Research and Development of Magnetorheological Spiral-flow Damper

半主動式磁黏滯阻尼器在無磁場時以被動控制振動；在外加磁場的作用下，磁黏滯液由一黏度較低的牛頓流體在幾毫秒內變為有較高黏度與剪切強度的Bingham流體，隨磁場變化改變阻尼性能以減少結構所承受的震動和衝擊。 本文應用磁黏滯液體設計螺旋流道式磁黏滯阻尼器並以法蘭式驅動模式與一般黏滯液體阻尼器整合為減振系統。將流道設計為螺旋狀可在固定體積內依不同螺旋節距來增減流道的長度、激磁後的磁黏滯液體流動方向與磁力線方面垂直、構造簡單等優點。採用模組化設計，使其能整合於其他阻尼器或是液壓缸中。模組化整合的優點在於容易置換與維護，節省維修與維護的時間。 使用MTS_810拉伸試驗控制系統來測試法蘭式螺旋流道阻尼器，測試項目包括位移、受力、速度等。實驗結果顯示當驅動電流為0A時，在速度6.28 mm/s時的阻尼力為149N；當驅動電流達2.0A時，阻尼力為521N，阻尼力已不再增加，表示磁黏滯液已達到磁飽和，可調阻尼力係數(dynamic ratio)為3.49(521/149)。因此,通過調節磁流變阻尼器的驅動電流,就能達到半主動控制阻尼力的目的。 本文利用數學分析軟體Matlab之系統分析模組Simulink建構Bouc-Wen磁流變阻尼器模型，再應用類神經網路演算法作為控制驅動電流產生磁場變化來改變阻尼特性，將系統的最大位移、平均位移、最大加速度、平均加速度和阻尼力作為訊號回饋，模擬一單自由度振動系統。由實驗結果可知螺旋流磁黏滯阻尼器(MR-SF damper)半主動控制減振的效果達41.6%。 本論文所開發之螺旋流道阻尼器具有高可調阻尼力範圍、結構簡單。不僅在無驅動電流狀態下也能產生基本的阻尼力，而且應用類神經網路控制驅動電流大小進而改變阻尼特性，可以有效的運用於結構半主動控制減振。Magnetorheological fluid, or MR fluid, can be transformed from low viscosity Newtonian fluid to Bingham fluid with higher viscosity and shear stress within only microseconds by applying external magnetic field. Thus, the damping characteristic of Magneto-Rheological fluid dampers, or MRF dampers, can be easily regulated to effectively absorb the impact or vibration energy. A semi-active magneto-rheological spiral-flow (MR-SF) damper has been developed by integrating the passive and active (damping principles) through the spiral MR fluid channel. The channel structure of damper can be easily varied by replacing the channel part with different spiral-pitches. The proposed modularization design is available for not only the commercial dampers but other hydraulic cylinders, which saves time for maintenance. The operating performance of developed MR-SF damper was tested and measured by using the MTS_810 tensioning equipment. The passive damping force is 149 N at the operating speed of 6.28 mm/s without magnetic field, and the active damping force reaches its maximum value as 521 N when driving current is 2.0 A. The damping force also comes to stagnate since the MR fluid becomes magnetically saturated with higher driving current. Its dynamic ratio is 3.49. Therefore, damping force can be easily regulated by the modulating driving current. For realizing the close-loop control algorithm, the damping hysteresis behavior of the MR-SF damper is modeled based on the Bouc-Wen hysteresis model with MATLAB and Simulink. Furthermore by using the artificial neural network (ANN) algorithm, the damping coefficient is controlled by regulating magnetic field. Through the developed close-loop control algorithm, the MR-SF damper model can effectively reduce 41.6% of stimulated vibration. Through the novel concept of the spiral-flow channel, the developed MR-SF damper can realize a wide range of dynamic ratio. The passive damping function provides the main damping effect by the viscous drag, and the active damping function delivers the adjustable damping effect