基于改进CQPSO算法的压电陶瓷建模研究

压电精密定位技术在高精度定位与控制领域起着关键作用。针对压电陶瓷迟滞非线性特性进行建模与分析是当前研究的一大热点。传统智能算法对迟滞模型的辨识易陷入局部最优,因此,提出一种改进混沌量子粒子群算法(Improved chaotic quantum particle swarm optimiztaion,ICQPSO)。以量子粒子群算法(QPSO)结合基于早熟系数的混沌映射跳出局部收敛。引入变尺度法缩小可行解空间提升收敛效率和精度,采用Bouc-Wen微分方程模型对多频输入动态迟滞现象进行建模。经实验验证,该算法对Bouc-Wen模型的辨识精度明显高于GA、MFA算法。

MODELING OF PIEZO-ELECTRIC CERAMICS BASED ON IMPROVED CQPSO

Piezo-electric precise positioning technology plays a key role in the field of high precision positioning and control.Modeling and analysis of hysteresis nonlinearity for piezo-electric ceramics is a hot topic in current research.Traditional intelligent algorithms have the disadvantage of easily falling into local optimum in identifying hysteretic models.Therefore,this paper proposed an improved chaotic quantum particle swarm optimization (ICQPSO).It jumped out of local convergence by quantum particle swarm optimization (QPSO) combined with chaotic mapping based on premature coefficient.Mutative scale space search (MSSS) was introduced to reduce the feasible solution space and improve the convergence efficiency and accuracy.The Bouc-Wen differential equation model was used to model the dynamic hysteresis of multi-frequency input.Experiments show that the recognition accuracy of this algorithm for Bouc-Wen model is significantly higher than that of GA and MFA.