Machine learning tensile strength and impact toughness of wheat straw reinforced composites
Yun Zhang, Xiaojie Xu
Abstract
The wheat straw reinforced composite is considered as a “green composite” as it utilizes biodegradable and recyclable food waste materials as reinforcing materials. Mechanical properties, such as tensile strength and impact toughness, are greatly dependent on the composite content and processing parameters. While compression molding is one of popular methods to fabricate the wheat straw/polypropylene composite, experimental trials to achieve targeted mechanical properties can be time-consuming and expensive. In this work, we develop the Gaussian process regression model to present the relationship among the fiber content, processing parameters of the compression molding, and mechanical performance of wheat straw reinforced polypropylene composites. The model achieves a correlation coefficient of 99.13% (95.68%), a root mean square error of 0.0857 (0.4369), and a mean absolute error of 0.0693 (0.3265) for tensile strength (impact toughness). The models are simple and fast to implement, produce predictions with high accuracy, and thus might be considered as efficient tools for mechanical property estimations. Should data become available, the model may be extended to include other descriptors, such as the wheat straw length, size distribution, and chemical treatment parameters.