Research On Structure Vibration Control Based On MR Damper

There are lots of earthquakes in China, with the development of urbanization in China, and the increasing high-rise buildings. How to reduce the losses caused by the earthquakes, and ensure the safety of people’s life and property is an increasingly serious problem. The performance of Magneto-rheological (MR) damper is excellent, when the earthquakes come, MR damper which is installed in the structure, used as an energy dissipation device, can effectively reduce the seismic response of the structure. While the control algorithm will affect the great performance of MR damper to a certain extent, therefore, how to design a better control algorithm is a hot issue. Hence, this thesis mainly explains it by the following parts.(1) Introducing the working principle of MR damper and the commonly used dynamic model, Comparing the pros and cons of various models, then we choose the modified Bouc-Wen model as the dynamics model of MR damper.(2) The traditional control algorithms need to establish an accurate mathematical model for the controlled system, human knowledge and logical thinking cannot be used effectively during these control algorithms. Therefore, this article will design two kinds of smart control algorithms which are suitable for the MR damper--fuzzy control algorithm and the improved fuzzy control algorithm. And this paper will use the Matlab software to construct models and do the simulation analysis for 4 story reinforced concrete structure,9 story and 20 story nonlinear steel frame structure which are installed MR damper. The result shows that the designed control algorithm can effectively suppress the seismic responses of building structures, and the control effect of improved fuzzy control algorithm is better than fuzzy control algorithm. That means, it is feasible to use the acceleration of each layer to adjust the voltage proportional factor timely under seismic action.(3)Lack of self-adjusting ability in the control process of fuzzy control algorithm, and neural network is not easy to utilize human logical thinking. Therefore, by using the adaptive neural fuzzy inference system (ANFIS), this article designs a smart control algorithm with strong earthquake randomness-adaptive neural fuzzy control algorithm. And this paper will use the Matlab software to construct models and do the simulation analysis for 9 story and 20 story nonlinear steel frame structures. The result also shows that, by comparing with the fuzzy control, the adaptive neural-fuzzy control algorithm can further restrain the seismic responses of the building structures, and with higher computational efficiency, it can better improve the safety performance of the structures.