Scholarlink

<p>Due to increasing sustainability concerns, wood filler-reinforced polymer (WFRP) composite materials have gained prominence as a potential material in various industries such as construction, automotive and consumer products. This real-world application of wood filler-reinforced polymer composites requires well understanding and prediction of important mechanical properties. The conventional experimental methods of material characterization are often resource intensive and time-consuming. Recently, the machine learning (ML) presented a novel and viable avenues for augmenting prediction models, enabling the accurate estimation of mechanical properties with fewer experiments and improved generalization. The current work presents the application of ML techniques for the prediction of tensile properties of WFRP composite. Various models like support vector machine, polynomial regression, and decision trees (DT) are explored for their potential to predict tensile properties based on input variables like filler content, and crosshead speed. These models are very effective and accurate in representing highly intricate and nonlinear interdependencies between material input parameters and its performance. Additionally, the artificial neural network model, in particular, exhibits an excellent capability of predicting tensile strength of composites with the lowest MSE value. The study highlights the efficiency of ML models, demonstrating their potential to enhance material property prediction.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Tensile Properties Prediction of WFRP Composite Based on Machine Learning and ANN Models: An Experimental and Statistical Approach

  • 期刊论文

  • 发表日期：

    2026年4月28日

查看全文
  • Rahul Kumar,
  • Jagadish,
  • Divya Zindani,
  • Sumit Bhowmik

摘要

Due to increasing sustainability concerns, wood filler-reinforced polymer (WFRP) composite materials have gained prominence as a potential material in various industries such as construction, automotive and consumer products. This real-world application of wood filler-reinforced polymer composites requires well understanding and prediction of important mechanical properties. The conventional experimental methods of material characterization are often resource intensive and time-consuming. Recently, the machine learning (ML) presented a novel and viable avenues for augmenting prediction models, enabling the accurate estimation of mechanical properties with fewer experiments and improved generalization. The current work presents the application of ML techniques for the prediction of tensile properties of WFRP composite. Various models like support vector machine, polynomial regression, and decision trees (DT) are explored for their potential to predict tensile properties based on input variables like filler content, and crosshead speed. These models are very effective and accurate in representing highly intricate and nonlinear interdependencies between material input parameters and its performance. Additionally, the artificial neural network model, in particular, exhibits an excellent capability of predicting tensile strength of composites with the lowest MSE value. The study highlights the efficiency of ML models, demonstrating their potential to enhance material property prediction.