PSI - Issue 78

Anastasios Drougkas et al. / Procedia Structural Integrity 78 (2026) 2102–2109

2108

Table 3. Analysis results for di ff erent combinations of contributing mechanisms. AR VF

CC

λ

49.04

✓

✓ ✓ ✓

✓

8.06 1.56 1.31

✗

✗ ✗ ✗

✓

✗

✗

3.2. Sensitivity analysis

A brief sensitivity analysis was performed for evaluating the influence of various geometrical, mechanical and loading parameters on the piezoresistive performance of the composite as predicted by the proposed model. The scope of the investigation includes: the Young’s modulus of the paste and aggregates ( E m and E a ), the crack closure stress ( σ cc ), the electrical resistivity of the paste and aggregate ( ρ m and ρ a ), the volume fraction of the pores, cracks and aggregate ( ω p , ω c and ω a ) and the aspect ratio of the pores and cracks through adjustment of their lengths ( a p 1 and a c 1 = a c 2 ). All these parameters are characterised by high uncertainty and technically demanding characterisation methods. When modifying volume fractions, the volume fraction of the paste matrix was adjusted in order for the sum of all fractions to be equal to 1 . 0. Each parameter is varied individually within the range [0 . 5 , 2 . 0]. For presentation of the results, the parameters are normalised over their reference value found in Table 1 and Table 2 and presented with a hat operator ( ˆ □ ). The resulting λ is similarly normalised over the reference value presented in Table 3. The sensitivity analysis results are illustrated in Fig. 2. The response of the composite was found to be very sensitive to σ cc , as was expected, due to a greater proportion of cracks closing for the same load when σ cc is lowered. Similarly, higher ω c resulted in higher λ due to a greater absolute volume of cracks closing for the same load. Conversely, the composite was nearly completely insensitive to changes in ω a , despite noted sensitivity to ρ a . This indicates that while for ordinary resistivity the aggregate e ff ectively functions as an insulating phase and does not participate in changes to the composite’s response, lowering its resistivity activates its participation in the piezoresistive mechanism. The calculated λ presents a nearly linear proportionality to both E m and E a . Increased composite sti ff ness results in larger strains in the more deformable inclusions, leading to more pronounced crack closure. Higher ρ m resulted in increased λ . This result appears counter-intuitive, but is consistent with the observation that more conducting materials are less piezoresistively sensitive due to their reduced capacity to form additional conductive networks. Lowering the porosity ω p and increasing the aspect ratio a p 1 / a p 2 of the pores along the direction of loading resulted in an increase of λ , as these geometric changes lead to a decrease in the role of the pores as an insulating phase. Finally, increasing a c 1 / a c 3 led to an increase in λ due to the increased tendency of the more slender cracks to close under compression.

D

E

c

E

m

p

E

a

a

cc

a

c 1

m

a

p 1

a

1RUPDOLVHG SDUDPHWHU

1RUPDOLVHG SDUDPHWHU

Fig. 2. Sensitivity analysis results: a) material properties; b) geometric properties.