压电作动器非对称迟滞模型的建立和参数辨识
发布时间：2018-12-27 07:45
【摘要】：现有的众多基于传统Bouc-Wen改进的压电陶瓷非对称迟滞模型存在参数冗余,降低了模型参数辨识的准确性,而且常用的粒子群算法(PSO)在辨识压电陶瓷非对称迟滞模型参数方面收敛慢且容易陷入局部最优值。为此,首先提出了一种归一化的非对称迟滞模型,采用两个多项式达到非对称效果,利用归一化Bouc-Wen消除参数冗余;然后采用参数和变异策略自适应的差分进化算法进行迟滞参数辨识;建立了相应的测试系统,对压电陶瓷作动器进行了实验研究。结果表明,相比于传统的Bouc-Wen模型,所提出的模型能更精确地描述压电陶瓷实际电压位移曲线,而且消除了参数的冗余,降低了参数辨识的难度。相比于粒子群算法和传统差分进化算法,自适应差分进化算法能更快更精准地找到最优参数值。
[Abstract]:Many existing piezoelectric ceramic asymmetric hysteresis models based on traditional Bouc-Wen have parameter redundancy, which reduces the accuracy of model parameter identification. Moreover, the commonly used particle swarm optimization (PSO) algorithm has slow convergence in identifying the parameters of piezoelectric ceramic asymmetric hysteresis model and is easy to fall into local optimal value. For this reason, a normalized asymmetric hysteresis model is proposed. Two polynomials are used to achieve asymmetric effects, and normalized Bouc-Wen is used to eliminate parameter redundancy. Then the parameter and mutation strategy adaptive differential evolution algorithm is used to identify the hysteresis parameters, and the corresponding testing system is established, and the experimental research on piezoelectric actuator is carried out. The results show that compared with the traditional Bouc-Wen model, the proposed model can more accurately describe the actual voltage displacement curve of piezoelectric ceramics, eliminate the redundancy of parameters, and reduce the difficulty of parameter identification. Compared with particle swarm optimization algorithm and traditional differential evolution algorithm, adaptive differential evolution algorithm can find the optimal parameter more quickly and accurately.
【作者单位】： 南京航空航天大学机电学院;盐城工学院汽车工程学院;
【基金】：国家自然科学基金(51375230、51405417) 江苏省科技成果转化专项(BA2015098) 研究生创新基地(实验室)开放基金(kfjj20160508)项目资助
【分类号】：TP18
本文编号：2392728
[Abstract]:Many existing piezoelectric ceramic asymmetric hysteresis models based on traditional Bouc-Wen have parameter redundancy, which reduces the accuracy of model parameter identification. Moreover, the commonly used particle swarm optimization (PSO) algorithm has slow convergence in identifying the parameters of piezoelectric ceramic asymmetric hysteresis model and is easy to fall into local optimal value. For this reason, a normalized asymmetric hysteresis model is proposed. Two polynomials are used to achieve asymmetric effects, and normalized Bouc-Wen is used to eliminate parameter redundancy. Then the parameter and mutation strategy adaptive differential evolution algorithm is used to identify the hysteresis parameters, and the corresponding testing system is established, and the experimental research on piezoelectric actuator is carried out. The results show that compared with the traditional Bouc-Wen model, the proposed model can more accurately describe the actual voltage displacement curve of piezoelectric ceramics, eliminate the redundancy of parameters, and reduce the difficulty of parameter identification. Compared with particle swarm optimization algorithm and traditional differential evolution algorithm, adaptive differential evolution algorithm can find the optimal parameter more quickly and accurately.
【作者单位】： 南京航空航天大学机电学院;盐城工学院汽车工程学院;
【基金】：国家自然科学基金(51375230、51405417) 江苏省科技成果转化专项(BA2015098) 研究生创新基地(实验室)开放基金(kfjj20160508)项目资助
【分类号】：TP18
【相似文献】
相关期刊论文 前1条
1 官铮;赵东风;;非对称轮询多址门限服务排队系统高阶特性分析[J];云南大学学报(自然科学版);2007年S2期
相关博士学位论文 前1条
1 陈少白;非对称度量空间及其在控制中的应用[D];华南理工大学;2006年
相关硕士学位论文 前1条
1 陈小丹;非对称离子空间[D];武汉科技大学;2011年
，本文编号：2392728
本文链接：https://www.wllwen.com/kejilunwen/zidonghuakongzhilunwen/2392728.html
