基于特征迁移的小样本增材制造疲劳寿命预测方法

doi:10.3901/JME.2025.17.255

摘要/Abstract

摘要： 数据驱动模型因其强大的高维非线性建模能力而被广泛用于预测增材制造材料的疲劳寿命。然而，受限于增材疲劳中广泛存在的小样本工况，该类模型在实际应用中往往无法获得足量的训练数据，适用范围因此仍十分有限。针对这一问题，通过引入迁移学习，在迁移成分分析框架下开展源域数据融合及基于聚类指标的超参数优化，提出了一种小样本增材制造疲劳寿命预测模型。该模型可协同利用不同工况下的实验数据对疲劳寿命进行建模，因此能够有效弥补单一工况下数据不足和多工况下数据分布不一致等问题。基于激光粉末床熔融成形316L不锈钢实验数据，开展了模型验证。结果显示：相较于未配置迁移学习和聚类指标以及未进行源域数据融合的几类退化模型，所提模型兼具更优的预测性能和更低的数据需求，对于处理增材疲劳寿命显示出良好的应用潜力。

关键词: 增材制造, 疲劳寿命预测, 数据驱动, 迁移学习, 聚类

Abstract: Data-driven models have been widely used to predict the fatigue life of additive manufactured materials because of their powerful capability in high-dimensional nonlinear modeling. However, owing to the high requirement of data support, these models are not economically feasible for many practical cases. To tackle this issue, a transfer learning-based model for fatigue life prediction of additive manufactured materials is proposed, through combining source and target domain data and introducing a clustering-based procedure for hyper parameter optimization. The proposed model can collaboratively model fatigue life using experimental data from different manufacturing conditions, effectively addressing the insufficiency of training data under a single condition and the inconsistency of data distributions under multiple conditions. Experimental results of 316L stainless steel processed by laser powder bed fusion are collected for model verification. The results show that, compared to several degradation models, the proposed model has better prediction performance and lower data requirement, exhibiting promising potential in predicting the fatigue life of additive manufactured materials.

Key words: additive manufacturing, fatigue life prediction, data-driven model, transfer learning, clustering

中图分类号:

O346

范志明, 甘磊, 淦志强, 王谙斌, 苏永辉, 吴昊. 基于特征迁移的小样本增材制造疲劳寿命预测方法[J]. 机械工程学报, 2025, 61(17): 255-265.

FAN Zhiming, GAN Lei, GAN Zhiqiang, WANG Anbin, SU Yonghui, WU Hao. Feature-transfer Learning Based Model for Fatigue Life Prediction of Additive Manufactured Materials Using Small Samples[J]. Journal of Mechanical Engineering, 2025, 61(17): 255-265.

参考文献

[1] LU B, WANG L. Development of additive manufacturing technology and industry in China[J]. Chinese Journal of Engineering Science, 2022, 24(4): 202-211.
[2] 杨冰, 廖贞, 吴圣川, 等. 增材制造技术发展和在先进轨道交通装备中的应用展望[J]. 交通运输工程学报, 2021, 21(1): 132-153. YANG Bing, LIAO Zhen, WU Shengchuan, et al. Development of additive manufacturing technology and its application prospect in advanced rail transit equipment[J]. Journal of Traffic and Transportation Engineering, 2021, 21(1): 132-153.
[3] SHAMSAEI N, YADOLLAHI A, BIAN L, et al. An overview of direct laser deposition for additive manufacturing;Part Ⅱ: Mechanical behavior, process parameter optimization and control[J]. Additive Manufacturing, 2015, 8: 12-35.
[4] 林鑫, 黄卫东. 高性能金属构件的激光增材制造[J]. 中国科学: 信息科学, 2015, 45(9): 1111-26. LIN Xin, HUANG Weidong. Laser additive manufacturing of high-performance metal components[J]. Scientia Sinica Information, 2015, 45(9): 1111-1126.
[5] EDWARDS P, RAMULU M. Fatigue performance evaluation of selective laser melted Ti-6Al-4V[J]. Materials Science and Engineering: A, 2014, 598: 327-337.
[6] 易敏, 常珂, 梁晨光, 等. 增材制造微结构演化及疲劳分散性计算[J]. 力学学报, 2021, 53(12): 3263-3273. YI Min, CHANG Ke, LIANG Chenguang, et al. Computational study of evolution and fatigue dispersity of microstructures by additive manufacturing[J]. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(12): 3263-3273.
[7] 代俊林, 吴世品, 张宇, 等. 增材制造金属材料的疲劳性能研究进展[J]. 精密成形工程, 2024, 16(1): 1-13. DAI Junlin, WU Shipin, ZHANG Yu, et al. Research progress on fatigue properties of additive manufactured metal materials[J]. Journal of Netshape Forming Engineering, 2024, 16(1): 1-13.
[8] 吴圣川, 胡雅楠, 杨冰, 等. 增材制造材料缺陷表征及结构完整性评定方法研究综述[J]. 机械工程学报, 2021, 57(22): 3-34. WU Shengchuan, HU Yanan, YANG Bing, et al. Review on defect characterization and structural integrity assessment method of additively manufactured materials[J]. Journal of Mechanical Engineering, 2021, 57(22): 3-34.
[9] 陈泽坤, 蒋佳希, 王宇嘉, 等. 金属增材制造中的缺陷、组织形貌和成形材料力学性能[J]. 力学学报, 2021, 53(12): 3190-3205. CHEN Zekun, JIANG Jiaxi, WANG Yujia, et al. Defects, microstructures and mechanical properties of materials fabricated by metal additive manufacturing[J]. Chinese Journal of Theoretical and Applied Mechanics, 2021, 53(12): 3190-3205.
[10] 奥妮, 何子昂, 吴圣川, 等. 激光增材制造AlSi10Mg合金的力学性能研究进展[J]. 焊接学报, 2022, 43(9): 1-19. AO Ni, HE Ziang, WU Shengchuan, et al. Recent progress on the mechanical properties of laser additive manufacturing AlSi10Mg alloy[J]. Transactions of the China Welding Institution, 2022, 43(9): 1-19.
[11] FATEMI A, MOLAEI R, PHAN N. Multiaxial fatigue of additive manufactured metals: Performance, analysis, and applications[J]. International Journal of Fatigue, 2020, 134: 105479.
[12] ZHAN Z, LI H, LAM K Y. Development of a novel fatigue damage model with AM effects for life prediction of commonly-used alloys in aerospace[J]. International Journal of Mechanical Sciences, 2019, 155: 110-124.
[13] SHERIDAN L, GOCKEL J E, SCOTT-EMUAKPOR O E. Stress-defect-life interactions of fatigued additively manufactured alloy 718[J]. International Journal of Fatigue, 2021, 143: 106033.
[14] 廉艳平, 王潘丁, 高杰, 等. 金属增材制造若干关键力学问题研究进展[J]. 力学进展, 2021, 51(3): 648-701. LIAN Yanping, WANG Panding, GAO Jie, et al. Fundamental mechanics problems in metal additive manufacturing: A state-of-art review[J]. Advances in Mechanics, 2021, 51(3): 648-701.
[15] CUTOLO A, LAMMENS N, VANDEN BOER K, et al. Fatigue life prediction of a L-PBF component in Ti-6Al-4V using sample data, FE-based simulations and machine learning[J]. International Journal of Fatigue, 2022, 167: 107276.
[16] BAO H, WU S, WU Z, et al. A machine-learning fatigue life prediction approach of additively manufactured metals[J]. Engineering Fracture Mechanics, 2021, 242: 107508.
[17] 詹志新, 高同州, 刘传奇, 等. 基于数据驱动的增材制造铝合金的疲劳寿命预测[J]. 固体力学学报, 2023, 44(3): 381-394. ZHAN Zhixin, GAO Tongzhou, LIU Chuanqi, et al. Data-driven fatigue life prediction of additive manufacturing aluminum alloys[J]. Chinese Journal of Solid Mechanics, 2023, 44(3): 381-394.
[18] ZHANG M, SUN C N, ZHANG X, et al. High cycle fatigue life prediction of laser additive manufactured stainless steel: A machine learning approach[J]. International Journal of Fatigue, 2019, 128: 105194.
[19] PAN S J, YANG Q. A survey on transfer learning[J]. IEEE Transactions on Knowledge and Data Engineering, 2009, 22(10): 1345-1359.
[20] SUN X, LI J, ZENG B, et al. Small-sample day-ahead power load forecasting of integrated energy system based on feature transfer learning[J]. Control Theory & Applications, 2021, 38(1): 63-72.
[21] 曹宁, 江志农, 高金吉. 小样本下基于迁移学习的轴承状态识别方法[J]. 噪声与振动控制, 2020, 40(5): 89-94, 132. CAO Ning, JIANG Zhinong, GAO Jinji. Rolling bearing state recognition based on transfer learning under small samples[J]. Noise and Vibration Control, 2020, 40(5): 89-94, 132.
[22] 高勋章, 刘迪阳, 杨宜. 基于特征迁移和原型网络的小样本雷达像识别[J]. 现代雷达, 2022, 44(4): 50-55. GAO Xunzhang, LIU Diyang, YANG Yi. Radar image recognition with few samples based on feature transfer and prototypical network[J]. Modern Radar, 2022, 44(4): 50-55.
[23] CHENG Longwei, WANG Kai, TSUNG Fugee. A hybrid transfer learning framework for in-plane freeform shape accuracy control in additive manufacturing[J], ⅡSE Trans, 2020, 53(3): 298-312.
[24] AKSHAY J T, MATEUSZ J, EDUARDO B, et al. Probabilistic physics-guided transfer learning for material property prediction in extrusion deposition additive manufacturing[J]. Computer Methods in Applied Mechanics and Engineering, 2024, 419(1): 116660.
[25] 李治文, 张志芬, 张帅, 等. 基于空气传播声发射与深度迁移学习的激光粉末床熔融缺陷在线监测[J]. 精密成形工程, 2023, 15(11): 76-88. LI Zhiwen, ZHANG Zhifen, ZHANG Shuai, et al. Online monitoring for laser powder bed fusion defects based on air-borne acoustic emission and deep transfer learning[J]. Journal of Netshape Forming Engineering, 2023, 15(11): 76-88.
[26] PAN S J, TSANG I W, KWOK J T, et al. Domain adaptation via transfer component analysis[J]. IEEE Transactions on Neural Networks, 2011, 22(2): 199-210.
[27] ROHANI Bastami A, BASHARI A. Rolling element bearing diagnosis using spectral kurtosis based on optimized impulse response wavelet[J]. Journal of Vibration and Control, 2019, 26(3-4): 175-185.
[28] LIU C, GRYLLIAS K. A semi-supervised support vector data description-based fault detection method for rolling element bearings based on cyclic spectral analysis[J]. Mechanical Systems and Signal Processing, 2020, 140: 106682.
[29] YANG S, ZHANG Y, ZHU Y, et al. Representation learning via serial autoencoders for domain adaptation[J]. Neurocomputing, 2019, 351: 1-9.
[30] LI X, ZHANG W, DING Q. Cross-Domain fault diagnosis of rolling element bearings using deep generative neural networks[J]. IEEE Transactions on Industrial Electronics, 2019, 66(7): 5525-5534.
[31] WEI D, HAN T, CHU F, et al. Weighted domain adaptation networks for machinery fault diagnosis[J]. Mechanical Systems and Signal Processing, 2021, 158: 107744.
[32] BORGWARDT K M, GRETTON A, RASCH M J, et al. Integrating structured biological data by Kernel maximum mean discrepancy[J]. Bioinformatics, 2006, 22(14): 49-57.
[33] CHEN S X, ZHOU L, NI Y Q. Wheel condition assessment of high-speed trains under various operational conditions using semi-supervised adversarial domain adaptation[J]. Mechanical Systems and Signal Processing, 2022, 170: 108853.
[34] LI W, XING X, FU L, et al. Application of improved grid search algorithm on SVM for classification of tumor gene[J]. International Journal of Multimedia and Ubiquitous Engineering, 2014, 9(11): 181-188.
[35] HAN P, WANG W, SHI Q, et al. A combined online-learning model with K-means clustering and GRU neural networks for trajectory prediction[J]. Ad Hoc Networks, 2021, 117: 102476.
[36] 崔佳旭, 杨博. 贝叶斯优化方法和应用综述[J]. 软件学报, 2018, 29(10): 3068-3090. CUI Jiaxu, YANG Bo. Survey on Bayesian optimization methodology and applications[J]. Journal of Software, 2018, 29(10): 3068-3090.
[37] SHI Tao, SUN Jingyu, LI Jianghua, et al. Machine learning based very-high-cycle fatigue life prediction of AlSi10Mg alloy fabricated by selective laser melting[J]. International Journal of Fatigue, 2023, 171: 107585.
[38] LEI Liming, LI Bo, WANG Haijie, et al. High-temperature high-cycle fatigue performance and machine learning-based fatigue life prediction of additively manufactured Hastelloy X[J]. International Journal of Fatigue, 2024, 178: 108012.
[39] ZHANG M, SUN C N, ZHANG X, et al. Predictive models for fatigue property of laser powder bed fusion stainless steel 316L[J]. Materials & Design, 2018, 145: 42-54.
[40] ZHANG M, SUN C N, ZHANG X, et al. Fatigue and fracture behaviour of laser powder bed fusion stainless steel 316L: Influence of processing parameters[J]. Materials Science and Engineering: A, 2017, 703: 251-261.
[41] ZHANG M, SUN C N, ZHANG X, et al. Elucidating the relations between monotonic and fatigue properties of laser powder bed fusion stainless steel 316L[J]. JOM, 2017, 70(3): 390-395.
[42] ZHANG M, SUN C N, ZHANG X, et al. High cycle fatigue and ratcheting interaction of laser powder bed fusion stainless steel 316L: Fracture behaviour and stress-based modelling[J]. International Journal of Fatigue, 2019, 121: 252-264.
[43] LIU Z, WU H. Data-driven fatigue life prediction method based on the influence of parameters[J]. Journal of Mechanical Engineering, 2023, 59(4): 71-79.
[44] ZHANG Y, YANG X, SHARDT Y A W, et al. A KPI-based probabilistic soft sensor development approach that maximizes the coefficient of determination[J]. Sensors (Basel), 2018, 18(9): 3058.
[45] LU H, MA X, MA M, et al. Energy price prediction using data-driven models: A decade review[J]. Computer Science Review, 2021, 39: 100356.