摘要
压电陶瓷驱动器电压位移之间的非线性特点严重影响着它的位移控制精度,建立压电陶瓷驱动器非线性模型是纳米级微位移测控中的关键环节。采用支持向量机回归的方法,通过引入核函数和损失函数将非线性回归转化成线性问题并提高回归精度,建立了一种新的压电陶瓷驱动器外环非线性模型,并就模型的准确性与其它建模方法进行了比较。试验证明,所建的基于支持向量回归的压电陶瓷驱动器非线性模型很好的描述了压电陶瓷驱动器外环非线性特点,误差控制在2%以内,并且建模过程简单,准确性高。
This paper is to model the nonlinearity of piezoelectric actuator. Due to the inherent nonlinearity, piezoelectric actuator always causes position error in the open-loop system and instability in the closed-loop system; to remedy this problem, a new nonlinear model of piezoelectric actuator using support vector machine for regression is presented by introducing a kernel function and a loss function to turn nonlinear regression into linear regression. For validating the nonlinear model based on support vector machines for regression, this paper also establishes another nonlinear model using the polynomial curve fitting based on the theory of least squares, and modeling experiments on nonlinear characteristic of a piezoelectric actuator are carried out; the comparison between presented model simulation results and experiment value shows that the nonlinear model based on support vector machine for regression commendably describes the nonlinear characteristic of piezoelectric actuator, the maximal error is less than 2%, the process of modeling is explicit, and the computation is simple.
出处
《传感技术学报》
CAS
CSCD
北大核心
2006年第4期1079-1082,共4页
Chinese Journal of Sensors and Actuators
基金
国家自然科学基金资助(50205025)
浙江省自然科学基金资助(502021)
关键词
压电陶瓷驱动器
非线性建模
支持向量机
支持向量回归
piezoelectric actuator
nonlinear model
support vector machine
support vector machine for regression
