Identification of Hyperelastic Material Parameters of Elastomers by Reverse Engineering Approach

Area of Science:

• Materials Science
• Mechanical Engineering
• Computational Mechanics

Background:

• Hyperelastic material models are crucial for simulating rubber's mechanical behavior.
• Traditional parameter determination involves multiple loading modes (uniaxial, equibiaxial, volumetric) via coupon testing.
• Alternative methods like artificial neural networks (ANNs) offer potential for streamlined parameter identification.

Purpose of the Study:

• To determine hyperelastic material parameters for neoprene, silicone, and natural rubbers using an ANN model.
• To investigate the feasibility of deriving material parameters directly from component performance data.
• To eliminate the need for conventional coupon material testing in hyperelastic model calibration.

Main Methods:

• Development and training of an artificial neural network (ANN) model.
• Simulation of O-ring tension and compression using finite element analysis to generate training data.
• Application of the trained ANN model to identify hyperelastic material parameters.

Main Results:

• The ANN model successfully identified hyperelastic material parameters for the selected rubbers.
• The study demonstrated that component experimental data can be used directly for parameter determination.
• Accurate hyperelastic material parameters were obtained without performing coupon tests.

Conclusions:

• Artificial neural networks provide an effective alternative for calibrating hyperelastic material models.
• Component-level data analysis can replace traditional coupon testing for rubber material characterization.
• This approach enhances efficiency and accuracy in simulating the mechanical behavior of elastomeric components.