Open AccessResearch on Automatic Control System of MR Damper Based on Neural Network Algorithm
Author(s) -
Yunhan Gan,
Yi X. Zhong
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1848/1/012152
Subject(s) - damper , artificial neural network , control theory (sociology) , computer science , vibration , current (fluid) , bridge (graph theory) , control system , algorithm , pulse width modulation , engineering , control engineering , control (management) , artificial intelligence , acoustics , physics , medicine , electrical engineering , voltage
The use of MR dampers can achieve the absorption of harmful vibrations such as seismic waves that affect the structure of the building itself. The current control of the MR damper generally uses a pulse width modulation system to control the current input to the MR damper, thereby controlling the MR damper. But this is a semi-automatic control method, and the error is relatively large. This article introduces an MR damper control system that introduces a neural network algorithm. The system uses a half-bridge drive circuit for output current control in the circuit part and uses the Shen BP network algorithm for numerical prediction and fitting. The experimental research and tests conducted in the project show that the use of this fuzzy algorithm can make the MR damper achieve good damping effects under different conditions, and can well improve the original semi-automatic control method.