Development of robust organic matrix composite cure cycles using predictive pareto genetic and constraint satisfaction algorithms

The process of developing a robust cure cycle for organic matrix composites remains a challenging task, due largely to variations in autoclave capabilities, part dimensions, and a lack of mature computational tools that can predict the evolution of the properties of the part. In this work, we describe a rigorous computational approach to link material process models with parametric optimization and constraint satisfaction algorithms. Specifically, the material property models for 5320-1/IM7, an out-of-autoclave organic matrix composite, are implemented in MATLAB to allow for predicting the viscosity, glass transition, and degree of cure as a function of time and temperature during the cure process. These models are integrated into a Parameterized Predictive Pareto Genetic Algorithm (P3GA), which provides a Pareto front of optimal processing conditions while accounting for fluctuations and uncertainty in the boundary conditions or system characteristics. Additionally, a constraint satisfaction algorithm is employed to identify a range of allowable processing conditions that are expected to produce the desired material properties. Such a computational framework is expected to provide significant insight to the development of a robust process cycle and also provide a foundation for more complex considerations, such as spatially-varying residual stresses.

Development of robust organic matrix composite cure cycles using predictive pareto genetic and constraint satisfaction algorithms Conference Paper