Analysis about the rolling bearing performance data

Discrete data and the error of some conditions brings a lot of difficult to data analysis, and severe damage data and error analysis results, under ideal conditions the phenomenon do exist. The median is - in data analysis A very robust parameters, when the data is not discrete data, its value is close to uniform value of the data set. When data are discrete data, uniform values will be away from the real value, and the median don't deviate from the real value, expounds the median is the steady. Thus, can mean the difference between the median war to identifying data can exist in the discrete data. Application of improved HuberM way dispose of data, even if the value is more and more close to the median, clarify data there is no discrete data; Even if the value is more and more deviated from the median of data is discrete data. Significant degree according to the data request, on the basis of modern statistics, introduce value range of 0 ~ 0. 1. Improved HuberM method to disposal the data, the discrete data affects overall data reduced, significant progress, provide strong guarantee for the progress of the data analysis. Some progress perfect theory of demand, such as median and uniform value of similar degree, such as data how significant is sure, is still a lack of theoretical basis. Based on improved HuberM method for rolling bearing vibration data and A, B C disposal result of rolling bearing friction torque data, to be able to see the significant degree of 0 ~ 0. Within 1 C of rolling bearing vibration data variance change complex, A and B of rolling bearing friction torque data variance monotone decreasing.