人工智能在精密铸造及铸件检测中的应用研究进展

Research progress on application of artificial intelligence in precision casting and casting inspection

摘要

精密铸造是航空发动机、燃气轮机涡轮叶片等关键高端装备核心部件的主要制造工艺，直接决定部件的质量和性能，影响高端装备的效率与可靠性。然而，精密铸造工艺仍面临铸件缺陷控制、尺寸精度与变形控制、铸件质量检测等挑战。人工智能技术的迅速发展，为精密铸造工艺优化及铸件检测提供新的技术路径。本工作系统综述人工智能技术在精密铸造领域的研究进展，主要涵盖人工智能在“型芯-蜡模-型壳”主要中间产物制备工艺环节的应用、人工智能在铸件凝固过程中辅助缺陷与性能控制和尺寸控制的进展、人工智能在铸件检测的应用包括表面和内部缺陷检测、金相组织分析等方面。最后，总结人工智能技术应用在精密铸造工艺与铸件检测中的优势与挑战，并展望未来研究方向，包括：（1）推动人工智能由单环节经验预测向全流程、多尺度统一建模发展；（2）推动精密铸造由“控形”或“控性”向“形性协同调控”转变；（3）发展适用于精密铸造过程的物理信息神经网络；（4）实现精密铸造过程从离线优化向自主闭环控制的转变；（5）大模型为精密铸造智能化发展提供新的技术路径。

Abstract

Precision casting is the primary manufacturing process for core components of high-end critical equipment such as turbine blades for aero-engines and gas turbines，which directly determines the quality and performance of components and affects the efficiency and reliability of high-end equipment. Nevertheless，the precision casting process still faces challenges including casting defect control，dimensional accuracy and deformation control，and casting quality inspection. The rapid advancement of artificial intelligence technology offers new technical approaches for investment-casting process optimization and casting inspection. This paper systematically reviews the research progress of artificial intelligence in the field of investment casting. It mainly covers the applications of artificial intelligence in the preparation of key intermediate products including ceramic cores，wax patterns，and ceramic shells；the advances of artificial intelligence in assisting defect-performance control and dimensional control during casting solidification；as well as its applications in casting inspection，involving surface and internal defect detection，metallographic microstructure analysis，and other aspects. Finally，the advantages and challenges of applying artificial intelligence to precision casting processes and casting inspection are summarized，and future research directions are prospected： (1) promote the evolution of artificial intelligence from empirical prediction for individual process links toward unified full-process and multi-scale modelling； (2) drive the transformation of investment casting from independent shape control or performance control toward coordinated shape-performance regulation； (3) develop physics-informed neural networks tailored for the investment-casting process； (4) realize the transition of investment-casting processes from offline optimization to autonomous closed-loop control； (5) large-language models provide new technical pathways for the intelligent development of precision casting.

关键词

人工智能 / 涡轮叶片 / 精密铸造 / 控形控性 / 铸件检测

Key words

artificial intelligence / turbine blade / precision casting / shape and property control / casting inspection

参考文献

[1] Biswas S，Ramachandra S，Hans P，et al. Materials for gas turbine engines：present status，future trends and indigenous efforts[J].Journal of the Indian Institute of Science，2022，102（1）：297-309.

[2] 张攀，汪东红，肖程波，等. 航空发动机高温合金铸件精密成形研究进展[J].航空制造技术，2025，68（18）：162-172.

[3] Zhang P，Wang D H，Xiao C B，et al. Research progress on precision forming of aero—engine superalloy castings[J].Aeronautical Manufacturing Technology，2025，68（18）：162-172.

[4] 刘浩，李国庆，张深，等. 机器学习在涡轮机械中的应用进展[J].工程热物理学报，2023，44（4）：938-951.

[5] Liu H，Li G Q，Zhang S，et al. Application progress of machine learning in turbomachinery[J].Journal of Engineering Thermophysics，2023，44（4）：938-951.

[6] 刘维伟. 航空发动机叶片关键制造技术研究进展[J].航空制造技术，2016，59（21）：50-56.

[7] Liu W W. Research progress on key manufacturing technology of aeroengine blades[J].Aeronautical Manufacturing Technology，2016，59（21）：50-56.

[8] Pattnaik S，Karunakar D B，Jha P K. Developments in investment casting process：a review[J].Journal of Materials Processing Technology，2012，212（11）：2332-2348.

[9] 蒋睿嵩，汪文虎，王增强，等. 航空发动机涡轮叶片精密成形技术及其发展趋势[J].航空制造技术，2016，59（21）：57-62.

[10] Jiang R S，Wang W H，Wang Z Q，et al. Precision forming technology and its development trend of aeroengine turbine blade[J].Aeronautical Manufacturing Technology，2016，59（21）：57-62.

[11] Xuan W，Du L，Han Y，et al. Investigation on microstructure and creep properties of nickel based single crystal superalloys PWA1483 during heat treatment under an alternating magnetic field[J].Materials Science and Engineering：A，2019，762：138087.

[12] Li X，Su H J，Dong D，et al. New approach to preparing near zero shrinkage alumina ceramic cores with excellent properties by vat photopolymerization[J].Journal of Materials Science & Technology，2024，93：61-72.

[13] 崔加裕，汪东红，肖程波，等. 航空发动机用高温合金复杂薄壁精密铸件尺寸精度控制技术研究进展[J].航空材料学报，2024，44（2）：31-44.

[14] Cui J Y，Wang D H，Xiao C B，et al. Research progress on dimensional accuracy control technologies of complex thin—walled superalloy investment castings for aero—engines[J].Journal of Aeronautical Materials，2024，44（2）：31-44.

[15] 孙宝德，王俊，康茂东，等. 高温合金超限构件精密铸造技术及发展趋势[J].金属学报，2022，58（4）：412-427.

[16] Sun B D，Wang J，Kang M D，et al. Investment casting technology and development trend of superalloy beyond—limit components[J].Acta Metallurgica Sinica，2022，58（4）：412-427.

[17] Li X，Su H J，Dong D，et al. Vat photopolymerization 3D printing of ceramic cores：advances，challenges，and prospects[J].China Foundry，2025，22（55）：493-506.

[18] Ren Z H，Fang F Z，Yan N，et al. State of the art in defect detection based on machine vision[J].International Journal of Precision Engineering and Manufacturing—Green Technology，2022，9（2）：661-691.

[19] Czimmermann T，Ciuti G，Milazzo M，et al. Visual—based defect detection and classification approaches for industrial applications：a survey[J].Sensors，2020，20（5）：1459.

[20] Chen Y J，Ding Y Y，Zhao F，et al. Surface defect detection methods for industrial products：a review[J].Applied Sciences，2021，11（16）：7657.

[21] 王璟怡，张悦，钟斌，等. 基于机器学习的航空材料疲劳寿命预测研究进展[J].航空材料学报，2026，46（3）：1-17.

[22] Wang J Y，Zhang Y，Zhong B，et al. Progress on fatigue life prediction of aeronautical materials based on machine learning[J].Journal of Aeronautical Materials，2026，46（3）：1-17.

[23] 张云，苏海军，李翔，等. 耐高温陶瓷基复合材料增材制造技术研究进展[J].航空材料学报，2026，46（1）：1-14.

[24] Zhang Y，Su H J，Li X，et al. Research progress of additive manufacturing technologies for high—temperature resistant ceramic matrix composites[J].Journal of Aeronautical Materials，2026，46（1）：1-14.

[25] Dong Y W，Yan W G，Wu Z P，et al. Modeling of shrinkage characteristics during investment casting for typical structures of hollow turbine blades[J].The International Journal of Advanced Manufacturing Technology，2020，110（5/6）：1249-1260.

[26] Cui K，Jing L，Jiang R S，et al. Core shift limitation in investment casting process of hollow turbine blade[J].Chinese Journal of Aeronautics，2024，37（8）：513-526.

[27] Rosnitschek T，Erber M，Alber—Laukant B，et al. Predicting the solidification time of low pressure die castings using geometric feature—based machine learning meta—models[J].Procedia CIRP，2023，118：1102-1107.

[28] Rachman F，Purnomo D A，Damayanti K A. Implementation of response surface method to minimize product defects on wax pattern injection process parameters[C]//MATEC Web of Conference. Paris：Atlantis Press，2022.

[29] Xu S Z，Yuan Q Y，Tang Z F，et al. A deformation prediction based on deep learning for sintering process of ceramic core[J].Advances in Manufacturing，2025，13（1）：147-162.

[30] Dong Y W，Guo X，Ye Q W，et al. Shrinkage during solidification of complex structure castings based on convolutional neural network deformation prediction research[J].The International Journal of Advanced Manufacturing Technology，2022，118：4073-4084.

[31] Ma Y F，Liu M，Zhang Y Q，et al. SPDP—net：a semantic prior guided defect perception network for automated aero—engine blades surface visual inspection[J].IEEE Transactions on Automation Science and Engineering，2025，22：2724-2733.

[32] Nikolić F，Štajduhar I，Čanađija M. Casting microstructure inspection using computer vision：dendrite spacing in aluminum alloys[J].Metals，2021，11（5）：756.

[33] Xu J H，Li L F，Liu X G，et al. Fast characterization framework for creep microstructure of a nickel—based SX superalloy with high—throughput experiments and deep learning methods[J].Materials Characterization，2022，187：111857.

[34] 官邦，汪东红，马洪波，等. 熔模铸件尺寸控制的数字孪生建模关键技术与应用[J].金属学报，2024，60（4）：548-558.

[35] Guan B，Wang D H，Ma H B，et al. Key technology and application of digital twin modeling for deformation control of investment casting[J].Acta Metallurgica Sinica，2024，60（4）：548-558.

[36] Wang D H，Yu J P，Yang C L，et al. Dimensional control of ring—to—ring casting with a data—driven approach during investment casting[J].The International Journal of Advanced Manufacturing Technology，2022，119（1/2）：691-704.

[37] Dai J，Xu S Z，Chen C Y，et al. A multi—objective optimization based on machine learning for dimension precision of wax pattern in turbine blade manufacturing[J].Advances in Manufacturing，2024，12（3）：428-446.

[38] Wilk—Kołodziejczyk D， Małysza M，Jaśkowiec K，et al. Modification of casting production parameters in order to obtain products with the assumed parameters with using machine learning[J].International Journal of Metalcasting，2023，17（4）：2680-2688.

[39] Mou S，Bu K，Ren S J，et al. Digital twin modeling for stress prediction of single—crystal turbine blades based on graph convolutional network[J].Journal of Manufacturing Processes，2024，116：210-223.

[40] Zong X，Wu S S，Lin K J，et al. Advanced ceramics with integrated structures and functions：machine learning prediction and experimental verification[J].Ceramics International，2024，50（13）：24126-24138.

[41] Yin X L，Li R P，Sun X，et al. Prediction of composite performance of refractory and ceramic materials based on deep learning[J].JOM，2025，77（5）：3793-3806.

[42] Jin S，Liu C H，Lai X M，et al. Bayesian network approach for ceramic shell deformation fault diagnosis in the investment casting process[J].The International Journal of Advanced Manufacturing Technology，2017，88（1/2/3/4）：663-674.

[43] Liu C H，Jiang C H，Zhou Z F，et al. Experimental study and GRNN modeling of shrinkage characteristics for wax patterns of gas turbine blades considering the influence of complex structures[J].Machines，2023，11（6）：645.

[44] Wang D H，Sun J Y，Dong A P，et al. Prediction of core deflection in wax injection for investment casting by using SVM and BPNN[J].The International Journal of Advanced Manufacturing Technology，2019，101（5/6/7/8）：2165-2173.

[45] Zhao D Y，Wang D H，Zhou L Y，et al. Effect of mold dwell time on shrinkage defects in investment casting of superalloy turbine blade[J].The International Journal of Advanced Manufacturing Technology，2024，132：4473-4487.

[46] Chen S K，Kaufmann T. Development of data—driven machine learning models for the prediction of casting surface defects[J].Metals，2022，12（1）：1.

[47] Wilk—Kolodziejczyk D，Regulski K，Gumienny G. Comparative analysis of the properties of the nodular cast iron with carbides and the austempered ductile iron with use of the machine learning and the support vector machine[J].The International Journal of Advanced Manufacturing Technology，2016，87（1/2/3/4）：1077-1093.

[48] 李杨，邸钰婷，张志坤，等. 基于机器学习的高温合金熔模铸造涡轮叶片工艺参数优化[J].特种铸造及有色合金，2025，45（8）：1160-1164.

[49] Li Y，Di Y T，Zhang Z K，et al. Optimization of process parameters for high temperature alloy investment casting blade based on machine learning[J].Special Casting & Nonferrous Alloys，2025，45（8）：1160-1164.

[50] Dong R Z，Wang W H，Zhang T R，et al. Ensemble learning—enabled early prediction of dimensional accuracy for complex products during investment casting[J].Journal of Manufacturing Processes，2024，113：291-306.

[51] 唐建朝，王丹，田锦娟，等. 基于机器学习的涡轮叶片荧光渗透检测图像缺陷识别[J].信息与电脑，2022，34（10）：90-93.

[52] Tang J C，Wang D，Tian J J，et al. Defect recognition for fluorescent penetrant inspection of turbine blade image based on machine learning[J].China Computer & Communication，2022，34（10）：90-93.

[53] Zhao Q C，Kang J W，Wu K. Study on the recognition of metallurgical graphs based on deep learning[J].Metals，2024，14（6）：732.

[54] Wang D H，Xiao H，Huang S Q. Automatic defect recognition and localization for aeroengine turbine blades based on deep learning[J].Aerospace，2023，10（2）：178.

[55] Wu B，Zhou J X，Yang H Q，et al. An ameliorated deep dense convolutional neural network for accurate recognition of casting defects in X—ray images[J].Knowledge—Based Systems，2021，226：107096.

[56] Richter A R，Scholz F，Eggeler G，et al. Microstructure informatics：using computer vision for the characterization of dendrite growth phenomena in Ni—base single crystal superalloys[J].Materials Characterization，2025，223：114878.

[57] 顾国红，曹腊梅. 熔模铸造空心叶片用陶瓷型芯的发展[J].铸造技术，2002，23（2）：81-83.

[58] Gu G H，Cao L M. Development of ceramic cores for investment casting hollow blades[J].Foundry Technology，2002，23（2）：81-83.

[59] 吴笑非，李鑫，许西庆，等. 不同粒度莫来石粉改性硅基陶瓷型芯的制备及性能[J].航空材料学报，2021，41（4）：128-133.

[60] Wu X F，Li X，Xu X Q，et al. Fabrication and properties of silica—based ceramic cores modified by mullite powders with different particle sizes[J].Journal of Aeronautical Materials，2021，41（4）：128-133.

[61] Zhu Q，Xie Y K，Xuan W D，et al. Prediction of cross—material properties of mineralized ceramic core of aluminosilicate system based on deep learning[J].Journal of the Ceramic Society of Japan，2025，133（4）：150-160.

[62] Li X，Su H J，Dong D，et al. Insight into the limitation of the Beer—Lambert law in stereolithography ceramic 3D printing：a study on irradiance parameters and slurry curing[J].Virtual and Physical Prototyping，2025，20（1）：e2544761.

[63] Huang Q Q，Chen C Y，Yin Y H，et al. Influence of final sintering temperature on properties of nano—ZrO₂ reinforced SiO₂—based ceramic cores via stereolithography additive manufacturing[J].China Foundry，2025，22（5）：519-533.

[64] Li X，Su H J，Dong D，et al. Strategic design and functional evaluation of oligomer components in vat photopolymerization 3D printing of alumina ceramics[J].Journal of the European Ceramic Society，2025，45（16）：117676.

[65] Nohut S，Schwentenwein M. Machine learning assisted material development for lithography—based additive manufacturing of porous alumina ceramics[J].Open Ceramics，2024，18：100573.

[66] Zhang Z L，Yang Z T，Sisson R D，et al. Improving ceramic additive manufacturing via machine learning—enabled closed—loop control[J].International Journal of Applied Ceramic Technology，2022，19（2）：957-967.

[67] Tarak F，Okoruwa L，Ozkan B，et al. AI for AM in zirconia ceramic vat photopolymerization：a machine learning approach to stabilising slurry suspension and maximizing dimensional accuracy of the 3D print process[J].Additive Manufacturing，2025，109：104893.

[68] 康海峰，李飞，赵彦杰，等. 高温合金空心叶片精密铸造用陶瓷型芯与型壳的研究现状[J].材料工程，2013（8）：85-91.

[69] Kang H F，Li F，Zhao Y J，et al. Research status on ceramic cores and shells for superalloy hollow blades investment casting[J].Journal of Materials Engineering，2013（8）：85-91.

[70] 闫学伟，毛爽，杨其航，等. 钛合金熔模铸造陶瓷型壳焙烧过程温度与变形数值模拟[J].航空材料学报，2025，45（5）：161-170.

[71] Yan X W，Mao S，Yang Q H，et al. Numerical simulation of temperature and deformation during sintering process of ceramic shell for titanium alloy investment casting[J].Journal of Aeronautical Materials，2025，45（5）：161-170.

[72] Sun Z H，Hu N Y，Ke L H，et al. Machine learning—assisted design of Al₂O₃—SiO₂ porous ceramics based on few—shot datasets[J].Ceramics International，2023，49（18）：29400-29408.

[73] Chen J C，Guo G J，Wang W N. Artificial neural network—based online defect detection system with in—mold temperature and pressure sensors for high precision injection molding[J].The International Journal of Advanced Manufacturing Technology，2020，110：2023-2033.

[74] Wang D H，He B，Li F，et al. Cavity pressure and dimensional accuracy analysis of wax patterns for investment casting[J].Materials and Manufacturing Processes，2013，28（6）：637-642.

[75] 张健，王莉，王栋，等. 镍基单晶高温合金的研发进展[J].金属学报，2019，55（9）：1077-1094.

[76] Zhang J，Wang L，Wang D，et al. Recent progress in research and development of nickel—based single crystal superalloys[J].Acta Metallurgica Sinica，2019，55（9）：1077-1094.

[77] 吴雨萌，佟海滨，丁方政，等. 镍基高温合金定向凝固过程数值模拟研究进展[J].航空材料学报，2025，45（5）：44-60.

[78] Wu Y M，Tong H B，Ding F Z，et al. Research progress on numerical simulation of directional solidification process for nickel—based superalloy[J].Journal of Aeronautical Materials，2025，45（5）：44-60.

[79] 李金国，孟祥斌，刘纪德，等. 单晶高温合金涡轮叶片的常见凝固缺陷及控制方法[J].特种铸造及有色合金，2021，41（11）：1321-1327.

[80] Li J G，Meng X B，Liu J D，et al. Common solidification defects and inhibition methods in single crystal superalloy turbine blades[J].Special Casting & Nonferrous Alloys，2021，41（11）：1321-1327.

[81] 贾玉亮，张勇佳，史泽楷，等. 高温合金定向凝固雀斑形成机理及控制技术研究进展[J].金属学报，2026，62（2）：309-327.

[82] Jia Y L，Zhang Y J，Shi Z K，et al. Review of the formation mechanism and control technology for freckle defects in directionally solidified superalloys[J].Acta Metallurgica Sinica，2026，62（2）：309-327.

[83] Xiong Y T，Liu Z Q，Li Y，et al. Control and optimization of stray grain defect during LMC directional solidification process of single—crystal blade[J].Journal of Alloys and Compounds，2025，1021：179742.

[84] Liu Y，Wang F，Ma D X，et al. Freckle prediction model incorporating geometrical effects for Ni—based single—crystal superalloy components[J].Acta Materialia，2024，266：119702.

[85] Zhao L，Liu Z Q，Zhang Y Q，et al. Effect of pouring and preheating temperature on shrinkage cavity and porosity defects of nickel—based cast superalloy[J].Metallurgical and Materials Transactions B，2025，56（2）：1316-1333.

[86] 王祥，张莹，隋大山，等. 熔模精铸 k4169 合金复杂薄壁件缺陷预测与实验研究[J].材料工程，2025，53（1）：131-142.

[87] Wang X，Zhang Y，Sui D S，et al. Defect prediction and experimental study on investment casting of K4169 alloy complex thin—walled parts[J].Journal of Materials Engineering，2025，53（1）：131-142.

[88] Yu J P，Wang D H，Li D Y，et al. Engineering computing and data—driven for gating system design in investment casting[J].The International Journal of Advanced Manufacturing Technology，2020，111：829-837.

[89] Warriner W E，Monroe C A. Locating solidification hot spots and feeder positions in casting geometries by image analysis[J].International Journal of Metalcasting，2018，12（2）：224-234.

[90] 李桌汉，有移亮，赵子华，等. 人工智能技术在失效分析领域的应用[J].航空材料学报，2024，44（5）：1-16.

[91] Li Z H，You Y L，Zhao Z H，et al. Application of artificial intelligence technology in failure analysis[J].Journal of Aeronautical Materials，2024，44（5）：1-16.

[92] Del Vecchio C，Fenu G，Pellegrino F A，et al. Support vector representation machine for superalloy investment casting optimization[J].Applied Mathematical Modelling，2019，72：324-336.

[93] 杜晓明，陆瑶，刘纪德. 基于机器学习的镍基单晶高温合金蠕变寿命预测模型研究[J].沈阳理工大学学报，2025，44（1）：44-50.

[94] Du X M，Lu Y，Liu J D. Research on prediction model of creep life of nickel—based single crystal superalloys based on machine learning[J].Journal of Shenyang Ligong University，2025，44（1）：44-50.

[95] 李春雨，刘纪德，张文颖，等. 基于人工神经网络的镍基单晶高温合金蠕变寿命预测模型及应用[J].铸造，2025，74（7）：905-914.

[96] Li C Y，Liu J D，Zhang W Y，et al. Creep life prediction model of nickel—based single crystal superalloy based on artificial neural network and its application[J].Foundry，2025，74（7）：905-914.

[97] Virdi J S，Peng W，Sata A. Feature selection with LASSO and VSURF to model mechanical properties for investment casting[C]//2019 2nd International Conference on Computational Intelligence in Data Science （ICCIDS）. Piscataway：IEEE，2019.

[98] Sata A，Ravi B. Mechanical property prediction of investment castings using artificial neural network and multivariate regression analysis[C]//63rd Indian Foundry Congress. Greater Noida：Institute of Indian Foundrymen，2015.

[99] Ding Z Y，Wang D H，Guan B，et al. Semi—solid plasticity and deformation control of superalloy investment casting[J].The International Journal of Advanced Manufacturing Technology，2024，130：3419-3429.

[100] Tian G L，Bu K，Zhao D Q，et al. A shrinkage prediction method of investment casting based on geometric parameters[J].The International Journal of Advanced Manufacturing Technology，2018，96（1/2/3/4）：1035-1044.

[101] 何玄，宋鹏，孔德昊，等. 机器学习技术在航空材料领域的应用[J].航空材料学报，2025，45（3）：85-104.

[102] He X，Song P，Kong D H，et al. Application of machine learning technology in field of aviation materials[J].Journal of Aeronautical Materials，2025，45（3）：85-104.

[103] 鹿星航，夏增龙，张海钲，等. 基于结构光的涡轮叶片在机三维测量技术研究[J].燃气涡轮试验与研究，2026，39（3）：112-122.

[104] Lu X H，Xia Z L，Zhang H Z，et al. Research of on—machine three—dimensional measurement technology for turbine blades based on structured light[J].Gas Turbine Experiment and Research，2026，39（3）：112-122.

[105] Li X B，Wang W Q，Sun L H，et al. Deep learning—based defects detection of certain aero—engine blades and vanes with DDSC—YOLOv5s[J].Scientific Reports，2022，12：13067.

[106] Yang X M，Song K C，Liu S N，et al. An edge—guided defect segmentation network for in—service aerospace engine blades[J].Engineering Applications of Artificial Intelligence，2025，154：110974.

[107] Oh S，Cha J，Kim D，et al. Quality inspection of casting product using CAE and CNN[C]//2020 4th International Conference on Imaging，Signal Processing and Communications （ICISPC）. Piscataway：IEEE，2020：34-38.

[108] Gupta R，Anand V，Gupta S，et al. Deep learning model for defect analysis in industry using casting images[J].Expert Systems with Applications，2023，232：120758.

[109] Thalagala S，Walgampaya C. Application of AlexNet convolutional neural network architecture—based transfer learning for automated recognition of casting surface defects[C]//2021 International Research Conference on Smart Computing and Systems Engineering （SCSE）. Piscataway：IEEE，2021：129-136.

[110] Błachnio J，Chalimoniuk M，Kułaszka A，et al. Exemplification of detecting gas turbine blade structure defects using the X—ray computed tomography method[J].Aerospace，2021，8（4）：119.

[111] García Pérez A，Gómez Silva M J，De La Escalera Hueso A. Automated defect recognition of castings defects using neural networks[J].Journal of Nondestructive Evaluation，2022，41：11.

[112] Wang D H，Xiao H，Wu D Y. Application of unsupervised adversarial learning in radiographic testing of aero—engine turbine blades[J].NDT & E International，2023，134：102766.

[113] Jiang L L，Wang Y X，Tang Z H，et al. Casting defect detection in X—ray images using convolutional neural networks and attention—guided data augmentation[J].Measurement，2021，170：108736.

[114] Shen C G，Wang C C，Huang M H，et al. A generic high—throughput microstructure classification and quantification method for regular SEM images of complex steel microstructures combining EBSD labeling and deep learning[J].Journal of Materials Science & Technology，2021，93：191-204.

[115] Biswas M，Pramanik R，Sen S，et al. Microstructural segmentation using a union of attention guided U—Net models with different color transformed images[J].Scientific Reports，2023，13：5722.

[116] Akbari M，Wang L，Wang Q G. Material characterization of aluminum castings using machine learning techniques[J].International Journal of Metalcasting，2026，20（2）：680-685.

[117] Bárcena M，Lloret Iglesias L，Ferreño D，et al. Classification of cast iron alloys through convolutional neural networks applied on optical microscopy images[J].Steel Research International，2024，95（12）：2400120.

基金资助

新材料重大专项(2024ZD0600800)