To enhance the acceleration performance of a superconducting linear accelerator (SLA) system used for fuel pellet injection in fusion reactors, a numerical investigation was carried out. For this purpose, a numerical code based on the finite element method (FEM) was developed to analyze the simultaneous behavior of shielding current density and dynamic motion of the high-temperature superconducting thin film. To optimize the current profile in the electromagnets, a method combining the normalized Gaussian network approach with a genetic algorithm was implemented in the code. The computational results demonstrated that the optimized current distribution resulted in a narrower profile compared to the uniform distribution. As a result, it was possible to increase the size of the thin film, leading to a significant reduction in the acceleration time required to reach the desired speed, approximately 2.9 times faster. These findings emphasize the effectiveness of current profile optimization in enhancing the acceleration performance of the SLA system.