Investigation On MR Dampers And Intelligent Control On Seismic Whiplash Effect Of Ship-lifting Structure

Over the past three decades, many interests have been generated to apply structural protective systems to mitigate the effects on civil engineering structures suffering from dynamic hazards, such as earthquake and strong wind. These systems always employ damping devices to intensify the energy dissipation ability of civil engineering structures. One of the most promising devices is magnetorheological (MR) fluid dampers. Because of their mechanical simplicity, high dynamic range, low power requirements and large force capacity, they have been shown to mesh well with application demands and constraints to offer an attractive means against severe earthquake and strong wind. The focus of this dissertation is the extensive research on MR dampers, including devices designing, improvement of the MR damper response time and intelligent control algorithm.New type of MR damper is designed and made based on the quasi-static model, so called parallel-plate model of MR dampers, which has the advantage of high dynamic range and stable performance. Effects of geometry on MR damper performance, controllable force and dynamic range are also discussed. Experimental results for the variable input current tests, amplitude-dependent tests, frequency-dependent tests are presented. The testing results imply that the MR damper resisting force increases as the applied current increases. The low carbon steel, which has a high permeability, increases the magnetic field in the gap and the saturation current resulting in an increased damping force. Due to relative high viscosity of MR fluid, eliminating air pockets in the damper and air dissolved in the fluid is very difficult. The trapped air results in a force lag in the MR damper responses. To reduce the effect of trapped air on damper performance, a pressurized spring accumulator is utilized, which has the advantage of mechanical simplicity over ordinary acridine accumulator.Magnetic hysteresis is the resisting force lag the controllable current due to the MR fluid response time and the resistance of magnetic material, which is the intrinsic characteristic of MR dampers. When the resistance force is large, magnetic hysteresis will have more serious effect on the performance of MR dampers. Therefore, resolving the problem of magnetic hysteresis is very significant to the structural controlling because of the large-scale MR dampers used in civil engineering structures. In order to decrease control infection resulting from the magnetic hysteresis of MR dampers, neural network prediction is proposed in this paper. Response prediction is to predict the future response through 4 to 5 past structural response, so that enough time is won to compute and generate the control force. The simulation results of this study imply that magnetic hysteresis of MR dampers can really influence the semi-active control whereas neural network prediction can successfully compensate for the response time of MR dampers.Taking into account of magnetic hysteresis of MR dampers, revised Bingham model for MR dampers is proposed in this dissertation. Training data is obtained from this model and is used to train the neural network. As a result, a inverse neural network model is established for MR dampers, which can be connected with LQR controlling method or fuzzy controlling method to form the intelligent control for MR dampers. The simulation results show that intelligent control can obtain the continuous controlling current for MRdampers, consequently, the resistance force of MR dampers can also be adjusted continuously.The application of intelligent controlling algorism on MR dampers is discussed in this dissertation. Ship-lifting structure whose main function is to lift the coming-and-going ships vertically to cut short the dam crossing time, is a kind of new structure in water conservancy works. Due to the great weight of the ship itself, it needs to be lifted to such a height that the ship-lifting structure has some particular features including huge lifting force, huge load and unusual struct...