PSI - Issue 41

N.A. Fountas et al. / Procedia Structural Integrity 41 (2022) 638–645 Author name / Structural Integrity Procedia 00 (2019) 000–000

642

5

By examining the results concerning the response of UTS (MPa), it can be observed that the highest value of 17.420 MPa is achieved for layer height equal to 0.1 mm, nozzle temperature 180 o C, raster deposition angle 0 deg., and printing speed 30 mm/sec (1 st experiment, Table 2). The second highest result for UTS (16.920 MPa) is observed in both 7 th and 10 th experiments. However, these two experimental runs suggest different settings for process-related variables under study. Moreover, at an early stage it can be asserted that strong interactions among

the levels of process-related parameters exist. 3. Statistical analysis and regression modeling

Reliable outputs by studying interactions among process-related parameters are offered by contour plots. Prior to the generation of contour plots, statistical investigation involving analysis of variance (ANOVA) and regression modeling should be conducted. MINITAB ® R17 statistical analysis environment was used for interpreting the results in terms of percentage contribution, significance checking and model fitting. It was found that the full quadratic model was the one that fits best the experimental results, thus accurately explaining the variation of FFF experiments. ANOVA decomposes the error per each variable on the overall error when a mathematical model is fitted on the results. Eq. 1 represents the generalized full quadratic regression equation implemented to generate the model for UTS (MPa).

k

k

i i       ii i i x x b x 

y

ij i j x x

0   



(1)

1

1

i

i

i j 





To check significance, F and p indicators are examined in the ANOVA table (Table 3). Probability of F greater than computed F owing to noise, is indicated by p -value.

Table 3. ANOVA results for UTS (MPa) with regard to FFF-related parameters. Source DF Seq. SS Contribution Adj. SS

Adj. MS 16.5928 29.1797 6.7489 96.8722 0.0028 3.8433 0.4399 7.8175 2.7837 3.3142 5.147

F-Value

P-Value

Model Linear

13

215.706 184.557

97.79% 83.67% 2.55% 2.99% 76.28% 1.84% 5.11% 0.12% 3.85% 1.14% 9.01% 3.96% 2.17% 0.16% 2.40% 0.13% 0.19% 2.21% 100.00%

215.706 116.719

13.6

0.011 0.005 0.109 0.078 0.001 0.964 0.148 0.581 0.065 0.205 0.176 0.056 0.948 0.104 0.651 0.587 0.19

4 1 1 1 1 3 1 1 1 6 1 1 1 1 1 1 4

23.91

5.627 6.599

5.147 6.749

4.22 5.53

LT NT DA

168.263

96.872

79.39

4.068

0.003 11.53

0

PS

Square

11.264

3.15 0.36 6.41 2.28 2.72 7.11 2.49 4.41 0.24 0.35 0

NT 2 DA 2 PS 2

0.258 8.483 2.523

0.44

7.818 2.784

2-Way LTxNT LTxDA LTxPS NTxDA NTxPS DAxPS

19.885

19.885

8.73 4.79

8.676 3.036 0.006 5.386 0.291 0.425 4.881

8.676

3.0364

0.363 5.288 0.291 0.425 4.881

0.006

5.3862 0.2908 0.4249 1.2202

Error Total

17

220.587