1. Vukašin Slavković, Univerzitet u Kragujevcu, Fakultet inženjerskih nauka, Sestre Janjić 6, Serbia
2. Nenad Grujović, Fakultet inženjerskih nauka Univerziteta u Kragujevcu, Serbia
3. Aleksandar Dišić, Univerzitet u Kragujevcu, Fakultet inženjerskih nauka, Sestre Janjić 6, Serbia
4. Vladimir Dunić, Univerzitet u Kragujevcu, Fakultet inženjerskih nauka, Sestre Janjić 6, Serbia
5. Vladimir Milovanović, Univerzitet u Kragujevcu, Fakultet inženjerskih nauka, Sestre Janjić 6, Serbia
Uniaxial compression tests at wide range of temperature and strain rates have been performed to characterize PLA shape-memory polymer mechanical response. The effects of strain rate and temperature on the inelastic response of a PLA have been studied. Deformation tests in uniaxial compression to strains of -0.68 were conducted over a range of temperatures and strain rates of -0.0001 and -0.001 providing nearly isothermal test conditions and thus documenting the temperature dependence yield, strain softening and strain hardening. A temperature and strain rate-dependent material model has been developed and implemented in a finite element code PAK to predict such a mechanical behavior. The experimental results have been simulated using a fully 3D constitutive model of the large strain inelastic response of shape-memory polymer PLA finite element analysis. The strain rate and temperature dependence of initial yield is included in the material model as well as temperature dependence and its associated hardening. The Parallel Network Model (PNM) an advanced modular constitutive model framework allows for specific isotropic hyperelastic and viscoplastic components to be combined to capture the experimentally observed response of PLA. The proposed Parallel Network Model (PNM) consists of two parts: elastic-viscoplastic response due to intermolecular resistance denoted part A and an entropic hyperelastic response due to re-orientation of molecular chains called part B. Both parts are developed within a framework for finite element strains. Excellent agreement between simulation and experiment is found when the stress-strain curves are well predicted over a range in strain rate and temperature.