Abstract: In view of the shortcomings of complex calculation and low accuracy of boom plate bending rebound, a hybrid model based on neural network was proposed. A mechanical model for the bending and rebound of boom plates was developed based on the theory of elastic-plastic deformation of medium-thick plates. A BP neural network optimized by genetic algorithm was employed to approximate the discrepancy between actual systems and mechanical models, thereby constructing a predictive model for boom plate bending rebound. Finally, the simulation model was trained and tested using experimental data, and the predicted results were compared with the BP neural network model. The accuracy of the model was evaluated using mean square error (MSE) and determination coefficient (R²) as test parameters. The results show that the maximum relative error of the rebound predicted by the hybrid model does not exceed 6%, and it has higher nonlinear goodness of fit and prediction accuracy, which can meet the high-precision production requirements.