance and layer thickness on surface roughness showed a monotonous increasing trend. The order of the influence of each process parameter on porosity was: scanning speed > laying thickness > laser power > hatch space, and the order of the influence of each process parameter on surface roughness was: hatch space > layer thickness > laser power > scanning speed. So the porosity of the part is most sensitive to scanning speed, and the surface roughness is the most sensitive to hatch space. The above conclusions are expected to provide process control basis for actual LPBF production of the 316L stainless steel alloy.
Mesoscopic-scale Numerical Simulation including the Influence of Process Parameters on SLM Single-layer Multi-pass Formation
Liu Cao*
引用：Liu Cao. Mesoscopic-scale Numerical Simulation including the Influence of Process Parameters on SLM Single-layer Multi-pass Formation[J]. Metallurgical and Materials Transactions A, 2020, 51: 4130-4145
摘要：Selective laser melting (SLM) is a metal additive manufacturing technology that directly forms three-dimensional complex components according to digital models via layer-by-layer addition. It has been widely used in medical personalization, aerospace, and other fields. In order to analyze the influence of different process parameters, such as line energy density and hatch space on the SLM single-layer multi-pass formation process, a random particle distribution of the powder bed was first obtained via the open-source discrete element method (DEM) code Yade. The prediction model of the molten pool dynamic behavior during the SLM formation process was then established based on the “metal-gas” two-phase flow model. The conservation equation considered thermal factors, such as Marangoni effect, the porosity in the mushy zone, and the gasification phenomenon. Laser energy was then applied by the body heat source model, which directly tracked the metal-phase surface affected by the laser in real time and applied energy to the metal-phase elements within a certain thickness. By analyzing the simulation results, it was found that in order to obtain a good formation zone in actual SLM production for 316L stainless steel, from the perspective of controlling line energy density, 200 J/m should be used to obtain a relatively flat solidified track and to establish a good connection with the substrate or the upper formation layer; from the perspective of controlling the hatch space, 45 μm should be used to ensure a good connection between adjacent solidified tracks, a relatively flat formation surface, and high production efficiency. These conclusions were consistent with the experimental results. This paper offers a scientific rationale for parameter selection during the SLM formation process of 316L stainless steel.
Mesoscopic-scale simulation of pore evolution during laser powder bed fusion process
Liu Cao*
期刊：Computational Materials Science（JCR2区）
引用：Liu Cao. Mesoscopic-scale simulation of pore evolution during laser powder bed fusion process[J]. Computational Materials Science, 2020, 179: 109686
摘要：Laser powder bed fusion (LPBF) is an advanced manufacturing technology that uses data-driven, layer-by-layer accumulation of materials to form metal components and has been widely applied in aerospace and other fields. Effectively controlling pore defects is a key scientific problem and technical difficulty in LPBF industrial production. Based on the open-source discrete element method code Yade, the particle distribution of the powder bed was obtained. Based on the open-source computational fluid dynamics code OpenFOAM, the pore evolution during the LPBF formation process at the mesoscopic scale was predicted. The thermal–force factors affecting the molten pool included the surface tension, Marangoni effect, gasification recoil force, and mushy drag force. The laser energy model used a body heat source based on interface tracking. First, dimensionless analysis of the molten pool evolution in the case of LPBF single-track formation was carried out. The molten pool evolution was mainly influenced by the gasification recoil force, Marangoni effect and surface tension, and the main influencing factors on different zones of the molten pool were different. To examine the influences of the laser power, scanning speed, powder bed thickness, and hatch space on the pore defect in the LPBF formation process, simulations were carried out and compared with experimental results. When the volumetric energy density was too small, pore defects occurred due to insufficient fusion of metal particles, and when the volumetric energy density was too large to cause the “keyhole” effect, pore defects occurred because the entrained gas could not escape in time. This paper is expected to provide theoretical guidance for the scientific regulation of pore defects in LPBF production.
Study on the numerical simulation of laying powder for selective laser melting process
Liu Cao*
引用：Liu Cao. Study on the numerical simulation of laying powder for selective laser melting process[J]. International Journal of Advanced Manufacturing Technology, 2019, 105: 2253-2269
摘要：Because the selective laser melting (SLM) formation process involves rapid melting and solidification of slices, the SLM process places high demands on the tightness, uniformity, and flatness of the powder layer. Based on the discrete element method (particle contact force model, particle collision judgment algorithm, and particle motion equation) and the SLM laying powder process, a numerical simulation of the SLM laying powder process was carried out. For the performance measurement experiment of the TC4 titanium alloy powder, the powder bulk density, tap density, and angle of repose were calculated and analyzed. It was found that the tap density increased by 7.5% compared to the bulk density, and the calculated average angle of repose (32.6°) was in good agreement with the experimental data (33.2°), thus verifying the accuracy of the calculation model used for the SLM laying powder. The influences of different scraping methods and scraping speeds on the quality of the laying powder were calculated and analyzed. It was found that the scraping method using a roller (not rotating) obtained the highest tightness and most uniform powder distribution, and, as the scraping speed increased, the laying tightness tended to decrease linearly. The results of the numerical simulation study of the SLM laying powder process can be used to guide the actual SLM laying powder process and, alternatively, provide basic data for the numerical simulation of SLM molten pool dynamics based on the particle scale.
Numerical simulation of the impact of laying powder on selective laser melting single-pass formation
Liu Cao*
引用：Liu Cao. Numerical simulation of the impact of laying powder on selective laser melting single-pass formation[J]. International Journal of Heat and Mass Transfer, 2019, 141: 1036-1048
摘要：Selective laser melting (SLM) is an additive manufacturing technology to directly form complex metal parts. It has been demonstrated in aerospace and other fields. In the SLM single-pass formation process, the SLM layer distribution was modeled here via the discrete element method model (particle contact force model and particle motion equation). A dynamic behavior model of molten pool based on the particle scale was also established. The governing equations considered the influence of thermodynamic factors such as Marangoni effect, gasification recoil, and gasification heat dissipation. The laser energy model used a Gaussian heat source based on interface tracking. To study the influence of laying powder on SLM single-pass formation, the corresponding simulations were performed from three aspects: particle size distribution, powder bed tightness, and lamination thickness. To obtain a good formation zone in the actual SLM process, metal powder with a smaller average particle size should be used on the basis of saving money, and the proportion of large-sized particles in the powder should be minimized. A higher powder bed tightness is needed on the basis of ensuring a good connection with the substrate or the previous printed layer. Finally, a larger lamination thickness should be used on the basis of ensuring that the powder bed is sufficiently melted in the formation zone. This paper is expected to provide some guidance for the actual SLM laying powder process.
Study on the Numerical Simulation of the SLM Molten Pool Dynamic Behavior of a Nickel-Based Superalloy on the Workpiece Scale
Liu Cao*, Xuefeng Yuan
期刊：Materials（JCR2区）
引用：Liu Cao, Xuefeng Yuan. Study on the Numerical Simulation of the SLM Molten Pool Dynamic Behavior of a Nickel-Based Superalloy on the Workpiece Scale[J]. Materials, 2019, 12: 2272
摘要：Nickel-based superalloys are one of the most industrially important families of metallic alloys at present. Selective Laser Melting (SLM), as one of the additive manufacturing technologies for directly forming complex metal parts, has been applied in the production of Inconel 718 components. Based on the more reasonable and comprehensive equivalent processing models (vaporization heat loss, equivalent physical parameters) for the nickel-based superalloy SLM process, an SLM molten pool dynamic behavior prediction model on the workpiece scale was established. Related equivalent processing models were customized by secondary development with the software Fluent. In order to verify the feasibility of the SLM molten pool dynamics model, the SLM single-pass employed to form the Inconel 718 alloy process was calculated. The simulated and experimental solidified track dimensions were in good agreement. Then, the influences of different process parameters (laser power, scanning speed) on the SLM formation of the Inconel 718 alloy were calculated and analyzed. The simulation and experimental solidified track widths were well-matched, and the result showed that, as a rule, the solidified track width increased linearly with the laser power and decreased linearly with the scanning speed. This paper will help lay the foundation for a subsequent numerical simulation study of the thermal-melt-stress evolution process of an SLM workpiece.
Numerical simulation of cold-lap defects during casting filling process
Liu Cao, Dunming Liao*, Fei Sun, Tao Chen
引用：Liu Cao, Dunming Liao, Fei Sun, Tao Chen. Numerical simulation of cold-lap defects during casting filling process[J]. International Journal of Advanced Manufacturing Technology, 2018, 97: 2419-2430
摘要：Cold lap is one of the common defects in casting productions, which could result in poor surface accuracy and mechanical properties of castings. The cold-lap formation process is difficult to observe directly using only experimental measures, since the casting filling process is in a state of high-temperature flow inside a mold. The keys to predicting defects accurately are searching the interface front of flow, judging the collision of fronts, and defining the formation model of cold laps. A formation model of cold lap, which is related to the solid fraction, velocity, and volume fraction of the metal phase, is developed in this paper, and a method for judging the interface front and its collision was adopted. On the basis of the open-source computational fluid dynamics software OpenFOAM, a solver for predicting cold lap defects during the casting filling process was developed. The filling process of an aluminum alloy benchmark test was simulated, and the simulation results were consistent with the experimental results, which indicates the accuracy of the adopted model. Two low-pressure die-casting copper alloy filling processes with different ingates were calculated, and the formation processes of cold lap were compared with each other. The simulation results showed that there was significantly lesser cold lap in the scheme with an ingate under the lower surface of casting, which was basically consistent with the experimental results; therefore, the practicability of the adopted model was confirmed.
Quantitative prediction of oxide inclusion defects inside the casting and on the walls during cast-filling processes
Liu Cao, Fei Sun*, Tao Chen, Yulong Tang, Dunming Liao
引用：Liu Cao, Fei Sun, Tao Chen, Yulong Tang, Dunming Liao. Quantitative prediction of oxide inclusion defects inside the casting and on the walls during cast-filling processes[J]. International Journal of Heat and Mass Transfer, 2018, 119: 614-623
摘要：Oxide inclusions are a common defect in casting production, but their formation and evolution processes are difficult to directly observe experimentally. To accurately predict these, one should calculate the formation rate of the oxide inclusion as well as its wall adhesion. Here, we propose a formation rate model that is relevant to the volume fraction of liquid metal, the temperature, and the current oxide inclusion density. A boundary condition was developed to handle the adhesion and accumulation on the wall; thus, a quantitative indicator is available to directly obtain the content of the oxide inclusion defects on different parts. We used the open source computational fluid dynamics (CFD) software OpenFOAM to predict oxide inclusion defects inside the casting and on the walls during cast-filling processes. A representative aluminum alloy-casting technology was simulated to analyze the effects of different runner systems on the oxide inclusion distributions inside casting and on the wall. The simulation results largely coincide with experimental data from the literature. A practical copper alloy-casting technology was then calculated, and the simulation results of the oxide inclusion defects are consistent with the actual casting defect distribution. This confirms the utility of the model.
Prediction of gas entrapment defects during zinc alloy high-pressure die casting based on gas-liquid multiphase flow model
Liu Cao, Dunming Liao*, Fei Sun, Tao Chen, Zihao Teng, Yulong Tang
引用：Liu Cao, Dunming Liao, Fei Sun, Tao Chen, Zihao Teng, Yulong Tang. Prediction of gas entrapment defects during zinc alloy high-pressure die casting based on gas-liquid multiphase flow model[J]. International Journal of Advanced Manufacturing Technology, 2018, 94(1): 807-815
摘要：Zinc alloy is the preferred material for high-pressure die casting (HPDC) production because of its good casting and mechanical properties, and the most common problem in zinc alloy die casting is gas entrapment. A gas-liquid multiphase flow model was here used to predict the gas entrapment defects during the zinc alloy HPDC filling process, and the continuum surface force (CSF) model was used to treat the surface tension of gas-liquid multiphase. In addition, finite volume method (FVM) was used for discretization equations; the pressure implicit with splitting of operator (PISO) algorithm was used for coupling pressure and velocity; and the volume of fluid (VOF) algorithm was used for interface tracking. A water filling experiment was simulated in an S-shaped channel and the simulative results were closely consistent with the experimental results, which indicates the accuracy of the adopted model. Two HPDC zinc alloy filling processes with different ingates were calculated, and the simulative results showed that gas entrapment defects associated with a single ingate were visibly more pronounced than with double ingates, which was basically consistent with the experimental results; therefore, the practicability of the adopted model was confirmed.
铸造充型过程中液固转变影响流动行为的数值计算
曹流, 孙飞*, 陈涛, 滕子浩, 唐玉龙, 廖敦明
期刊：金属学报（JCR3区）
引用：曹流, 孙飞, 陈涛, 滕子浩, 唐玉龙, 廖敦明. 铸造充型过程中液固转变影响流动行为的数值计算[J]. 金属学报, 2017, 53(11): 1521-1531
摘要：为准确预测浇不足及冷隔，在已有的处理液固转变方法基础上，提出基于固相率变化的糊状区流动行为计算模型，该模型可以有效地处理液固转变过程中糊状区不同阶段的流动行为，即高固相率糊状区采用临界固相率方法，低固相率糊状区采用变黏度方法，中等固相率糊状区采用多孔介质拖拽模型。计算了S型铸型水模拟实验，模拟结果与实验结果吻合很好，验证了不考虑液固转变时所采用模型的准确性。针对简单形状的底注式铸造工艺，对比分析了处理液固转变过程中采用不同控制参数的计算效果，证明了糊状区流动行为计算模型的合理性。
铸造宏观过程数值模拟技术的研究现状与展望
廖敦明*, 曹流, 孙飞, 陈涛
期刊：金属学报（JCR3区）
引用：廖敦明, 曹流, 孙飞, 陈涛. 铸造宏观过程数值模拟技术的研究现状与展望[J]. 金属学报, 2017, 54(2): 161-173
摘要：本文综述了铸造宏观过程模拟技术的研究现状。对比了不同流动场模型下铸造充型过程的差别，其中，两相流模型可准确地考虑气相对充型过程的影响。分析了不同应力场模型对铸造应力演变过程的适用程度，并阐明了其发展趋势。说明了适用于铸造宏观过程模拟的物性参数的获取及修正方法，即采用实验手段测量合金成分和液固相线温度，通过物性参数计算软件计算得到合金物性参数并做适当调整，进而结合测温实验对相关参数进行修正。对比了不同铸造工艺下的边界条件，并对高压铸造工艺(速度进口边界)和定向凝固工艺(辐射换热边界)的边界条件进行了说明。对比了不同网格类型的区别，并结合不同网格类型说明了不同数值求解算法的区别，本文认为自适应六面体网格和混合网格类型更适合于有限体积法(充型过程计算)和有限元法(凝固过程和应力计算)。针对各种铸造缺陷，对其预测模型和分析方法进行了说明，并阐明了其发展趋势。
Heat transfer model of directional solidification by LMC process for superalloy casting based on finite element method
Liu Cao, Dunming Liao, Yuzhang Lu*, Tao Chen
引用：Liu Cao, Dunming Liao, Yuzhang Lu, Tao Chen. Heat transfer model of directional solidification by LMC process for superalloy casting based on finite element method[J]. Metallurgical and Materials Transactions A, 2016, 47(9): 4640-4647
摘要：With the rapid development of the aviation industry, the turbine blade, a critical component of the aeronautical engine, has come to be widely produced by liquid-metal cooling (LMC) process. A temperature- and time-dependent heat transfer coefficient was used to represent the heat convection between the shell and the cooling liquid, and an improved Monte Carlo ray tracing approach was adopted to handle the boundary of radiation heat transfer. Unstructured mesh was used to fit the irregular shell boundary, and the heat transfer model of directional solidification (DS) by LMC process based on finite element method (FEM) was established. The concept of local matrix was here proposed to guarantee computational efficiency. The pouring experiments of directional solidification by LMC process were carried out, then simulation and experimental results were here compared. The accuracy of the heat transfer model was validated by the cooling curves and grain morphology, and the maximum relative error between simulation and experimental cooling curve was 2%. The withdrawal rate showed an important influence on the shape of solidification interface, and stray grain is liable to be generated on the bottom of platform at an excessive withdrawal rate.
Radiation heat transfer model for complex superalloy turbine blade in directional solidification process based on finite element method
Dunming Liao, Liu Cao, Tao Chen*, Fei Sun, Yongzhen Jia, Zihao Teng, Yulong Tang
期刊：China Foundry（JCR4区）
引用：Dunming Liao, Liu Cao, Tao Chen, Fei Sun, Yongzhen Jia, Zihao Teng, Yulong Tang. Radiation heat transfer model for complex superalloy turbine blade in directional solidification process based on finite element method[J]. China Foundry, 2016, 13(2): 123-132
摘要：For the sake of more accurate shell boundary and calculation of radiation heat transfer in Directional Solidification (DS) process, a radiation heat transfer model based on Finite Element Method (FEM) is developed in this study. The key technologies, such as distinguishing boundaries automatically, local matrix and lumped heat capacity matrix, are also stated. So as to analyze the effect of withdrawing rate on DS process, the solidification processes of a complex superalloy turbine blade in High Rate Solidification (HRS) process with different withdrawing rates are simulated, meanwhile, the most suitable withdrawing rate is supposed to be 5.0 mm·min-1 by comparing the simulation results. Eventually, the correctness and reliability of the radiation heat transfer model are verified, because of the accordance of simulation results with practical process.
铸造充型过程中浇不足缺陷的数值计算研究
曹流, 廖敦明*, 胡柏乐, 张影, 陈涛
期刊：铸造（中文核心）
引用：曹流, 廖敦明, 胡柏乐, 张影, 陈涛. 铸造充型过程中浇不足缺陷的数值计算研究[J]. 铸造, 2018, 67(6): 487-491
摘要：浇不足是铸造生产中常见的缺陷，目前实验手段只能测量和分析铸件最终的浇不足缺陷分布，难以直接观测浇不足缺陷的形成和演变过程，准确预测浇不足缺陷的关键在于处理凝固过程中流动行为的转变。在已有的处理流动行为转变的基础上，提出一随固相率变化的浇不足缺陷计算模型，即高固相率区采用临界固相率方法，低固相率区采用变粘度方法，中等固相率区采用多孔介质拖拽模型。计算了一组多壁厚铜合金低压铸造工艺，对比分析了模拟和实验铸件充型高度，结果对应较好，证明了所提出的浇不足缺陷计算模型的准确性。
基于有限元法的多材质铸造温度场模拟软件开发
曹流, 廖敦明*, 周聪, 陈涛, 庞盛永, 周建新
期刊：特种铸造及有色合金（中文核心）
引用：曹流, 廖敦明, 周聪, 陈涛, 庞盛永, 周建新. 基于有限元法的多材质铸造温度场模拟软件开发[J]. 特种铸造及有色合金, 2015, 35(11): 1163-1166
摘要：本文以瞬态的傅里叶导热微分方程为基础，提出了一种处理多材质铸造过程边界条件的通用化方法，自主开发了基于有限元法（FEM）的多材质铸造过程温度场数值模拟程序，通过计算典型工字件空冷过程温度场以及多材质复杂铝合金铸件凝固过程温度场，并与ANSYS以及华铸CAE的模拟结果进行了对比，对比结果吻合良好，其中空冷过程温度场的相对误差在1‰数量级，凝固过程温度场的相对误差在1％数量级，验证了温度场程序的正确性以及可靠性。
基于有限元法的熔模铸造过程温度场模拟软件自主开发
曹流, 廖敦明*, 曹腊梅, 谷怀鹏, 陈涛, 庞盛永
期刊：铸造（中文核心）
引用：曹流, 廖敦明, 曹腊梅, 谷怀鹏, 陈涛, 庞盛永. 基于有限元法的熔模铸造过程温度场模拟软件自主开发[J]. 铸造, 2014, 63(12): 1235-1240
摘要：针对熔模铸造中型壳的随型特点，提出了一种处理不规则铸型边界的方法，并以瞬态傅里叶导热微分方程为基础，自主开发了基于有限元法（FEM）的熔模铸造温度场数值模拟软件，并阐述了潜热以及集中热容的处理等关键技术。通过一个复杂熔模缸体铸件铸造过程温度场的模拟计算，并与ProCAST的模拟结果进行了对比，发现二者模拟结果基本相近，凝固过程温度场的最大相对误差为1.113％，验证了所开发的温度场程序的正确性和可靠性。
发明专利：
毕业论文：
铸造充型过程中成形类缺陷演化机理及数值模拟研究
博士毕业论文
关键词：铸造充型；成形类缺陷；数值模拟；冷隔；氧化夹渣；OpenFOAM
引用：曹流. 铸造充型过程中成形类缺陷演化机理及数值模拟研究[D]. 武汉: 华中科技大学, 2018
主要内容：缺陷预测及工艺优化是铸造宏观过程模拟的最终目的，而铸造充型过程中可能产生的成形类缺陷(如卷气、浇不足、冷隔和氧化夹渣等)，往往只能借助充型模拟结果间接地分析，未能充分利用数值模拟技术所带来的数字化分析优势。且随着对铸造模拟技术实用化要求的逐步提高，相关缺陷分析方法已满足不了实际生产需求。采用数值模拟手段对铸造充型过程中成形类缺陷进行深入研究，对提高铸件性能和优化铸造工艺有着十分巨大的价值。本文对铸造充型过程成形类缺陷数值模拟所涉及的各个方面进行了深入的研究和探讨，包括铸造充型过程多相流数学建模、冷隔缺陷演化机理及预测和氧化夹渣缺陷演化机理及预测等。同时，本文基于计算流体力学开源代码完成了相关求解器的开发工作，有助于广大科研工作者深入研究铸造充型过程数值模拟技术。
基于有限元法的定向凝固过程温度场数值模拟的研究
硕士毕业论文
关键词：定向凝固；数值模拟；有限元法；高速凝固法；液态金属冷却法；辐射换热；盒子树法；G/L判据
引用：曹流. 基于有限元法的定向凝固过程温度场数值模拟的研究[D]. 武汉: 华中科技大学, 2015
主要内容：在航空航天领域中，主要采用定向凝固工艺生产涡轮叶片，而高性能的涡轮叶片一直是阻碍我国实现“大飞机”梦的最大障碍。实际定向凝固过程是十分复杂的材料成形过程，而且需要特别考虑辐射换热的影响。采用定向凝固温度场模拟技术，能够对铸件在凝固过程中的温度场演变过程进行有效的分析，预测可能出现的缺陷，为优化定向凝固工艺提供理论指导。本文对有限元定向凝固温度场数值模拟所涉及的各个方面进行了深入研究，包括定向凝固温度场数学建模、有限元数值求解、液态金属冷却液换热处理和定向凝固温度场分析判据等，开发了有限元定向凝固温度场数值模拟系统。
其他期刊论文：
学术会议：
指导学生(研究生)：
指导学生(本科生)：
指导学生(本科毕设)：
教学：
开源代码及云计算平台：
OpenFOAM is the leading free, open source software for CFD, owned by the OpenFOAM Foundation and distributed exclusively under the General Public Licence (GPL). OpenFOAM was created by Henry Weller in 1989 under the name "FOAM" and was released open source as "OpenFOAM" by Henry Weller, Chris Greenshields and Mattijs Janssens in December 2004.
PRISMS-PF is a powerful, massively parallel finite element code for conducting phase field and other related simulations of microstructural evolution. The phase field method is commonly used for predicting the evolution if microstructures under a wide range of conditions and material systems. PRISMS-PF provides a simple interface for solving customizable systems of partial differential equations of the type commonly found in phase field models, and has 24 pre-built application modules, including for precipitate evolution, grain growth, dendritic solidification, and spinodal decomposition.
Yade is an extensible open-source framework for discrete numerical models, focused on Discrete Element Method. The computation parts are written in c++ using flexible object model, allowing independent implementation of new alogrithms and interfaces. Python is used for rapid and concise scene construction, simulation control, postprocessing and debugging.
Elmer is an open source multiphysical simulation software mainly developed by CSC - IT Center for Science (CSC). Elmer development was started as national collaboration with Finnish Universities, research institutes and industry. After it's open source publication, the use and development of Elmer has become largely international. Elmer includes physical models of fluid dynamics, structural mechanics, electromagnetics, heat transfer and acoustics, and beyond. These are described by partial differential equations which Elmer solves by the Finite Element Method (FEM). Elmer supports parallel computing. For many problems good scalability over thousands of cores is reached.
OpenFVM is a general CFD solver released under the GPL license. It was developed to simulate the flow in complex 3D geometries. Therefore, the mesh can be unstructured and contain control volumes with arbitrary shape. The code uses the finite volume method to evaluate the partial differential equations. As well as solving the velocity and pressure fields, the code is capable of solving non-isothermal multiphase flow.
Computational analysis tools have revolutionized the way we design aerospace systems, but most established codes are proprietary, unavailable, or prohibitively expensive for many users. The SU2 team is changing this, making computational analysis and design freely available as open-source software and involving everyone in its creation and development.
Fluidity is an open source, general purpose, multiphase computational fluid dynamics code capable of numerically solving the Navier-Stokes equation and accompanying field equations on arbitrary unstructured finite element meshes in one, two and three dimensions. It is parallelised using MPI and is capable of scaling to many thousands of processors.
A software tool for classical CFD, particle-based models and complex physical interaction, Palabos offers a powerful environment for your fluid flow simulations.
MFiX is an open-source multiphase flow solver, providing a Two-Fluid Model (TFM), a Particle In Cell (PIC) model, a Discrete Element Model (DEM), and a Hybrid model. These models can be used to simulate a variety of multiphase flow devices including gasifiers, circulating fluid bed combusters, fluidized beds, fluid catalytic crackers, and chemical looping combustion systems.
GPUSPH was the first implementation of Weakly-Compressible Smoothed Particle Hydrodynamics (WCSPH) to run entirely on GPU with CUDA. The initial version was developed by Alexis Hérault (INGV, CNAM), Giuseppe Bilotta (DMI, INGV) and Robert A. Dalrymple (JHU).
DualSPHysics is based on the Smoothed Particle Hydrodynamics model named SPHysics (www.sphysics.org). The code is developed to study free-surface flow phenomena where Eulerian methods can be difficult to apply, such as waves or impact of dam-breaks on off-shore structures. DualSPHysics is a set of C++, CUDA and Java codes designed to deal with real-life engineering problems.
Solidity is an Open Source general purpose, two and three dimension finite element – discrete element solid mechanics code. Solidity solves highly non-linear problems for continuum and discontinuous domains. Features include fracture and fragmentation without the need to seed, a range of finite strain large deformation constitutive models including elasto-plasticity and thermal fields. Parallelisation and fast contact detection and interaction algorithms are included.
TetGen is a program to generate tetrahedral meshes of any 3D polyhedral domains. TetGen generates exact constrained Delaunay tetrahedralizations, boundary conforming Delaunay meshes, and Voronoi partitions. The following pictures respectively illustrate a 3D polyhedral domain (left), a boundary conforming Delaunay tetrahedral mesh (middle), and its dual - a Voronoi partition (right).
simFlow is a powerful general-purpose CFD software. It combines an intuitive graphical user interface with the advantages of the open-source OpenFOAM libraries.
SimScale is a browser-based online-platform that provides powerful modeling and simulation capabilities. With in-browser 3D visualization, scalable on-demand computing capacity and a pay-per-use pricing model, the SimScale platform enables a new way of using simulation technology: efficient, flexible and powerful.
Weka is a collection of machine learning algorithms for data mining tasks. The algorithms can either be applied directly to a dataset or called from your own Java code. Weka contains tools for data pre-processing, classification, regression, clustering, association rules, and visualization. It is also well-suited for developing new machine learning schemes.
The Robot Operating System (ROS) is a flexible framework for writing robot software. It is a collection of tools, libraries, and conventions that aim to simplify the task of creating complex and robust robot behavior across a wide variety of robotic platforms.
TensorFlow makes it easy for beginners and experts to create machine learning models for desktop, mobile, web, and cloud. See the sections below to get started.
学术背景：
广州大学是以中国重要中心城市“广州”命名的综合性大学，于2000年合并组建，有着近90年的办学传统。学校立足珠三角区位优势，强化大学的使命与担当，锐意改革，开拓进取，各项事业实现跨越式发展，成为一所快速发展、充满活力的大学，是广东省和广州市高水平大学建设高校。广州大学正在向国内一流、国际知名的高水平大学迈进。
2018年6月至今，我于广州大学机械与电气工程学院担任讲师。
华中科技大学是国家教育部直属的全国重点大学，由原华中理工大学、同济医科大学、武汉城市建设学院于2000年5月26日合并成立，是首批列入国家“211工程”重点建设和国家“985工程”建设高校之一。
在2009年至2018年期间，我于华中科技大学攻读学士、硕士以及博士学位（2009-2013 学士；2013-2015 硕士；2015-2018 博士）。