Surrogate Model-Based Control Considering Uncertainties for Composite Fuselage Assembly

Abstract

Shape control of composite parts is vital for large-scale production and integration of composite materials in the aerospace industry. The current industry practice of shape control uses passive manual metrology. This has three major limitations: (i) low efficiency: it requires multiple trials and a longer time to achieve the desired shape during the assembly process; (ii) nonoptimal: it is challenging to reach optimal deviation reduction; and (iii) experience-dependent: highly skilled engineers are required during the assembly process. This paper describes an automated shape control system that can adjust composite parts to an optimal configuration in a manner that is highly effective and efficient. The objective is accomplished by (i) building a finite element analysis (FEA) platform, validated by experimental data; (ii) developing a surrogate model with consideration of actuator uncertainty, part uncertainty, modeling uncertainty, and unquantified uncertainty to achieve predictive performance and embedding the model into a feed-forward control algorithm; and (iii) conducting multivariable optimization to determine the optimal actions of actuators. We show that the surrogate model considering uncertainties (SMU) achieves satisfactory prediction performance and that the automated optimal shape control system can significantly reduce the assembly time with improved dimensional quality.

On the Development of a Surrogate Modelling Toolbox for Virtual Assembly

“…For further research however, it might be promising to directly evaluate the R-value at the datum features. Patch Selection (CAD based) A Determination of subsets corresponding to a maximal distance to a nominal geometry such as a CAD file [74] Point set morphology Description and manipulation of the surface morphology Definition of Offset and Intersection A Definition of a certain offset between faces (e.g., to simulate adhesive gap) or of an intersection (virtual penetration) to simulate surface flattening or material loss (e.g., due to melting welding bead) [41] Morphological Filtering A Manipulate surfaces locally to simulate surface flattening due to mechanical load [51,52] Hertzian Contact Formulation A [41] Method of Influence Coefficients (MIC) A Linearized computation of elastic deformation due to mechanical load [50,57] Linear Finite Element Analysis (FEA) A Computation of elastic deformation due to mechanical load [75,76] Nonlinear FEA A Computation of elastic and plastic deformation due to mechanical load [65,66] Objective Function Searching of correspondences and the assembly position Modalities for Distance Computation Approaches to compute distances between S 1 and S 2…”

Geometric Deep Learning for Shape Correspondence in Mass Customization by Three-Dimensional Printing

“…These methods did not consider the deformation of the products. In contrast, many other studies imply the necessity to investigate the non-rigid bodies in manufacturing [9,10]. For instance, Camelio et al studied the geometrical variation propagation at the discrete measurement vertices in the automotive body assembly process with a compliant assemble system [9].…”

StressNet - Deep learning to predict stress with fracture propagation in brittle materials

“…Although this method gives accurate predictions of the fracture growth and the dynamics of stress distribution, it is computationally intensive, especially when multiple runs are needed to obtain the statistical variability naturally existent in real-world materials. Machine learning (ML) techniques are becoming popular 12 in modeling complex systems because they can serve as lower-order surrogates to approximate higher-fidelity models, which significantly reduces the model complexity and computation time, as shown in Fig. 1.…”