Riveting process parameter optimization and path planning for aircraft thin-walled parts

Increasing aerodynamic performance requirements for new aircraft impose increasingly stringent standards on the assembly accuracy of thin-walled aircraft parts. Improving the riveting assembly accuracy of these parts is therefore critical. Non-uniform distribution of riveting interference can cause deformation of thin-walled parts. Consequently, it is necessary to first optimize the single-rivet assembly process parameters to enhance the uniformity of single-rivet interference. On this basis, multi-rivet assembly path planning is carried out to reduce the overall deformation of thin-walled parts. According to practical engineering requirements and conditions, riveting force, rivet hole diameter, riveting process time, and riveting dwell time are selected as design variables. Sample point sets and their corresponding finite element simulation results are obtained through Latin hypercube sampling experiments. A radial basis function model is then constructed to predict the deformation indices of thin-walled parts. Based on this prediction model, the NSGA-III algorithm is employed to search for the pareto optimal frontier, and the entropy-weighted TOPSIS method is used to determine the optimal combination of process parameters. Finally, a genetic – ant colony algorithm is applied to optimize the riveting assembly path. Experimental results show that the proposed method can significantly improve interference uniformity and effectively reduce the deformation.