The evaluation index of the data is the visual analysis chart of the influence of various factors on the thickness of the sample, which can clearly show the most influential factors among the various factors, and the results are shown in the figure. The thickness of the surface layer after nitrocarburizing determines the wear resistance, hardness and other properties of the workpiece. Therefore, it is necessary to select the level with the largest thickness of the infiltration layer under various factors. It can be seen from the figure that the prolongation of the nitriding time has a positive effect on the thickness of the three parts of the sample. Taking the thickness of the sample at 1 / 2R as the standard, the influence of the gas proportion on the index is slightly less than that of the nitriding time, but at the same time, increasing the ratio of NH3 to CO2, i.e. increasing the nitrogen content, also has a certain positive significance on the increase of the thickness of the layer. In a certain range, the increment of active carbon atoms penetrating into the materials is more than that of active nitrogen atoms, and carbon atoms are absorbed by nitrides to form nitrogen carbon compounds, which makes the nitride content decrease and the thickness of compound layer decrease. With the increase of temperature and nitrogen content, the penetration of carbon atoms promotes the infiltration of active nitrogen atoms, which makes the thickness of compound layer increase rapidly.

Although the range analysis method has less calculation and is easy to use, the influence degree can be obtained by comparing the R values at different levels among the three parts of lzqt600-3 ductile iron profile, but the range analysis can not estimate the size of the error between the test period and the test results. In short, the range analysis method can not be sure that the error is caused by the different factor level or the error in the test, and the accuracy of the analysis results is difficult to determine. Variance analysis can make up for this defect.
The confidence level of α = 0.05 is selected for the analysis of variance of the results of the orthogonal test. The critical value f0.95 (2,2) can be obtained by looking up the F distribution table to judge the significance level of each factor.
ANOVA was used to test the significance of the permeability layer thickness data at the edge, 1 / 2R and the center. The calculation results are summarized in table 4-3. From the table, the factors t and T are more significant, but the factor s is not significant, which is consistent with the result of range analysis.
Through the range analysis of orthogonal test results, the best level under the most significant factors is selected, which is that the level change will directly lead to the difference in the significance of the indicators, but there is no limit for the level selection under the insignificant factors. The most appropriate selection is based on the analysis of variance and the reasons of reducing the capital investment of raw materials and equipment in actual production The level of. Therefore, the best scheme to take the penetration layer thickness of lzqt600-3 ductile iron profile as the index is: t3t3s
3, that is, the temperature is 580 ℃, the time is 6 h, the ratio of NH3 to CO2 is 4:1.