Skip to main content
Log in

A hybrid Modified Artificial Bee Colony and extended Kalman filter algorithm for structural system identification

  • Research
  • Published:
Asian Journal of Civil Engineering Aims and scope Submit manuscript

A Publisher Correction to this article was published on 29 July 2023

This article has been updated

Abstract

The continuous monitoring of structural condition is extremely important for sustaining and preserving the service life of civil structures. It provides resolute and staunch information on the health, serviceability, integrity and safety of structures. System Identification (SI) is a process of extracting a “vibration signature” and comparing the present signature to that of the reference (i.e., undamaged) state. Any notable difference in the vibration signature of a system can be attributed to a certain type of degradation. However, developing efficient SI techniques for civil structures is a critical challenge that needs to be addressed. Hence, an effort is made to formulate SI as a non-convex optimization problem. The Modified Artificial Bee Colony (MABC) is assimilated with extended Kalman filter (EKF) for damage identification using modal data. An objective function obtained by the fractional changes of damaged and undamaged structure is controlled by the metaheuristic optimization algorithm embedded in the proposed damage-detection framework. The residual error that arises in modeling is minimized, creating a more accurate prediction for the damage severity and location. The effectiveness of the proposed SI method is demonstrated via numerical simulation and real-time experimental study. The results of the study identify the damage (both location and severity in a single step of optimization) with lesser computational time and faster convergence when compared with Artificial Bee colony (ABC) and Particle Swarm Optimization (PSO) algorithm. This article emphasizes the importance of continuous structural condition monitoring of civil engineering structures.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

Data availability

Raw data were generated at National Institute of Technology Raipur. Derived data supporting the findings of this study are available from the corresponding author R.B.Malathy on request.

Change history

Abbreviations

\(x\) :

Vector of displacement

f :

Objective value

Fit :

Fitness value

D :

Dimension

\(\left[{K}_{n}^{d}\right]\) :

Global stiffness matrices

\(\left[{K}_{e}^{d}\right]\) :

Element stiffness matrices

Φ :

Mode shapes

ω :

Natural frequencies

[M]:

Global mass matrix

[K]:

Global stiffness matrix

\({N}_{\mathrm{m}}\) :

Number of modes

\({N}_{\mathrm{p}}\) :

Number of measured nodal displacement

\({\omega }_{i}\) :

Natural frequency at ith mode

\(\delta {\omega }_{i}\) :

Fractional change of the experimental natural frequency

\(\delta {\varnothing }_{i j}\) :

Fractional change of the analytical natural frequency

\(\alpha\) :

Stiffness reduction factor

\({N}_{L}\) :

Noise level

References

Download references

Acknowledgements

The author is grateful to the NIT, Raipur Management for their support in completing this research. Also, thanks to those directly or indirectly involved and helped in this study.

Funding

The author declare that no funds, grants, or other support were received during the preparation of this manuscript.

Author information

Authors and Affiliations

Authors

Contributions

Dr. R. B. Malathy : Conceptualization, Methodology, Software, Validation, Formal analysis, Investigation, Resources, Data curation, Writing - original draft, review & editing.

Corresponding author

Correspondence to R. B. Malathy.

Ethics declarations

Conflict of interest

The author declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

The original online version of this article was revised: Abstract, table 4 and 5 and figure 5 were not correct.

Appendix

Appendix

Pseudo-code structure of the SI algorithm

Calculation of stiffness matrix

  1. 1.

    Calculate the value of L1, L2, and L3 initially.

  2. 2.

    Calculate the value of G and J as

    $$J=\left[\frac{WH\left({W}^{2}+{H}^{2}\right)}{12}\right],$$
    $$G=\left[\frac{W{H}^{2}}{3}+\frac{1.8H}{W}\right].$$
  3. 3.

    Determine K1, K2, and K3…K10 values as shown below:

    $$\begin{aligned} k1 & = \frac{EA}{L}; k2 = \frac{GI_x }{L};k3 = \frac{12EI_y }{{L3}} ; k4 = \frac{12EI_z }{{L3}};k5 = \frac{6EI_y }{{L2}};k6 = \frac{6EI_z }{{L2}}; \\ k7 & = \frac{4EI_y }{L};k8 = \frac{4EI_z }{L};k9 = \frac{2EI_y }{L};k10 = \frac{2EI_z }{L}. \\ \end{aligned}$$
  4. 4.

    Generate the stiffness matrix as

    $$\left[\begin{array}{cccccccccccc}k1& 0& 0& 0& 0& 0& -k1& 0& 0& 0& 0& 0\\ 0& k4& 0& 0& 0& k6& 0& -k4& 0& 0& 0& k6\\ 0& 0& k3& 0& -k5& 0& 0& 0& -k3& 0& k5& 0\\ 0& 0& 0& k2& 0& 0& 0& 0& 0& -k2& 0& 0\\ 0& 0& -k5& 0& k7& 0& 0& 0& k5& 0& k9& 0\\ 0& k6& 0& 0& 0& k8& 0& -k6& 0& 0& 0& k10\\ -k1& 0& 0& 0& 0& 0& k1& 0& 0& 0& 0& 0\\ 0& -k4& 0& 0& 0& -k6& 0& k4& 0& 0& 0& -k6\\ 0& 0& -k3& 0& k5& 0& 0& 0& k3& 0& k5& 0\\ 0& 0& 0& -k2& 0& 0& 0& 0& 0& k2& 0& 0\\ 0& 0& -k5& 0& k9& 0& 0& 0& k5& 0& k7& 0\\ 0& k6& 0& 0& 0& k10& 0& -k6& 0& 0& 0& k8\end{array}\right].$$
  5. 5.

    Check if the member axes coincide with the global axes, if yes then [K] is the elemental stiffness matrix.

  6. 6.

    If not establish a relation between the local and global systems by using transformation matrix.

  7. 7.

    Determine

    $$r=\left[\begin{array}{ccc}CXx& CYx& CZx\\ CXy& CYy& CZy\\ CXz& CYz& CZz\end{array}\right],$$

    where \(CXx=\frac{({x}_{2}-{x}_{1})}{L}\);\(CYx=\frac{\left({y}_{2}-{y}_{1}\right)}{L}\); \(CZx=\frac{\left({z}_{2}-{z}_{1}\right)}{L}\); \(D=sqrt(CXx*CXx+CYx*CYx)\);\(CXy=\frac{-CYx}{D}\); \(CYy=\frac{CXx}{D}\); \(CZy=0\); \(CXy=\frac{-CXx*CZx}{D}\); \(CYz=\frac{-CYx*CZx}{D}\); \(CZz=D.\)

  8. 8.

    Determine R′.

  9. 9.

    End.

  10. 10.

    Calculate the elemental equation of space frame element as

    $$KKK=R*KK*{R}^{^{\prime}}.$$

    Calculation of Mass Matrix

    1. 1.

      Calculate the value of L, L2 and rx2.

    2. 2.

      Determine m1, m2, and m3…m10 values as shown below

      m1 = 70; m2 = 35; m3 = 78; m4 = 70*rx2; m5 = 35*rx2; m6 = 27; m7 = 22*L; m8 = 13*L; m9 = 8*L2; m10 = 6*L2;

    3. 3.

      Generate the mass matrix as shown below 

      $$MM = \left[ {\frac{(rho*A*L)}{{105}}} \right]*\left[ {\begin{array}{*{20}c} {m1} & 0 & 0 & 0 & 0 & 0 & {m2} & 0 & 0 & 0 & 0 & 0 \\ 0 & {m3} & 0 & 0 & 0 & {m7} & 0 & {m6} & 0 & 0 & 0 & { - m8} \\ 0 & 0 & {m3} & 0 & { - m7} & 0 & 0 & 0 & {m6} & 0 & {m8} & 0 \\ 0 & 0 & 0 & {m4} & 0 & 0 & 0 & 0 & 0 & { - m5} & 0 & 0 \\ 0 & 0 & 0 & 0 & {m9} & 0 & 0 & 0 & { - m8} & 0 & { - m10} & 0 \\ 0 & 0 & 0 & 0 & 0 & {m9} & 0 & {m8} & 0 & 0 & 0 & { - m10} \\ {m2} & 0 & 0 & 0 & 0 & 0 & {m1} & 0 & 0 & 0 & 0 & 0 \\ 0 & {m6} & 0 & 0 & 0 & {m8} & 0 & {m3} & 0 & 0 & 0 & { - m7} \\ 0 & 0 & {m6} & 0 & { - m8} & 0 & 0 & 0 & {m3} & 0 & {m7} & 0 \\ 0 & 0 & 0 & { - m5} & 0 & 0 & 0 & 0 & 0 & {m4} & 0 & 0 \\ 0 & 0 & {m8} & 0 & {m10} & 0 & 0 & 0 & {m7} & 0 & {m9} & 0 \\ 0 & { - m8} & 0 & 0 & 0 & { - m10} & 0 & { - m7} & 0 & 0 & 0 & {m9} \\ \end{array} } \right]$$
  11. 11.

    Check if the member axes coincide with the global axes, if yes then [K] is the elemental mass matrix.

  12. 12.

    If not establish a relation between the local and global systems by using transformation matrix and determine the mass matrix.

  13. 13.

    Determine R, R′.

  14. 14.

    Calculate the elemental equation of space frame element as

    $$MMM = R*MM*R^{\prime}.$$

Determination of the objective function

  1. 1.

    Calculate natural frequency ωi of the ith mode of the degraded frame based on numerical analysis.

  2. 2.

    Introduce a reduction factor and calculate the natural frequency of the damaged frame.

$${\text{num}}\_{\text{del}}\_{{w}}\left( {{i}} \right) = \left( {\left( {{\text{abs}}\left( {{\text{num}}\_{\text{fr}}\_{\text{udam}}\left( {{i}} \right) - {\text{num}}\_{\text{fr}}\_{\text{dam}}\left( {{i}} \right)} \right)*{\text{reduction}}\_{\text{alpha}}\left( {{i}} \right)} \right)/{\text{num}}\_{\text{fr}}\_{\text{udam}}\left( {{i}} \right)} \right).$$
  1. 3.

    Similarly calculate using FEM analysis as,

    $${\text{fem}}\_{\text{del}}\_{\text{w}}\left( {\text{i}} \right) = \left( {\left( {{\text{abs}}\left( {{\text{fem}}\_{\text{fr}}\_{\text{udam}}\left( {\text{i}} \right) - {\text{fem}}\_{\text{fr}}\_{\text{dam}}\left( {\text{i}} \right)} \right)} \right)/{\text{fem}}\_{\text{fr}}\_{\text{udam}}\left( {\text{i}} \right)} \right).$$
  1. 4.

    The fractional changes in natural frequency of the undamaged and damaged frame is given as, fr_fractional_changes = (num_del_w(i)-fem_del_w(i))^2.

  2. 5.

    Calculate mode shape of the ith mode of the undamaged frame based on numerical analysis.

  3. 6.

    Introduce a reduction factor and calculate the mode shape of the damaged frame. num_del_shape(i) = ((abs(num_shape_undam(i)-num_shape_dam(i))*reduction_alpha(i)) )

  4. 7.

    Similarly calculate for FEM analysis as,

    $${\text{fem}}\_{\text{del}}\_{\text{shape}}\left( {\text{i}} \right) = \left( {\left( {{\text{abs}}\left( {{\text{fem}}\_{\text{shape}}\_{\text{udam}}\left( {\text{i}} \right) - {\text{fem}}\_{\text{shape}}\_{\text{dam}}\left( {\text{i}} \right)} \right)} \right)} \right)$$
  1. 8.

    The fractional changes in natural frequency of the undamaged and damaged frame is given as, shape_fractional_changes(i) = (num_del_w(i)-fem_del_w(i))^2.

  2. 9.

    Determine the objective function as, sum_shape = fr_fractional_changes + shape_fractional_changes(i).

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Malathy, R.B. A hybrid Modified Artificial Bee Colony and extended Kalman filter algorithm for structural system identification. Asian J Civ Eng 25, 385–396 (2024). https://doi.org/10.1007/s42107-023-00782-3

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s42107-023-00782-3

Keywords

Navigation