Processing of Gross Error in Small Samples Based on Measurement Information Theory
Zhi Li · Mechanical Engineering & Automation · 2009
In the case of small sample,the traditional methods of error processing based on statistical theory can not correctly distinguish the gross error.This paper uses a new method based on measurement information theory to process the gross error.Firstly,it introduces the concepts of order statistics,order and information entropy into the new method,then it uses this method to find the gross error in the measurement data.Three methods are used to process and analyze the experimental data,and the result shows that the method based on measurement information theory is better at processing gross error in small samples than the methods based on statistical theory.