International Journal of Structural Engineering (5 papers in press)

Regular Issues

• Structural condition assessment and prediction using machine learning-based EKF  
  by Santosh Bisoyi, Amit Kumar Rathi 
  Abstract: This paper aims to improve the machine learning (ML)-based extended-Kalman filter (EKF) technique for condition assessment of structures. EKF is widely used for estimating structural properties, aiding in the detection and assessment of damage. However, it lacks accuracy in the prediction of aging and remaining service life. This study improves the prediction of structural conditions by incorporating ML algorithms, thus facilitating timely predictive maintenance and disaster prevention strategies. The proposed method employs EKF for the parameter estimation from vibration responses to train ML models for accurate prediction. The framework is validated based on multi-storey framed structures subjected to seismic excitation through numerical simulations and experimental tests. The results illustrate the accuracy and robustness of the proposed method up to 20% Gaussian and non-Gaussian noises. The ML model such as extreme gradient boosting achieves R2 score of 0.99 and RMSE of approx. 0.01 in predicting the stiffness parameters. The proposed approach demonstrates potential in real-time condition monitoring and predictive maintenance.
  Keywords: extended-Kalman filter; EKF; machine learning; prediction modelling; structural condition assessment.
  DOI: 10.1504/IJSTRUCTE.2025.10073206
   
  
• Deformation approach for the calculation of bent wooden elements  
  by Dzmitry N. Lazovski, Arthur I. Hil, Dmitry O. Glukhov 
  Abstract: The article presents the results of calculating the parameters of the stress-strain state of bent wooden elements at any stage of its work under the action of short-term static loading. The modelling took into account the complete diagrams of wood deformation under compression and tension, the dependences of their approximation were proposed, and verification was performed based on our own experimental data and data from other authors. The peculiarity of the destruction of wooden elements under the action of a bending moment has been experimentally established, which begins with the crumpling of wood fibres in a compressed zone and the formation of a plastic hinge followed by a sequential rupture of longitudinal fibres in a stretched zone along the height of the cross section, which corresponded to the simulation results. A criterion is proposed for calculating the bending resistance of wooden elements at the limiting stage in terms of strength in the form of the maximum force perceived by them, at which the conditions of equilibrium of forces in the cross section are fulfilled. The proposed criterion, in contrast to limiting the maximum deformations of wood during compression and stretching, allows us to more fully take into account the redistribution of forces between the longitudinal fibres in the cross-section of wood in the plastic stage of work before destruction.
  Keywords: wooden elements; wood deformation diagram; flat section hypothesis; deformation calculation approach.
  DOI: 10.1504/IJSTRUCTE.2025.10073508
   
  
• A stacking ensemble learning model for buckling response of the functionally graded carbon nanotube reinforced composite plates  
  by Lalit Kumar Sharma, Jyotindra Narayan, Naveen Sharma 
  Abstract: This study proposed a stacking ensemble learning model for the buckling response of functionally graded carbon nanotube-reinforced composite (FG-CNTRC) plates using an inverse hyperbolic shear deformation theory (IHSDT). Firstly, the mathematical for the FG-CNTRC plate has been developed in the framework of IHSDT, then a stacking ensemble learning model is employed to estimate critical buckling loads under various conditions, including different CNT distributions, span-to-thickness ratios, volume fractions, and temperature environments and compared with the performance index of random forest (RF) and support vector regression (SVR). Input datasets were generated using MATLAB, with an 85:15 train-test split and five-fold validation to ensure robustness. Numerical results validate the predictive accuracy of ML models, with stacking ensemble learning outperforming RF and SVR in terms of mean squared error (MSE), and the coefficient of determination (R2). This work demonstrates the effectiveness of machine learning techniques for predicting the mechanical behaviour of FG-CNT materials.
  Keywords: functionally graded materials; buckling; machine learning; stacking ensemble; random forest; support vector regression; SVR; shear deformation theory.
  DOI: 10.1504/IJSTRUCTE.2025.10073739
   
  
• A novel friction damper with an interference fit joint: an experimental and finite element study  
  by Jamshid Sabouri, Farid Khalajzadeh 
  Abstract: The present study proposes a novel friction damper with an interference fit joint for seismic energy dissipation in structures. In this model, friction is produced between steel parts by creating a crust via an interference fit joint in the middle of the damper. In fact, the part begins to slide due to the designed axial force, causing frictional energy dissipation in the process. This idea can lead to the desired behaviour in frictional dampers. For this purpose, three laboratory samples were prepared and subjected to displacement-control cyclic loading according to the ATC-24 loading protocol. The results from both laboratory tests and finite element simulations showed that Sample 3 exhibited symmetric and stable hysteresis loops. The equivalent viscous damping ratio reached approximately 30%, demonstrating high energy dissipation capacity. This favourable performance was attributed to the sufficient engagement length, identified as a key parameter in frictional dampers with interference fit joints.
  Keywords: friction damper; energy dissipation; cyclic behaviour; slipping; interference fit joint; press and shrink fits; equivalent viscous damping ratio; cyclic loading; experimental study; hysteresis curve; finite element.
  DOI: 10.1504/IJSTRUCTE.2025.10073891
   
  
• Seismic performance assessment of low-rise reinforced concrete moment frames with different slab types using nonlinear time history analysis [performance of LR RC frames with different slabs by NLTHA]  
  by Mahmoud Eissa, Murude Celikag 
  Abstract: This study explores the global influence of different slab types on the seismic behaviour of reinforced concrete (RC) frames, focusing on an innovative RC frame supporting a composite slab of galvanised steel deck and concrete. The aim is to provide key insights for improved seismic analysis and design in structural engineering. A rigorous nonlinear time history analysis was employed, considering bar slip, shear, and flexural deformation. These were modelled using a distributed plasticity framework with fibre-section techniques in OpenSees, using 51 ground motion records. Base shear from 2D frames with composite slabs was 30% less than other cases. Per Hazus-MH MR5, composite and solid one-way slab frames showed up to 27% lower inter-story drift, performing best among all cases. Achieving a 40% reduction in building weight, material use, and improved seismic performance under NTHA makes the composite slab a strong competitor to conventional slabs in RC construction.
  Keywords: composite slab with RC frame; seismic performance NTHA; RC frame-slab type; nonlinear moment frame performance.
  DOI: 10.1504/IJSTRUCTE.2025.10074119