PSI - Issue 14

S Usha Rani et al. / Procedia Structural Integrity 14 (2019) 142–149 Author name / Structural Integrity Procedia 00 (2018) 000–000

147

6

Table 3: Taguchi based grey relational analysis for all trials.

Temperature (°C)

Time (minutes)

Hardness (HRC)

Data pre processing

Deviation sequence

GR coefficient

GR-grade

Rank

400 400 400 450 450 450 500 500 500

15 30 60 15 30 60 15 30 60

42.8 44.5 40.8 37.6 37.4 34.4 31.6 31.8 39

1

0

1

1

2 1 3 5 4 6 7 9 8

1.154545 0.818182 0.527273 0.654545 0.509091 0.236364 -0.01818

-0.15455 0.181818 0.472727 0.345455 0.490909 0.763636 1.018182

1.447368 0.733333 0.514019 0.591398 0.504587 0.395683 0.329341 0.333333

1.447368 0.733333 0.514019 0.591398 0.504587 0.395683 0.329341 0.333333

0

1

3.3.1. Taguchi based Grey relational analysis The Taguchi method is a DoE method which is combined with grey relational (GR) analysis to optimize the performance characteristic (Raghuraman et al. (2013)). This method involves the following sequence: data preprocessing, deviation sequence, GR coefficient and GR grade as given in the Table 3. For normalizing the data, hardness (response) is selected as ‘larger-the-better’. The GR coefficient was calculated with the pre-processed sequence. The GR grade is calculated by averaging the grey relational coefficient corresponding to each trial. The average of the grey relational grade is 0.649 and the mean of grey relational grade is shown by the horizontal line as shown in Fig. 6. The higher grey relational grade represents most optimized combination. The trial 2 has the best performance characteristics among nine experiments as it has the highest grey relational grade. Thus, the combination of tempering temperature of 400°C and tempering time of 30 minutes were optimized as per the Grey relational grade.

Fig. 6: Effect of process parameter on Grey relational grade.

The Taguchi method using MINITAB software shows the main effect of the tempering parameter on hardness. The response table for mean and interaction plot is given in Table 4. The tempering temperature at 400°C has larger mean effect than time. That is, the line connecting the mean responses for temperature from 400°C to 500°C has a steeper slope than the line connecting the mean responses for tempering time from 15 to 60 minutes. Although the tempering temperature appears to affect the tensile strength and fatigue strength more than tempering time, it is very important to look at the interaction effect because an interaction among factors can magnify or nullify a main effect. The mean (hardness) at tempering temperature of 400°C is greater at 30 minutes of tempering time. However, it can