基于Wiener模型的传感器动态非线性辨识研究  

Study on Identification of Dynamic Nonlinear Sensor Based on Wiener Model

作　　者：雷超[1] 韩华亭[1] 刘滔[1] 

机构地区：[1]空军工程大学防空反导学院,西安710051

出　　处：《计算机测量与控制》2014年第7期2331-2334,共4页Computer Measurement &Control

摘　　要：针对实际测量中传感器存在较大非线性的缺点,提出利用改进型Wiener模型描述传感器动态非线性模型;将Wiener模型的动态线性环节和静态非线性环节分别利用Laguerre函数和最小二乘支持向量机进行辨识,最终实现传感器模型的建立;通过仿真实验验证比较不同方法的辨识误差与速度,最终结果表明该方法在非线性动态传感器模型辨识方面具有明显的速度和精度优势。In the view of the actual measurement sensor is nonlinear, The method of using modified Wiener model to describe the dynamic nonlinear sensor model is proposed in this paper. Respectively using Laguerre function and least squares support vector machine for the identification of dynamic linear part and nonlinear static link of the Wiener model. Through simulation experiment to compare the speed and identification error of different methods, the results show that this method in the identification of nonlinear dynamic sensors has an obvious advantage.

关 键 词：传感器 WIENER模型 最小二乘支持向量机 

分 类 号：TP212[自动化与计算机技术—检测技术与自动化装置]