基于系统识别理论的磁流变阻尼器模型

磁流变阻尼器(MR damper)是一种新型的半主动结构振动控制装置。只要给该阻尼器输入很小的能量,它就能在极短的时间内(毫秒级)产生很大的力。这种阻尼器的性能可通过一组非线性微分方程来描述,物理参数包括位移、电压和力。给阻尼器输入位移和电压,阻尼器能产生一定的力。基于系统识别理论,采用ARX模型和优化神经网络技术对磁流变阻尼器的性能进行了仿真。训练后的神经网络能分别通过前向模型和反向模型精确地预测磁流变阻尼器的力和电压。把这样的神经网络用于控制系统,还能实现结构的主动控制。

MODEL OF MAGNETORHEOLOGICAL DAMPER BASED ON SYSTEM IDENTIFICATION

Magnetorheological (MR) damper is one of the more promising new devices for semi-active control of structures. External energy required by the adjustable fluid damper is minuscule while the damper can produce great force on the order of milliseconds. The characteristics of MR damper have been described by a set of nonlinear differential equations including three physical parameters such as displacement, voltage and force. When displacement and voltage are input into MR damper, the device can generate force. In this paper, the performance of MR damper is simulated based on system identification with optimal multi-layer perceptron neural networks and ARX model. The trained optimal networks can accurately predict force by a forward model and voltage by an inverse model. If the neural networks are used in a control system, the semi-active MR damper can be easily used for semi-active control of structures.