The friction torque data of A and B rolling bearings The variance decreases monotonically. The degree of significance is determined according to the data request. According to modern statistics, the recommended value range is 0~0. According to the improved HuberM method on the vibration data of C rolling bearing and the friction torque data of A and B rolling bearings, it can be seen that the variance of the vibration data of C rolling bearing is complicated within the significance level of 0 to 0.

When the data has no discrete data, its value is close to the uniform value of the data set. This phenomenon does exist under ideal conditions.Discrete data and the errors of some conditions bring great difficulties to data analysis, and even severely destroy the data and bring wrong analysis results.. When the data has discrete data, the uniform value will be far away from the true value, and the median does not deviate from the true value, clarifying that the median is robust.

The progress analysis provides a reliable guarantee. For data processed by the improved HuberM method, the impact of discrete data on the overall data is reduced, and the significance degree is improved.1. Some of these theories need to be improved and perfected, such as the similarity between the median and the average value, and the degree of significance of the data, which still lack theoretical basis.

Therefore, the difference between the median and the mean value can be used to judge whether there is discrete data in the data. The improved HuberM method is used to process the data. The median is a very robust non-standard bearings parameter in data analysis. If the average value is getting closer and closer to the median, it means that there is no discrete data in the data; if the average value is more and more deviating from the median, it means that there is discrete data in the data