| Autor: |
Pendyala, Shravya, Pinjala, Devikiran, Puneet, N. P., Kumar, Hemantha |
| Předmět: |
|
| Zdroj: |
AIP Conference Proceedings; 2023, Vol. 2786 Issue 1, p1-6, 6p |
| Abstrakt: |
The capability of providing large and controllable damping force has increased the implementation of Magneto-rheological (MR) dampers in the semi active suspension systems of vehicles over the past few decades. The Magneto-rheological damper's nonlinear nature has made the modelling and control of these devices a difficult task. This work, presents a non-parametric mathematical model of custom-made MR damper using Radial basis function (RBF) network in MATLAB. The neural network used here has four input neurons that takes displacement, velocity, current and previous force values, forty hidden neurons, one output neuron which gives the predicted force. The network is trained and validated with the data obtained by experimental testing of the damper using Damper testing Machine (DTM). The comparison of experimental and predicted force vs displacement curves show that the dynamic characteristics of damper are well represented by the neural network. [ABSTRACT FROM AUTHOR] |
| Databáze: |
Complementary Index |
| Externí odkaz: |
|
