Three-dimensional (3D) printing technology has revolutionized manufacturing processes by enabling the production of complex geometries with high accuracy and precision. However, ensuring the quality of 3D printed parts remains a challenging task, especially when high-performance materials are used. Simulation tools offer a powerful approach to understand the behaviour of materials during the printing process and predict the properties of the final product. In this study, we investigate simulations for a pellet-based extrusion process with high-performance materials for 3D printing. The utilization of a robotic arm equipped with an extruder mounted on it has opened up new avenues for 3D printing, such as printing on curved surfaces or large overhangs without requiring additional support structures. In this research, our focus is on a doubly curved Carbon Fiber Reinforced Polymer (CFRP) structure, which has been strengthened by adding short fiber-reinforced ribs. These ribs are printed onto the curved surface of the structure to enhance its stiffness. However, during the high-temperature printing process with high-performance materials, significant temperature gradients occur, leading to eigenstrains and deformation of the ribs. To predict the impact of these effects on the stiffness of the structure, a process simulation is conducted by using the GENOA 3DP software suite from Alpha STAR Corp. The proposed AM process simulation takes into account AM parameters such as the toolpath (Gcode), deposition rate and extrusion temperature. The thermophysical process is modelled by an element-by-element activation based on the toolpath so that local temperature gradients are captured. This allows the consideration of manufacturing related effects for the prediction of process-dependent material and component properties. The primary objectives of this study are to simulate the intricate extrusion process and to estimate the eigenstrains and deformations that occur during the process as well as to gain an in- depth understanding of the process for further optimisation. Finally, the predicted part deformations are compared with 3D-measurements of the real part in order to evaluate the prediction accuracy.