摘要
误差修正技术是提高测量仪器精度的重要途径 ,本文详细介绍了具有时变系统误差与随机误差的非稳定动态测量过程实时误差预测修正方法 ,即 :用标准量插入法实时分离误差 ,采取神经网络建模法对分离出来的误差进行实时建模 ,跟踪误差的变化 ,对其进行预报修正。通过自行研制的动态测角仪对这种误差修正技术进行了验证 ,使该测角仪的测量精度提高近
A major method to improve measuring instrument precision with error correction is described. This paper details the technology of real-time error prediction correction for time-variant systematic error and random error in non-stationary dynamic measurement. That is to separate the errors real-timely with standard value interpolation, and then to predict and correct them with neural network model following their errors. The error correction technology has been verified in the experiment with an angle measuring instrument. By the error correction method the instrument precision increases nearly by ten times than before.
出处
《计量学报》
EI
CSCD
北大核心
2001年第2期81-86,共6页
Acta Metrologica Sinica
基金
国家自然科学基金 !(5 973 5 12 0 )
