<p>The process parameters have a significant impact on the single track during the manufacturing of AlCoCrFeNi_2.1 eutectic high-entropy alloy by laser directed energy deposition (LDED). As the fundamental manufacturing unit, the morphology of a single track plays a decisive role in the quality and dimensional accuracy of manufactured product. Due to the LDED process involving complex physicochemical interactions, it is challenging to establish the highly nonlinear relationships between process parameters and the morphology of single track. This study proposes an optimization method that combines artificial neural network (ANN) and genetic algorithm (GA), aiming to explore the relationships and the optimal process parameters to obtain a sound morphology of single track and improved performance of the sample. Firstly, experiments with three-factor and four-level designs are designed by the Taguchi method, and corresponding LDED single track experiments are performed to obtain the morphology features. Secondly, the correlations between the process parameters (scanning speed, laser power, and powder feeding rate) and the morphology features of single track were established using ANN model. The model was used to predict the morphology of the single track. Subsequently, based on the constructed ANN model, the GA is used to optimize process parameters. The optimization results are finally verified by a validation experiment. It illustrates that the optimization method can significantly improve the quality of the single track.</p>