Next, analyze the frequency distribution of rolling bearing vibration performance data, and the results are shown in Figure 7-2. The following further analyzes the comparison between the rolling bearing vibration data and the normal distribution, and the results are shown in Figure 7-3.

The following analysis of the data changes. Therefore, when the data does not obey the normal distribution or deviates far from the normal distribution, it is considered that the vibration performance of the rolling bearing has changed. It can be seen that the vibration performance data of rolling bearings is discrete. This variety and complexity shows that the actual rolling bearing vibration performance data does not obey the normal distribution.

According to Figure 7-3, it can be seen that the smaller and larger values ​​of the rolling bearing vibration performance data deviate significantly from the normal distribution. According to the histogram of the vibration frequency of the rolling bearing at different time periods in Figure 7-2, it can be seen that at different time periods, some of the rolling bearing vibration data are wheel bearing kits factory unimodal, some are bimodal; some are left and some are right. From the perspective of the performance degradation of rolling bearings, it means that the bearing performance has changed.

Variations are diverse, showing the complexity of variation.Robustness analysis of experimental data of needle roller bearings: According to the assumptions of modern statistics, the ideal rolling bearing vibration performance data should obey the normal distribution. The specific characteristics are that some are upper and some are lower; some have more small data, and some have less data; some have larger data and some have less data.