Mixed-integer, multi-objective layerwise optimization of variable-stiffness composites with gaps and overlaps.

Area of Science:

• Composite Materials Science
• Aerospace Engineering
• Structural Optimization

Background:

• Automated fiber placement (AFP) enables variable-angle tow (VAT) composites for enhanced mechanical performance and weight reduction in aerospace.
• Manufacturing defects like gaps and overlaps in VAT laminates present challenges for mass and structural optimization.
• Optimizing VAT structures involves complex discrete (number of layers) and continuous (fiber path) design variables.

Purpose of the Study:

• To develop a mixed-integer strategy for optimizing VAT laminate designs produced by AFP.
• To minimize the weight of VAT laminates while satisfying natural frequency and buckling load constraints.
• To incorporate the impact of manufacturing defects and the AFP process signature into the optimization.

Main Methods:

• Combines the Carrera unified formulation (CUF) for laminate modeling with the defect layer method (DLM) to account for fabrication defects.
• Employs a mixed-integer strategy to handle discrete and continuous design variables.
• Integrates the manufacturing signature of the AFP process into the optimization framework.

Main Results:

• Determines the minimum number of layers required for VAT laminates to meet frequency and buckling constraints, considering manufacturing defects.
• Identifies optimal fiber path parameters that minimize laminate weight.
• Evaluates the influence of different structural theories on the resulting optimal designs.

Conclusions:

• The proposed mixed-integer strategy effectively optimizes VAT laminate designs for weight and structural performance.
• Accounting for AFP manufacturing defects is crucial for accurate structural optimization of VAT composites.
• This approach facilitates the development of lighter and more efficient aerospace structures.