Buy article PDF
The purchased file will be sent to you
via email after the payment is completed.
US$ 35
|
Structural Engineering and Mechanics Volume 62, Number 2, April25 2017 , pages 237-246 DOI: https://doi.org/10.12989/sem.2017.62.2.237 |
|
|
|
Development of energy based Neuro-Wavelet algorithm to suppress structural vibration |
||
Yasser Bigdeli and Dookie Kim
|
||
| Abstract | ||
| In the present paper a new Neuro-Wavelet control algorithm is proposed based on a cost function to actively control the vibrations of structures under earthquake loads. A wavelet neural network (WNN) was developed to train the control algorithm. This algorithm is designed to control multi-degree-of-freedom (MDOF) structures which consider the geometric and material non-linearity, structural irregularity, and the incident direction of an earthquake load. The training process of the algorithm was performed by using the El-Centro 1940 earthquake record. A numerical model of a three dimensional (3D) three story building was used to accredit the control algorithm under three different seismic loads. Displacement responses and hysteretic behavior of the structure before and after the application of the controller showed that the proposed strategy can be applied effectively to suppress the structural vibrations. | ||
| Key Words | ||
| wavelet neural network; 3D building model; structural response; non-linearity; hysteretic behavior | ||
| Address | ||
| Yasser Bigdeli and Dookie Kim: Department of Civil Engineering, Kunsan National University, Jeonbuk, 573-701, Republic of Korea | ||