PSI - Issue 77

Alireza Shadmani et al. / Procedia Structural Integrity 77 (2026) 221–228 Shadmani et al. / Structural Integrity Procedia 00 (2026) 000–000

3

223

(a)

(b)

Fig. 1: (a) Normal and (b) shear stress distributions resulted from raindroplet impact

To characterize the mechanical response of the PU coating, a hyper-viscoelastic material model was implemented in the finite element model, where the material’s stress-strain behavior under uniaxial loading condition must be char acterized. Fig. 2(a) shows the stress-strain curve of the PU coating obtained from a uniaxial tensile test (Jespersen et al. (2023)). The material exhibits a non-linear stress-strain response, typical of elastomeric polymers, with a pronounced strain-hardening behavior at higher strains. To capture this behavior, a material model calibration was performed using ABAQUS to fit the experimental data to a hyperelastic model. The result of calibration shows that the Marlow model can e ff ectively capture the stress-strain behavior of the material compared to the other hyperelastic models available in ABAQUS, as shown in Fig. 2(b). The Marlow model does not require a predefined functional form relating strain energy density to strain invariants or stretch ratios, compared to other hyperelastic models, such as Ogden, Mooney Rivlin, etc., which have several coe ffi cients to formulate the strain energy density function. Instead, it uses direct interpolation of experimental data. It is assumed that the strain energy density is only a function of the first invariant of the strain tensor ( I 1 ), and is expressed as:

W = W dev ( ¯ I 1 ) + W vol ( J )

(1)