Pre-joining process planning model for a batch of skin–stringer panels based on statistical clearances

• Published in: International journal of advanced manufacturing technology 2015-04, Vol.78 (1-4), p.41-51
• Main Authors: Liu, Gang, Tang, Wei, Ke, Ying-Lin, Chen, Qing-Liang, Chen, Xue-Mei
• Format: Article
• Language: English
• Subjects: Assembly; CAE) and Design; Clearances; Computer simulation; Computer-Aided Engineering (CAD; Criteria; Curves; Drilling machines (tools); Engineering; Hypercubes; Industrial and Production Engineering; Latin hypercube sampling; Mathematical models; Mechanical Engineering; Media Management; Original Article; Panels; Process parameters; Process planning; Riveting; Stiffness matrix; Stringers; Workload
• ISSN: 0268-3768; 1433-3015
• DOI: 10.1007/s00170-014-6629-2

The residual clearances after pre-joining the panels directly affect the subsequent processes with the automatic drilling and riveting machine, so that these clearances could be used a criterion of pre-joining processes. Moreover, the pre-joining process model for single panel based on clearances criterion is not applicable to determine the processes for batches of panels, which leads to the non-identical pre-joining parameters for every panel, the inconvenient operations for workers, the huge measurement workload of clearances, and the low assembly efficiency. To solve the aforementioned problems, this paper proposes a pre-joining process planning model for a batch of panels based on statistical clearances. The proposed model takes both the stiffness matrices of key points of panels and the measured distribution parameters of sample of initial clearances as an input, regards the weighted sum of mean and variance of residual clearances after pre-joining as an evaluation criterion, thinks the pre-joining parameters adapted to all panels including the number, location, and sequence as an objective, employs the random Bezier curves to fit the clearance distributions of key points of panels and applies the method of Latin hypercube sampling to simulate processes rapidly. Hence, this model may be used to exponentially reduce both the number of degrees of freedom of nodes and the number of statistical simulation, widely adapt the differences of clearance distributions of panels, quickly determine the standardized parameters of pre-joining process and dramatically improve the assembly efficiency. Lastly, the model has been verified with an example consisting of 30 pairs of skin–stringers with the same configuration.