首页 | 本学科首页   官方微博 | 高级检索  
     

基于支持向量机的传感器的非线性校正
引用本文:彭继慎,于精哲,夏乃钦. 基于支持向量机的传感器的非线性校正[J]. 计算机测量与控制, 2011, 19(1)
作者姓名:彭继慎  于精哲  夏乃钦
作者单位:辽宁工程技术大学电气与控制工程学院,辽宁,葫芦岛,125105
基金项目:辽宁省高等学校创新团队支持计划
摘    要:铜热电限的非线性影响到它的测温准确度和测温范围,针对这个问题提出了采用支持向A机(SVM),建立了铜热电阻传感器的逆模型进行非线性校正,并且与以往采用的BP网络和最小二乘校正方法进行了比较;结果表明,采用支持向f机的非线性校正方法的最大误差为±0.0287%左右,与BP人工神经网络取得的结果(最大误差为± 0.0523%左右)和最小二乘法取得的结果(最大误差为±0.0865%左右)相比,精度高于以上2种校正方法;同时,SVM方法有较好的泛化能力,在很大程度上提高了传感器的线性度,并且补偿曲线更顺滑,预测性更强,它为铜热电阻传感器的非线性动态补偿提供了一种新方法.

关 键 词:铜热电阻传感器  支持向量机  非线性校正

Approaches to Non-linearity Compensation of Copper Resistor Based on Neural Network in Temperature Measurement
Peng Jishen,Yu Jingzhe,Xia Naigin. Approaches to Non-linearity Compensation of Copper Resistor Based on Neural Network in Temperature Measurement[J]. Computer Measurement & Control, 2011, 19(1)
Authors:Peng Jishen  Yu Jingzhe  Xia Naigin
Affiliation:Peng Jishen,Yu Jingzhe,Xia Naigin(Faculty of Electrical and Control Engineering,Liaoning Technical University,Huludao 125105,China)
Abstract:The copper thermal resistance's nonlinearity affects its measurement accuracy and temperature measurement range,in order to resolve this problem,support vector machine has been applied to the development of the nonlinear calibration inverse model of copper thermal resistance sensor,this method is compared with some commonly used calibration methods,such as BP neural network and Least square method.The result of experiment shows that the maximum error of the nonlinear calibration method based on SVM is about...
Keywords:copper thermal resistance sensor  support vector machine(SVM)  nonlinear calibration  
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号