• Structural Engineering and Mechanics
• /
• 제85권4호
• /
• pp.469-484
• /
• 2023

A deep recursive bidirectional Cuda Deep Neural Network Long Short Term Memory (Bi-CuDNNLSTM) layer is recruited in this paper to predict the entire force time histories, and the corresponding hysteresis and backbone curves of reinforced concrete (RC) bridge piers using experimental fast and slow cyclic tests. The proposed stacked Bi-CuDNNLSTM layers involve multiple uncertain input variables, including horizontal actuator displacements, vertical actuators axial loads, the effective height of the bridge pier, the moment of inertia, and mass. The functional application programming interface in the Keras Python library is utilized to develop a deep learning model considering all the above various input attributes. To have a robust and reliable prediction, the dataset for both the fast and slow cyclic tests is split into three mutually exclusive subsets of training, validation, and testing (unseen). The whole datasets include 17 RC bridge piers tested experimentally ten for fast and seven for slow cyclic tests. The results bring to light that the mean absolute error, as a loss function, is monotonically decreased to zero for both the training and validation datasets after 5000 epochs, and a high level of correlation is observed between the predicted and the experimentally measured values of the force time histories for all the datasets, more than 90%. It can be concluded that the maximum mean of the normalized error, obtained through Box-Whisker plot and Gaussian distribution of normalized error, associated with unseen data is about 10% and 3% for the fast and slow cyclic tests, respectively. In recapitulation, it brings to an end that the stacked Bi-CuDNNLSTM layer implemented in this study has a myriad of benefits in reducing the time and experimental costs for conducting new fast and slow cyclic tests in the future and results in a fast and accurate insight into hysteretic behavior of bridge piers.

• Journal of Mechanical Science and Technology
• /
• 제15권11호
• /
• pp.1548-1554
• /
• 2001

Flow patterns and dynamic properties of two-stage Weis-Fogh type ship propulsion mechanism are studied by a discrete vortex method. To study mutual interference between two wings, two cases are con sidered - wing motions with the same and reverse phases. The predicted flow patterns correspond to the available flow visualization results. Time histories of thrust and drag coefficients are also calculated, and the interference between the two wings are numerically clarified.

• Nuclear Engineering and Technology
• /
• 제50권6호
• /
• pp.971-980
• /
• 2018

Using the concept of quasi-static decomposition and using three-noded isoparametric locking-free element, this article presents a formulation of the finite element method for Timoshenko beam subjected to spatially different time-dependent motions at supports. To verify the validity of the formulation, three fixed-hinged beams excited by the real seismic motions are examined; one is a slender beam, another is a stocky one, and the other is an intermediate one. The numerical results of time histories of motions of the three beams are compared with corresponding analytical solutions. The internal loads such as bending moment and shearing force at a specific time are also compared with analytic solutions. These comparisons show good agreements. The comparisons between static components of the internal loads and the corresponding total internal loads show that the static components predominate in the stocky beam, whereas the dynamic components predominate in the slender one. Thus, the total internal loads of the stocky beam, which is governed by static components, can be predicted simply by static analysis. Careful numerical experiments indicate that the fundamental frequency of a beam can be used as a parameter identifying such a stocky beam.

• Steel and Composite Structures
• /
• 제27권3호
• /
• pp.273-283
• /
• 2018

Nonlinear low velocity impact response of sandwich beam with laminated composite face sheets and soft core is studied based on Extended High Order Sandwich Panel Theory (EHSAPT). The face sheets follow the Third order shear deformation beam theory (TSDT) that has hitherto not reported in conventional EHSAPT. Besides, the two dimensional elasticity is used for the core. The nonlinear Von Karman type relations for strains of face sheets and the core are adopted. Contact force between the impactor and the beam is obtained using the modified Hertz law. The field equations are derived via the Ritz based applied to the total energy of the system. The solution is obtained in the time domain by implementing the well-known Runge-Kutta method. The effects of boundary conditions, core-to-face sheet thickness ratio, initial velocity of the impactor, the impactor mass and position of the impactor are studied in detail. It is found that each of these parameters have significant effect on the impact characteristics which should be considered. Finally, some low velocity impact tests have been carried out by Drop Hammer Testing Machine. The contact force histories predicted by EHSAPT are in good agreement with that obtained by experimental results.