智能作动器Preisach迟滞参数辨识及补偿研究  

Parameter identification and compensation of Preisach hysteresis in smart actuators

作　　者：王萍萍[1] 刘磊[2] 栾晓娜[1] 

机构地区：[1]哈尔滨工业大学航天学院,哈尔滨150001 [2]大连理工大学航空航天学院,辽宁大连116024

出　　处：《振动与冲击》2014年第20期91-94,122,共5页Journal of Vibration and Shock

基　　金：国家自然科学基金(11302132);博士后科学基金(2013M530123)

摘　　要：针对实现智能作动器精确运动需对迟滞非线性进行辨识、补偿问题,采用奇异值分解方法辨识Preisach模型参数,构造变幅值谐波输入信号满足持续激励条件;基于辨识模型设计逆模型前馈补偿Preisach迟滞非线性;通过压电平台检验该辨识、补偿方法。实验结果表明,奇异值分解法可有效辨识迟滞效应非线性模型,基于辨识结果的逆模型补偿可减小跟踪误差89.5%。The working performance of smart actuators is affected by the hysteretic effect nonlinearity.Thus,it is necessary to identify and compensate the hysteresis nonlinearity to achieve precise motion.By virtue of the singular value decomposition （SVD）method,the parameters of the Preisach model were identified,taking designed harmonic signals with varying amplitudes as inputs to satisfy PE condition.Then,based on the identified Preisach model,the model-inversion feedforward was designed to compensate the hysteresis nonlinearity.A piezoelectric stage was used to validate the proposed identification and compensation methods.The experimental results indicate that the hysteresis nonlinearity can be identified using the SVD method,and the tracking error is reduced by 89.5% using the model-inversion compensator.

关 键 词：智能作动器 Preisach迟滞效应 奇异值分解 逆模型 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]