Steel and Composite Structures

Techno-Press (테크노프레스)
권호 표지
DOI QR코드

DOI QR Code

A hybrid singular value decomposition and deep belief network approach to detect damages in plates

  • Jinshang Sun (College of Mechanical and Electrical Engineering, Wenzhou University) ;
  • Qizhe Lin (College of Mechanical and Electrical Engineering, Wenzhou University) ;
  • Hu Jiang (College of Mechanical and Electrical Engineering, Wenzhou University) ;
  • Jiawei Xiang (College of Mechanical and Electrical Engineering, Wenzhou University)
  • Received : 2022.03.28
  • Accepted : 2024.06.20
  • Published : 2024.06.25
⟨ Previous Next ⟩

Abstract

Damage detection in structures using the change of modal parameters (modal shapes and natural frequencies) has achieved satisfactory results. However, as modal shapes and natural frequencies alone may not provide enough information to accurately detect damages. Therefore, a hybrid singular value decomposition and deep belief network approach is developed to effectively identify damages in aluminum plate structures. Firstly, damage locations are determined using singular value decomposition (SVD) to reveal the singularities of measured displacement modal shapes. Secondly, using experimental modal analysis (EMA) to measure the natural frequencies of damaged aluminum plates as inputs, deep belief network (DBN) is employed to search damage severities from the damage evaluation database, which are calculated using finite element method (FEM). Both simulations and experimental investigations are performed to evaluate the performance of the presented hybrid method. Several damage cases in a simply supported aluminum plate show that the presented method is effective to identify multiple damages in aluminum plates with reasonable precision.

Keywords

Acknowledgement

The authors are grateful for the support of the National Natural Science Foundation of China (No. 52375116), and the Wenzhou Major Science and Technology Innovation Project of China (No. ZG2023029).

References

  1. Adams, R.D., Cawley, P., Pye, C.J. and Stone, B.J. (1978), "A vibration technique for non-destructively assessing the integrity of structures", J. Mech. Eng. Sci., 20(2), 93-100.
  2. Arefi, S.L. and Gholizad, A. (2020), "Damage identification of structures by reduction of dynamic matrices using the modified modal strain energy method", Struct. Monit. Maint., 7(2), 125-147.
  3. Bernal, D. (2006), "Flexibility-based damage localization from stochastic realization results", J. Eng. Mech-Asce., 132(6), 651-658.
  4. Brahim, A.O., Capozucca, R., Khatir, S., Magagnini, E., Benaissa, B., Wahab, M.A. and Cuong-Le, T. (2024), "Artificial neural network and YUKI algorithm for notch depth prediction in X70 steel specimens", Theor Appl Fract Mec., 129, 104227.
  5. Benaissa, B., Hocine, N.A., Khatir, S., Riahi, M.K. and Mirjalili, S. (2021), "YUKI Algorithm and POD-RBF for Elastostatic and dynamic crack identification", J. Comput. Sci.-Neth., 55, 101451.
  6. Chen, B., Shen, B., Zhang, F., Xiao, W., Chen, F., Tian, H. and Chen, S. (2019), "Operation reliability evaluation of cutting tools based on singular value decomposition transform and support vector space", P. I. Mech. Eng. O-J. Ris., 233(2), 175-185.
  7. Chen, H., Wang, J., Tang, B., Xiao, K. and Li, J. (2017), "An integrated approach to planetary gearbox fault diagnosis using deep belief networks", Meas. Sci. Technol., 28(2), 025010.
  8. Fallah, N., Vaez. S.R.H. and Fasihi, H. (2018), "Damage identification in laminated composite plates using a new multi-step approach", Steel. Compos. Struct., 29(1), 139-149.
  9. Golub, G.H. (1970), "Singular value decomposition and least squares solutions", Numer. Math., 14(5), 403-420.
  10. He, J. and Zhou, Y. (2019), "A novel mode shape reconstruction method for damage diagnosis of cracked beam", Mech. Syst. Signal. Pr., 122, 433-447.
  11. Hinton, G.E., Osindero, S. and Teh, Y.-W. (2006), "A fast learning algorithm for deep belief nets", Neural. Comput., 18(7), 1527-1554.
  12. Huang, X., Zhang, X., Xiong, Y., Liu, H. and Zhang, Y. (2021), "A novel intelligent fault diagnosis approach for early cracks of turbine blades via improved deep belief network using three-dimensional blade tip clearance", IEEE. Access., 9, 13039-13051.
  13. Jiang, Y., Xiang, J., Li, B., Chen, X. and Lin, L. (2017), "A hybrid multiple damages detection method for plate structures", Sci. China. Technol. Sc., 60(5), 726-736.
  14. Jiang, Y., Wang, N. and Zhong, Y. (2021), "A two-step damage quantitative identification method for beam structures", Measurement., 168, 108434.
  15. Jiang, Z.S. and Xiang, J.W. (2020), "Method using XFEM and SVR to predict the fatigue life of plate-like structures", Struct. Eng. Mech., 73(4), 455-462.
  16. Kang, M. and Kim, J.-M. (2013), "Singular value decomposition based feature extraction approaches for classifying faults of induction motors", Mech. Syst. Signal. Pr., 41(1-2), 348-356.
  17. Khoshnoudian, F. and Bokaeian, V. (2017), "Damage detection in plate structures using frequency response function and 2d-pca", Smart. Struct. Syst., 20(4), 427-440.
  18. Kindova-Petrova, D.D. (2021), "A new crack size prediction method using damaged beam mode shape", Struct. Eng. Mech., 80(4), 391-399.
  19. Lee, K., Byun, N. and Shin, D.H. (2020), "A damage localization approach for rahmen bridge based on convolutional neural network", KSCE. J. Civ. Eng., 24(1), 1-9.
  20. Luo, B. and Hancock, E.R. (2001), "Structural graph matching using the em algorithm and singular value decomposition", IEEE. T. Pattern. Anal., 23(10), 1120-1136.
  21. Nanda, B., Maity, D. and Maiti, D.K. (2014), "Damage assessment from curvature mode shape using unified particle swarm optimization", Struct. Eng. Mech., 52(2), 307-322.
  22. Nobahari, M., Ghasemi, M.R. and Shabakhty, N. (2017), "Truss structure damage identification using residual force vector and genetic algorithm", Steel. Compos. Struct., 25(4), 485-496.
  23. Pan, Q., Liu, Y., Zhou, R., Wang, H., Chen, H. and He, T. (2019), "An automatic abrupt signal extraction method for fault diagnosis of aero-engines", J. Mech. Sci. Technol., 33(4), 1633-1640.
  24. Rebenciuc, M., Bibic, S.M. and Toma, A. (2021), "Assessment of structural monitoring by analyzing some modal parameters: An extended inventory of methods and developments", Arch. Comput. Method. E., 28(3), 1575-1590.
  25. Seyedpoor, S.M. and Montazer, M. (2016), "A two-stage damage detection method for truss structures using a modal residual vector based indicator and differential evolution algorithm", Smart. Struct. Syst., 17(2), 347-361.
  26. Shao, H., Jiang, H., Zhang, X. and Niu, M. (2015), "Rolling bearing fault diagnosis using an optimization deep belief network", Meas. Sci. Technol., 26(11), 115002-1-115002-17.
  27. Slimani, M., Khatir, T., Boutchicha, D., Tiachacht, S., Benaissa, B. (2022), "Experimental sensitivity analysis of sensor placement based on virtual springs and damage quantification in CFRP composite", J. Mater. Eng. Struct., 9(2), 207-220.
  28. Song, W.L., Xiang, J.W. and Zhong, Y.T. (2017), "Mechanical parameters detection in stepped shafts using the FEM simulation based impulse excitation technique", Smart. Struct. Syst., 20(4), 473-481.
  29. Su, Z., Zhang, Y., Jia, M., Xu, F. and Hu, J. (2011), "Gear fault identification and classification of singular value decomposition based on Hilbert-huang transform", J. Mech. Sci. Technol., 25(2), 267-272.
  30. Ubertini, F. (2013), "On damage detection by continuous dynamic monitoring in wind-excited suspension bridges", Meccanica., 48(5), 1031-1051.
  31. Wang, G. and Xiang, J.W. (2021), "Railway sleeper crack recognition based on edge detection and CNN", Smart. Struct. Syst., 28(6), 779-789
  32. Wang, N., Jiang, Y., Zhong, Y. and Shao, L. (2021), "An adaptive damage detection method based on differential evolutionary algorithm for beam structures", Measurement., 178, 109227.
  33. Wang, S.Q., Jiang, Y.F., Xu, M.Q., Li, Y.C. and Li, Z.X. (2020), "Structural damage identification using an iterative two-stage method combining a modal energy based index with the BAS algorithm", Steel. Compos. Struct., 36(1), 31-45.
  34. Wang, S.H., Xiang, J.W., Zhong, Y.T. and Tang, H.S (2018), "A data indicator-based deep belief networks to detect multiple faults in axial piston pumps", Mech. Syst. Signal Pr., 112, 154-170.
  35. Xiang, J.W., Chen, X.F., Li, B., He, Y.M. and He, Z.J. (2006), "Identification of crack in a beam based on finite element method of B-spline wavelet on the interval", J. Sound Vib., 296(4-5), 1046-1052.
  36. Xiang, J.W. and Liang, M. (2012a), "A two-step approach to multi-damage detection for plate structures", Eng. Fract. Mech., 91, 73-86.
  37. Xiang, J.W., Liang, M. and He, Y.M. (2014), "Experimental investigation of frequency-based multi-damage detection for beams using support vector regression", Eng. Fract. Mech., 131, 257-268.
  38. Xiang, J.W., Matsumoto, T., Wang, Y.X. and Jiang, Z.S. (2012b), "A simple method to detect cracks in beam-like structures", Smart. Struct. Syst., 9(4), 335-353.
  39. Xiang, J.W., Matsumoto, T., Wang, Y.X. and Jiang, Z.S. (2013), "Detect damages in conical shells using curvature mode shape and wavelet finite element method", Int. J. Mech. Sci., 66, 83-93.
  40. Yang, R.G., Wang, N.G. and Xiang, J.W. (2022), "No-backlash transmission characteristics analysis of a cycloidal ball planetary under axial pre-tightening", Struct. Eng. Mech., 81(4), 481-492.
  41. Yang, Z.B, Chen, X.F., Yu, J., Liu, R., Liu, Z.H. and He, Z.J. (2013), "A damage identification approach for plate structures based on frequency measurements", Nondestruct. Test. Eva., 28(4), 321-341.
  42. Yang, Z.B., Radzienski, M., Kudela, P. and Ostachowicz, W. (2017), "Fourier spectral-based modal curvature analysis and its application to damage detection in beams", Mech. Syst. Signal. Pr., 84, 763-781.
  43. Yang, Z.B., Yu, J.T., Tian, S.H., Chen, X.F. and Xu, G.J. (2018), "A damage localization method based on the singular value decomposition (SVD) for plates", Steel Compos. Struct., 22(5), 621-630.
  44. Yoon, M.K., Heider, D., Gillespie, J.W., Ratcliffe, C.P. and Crane, R.M. (2009), "Local damage detection with the global fitting method using mode shape data in notched beams", J. Nondestruct. Eval., 28(2), 63-74.
  45. Yu, L., Zhou, R., Tang, L. and Chen, R. (2018), "A dbn-based resampling svm ensemble learning paradigm for credit classification with imbalanced data", Appl. Soft. Comput., 69, 192-202.
  46. Zhang, Z. and Zhao, J. (2017), "A deep belief network based fault diagnosis model for complex chemical processes", Comput. Chem. Eng., 107, 395-407.
  47. Zhao, X. and Ye, B. (2016), "Singular value decomposition packet and its application to extraction of weak fault feature", Mech. Syst. Signal. Pr., 70-71, 73-86.
  48. Zhao, H., Liu, H., Xu, J., Guo, C. and Deng, W. (2019), "Research on a fault diagnosis method of rolling bearings using variation mode decomposition and deep belief network", J. Mech. Sci. Technol., 33(9), 4165-4172.
  49. Zhao, L.C., Chen, S.S., Xu, Y.P. and Tahouneh, V. (2021), "Vibration analysis of damaged core laminated curved panels with functionally graded sheets and finite length", Steel Compos. Struct., 38(5), 477-496.
  50. Zhong, Y.T. and Xiang, J.W. (2019), "Impact location on a stiffened composite panel using improved linear array", Smart. Struct. Syst., 24(2), 173-182.