Prediction of the Young's Modulus of Polylactic Acid Specimens Manufactured by Fused Deposition Modeling Using...

Area of Science:

• Materials Science and Engineering
• Additive Manufacturing
• Artificial Intelligence

Background:

• Fused Deposition Modeling (FDM) is a widely used additive manufacturing technique for producing polylactic acid (PLA) parts.
• Accurate prediction of material properties like Young's modulus is crucial for FDM part design and application.
• Traditional experimental testing for material properties is time-consuming and resource-intensive.

Purpose of the Study:

• To develop a machine learning model for predicting the Young's modulus of FDM-manufactured PLA specimens.
• To investigate the influence of process parameters and material strengths on Young's modulus.
• To reduce the reliance on extensive experimental testing through AI-based prediction.

Main Methods:

• A stacked ensemble machine learning architecture was employed, integrating nine base models with a linear meta-model.
• Input features included FDM process parameters (fill degree, printing speed, filling pattern) and material properties (yield strength, tensile strength).
• Feature engineering was utilized to derive additional predictive variables.

Main Results:

• The proposed model achieved a Mean Squared Error (MSE) of 7.31 and a Coefficient of Determination (R²) of 0.99 on the test dataset.
• The model demonstrated superior predictive performance compared to individual base models.
• Experimental validation showed an average difference of approximately 1% between measured and predicted Young's modulus.

Conclusions:

• The developed AI-based stacked ensemble model accurately predicts the Young's modulus of FDM-printed PLA.
• The model offers a rapid and reliable method for estimating Young's modulus, significantly reducing experimental effort.
• This AI approach facilitates optimized material selection and process design in additive manufacturing.