基于Bouc-wen模型的阻尼器建模仿真

Modeling and simulation of a damper based on Bouc-wen model

阻尼器是一个强非线性迟滞系统,难以用简单的数学模型对其进行表达。为解决这一问题,应用Bouc-wen模型对阻尼器进行建模用以描述阻尼器的特性。Bouc-wen模型已经广泛应用于描述材料和结构对作用力的迟滞响应。首先,为确定阻尼器的非线性特性,通过实验测试,得到阻尼器在不同频率和振幅激励下的迟滞特性曲线。通过这些数据Bouc-wen模型的参数能够被辨识出来。分别应用遗传算法与神经网络这两种不同的算法对Bouc-wen模型的参数进行辨识。比较这两种算法,仿真结果表明神经网络的逼近误差更小,比遗传算法的效率更高。

The damper,which is a strong nonlinear hysteretic system,is hard to be considered as a easy mathematical model.In order to solve this problem,the Bouc-wen model is applied to build a model to describe the characteristic of damper.The Bouc-wen model has been widely adopted to describe the responses of hysteretic materials and structures to applied force.Firstly,to determine the nonlineaer characteristic of damper,the curves of hysteretic characteristic are obtained through the physical tests with different frequencies and amplitudes.Using the data,the parameters of Boucwen model could be identified.To obtain the parameters,two different algorithms,genetic algorithm and neural network,are used to identified the parameters respectively.Comparing genetic algorithm and neural network,the simulation results show that neural network is able to achieve a smaller approximation error,and more effcint than genetic algorithm.