Process modeling and parameter optimization based on assumed inherent sensor inversion for composite automated placement

Abstract

Composite automated placement shows great potential for efficient manufacturing of large composite structures. In order to realize online layup quality detection and parameter optimization with high speed and desired layup quality, a methodology is developed based on assumed inherent sensor inversion. First, it is necessary to conduct sensitive analysis in order to analyze the importance of process parameters and their changes. Then the relationship between these process parameters and the layup quality could be established by assumed inherent sensor inversion, which is considered as the basis of parameter optimization. Simultaneously, genetic algorithm combined with the multi-objective optimization theory is applied to determine the optimum set for obtaining desired composite components with high speed and best layup quality. A series of experiments had been conducted to verify the feasibility of the developed approach. Results demonstrate that the model has high precision, and significant improvement could be achieved through parameter optimization.