Iterative experimental design and identifiability analysis of composite material failure models.

Area of Science:

• Materials Science
• Mechanical Engineering
• Computational Mechanics

Background:

• Parameter identification for composite ply failure models (e.g., Tsai-Wu, Puck) is complex due to brittle fracture effects.
• Existing methods may require extensive experimental data, increasing costs and time.
• Accurate failure models are crucial for predicting composite material performance and safety.

Purpose of the Study:

• To develop an efficient, iterative design of experiments (DoE) approach for identifying composite failure model parameters.
• To minimize the number of experiments required for accurate parameter identification.
• To validate and rank different composite failure criteria based on experimental data.

Main Methods:

• An iterative, nonlinear design of experiments (DoE) was employed.
• Parameter identification for multiple failure models (Tsai-Wu, Tsai-Hill, Hoffman, Hashin, max stress, Puck) was performed.
• Validation involved assessing the Euclidean distance between measured points and model surfaces, followed by sensitivity analysis and iterative experiment selection.

Main Results:

• The proposed DoE method successfully identified optimal experiments for parameter identification from generated data.
• The method demonstrated robustness and accuracy when sufficient information was present in the dataset.
• Sensitivity analysis guided the selection of experiments, enabling parameter identification with minimal data.

Conclusions:

• The iterative DoE approach significantly reduces the number of experiments needed for composite material failure model parameter identification.
• The method provides a robust framework for selecting the most informative tests, even suggesting novel tests when data is insufficient.
• This work enhances the efficiency and accuracy of characterizing composite material behavior under various loading conditions.