基于混沌粒子群算法的压电微纳平台迟滞模型参数辨别

Piezoelectric Micro-nano Platform Hysteresis Model Parameter Discrimination with Chaotic Particle Swarm Algorithm

针对压电微纳平台存在迟滞非线性问题,建立改进的Bouc-Wen模型描述压电微纳平台迟滞现象并使用混沌粒子群算法对模型参数进行参数辨识。通过引入迟滞非线性项对原始Bouc-Wen模型进行改进来获得更好的迟滞拟合曲线;为了解决辨识精度低的问题,使用混沌粒子群算法来增强算法在局部的寻优能力。结果表明,以频率为1 Hz驱动电压为例,改进的Bouc-Wen模型相较于传统Bouc-Wen模型能更加准确地描述其迟滞现象。同时使用混沌粒子群优化算法参数拟合的平均误差以及均方根误差均优于传统粒子群算法,有效提高了位移迟滞数据的拟合精度。

An improved Bouc-Wen model was established to describe the hysteresis of the piezoelectric micro-nano platform and the model parameters were identified using the chaotic example group algorithm.Firstly,the hysteretic nonlinear term was introduced to improve the original Bouc-Wen model to obtain a better hysteretic fitting curve.Secondly,in order to solve the problem of low identification accuracy,chaotic particle swarm optimization algorithm was used to enhance the local optimization ability of the algorithm.The results showed that the improved Bouc-Wen model could describe the hysteresis more accurately than the traditional Bouc-Wen model when the frequency was 1 Hz driving voltage.Meanwhile,the mean error and root mean square error of parameter fitting using chaotic particle swarm optimization algorithm were better than traditional particle swarm optimization algorithm,which effectively improved the fitting accuracy of displacement hysteresis data.