Smart Structures and Systems

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

lp-norm regularization for impact force identification from highly incomplete measurements

  • Yanan Wang (National Key Lab of Aerospace Power System and Plasma Technology, Xi'an Jiaotong University) ;
  • Baijie Qiao (National Key Lab of Aerospace Power System and Plasma Technology, Xi'an Jiaotong University) ;
  • Jinxin Liu (National Key Lab of Aerospace Power System and Plasma Technology, Xi'an Jiaotong University) ;
  • Junjiang Liu (National Key Lab of Aerospace Power System and Plasma Technology, Xi'an Jiaotong University) ;
  • Xuefeng Chen (National Key Lab of Aerospace Power System and Plasma Technology, Xi'an Jiaotong University)
  • Received : 2022.06.03
  • Accepted : 2024.09.23
  • Published : 2024.08.25
⟨ Previous Next ⟩

Abstract

The standard l1-norm regularization is recently introduced for impact force identification, but generally underestimates the peak force. Compared to l1-norm regularization, lp-norm (0 ≤ p < 1) regularization, with a nonconvex penalty function, has some promising properties such as enforcing sparsity. In the framework of sparse regularization, if the desired solution is sparse in the time domain or other domains, the under-determined problem with fewer measurements than candidate excitations may obtain the unique solution, i.e., the sparsest solution. Considering the joint sparse structure of impact force in temporal and spatial domains, we propose a general lp-norm (0 ≤ p < 1) regularization methodology for simultaneous identification of the impact location and force time-history from highly incomplete measurements. Firstly, a nonconvex optimization model based on lp-norm penalty is developed for regularizing the highly under-determined problem of impact force identification. Secondly, an iteratively reweighed l1-norm algorithm is introduced to solve such an under-determined and unconditioned regularization model through transforming it into a series of l1-norm regularization problems. Finally, numerical simulation and experimental validation including single-source and two-source cases of impact force identification are conducted on plate structures to evaluate the performance of lp-norm (0 ≤ p < 1) regularization. Both numerical and experimental results demonstrate that the proposed lp-norm regularization method, merely using a single accelerometer, can locate the actual impacts from nine fixed candidate sources and simultaneously reconstruct the impact force time-history; compared to the state-of-the-art l1-norm regularization, lp-norm (0 ≤ p < 1) regularization procures sufficiently sparse and more accurate estimates; although the peak relative error of the identified impact force using lp-norm regularization has a decreasing tendency as p is approaching 0, the results of lp-norm regularization with 0 ≤ p ≤ 1/2 have no significant differences.

Keywords

Acknowledgement

This work is supported by National Natural Science Foundation of China (Nos. 52305127 & 52075414), and China Postdoctoral Science Foundation (No.2021M702595).

References

  1. Candes, E.J., Wakin, M.B. and Boyd, S.P. (2008), "Enhancing sparsity by reweighted ℓ1 minimization", J. Fourier Anal. Appl., 14(5), 877-905. https://doi.org/10.1007/s00041-008-9045-x
  2. Chartrand, R. (2007), "Exact reconstruction of sparse sig-nals via nonconvex minimization", IEEE Signal Proc. Let., 14(10), 707-710. https://doi.org/10.1109/LSP.2007.898300
  3. Chartrand, R. and Staneva, V. (2008), "Restricted isometry properties and nonconvex compressive sensing", Inverse Probl., 24(3), 035020. https://doi.org/10.1088/0266-5611/24/3/035020
  4. Chen, Y., Joffre, D. and Avitabile, P. (2018), "Underwater dynamic response at limited points expanded to full-field strain response", J. Vib. Acoust., 140(5), 051016. https://doi.org/10.1115/1.4039800
  5. Donoho, D.L. (2006), "For most large underdetermined sys-tems of linear equations the minimal ℓ1-norm solution is also the sparsest solution", Commun. Pur. Appl. Math., 59(6), 797-829. https://doi.org/10.1002/cpa.20132
  6. Foucart, S. and Lai, M.J. (2009), "Sparsest solutions of underdetermined linear systems via ℓq-minimization for 0 < q⩽ 1", Appl. Comput. Harmon. A., 26(3), 395-407. https://doi.org/10.1016/j.acha.2008.09.001
  7. Inoue, H., Harrigan, J.J. and Reid, S.R. (2001), "Review of inverse analysis for indirect measurement of impact force", Appl. Mech. Rev., 54(6), 503-524. https://doi.org/10.1115/1.1420194
  8. Jacquelin, E., Bennani, A. and Hamelin, P. (2003), "Force reconstruction: analysis and regularization of a decon-volution problem", J. Sound Vib., 265(1), 81-107. https://doi.org/10.1016/S0022-460X(02)01441-4
  9. Kalhori, H., Alamdari, M.M. and Ye, L. (2018), "Automated algorithm for impact force identification using cosine similarity searching", Measurement, 122, 648-657. https://doi.org/10.1016/j.measurement.2018.01.016
  10. Khoo, S.Y., Ismail, Z., Kong, K.K., Ong, Z.C., Noroozi, S., Chong, W.T. and Rahman, A.G.A. (2014), "Impact force identification with pseudo-inverse method on a light-weight structure for under-determined, even-determined and overdetermined cases", Int. J. Impact Eng., 63, 52-62. https://doi.org/10.1016/j.ijimpeng.2013.08.005
  11. Koh, K., Kim, S.J. and Boyd, S. (2007), "An interior-point method for large-scale l1-regularized logistic regression", J. Mach. Learn. Res., 8, 1519-1555.
  12. Lai, M. and Wang, J. (2011), "An Unconstrained lq Minimization with 0 < q ≤ 1 for Sparse Solution of Underdeter-mined Linear Systems", SIAM J. Optimiz., 21(1), 82-101.
  13. Li, Q. and Lu, Q. (2018), "A hierarchical Bayesian method for vibration-based time domain force reconstruction problems", J. Sound Vib., 421, 190-204. https://doi.org/10.1016/j.jsv.2018.01.052
  14. Liu, J., Sun, X., Han, X., Jiang, C. and Yu, D. (2014), "A novel computational inverse technique for load identifi-cation using the shape function method of moving least square fitting", Comput. Struct., 144, 127-137. https://doi.org/10.1016/j.compstruc.2014.08.002
  15. Liu, J., Qiao, B., He, W., Yang, Z. and Chen, X. (2020), "Impact force identification via sparse regularization with generalized minimax-concave penalty", J. Sound Vib., 484, 115530. https://doi.org/10.1016/j.jsv.2020.115530
  16. Liu, J., Qiao, B., Chen, Y., Zhu, Y., He, W. and Chen, X. (2022), "Impact force reconstruction and localization using nonconvex overlapping group sparsity", Mech. Syst. Signal Process., 162, 107983. https://doi.org/10.1016/j.ymssp.2021.107983
  17. Liu, J., Qiao, B., Wang, Y., He, W. and Chen, X. (2023), "Nonconvex sparse regularization via convex optimiza-tion for impact force identification", Mech. Syst. Signal Prcess., 191, 110191. https://doi.org/10.1016/j.ymssp.2023.110191
  18. Mahzan, S., Staszewski, W.J. and Worden, K. (2010), "Experimental studies on impact damage location in compo-site aerospace structures using genetic algorithms and neural networks", Smart Struct. Syst., Int. J., 6(2), 147-165. https://doi.org/10.12989/sss.2010.6.2.147
  19. Martin, M.T. and Doyle, J.F. (1996), "Impact force identification from wave propagation responses", Int. J. Impact Eng., 18(1), 65-77. https://doi.org/10.1016/0734-743X(95)00022-4
  20. Pan, C.D., Yu, L., Liu, H.L., Chen, Z.P. and Luo, W.F. (2018), "Moving force identification based on redundant concatenated dictionary and weighted l1-norm regularization", Mech. Syst. Signal P rocess., 98, 32-49. https://doi.org/10.1016/j.ymssp.2017.04.032
  21. Park, J., Ha, S.F. and Chang, K. (2009), "Monitoring impact events using a system-identification method", AIAA J., 47(9), 2011-2021. https://doi.org/10.2514/1.34895
  22. Qiao, B., Zhang, X., Gao, J., Liu, R. and Chen, X. (2016a), "Impact-force sparse reconstruction from highly incomplete and inaccurate measurements", J. Sound Vib., 376, 72-94. https://doi.org/10.1016/j.jsv.2016.04.040
  23. Qiao, B., Zhang, X., Wang, C., Zhang, H. and Chen, X. (2016b), "Sparse regularization for force identification using dictionaries", J. Sound Vib., 368, 71-86. https://doi.org/10.1016/j.jsv.2016.01.030
  24. Qiao, B., Zhang, X., Gao, J., Liu, R. and Chen, X. (2017), "Sparse deconvolution for the large-scale ill-posed inverse problem of impact force reconstruction", Mech. Syst. Signal Process., 83, 93-115. https://doi.org/10.1016/j.ymssp.2016.05.046
  25. Qiao, B., Liu, J., Liu, J., Yang, Z. and Chen, X. (2019), "An enhanced sparse regularization method for impact force identification", Mech. Syst. Signal Process., 126, 341-367. https://doi.org/10.1016/j.ymssp.2019.02.039
  26. Qiao, B., Ao, C., Mao, Z. and Chen, X. (2020a), "Non-convex sparse regularization for impact force identification", J. Sound Vib., 477, 115311. https://doi.org/10.1016/j.jsv.2020.115311
  27. Qiao, B., Mao, Z., Sun, H., Chen, S. and Chen, X. (2020b), "Sparse reconstruction of guided wavefield from limited measurements using compressed sensing", Smart Struct. Syst., Int. J., 25(3), 369-384. https://doi.org/10.12989/sss.2020.25.3.369
  28. Qiu, L. and Yuan, S. (2011), "A phase synthesis time reversal impact imaging method for on-line composite structure monitoring", Smart Struct. Syst., Int. J., 8(3), 303-320. https://doi.org/10.12989/sss.2011.8.3.303
  29. Saleem, M.M. and Jo, H. (2019), "Impact force localization for civil infrastructure using augmented Kalman Filter optimization", Smart Struct. Syst., Int. J., 23(2), 123-139. https://doi.org/10.12989/sss.2019.23.2.123
  30. Selesnick, I.W. and Bayram, I. (2014), "Sparse signal estimation by maximally sparse convex optimization", IEEE Trans. Signal Process, 62(5), 1078-1092. https://doi.org/10.1109/TSP.2014.2298839
  31. Thite, A.N. and Thompson, D.J. (2003), "The quantification of structure-borne transmission paths by inverse methods. Part 2: Use of regularization techniques", J. Sound Vib., 264(2), 433-451. https://doi.org/10.1016/S0022-460X(02)01203-8
  32. Wambacq, J., Maes, K., Rezayat, A., Guillaume, P. and Lombaert, G. (2019), "Localization of dynamic forces on structures with an interior point method using group sparsity", Mech. Syst. Signal Process., 115, 593-606. https://doi.org/10.1016/j.ymssp.2018.06.006
  33. Wipf, D. and Nagarajan, S. (2010), "Iterative Reweighted l1 and l2 methods for finding sparse solutions", IEEE J-STSP, 4(2), 317-329.
  34. Xu, Z., Chang, X. Xu, F. and Zhang, H. (2012), "L1/2 regularization: A thresholding representation theory and a fast solver", IEEE T. Neur. Net. Lear., 23(7), 1013-1027. https://doi.org/10.1109/TNNLS.2012.2197412
  35. Yan, G., Sun, H. and Buyukozturk, O. (2017), "Impact load identification for composite structures using Bayesian regularization and unscented Kalman filter", Struct. Control Health Monitor., 24(5), e1910. https://doi.org/10.1002/stc.1910
  36. Zhong, Y. and Xiang, J. (2016), "A two-dimensional plum-blossom sensor array-based multiple signal classi-fication method for impact localization in composite structures", Comput.-Aided Civil Infrastr. Eng., 31(8), 633-643. https://doi.org/10.1111/mice.12198
  37. Zhong, Y. and Xiang, J. (2019), "Impact location on a stiffened composite panel using improved linear array", Smart Struct. Syst., Int. J., 24(2), 173-182. https://doi.org/10.12989/sss.2019.24.2.173
  38. Zhou, R., Wang, Y., Qiao, B., Zhu, W., Liu, J. and Chen, X. (2024), "Impact force identification on composite panels using fully overlapping group sparsity based on Lp-norm regularization", Struct Health Monitor., 23(1), 137-161. https://doi.org/10.1177/1475921723116570