PSI - Issue 64

Niloofar Heirani et al. / Procedia Structural Integrity 64 (2024) 6–13 Author name / Structural Integrity Procedia 00 (2019) 000–000

11

6

Fig. 8. Response of a sensor sample with 15% CF, GU with 20% SF, and mixed with Hobart mixer following method A over 30 compressive loading cycles

Since the sensor GFs are well above traditional strain sensors and therefore adequate, the Taguchi optimization was performed on the sensing repeatability. For better repeatability, the difference between GFs from the first and last cycles was minimized; therefore, the signal-to-noise ratio (S/N) variance index was described by the ‘the smaller the better’ quality characteristic (Roy, 2010): = − 10×log 10 ∑ 2 (1) The statistical software MINITAB was used to calculate the S/N measures. Figure 9 presents the main effect plot for all five factors and their corresponding levels. To minimize the uncontrollable factors, a higher S/N is desired. Control factors were then ranked by significance. The outcome recommends the following for improved repeatability: 1) higher CF contents; 2) centrifugal mixers; 3) supplementary cementitious materials (silica fume and slag); 4) fibers added at the beginning of the mixing procedure; and 5) higher mixing speeds.

Fig. 9. Main effect plot for the Taguchi optimization of CFCP sensor response repeatability

4. Concluding remarks In this preliminary study, CFCP cementitious sensors were analyzed using microscopy and the Taguchi method to rank fabrication factors based on their impact on fiber breakage and dispersion as well as sensor response repeatability. The findings revealed a clear hierarchy of influence: higher CF content significantly enhanced repeatability, while mixing with a centrifugal mixer minimized fiber breakage and improved uniformity and dispersion, further enhancing