Composite materials can bring significant weight saving in the design of a new component. These materials are one of the solutions offered to designers to achieve new fuel efficiency regulation. New challenge arises in term of design optimization and manufacturing. Shifting from a metal to composite paradigm requires a dedicated tool for composite design in order to take into account the specific composite behavior. Material performance varies widely over the entire part mainly due to the manufacturing process and the corresponding microstructure. Classical design tools are not able to describe accurately the local composite material behavior, leading to the introduction of safety factors and lack of confidence in the design.
Accurate modelling of composites require the use of a multi-scale approach. The composite is not seen as a homogeneous material anymore but as a heterogeneous material made of several constituents. The mechanical performance of this composite depends on the performance of each constituent and on its microstructure. Fiber can be chopped or continuous and their orientations are controlled by the manufacturing process: injection, compression, drappage, etc. A bridge between the manufacturing process and the prediction of the structural performance of the component is therefore necessary.
This paper presents the benefits of a nonlinear multi-scale approach applied to composite material in the Finite Element prediction of the performance of an engine cover.