SVM法在虚拟仪器非线性校准中的应用

Application of SVM in Calibration of Nonlinear Error in VI

在构建虚拟仪器系统时,非线性误差是值得考虑的一个问题,本文主要是为了解决这方面的问题。文中首先介绍了虚拟仪器非线性校准的基本原理,接着分析了支持向量机回归模型,并通过采用遗传算法来优化支持向量机的参数,以便建立基于优化的支持向量机非线性校准模型。最后,通过构建一个简单的虚拟仪器数据采集校准系统的实验来验证这种校准方法的有效性,实验结果表明,这种方法是可行的,而且测量精度有了显著提高。

The nonlinear errors would be considered in constructing VI systems. The paper was to solve the problem. Firstly the fundamental of nonlinear calibration was introduced; secondly the support vector regression algorithm model was analyzed and the parameters of SVM were optimized by GA in order to build a nonlinear calibration model based on an improved SVM; lastly the validity was validated by an experimentation of building a simple VI DAQ and Calibration system. The result indicated the method was feasible and the measure precision was increased largely.