间接测量的神经网络解析模型及灵敏系数的计算
详细信息 本馆镜像全文    |  推荐本文 | | 获取馆网全文
摘要
针对无测量模型的间接测量不确定度评定问题,以径向基函数神经网络模型为基础,提出了建立间接测量解析模型和利用该模型计算灵敏系数的方法,导出了间接测量模型的解析表达式和灵敏系数的计算公式。仿真结果表明:神经网络解析模型的建模方法及灵敏系数的计算方法具有较高的精度,从而保证了间接测量结果的估计精度及测量不确定度的评定精度。
A new method is developed to solve the problem of evaluating the measurement uncertainty in indirect measurement with no given measurement model.An analytical model based on neural network using Radial Basis Function is set up.The expression of indirect measurement model and the calculation formula of the sensitivity coefficient are derived.The simulation results showed that the analytical model based on neural network and the method of sensitivity coefficient calculation has high precision,thus effectively guarantees the precision of measurement result and its uncertainty estimation in indirect measurement.
引文
1施文康.测试技术[M].上海:上海交通大学出版社,1996.
    2 ISO/TAG 4/W G 3.G u ide to the express ion of uncerta in ty in m easurem en t[S].Sw itzerland,1993.
    3李敏远,都延丽.一类过程控制对象的神经网络建模及仿真[J].系统仿真学报,2003,15(11):1 533~1 536.
    4杨会菊,彭熙伟,谭日飞.基于人工神经网络的液压泵工作特性建模[J].仪器仪表学报,2003,24(增刊):617~618.
    5徐秉铮.神经网络理论与应用[M].广州:华南理工大学出版社,1994.
    6王碧泉.模式识别理论、方法和应用[M].北京:地震出版社,1989.
    7 Chen S,Cow an C F N,G ran t P M.O rthogona l least squares learn ing a lgorithm for rad ia l bas is function netw orks[J].IEEE T ransactions on N eura l N etw orks,1991,2(2):302~309.

版权所有:© 2023 中国地质图书馆 中国地质调查局地学文献中心