The national standard GB/T7141-2008 specifies that when plastic materials are compared at a single temperature, the average of the measured performance data of each material at each exposure time should be compared using analysis of variance. Analysis of variance was performed using each set of parallel measurements for each of the compared materials. It is recommended to use the F statistic with a 95% confidence level to determine the validity of the ANOVA results. When comparing materials at a range of different temperatures, the following methods should be used to analyze the data and estimate the exposure time required to reach a predetermined level of performance change at lower temperatures.
1. Draw a function curve of all the exposure time at the temperature to the measured performance, the abscissa is the logarithm of time, and the ordinate is the measured performance value;
2. Use regression analysis to determine the relationship between the logarithm of exposure time and the measured performance, and use regression equations to determine the exposure time required to achieve a predetermined level of performance change. An acceptable regression equation should satisfy r2 ≥ 80%. The residual relative to the thermal aging time (the performance retention value predicted by the regression equation minus the measured value) curve should be a random distribution;
3. A plot of the logarithm of the time required to reach the predetermined level of performance change and the inverse of the absolute temperature used for each exposure (1/T, temperature unit K);
4. Determine the time to achieve this performance change at the preselected temperature agreed by all interested parties using a functional equation of the logarithm of the time required to reach the specified level of performance change and the reciprocal of the absolute temperature;
5. Use the 95% confidence interval of time to calculate the amount of change in specific performance. The standard error is obtained by regression analysis of the estimated time at a certain temperature. Regression analysis is available in most application software packages, and the 95% confidence interval can be determined by the calculation time ± (2 × standard error of estimated time).
Lens Cloth,Eyeglass Cleaning Cloth,Silk Screen Printing Lens Cloth,Digital Printing Cloth
Shenzhen Xinmingcai Clothing Co,.Ltd , https://www.xmcsz.com