zhpsЖурнал прикладной спектроскопииZhurnal Prikladnoii Spektroskopii0514-7506B. I. Stepanov Institute of Physics of the National Academy of Scienceszhps-1874Research ArticleАННОТАЦИИ АНГЛОЯЗЫЧНЫХ СТАТЕЙABSTRACTS ENGLISH-LANGUAGE ARTICLESБыстрое выявление дефектов в металлических компонентах аддитивного производства с помощью лазерно-искровой эмиссионной спектроскопии в сочетании с нейронной сетью обратного распространения и алгоритмом случайного лесаRapid Identification of Defects in Metal Additive Manufacturing Components by LIBS Combined with BP-Neural Network and Random Forest MethodGaoS.GaoS.

Цюаньчжоу, Фуцзянь

Shanping Gao

Quanzhou, Fujian

LinX.LinX.

Цюаньчжоу, Фуцзянь; Чанчунь, Цзилинь

Xiaomei Lin

Quanzhou, Fujian; Changchun, Jilin

linxiaomei4427@163.comHuangY.HuangY.

Цюаньчжоу, Фуцзянь

Yixiang Huang

Quanzhou, Fujian

ChenZ.ChenZ.

Цюаньчжоу, Фуцзянь

Zongxu Chen

Quanzhou, Fujian

ChenH.ChenH.

Цюаньчжоу, Фуцзянь

Huijin Chen

Quanzhou, Fujian

Университет информационной инженерии ЦюаньчжоуQuanzhou University of Information EngineeringУниверситет информационной инженерии Цюаньчжоу; Чанчуньский технологический университетQuanzhou University of Information Engineering; Changchun University of Technology202520052025923414414Copyright © Gao S., Lin X., Huang Y., Chen Z., Chen H., 20252025Gao S., Lin X., Huang Y., Chen Z., Chen H.Gao S., Lin X., Huang Y., Chen Z., Chen H.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://zhps.ejournal.by/jour/article/view/1874

Лазерно-искровая эмиссионная спектроскопия (LIBS) использована для быстрой идентификации дефектов в металлических компонентах аддитивного производства и контрольных групп без дефектов. По характеристическим эмиссионным линиям четырех элементов (Fe, Cr, Mn и Ti), выбранным из базы данных методом построения “случайного леса” (RF) классификационных деревьев, проведена классификация дефектов и бездефектных образцов методом k-ближайших соседей (KNN) и нейронной сетью с обратным распространением ошибки (BPNN). Сравнение эффективности классификации и результатов верификации моделей KNN, RF-KNN и RF-BPNN показало преимущество модели RF-BPNN, которая продемонстрировала в обучающем, тестовом и проверочном наборах наилучшую точность (99.4, 97.2 и 96.67% соответственно). Показано, что LIBS в сочетании с RF-BPNN можно использовать для обнаружения дефектов при 3D-печати металлических изделий.

Rapid detection of defects in metal additive manufacturing (AM) components remains a challenge. In this paper, laser-induced breakdown spectroscopy (LIBS) technology was used to establish a rapid identification of metal AM defects and defect-free control groups. The corresponding spectral acquisition of metal AM components with defects and without defects was carried out, and the research elements (Fe, Cr, Mn, and Ti) and corresponding spectral lines were obtained in combination with the NIST database. The spectral lines with features of importance greater than the average value are selected by random forest (RF). The selected spectral lines were used as the input variables of the k-nearest neighbor (KNN) model and the backpropagation neural network (BPNN) model. The classification performance and verification results of KNN, RF-KNN, and RF-BPNN models were compared. The results showed that the RF-BPNN model exhibited the best accuracy, sensitivity, and specificity in the training set, test set and validation set, with accuracies of 99.4, 97.2, and 96.67%, respectively. This indicates that LIBS combined with RF-BPNN can be used for the detection of defects in metal AM components.

лазерно-искровая эмиссионная спектроскопияаддитивное производствобыстрая идентификацияlaser-induced breakdown spectroscopyadditive manufacturingrapid identificationThis paper was supported by the National Natural Science Foundation of China (NSFC: 62405029), the Department of Science and Technology of Jilin Province of China (grant No. 20240302043GX), and the Department of Science and Technology of Jilin Province of China (grant No. YDZJ202401296ZYTS)This paper was supported by the National Natural Science Foundation of China (NSFC: 62405029), the Department of Science and Technology of Jilin Province of China (grant No. 20240302043GX), and the Department of Science and Technology of Jilin Province of China (grant No. YDZJ202401296ZYTS)ReferencesF. Calignano, D. Manfredi, E. P. Ambrosio, et al., Proc. IEEE, 105, No. 4, 593–612 (2017).F. Calignano, D. Manfredi, E. P. Ambrosio, et al., Proc. IEEE, 105, No. 4, 593–612 (2017).A. N. Guojin, Modern Machinery, Issue 3 (2019).A. N. Guojin, Modern Machinery, Issue 3 (2019).M. Dadkhah, M. H. Mosallanejad, L. Luliano, et al., Acta Metall. Sin. (Engl. Lett.), 34, 1173–1200 (2021), doi: 10.1007/s40195-021-01249-7.M. Dadkhah, M. H. Mosallanejad, L. Luliano, et al., Acta Metall. Sin. (Engl. Lett.), 34, 1173–1200 (2021), doi: 10.1007/s40195-021-01249-7.D. Gu, D. Dai, M Xia, et al., J. Nanjing University of Aeronautics & Astronautics, 49, No. 5, 645–652 (2017).D. Gu, D. Dai, M Xia, et al., J. Nanjing University of Aeronautics & Astronautics, 49, No. 5, 645–652 (2017).H. Masuo, Y. Tanaka, S. Morokoshi, et al., Proc. Struct. Integrity, 7, 19–26 (2017), doi: 10.1016/j.prostr.2017.11.055.H. Masuo, Y. Tanaka, S. Morokoshi, et al., Proc. Struct. Integrity, 7, 19–26 (2017), doi: 10.1016/j.prostr.2017.11.055.D. S. Ertay, M. A. Naiel, M. Vlasea, et al., CIRP J. Manufacturing Science and Technology, 35, No. 2, 298–314 (2021), doi: 10.1016/j.cirpj.2021.06.015.D. S. Ertay, M. A. Naiel, M. Vlasea, et al., CIRP J. Manufacturing Science and Technology, 35, No. 2, 298–314 (2021), doi: 10.1016/j.cirpj.2021.06.015.J. Lee, H. J. Park, S. Chai, et al., Appl. Sciences, 11, No. 4, 1966 (2021), doi: 10.3390/app11041966.J. Lee, H. J. Park, S. Chai, et al., Appl. Sciences, 11, No. 4, 1966 (2021), doi: 10.3390/app11041966.Z. Guo, P. Ni, Y. Dai, et al., J. Phys.: Conf. Ser., 1827, No. 1, 012039(1–7) (2021).Z. Guo, P. Ni, Y. Dai, et al., J. Phys.: Conf. Ser., 1827, No. 1, 012039(1–7) (2021).W. Du, Q. Bai, Y. Wang, et al., Int. J. Adv. Manufacturing Technology, 95, 3185–3195 (2018).W. Du, Q. Bai, Y. Wang, et al., Int. J. Adv. Manufacturing Technology, 95, 3185–3195 (2018).H. Rieder, A. Dillhöfer, M. Spies, et al., Proc. 11th Eur. Conf. Non-Destructive Testing, 1, 2194–2201 (2014).H. Rieder, A. Dillhöfer, M. Spies, et al., Proc. 11th Eur. Conf. Non-Destructive Testing, 1, 2194–2201 (2014).D. W. Yee, M. L. Lifson, B. W. Edwards, et al., Adv. Materials, 31, No. 33, 1901345 (2019), doi: 10.1002/adma.201901345.D. W. Yee, M. L. Lifson, B. W. Edwards, et al., Adv. Materials, 31, No. 33, 1901345 (2019), doi: 10.1002/adma.201901345.R. Hama-Saleh, A. Weisheit, J. H. Schleifenbaum, et al., Proc. Manufacturing, 47, 1023–1028 (2020), doi: 10.1016/j.promfg.2020.04.317.R. Hama-Saleh, A. Weisheit, J. H. Schleifenbaum, et al., Proc. Manufacturing, 47, 1023–1028 (2020), doi: 10.1016/j.promfg.2020.04.317.S. Doshvarpassand, X. Wang, X. Zhao, Struct. Health Monitoring, 21, No. 2, 354–369 (2022), doi: 10.1177/1475921721999599.S. Doshvarpassand, X. Wang, X. Zhao, Struct. Health Monitoring, 21, No. 2, 354–369 (2022), doi: 10.1177/1475921721999599.C. Gu, Y. Lu, M. Chen, et al., Measurement, 187 (2022), doi: 10.1016/j.measurement.2021.110166C. Gu, Y. Lu, M. Chen, et al., Measurement, 187 (2022), doi: 10.1016/j.measurement.2021.110166S. Felix, S. R. Majumder, H. K. Mathews, et al., Sci. Rep., 12, No. 1, 8503 (2022), doi: 10.1038/s41598022-12381-4.S. Felix, S. R. Majumder, H. K. Mathews, et al., Sci. Rep., 12, No. 1, 8503 (2022), doi: 10.1038/s41598022-12381-4.V. Detalle, X. Bai, Spectrochim. Acta B: At. Spectroscopy, 191, 106407 (2022), doi: 10.1016/j.sab.2022.106407.V. Detalle, X. Bai, Spectrochim. Acta B: At. Spectroscopy, 191, 106407 (2022), doi: 10.1016/j.sab.2022.106407.L. B. Guo, D. Zhang, L. X. Sun, et al., Front. Physics, 16, No. 2 (2021), doi: 10.1007/s11467-020-1007-z.L. B. Guo, D. Zhang, L. X. Sun, et al., Front. Physics, 16, No. 2 (2021), doi: 10.1007/s11467-020-1007-z.Y. Zhang, T. Zhang, H. Li, Spectrochim. Acta B: At. Spectroscopy, 181, 106218 (2021), doi: 10.1016/j.sab.2021.106218.Y. Zhang, T. Zhang, H. Li, Spectrochim. Acta B: At. Spectroscopy, 181, 106218 (2021), doi: 10.1016/j.sab.2021.106218.D. A. Gonalves, G. S. Senesi, G. Nicolodelli, Trends Environ. Anal. Chem., 30C, e00121 (2021), doi: 10.1016/j.teac.2021.e00121.D. A. Gonalves, G. S. Senesi, G. Nicolodelli, Trends Environ. Anal. Chem., 30C, e00121 (2021), doi: 10.1016/j.teac.2021.e00121.K. Zhang, Z. Xu, F. Fang, Spectroscopy and Spectral Analysis, 41, No. 06 (2021), doi: 10.3964/j.issn.1000-0593(2021)06-1961-05.K. Zhang, Z. Xu, F. Fang, Spectroscopy and Spectral Analysis, 41, No. 06 (2021), doi: 10.3964/j.issn.1000-0593(2021)06-1961-05.H. Kim, J. Lee, E. Srivastava, et al., Spectrochim. Acta B: At. Spectroscopy, 184, No. 7, 106282 (2021), doi: 10.1016/j.sab.2021.106282.H. Kim, J. Lee, E. Srivastava, et al., Spectrochim. Acta B: At. Spectroscopy, 184, No. 7, 106282 (2021), doi: 10.1016/j.sab.2021.106282.P. A. Sdvizhenskii, V. N. Lednev, R. D. Asyutin, et al., J. Anal. At. Spectrometry, 35 (2020), doi: 10.1039/C9JA00343F.P. A. Sdvizhenskii, V. N. Lednev, R. D. Asyutin, et al., J. Anal. At. Spectrometry, 35 (2020), doi: 10.1039/C9JA00343F.V. N. Lednev, P. A. Sdvizhenskii, R. Asyutin, et al., Opt. Express, 27, No. 4, 4612–4628 (2019), doi: 10.1364/OE.27.004612.V. N. Lednev, P. A. Sdvizhenskii, R. Asyutin, et al., Opt. Express, 27, No. 4, 4612–4628 (2019), doi: 10.1364/OE.27.004612.

The authors declare that there are no conflicts of interest present.