Abstract: Digital modeling and disturbance control of the die forging process are significant challenges for realizing high-quality intelligent forging of components. Using the die forging of the AA2014 aluminum alloy as a case study, a machine learning assisted method for digital modeling of forging force and autonomous control in response to forging parameter disturbances was proposed in this paper. Firstly, finite element simulations of forging processes were carried out under varying friction factors, die temperatures, billet temperatures and forging velocities, and the sample data including process parameters and forging force under different forging strokes were garnered. The prediction models of the forging force were established using the support vector regression algorithm. The prediction error of the Ff, i.e., the forging force required to fully fill the die cavity can go as low as 4.1%. In order to further improve the prediction accuracy of the model on the actual Ff, this study conducted two rounds iterative forging experiments by the Bayesian optimization algorithm, and the prediction error of Ff in forging experiments was reduced from 6.0% to 1.5%. Finally, the prediction model of Ff combined with a genetic algorithm was used to establish an autonomous optimization strategy for the forging velocity in each stage of the forging stroke when the billet and die temperatures were disturbed, which realized the autonomous control in response to disturbances. In the case of -20°C or -40°C disturbance in the die and billet temperatures, conducting forging experiments with the autonomous optimization strategy can control the measured Ff around the target value of 180 t, with the relative error ranging from -1.3% to +3.1%. The present work could provide references for the study of digital modeling and autonomous optimization control of quality factors in the forging process.