Nonparametric modeling of magnetorheological elastomer base isolator based on artificial neural network optimized by ant colony algorithm
- Publication Type:
- Journal Article
- Journal of Intelligent Material Systems and Structures, 2015, 26 (14), pp. 1789 - 1798
- Issue Date:
Files in This Item:
|Revision-JIM-14-255 R1-Nonparametric modeling of MRE islator based on artificial neural network optimized by ant colony algorithm_v3.docx||Accepted Manuscript Version||2.51 MB|
Microsoft Word XML
Copyright Clearance Process
- Recently Added
- In Progress
- Open Access
This item is open access.
© SAGE Publications. Laminated magnetorheological elastomer base isolator is regarded as one of the most promising candidates for realizing adaptive base isolation for civil structures. However, the intrinsic hysteretic and nonlinear behavior of magnetorheological elastomer base isolators imposes challenge for adopting the device to accomplish high-accuracy performance in structural control. Therefore, it is essential to develop an accurate model for symbolizing this unique characteristic before designing a feedback controller. So far, some classical parametric models, such as Bouc-Wen, Dahl, and LuGre, have been proposed to depict the hysteretic response of magnetorheological devices, that is, magnetorheological damper, which may also be used for describing the nonlinear behavior of magnetorheological elastomer base isolator. However, the parameter identification is difficult to implement due to the nonlinear differential equations existing in these models. Considering this problem, this article proposes a nonparametric model, that is, an artificial neural network-based model with 3 input neurons, 18 hidden neurons, and 1 output neuron, to predict the magnetorheological elastomer isolator behavior. In this model, the ant colony algorithm is employed for model training to obtain the optimal weights based on the force-displacement/velocity data sampled from the magnetorheological elastomer isolator. Finally, experimental data are used to validate the effectiveness of the proposed artificial neural network-based model with the good forecasting results.
Please use this identifier to cite or link to this item: