基于支持向量机的传感器动态补偿新方法

提出一种基于支持向量机(SVM)的传感器动态补偿新方法,给出相应的补偿过程及学习算法。与常用的神经网络补偿方法比较,其优点是明显的。它采用了结构风险最小化准则,在最小化样本误差的同时减小模型泛化误差的上界,提高了模型的泛化能力;而且将学习算法转换为求解二次规划问题,使得在整个学习过程中有且仅有一个全局极值点,确定了所构造补偿器的唯一性。仿真和实验结果均表明,经过SVM动态补偿器可极大地缩短传感器达到稳定的时间,应用SVM模型对传感器进行动态补偿方法有效。

Dynamic Compensation Method for Sensors Based on Support Vector Machine

A new method based on support vector machine model to correct dynamic measurement errors of sensors is presented,the corresponding design processes and learning algorithm are given.Compared with conventional compensation methods,this method possesses prominent advantages:over fitting is unlikely to occur by employing structural risk minimization criterion to minimize the errors at the samples and decrease simultaneously the upper bound of the predict error of the models.Furthermore,the global optimal solution can be uniquely obtained owing to the learning algorithm converts machine learning into quadratic programming.Simulations and experimental result show the dynamic compensation method is effective and precision.