1,536
Views
1
CrossRef citations to date
0
Altmetric
Research Article

Optimization of Characteristics Polymer Composite Reinforced Kenaf and Jute Fiber Using Taguchi-Response Surface Methodology Approach

, & ORCID Icon

Figures & data

Figure 1. Kenaf fiber.

Figure 1. Kenaf fiber.

Figure 2. Jute fiber.

Figure 2. Jute fiber.

Table 1. Bisphenol-A epoxy resin property.

Table 2. CMNFR fabrication parameters.

Table 3. Design matrix of orthogonal array L273 for the experimental runs.

Table 4. Mechanical strength of CMNFR.

Figure 3. Vacuum infusion laminates steps.

Figure 3. Vacuum infusion laminates steps.

Figure 4. VARI process.

Figure 4. VARI process.

Figure 5. Drying process.

Figure 5. Drying process.

Figure 6. Specimen release.

Figure 6. Specimen release.

Figure 7. Specimen cutting results.

Figure 7. Specimen cutting results.

Figure 8. A schematic diagram of this research.

Figure 8. A schematic diagram of this research.

Table 5. ANOVA analysis of impact strength and flexural strength.

Figure 9. Plots of mean and SN ratios of composite material strength: (a) impact; (b) flexural.

Figure 9. Plots of mean and SN ratios of composite material strength: (a) impact; (b) flexural.

Table 6. CI calculated confidence interval.

Table 7. Comparisons of results of the experimental and Taguchi predicted value.

Table 8. ANOVA analysis of impact strength and flexural strength.

Figure 10. 3D RSM plot of the strength of natural fiber composite materials: (a) NaOH vs. post-curing temperature for impact strength; (b) NaOH vs. post-curing temperature for flexural strength.

Figure 10. 3D RSM plot of the strength of natural fiber composite materials: (a) NaOH vs. post-curing temperature for impact strength; (b) NaOH vs. post-curing temperature for flexural strength.

Figure 11. Plots of response optimization (D= composite desirability; d= individual desirability; High = highest value parameter; Cur= optimal current value of control parameter; Low = lowest value parameter, y= response parameter, A= NaOH concentration; B= post-curing temperature; C= fiber type).

Figure 11. Plots of response optimization (D= composite desirability; d= individual desirability; High = highest value parameter; Cur= optimal current value of control parameter; Low = lowest value parameter, y= response parameter, A= NaOH concentration; B= post-curing temperature; C= fiber type).

Figure 12. Micro photos of composite specimens; (a) impact test specimen, (b) flexural test specimen.

Figure 12. Micro photos of composite specimens; (a) impact test specimen, (b) flexural test specimen.