摘要
本文探讨了AI Graf Compounder软件在橡胶配方开发中的应用。该系统基于前馈神经网络,能够根据成分预测材料性能,显著减少物理测试需求并加快研发进程。研究通过多个案例验证了其在EPDM、天然橡胶等配方中的预测准确性,强调高质量结构化数据(尤其是实验设计数据)对模拟结果的重要性。人工智能与结构化实验设计的结合,为橡胶行业提供了更高效、数据驱动的开发路径。
This article explores the application of AI Graf Compounder software in rubber compound development.Based on a feedforward neural network,this system can predict material properties based on its components,significantly reducing the need for physical testing and accelerating the research and development process.The study verifies its predictive accuracy in formulations such as EPDM and natural rubber through multiple cases,emphasizing the importance of high-quality structured data,especially experimental design data,for simulation results.The combination of artificial intelligence and structured experimental design provides a more efficient,data-driven development path for the rubber industry.
作者
章羽(编译)
Zhang Yu(compiler)(National Machinery Information Center of Rubber&Plastics,Beijing 100143,China)
出处
《橡塑技术与装备》
2026年第1期76-81,共6页
China Rubber/Plastics Technology and Equipment
关键词
人工智能
橡胶配方开发
神经网络
实验设计
数据驱动模型
artificial intelligence
rubber formula development
neural network
experimental design
data-driven model
