期刊文献+

基于RBF神经网络集成-模糊加权输出的数字温度传感器误差补偿 被引量:20

Error compensation for digital temperature sensor based on RBF neural network ensembles-fuzzy weighing output
下载PDF
导出
摘要 数字温度传感器存在非线性误差,在高精度测温系统中需要进行误差补偿。提出了一种基于径向基函数神经网络集成-模糊加权输出(RBFNNE-FWO)的数字温度传感器误差补偿方法:首先根据数字温度传感器的误差特征,提取特征阈值,构造三个相互独立的成员RBFNN;考虑到成员网络之间边界误差补偿问题,构建一种RBFNN集成输出权值模糊调节器,获得RBFNN集成输出权值,从而完成数字温度传感器的全量程误差补偿。与多种方法的比较仿真实验表明,这种RBFNNE-FWO方法的性能最佳、各成员网络边界误差最小,补偿后的数字温度传感器误差减少了两个数量级,大大提高了测温准确度。 Nonlinear error compensation for digital temperature sensor is necessary in high accurate temperature measurement system.An error compensation method based on radial basis function neural network ensembles-fuzzy weighing output(RBFNNE-FWO) is proposed.The characteristic threshold values are obtained according the error characteristics of the digital temperature sensor and then three independent member RBFNNs are founded.Considering the compensation issue for the boundary errors among the member networks,a fuzzy weighing adjuster is established,and then the output weights of the RBFNN ensembles are obtained.Finally,the error compensation for the whole measuring range of the digital temperature sensor is realized.Simulation experiments were conducted and the proposed method was compared with different kinds of other methods;experiment results show that the performances of the RBFNNE-FWO method is the best,its boundary error is the smallest,and with the RBFNNE-FWO method the measuring error of digital temperature sensor decreases by two orders of magnitude.
出处 《仪器仪表学报》 EI CAS CSCD 北大核心 2011年第7期1675-1680,共6页 Chinese Journal of Scientific Instrument
基金 商务部优化机电和高新技术产品进出口结构(No.财企[2007]301号) 湖南师范大学青年优秀人才培养计划(No.ET61107)资助项目
关键词 数字温度传感器 误差补偿 径向基函数神经网络集成-模糊加权输出 边界误差 digital temperature sensor error compensation radial basis function neural network ensembles-fuzzy weighing output boundary error
  • 相关文献

参考文献15

二级参考文献72

  • 1戴先中,殷铭,王勤.传感器动态补偿的神经网络逆系统方法[J].仪器仪表学报,2004,25(5):593-596. 被引量:26
  • 2王正群,陈世福,陈兆乾.并行学习神经网络集成方法[J].计算机学报,2005,28(3):402-408. 被引量:36
  • 3刘清.基于CMAC的非线性逆滤波器补偿热敏电阻的测量误差[J].仪器仪表学报,2005,26(10):1077-1080. 被引量:3
  • 4李凯,黄厚宽.小规模数据集的神经网络集成算法研究[J].计算机研究与发展,2006,43(7):1161-1166. 被引量:10
  • 5徐昕 李涛 等.MATLAB工具箱应用指南[M].北京:电子工业出版社,2000..
  • 6司端锋.基于BP网络的传感器特性补偿新方法的研究[J].传感器技术,1999,.
  • 7SIEMENS.S7-200用户指南[M].,1999..
  • 8HUANG J Q, HOU L SH, JI P. Design and experiment study methods for dynamic compensated digital filter and its application [ C ]. Proceeding of the 9th IEEE International on Instrumentation and Measurement Technology, Metropolitan, NY, USA, 1992:448-452.
  • 9WANG L, VOLKER H. Improving dynamic performance of temperature sensors with fuzzy control technique [ C ]. Proceeding of the IEEE International Symposium on Industrial Electronic, Xian, China, 1992(2) :872-873.
  • 10TAKENS K On the numerical determination of the dimension of an attractor [A]. Dynamical Systems and Turbulence, Lecture Notes in Mathematics, 1981,898: 230-241.

共引文献353

同被引文献278

引证文献20

二级引证文献161

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部