The main objectives of the present dissertation are the validation of the Palmyra software method to simulate the linear mechanical and linear thermo-mechanical behavior of particle reinforced polypropylene PP(H) based composite systems, as well as the verification of modeling parameters affecting the accuracy of the simulation results. The verification of the simulation accuracies is based on composite properties measured under various testing conditions (temperature, loading velocity). Based on a comprehensive set of experimental results comprising the characterization of the constituent properties and the characterization of the composite structures, it was shown, that the reinforcing efficiencies of different particles (glass-beads, talcum) are to a large extend determined by the specific particle/matrix interfaces. These findings were also confirmed by a simulation study considering composite microstructure unit models of different accurately modeled particle surfaces. In contrast, hardly any effect of the particle arrangements could be found when simulating the macroscopic composite properties of various multi-inclusion composite reference volume models. The amount of single data - the study considered in total 256 independently determined and simulated composite properties - and the achieved simulation accuracy allowed for a good verification of the simulation accuracy. The Palmyra software method computes the composites mechanics and thermo-mechanics in the linear regime to accuracies better than 5%. Moreover, a simulation method for the nonlinear composite mechanics is discussed, which indirectly allows the determination of the initiation and propagation of irreversible damage mechanisms at higher macroscopic stress loads. Simulated results in comparison with experimental data determine a critical macroscopic stress state at which the damage initiation of the particle/matrix interfaces occurs.