Rolling bearing data robustness of judgment

Rolling bearing data robustness: 1. The frequency histogram of data frequency histogram can reflect the distribution characteristics of the data, in statistics, when the sample size is more than a certain number of the number of training data, can be thought of data distribution for asymptotic normal distribution and normal distribution or is the data on both sides of the histogram for middle high low symmetric or nearly symmetric graphics; If the histogram of the data conforms to normal distribution or asymptotic normal distribution, can be thought of as the data is reasonable, there is no discrete values; If the histogram of the data is not in conformity with the asymptotic normal distribution, normal distribution or discrete values exist in the data. Use data histogram with a normal distribution or asymptotic normal distribution characteristics can be preliminary judging the rationality of the data. 2. Data of normality test data using histogram asymptotic normal distribution and normal distribution or similarities, can preliminary judgment data are discrete values; Then further test the performance data of normality, deviating from the normal distribution of data is discrete values. This method can be used to judge data discrete values, lay a foundation for the subsequent data robust treatment. 3. Rolling bearing performance data of the frequency histogram frequency histogram of the rolling bearing performance data can reflect the distribution of the rolling bearing performance data. In statistics, when the sample size is more than 800 the number of training data, can be thought of data distribution for normal distribution or asymptotic normal distribution, which is the data on both sides of the histogram for middle high low symmetric or nearly symmetric graphics; If rolling bearing performance data histogram conform to the normal distribution or asymptotic normal distribution, discrete values can be thought of rolling bearing performance data does not exist; If rolling bearing performance data do not accord with normal distribution or asymptotic normal distribution, can be thought of as discrete values exist in the rolling bearing performance data, using the data histogram asymptotic normal distribution and normal distribution or can preliminary judgment performance data of the rolling bearing the characteristics of robustness. 4. Rolling bearing performance data of normality test using rolling bearing performance data histogram asymptotic normal distribution and normal distribution or the similarity of the preliminary judging rolling bearing performance data are discrete values, and then further tested rolling bearing performance data of normality, deviates from the value of the normal distribution is the discrete values. Using this method can determine scope, the discrete values of rolling bearing performance data for subsequent rolling bearing performance data sound of lay the foundation.