Prediction of aircraft panel assembly deformation using a combined prediction model

Large size is one of the typical characteristics of aircraft panel components. These characteristics are prone to deformation during assembly process, which affects the assembly quality of aircraft panel seriously. Therefore, to solve this problem, a combined prediction model is proposed for predicting panel assembly deformation. Firstly, the prediction model of skin assembly deformation based on Adam optimized BP neural network algorithm is proposed to solve the normal deformation of key characteristics in skin assembly stage. Then, a panel assembly deformation prediction model based on substructure is established. The prediction results derived from the skin assembly stage are taken as the input of the prediction model of stringer assembly stage. Thereby, the combined prediction model is established. Finally, a panel assembly experiment is conducted to validate the combined model. By comparing the predicted results coming from the model and the results from assembly experiment, it can be found out that the average absolute error of aircraft panel assembly deformation is 0.155 mm. Therefore, this study provides a reliable and efficient new approach for the assembly deformation prediction of aircraft panel.