A Markov chain model was compared to experimental data in the testing of a natural‐fiber‐reinforced composite material proposed for low‐torque gearing applications. The material is tested on a single gear tooth bench to assess multiple damage measurement methods. The model was coupled to three indices of single gear tooth damage: cracking monitored with a high resolution camera, residual load measured with the fatigue machine load cell, and acoustic emission detected by piezoelectric sensors. The model has predictive skill and contributed to better understanding of how damage to the material evolved during the fatigue tests.

中文翻译：

---

天然短纤维增强高密度聚乙烯的疲劳：单齿轮弯曲的随机模型

在针对低扭矩齿轮应用提出的天然纤维增强复合材料的测试中，将马尔可夫链模型与实验数据进行了比较。该材料在单个齿轮齿台上进行测试，以评估多种损伤测量方法。该模型与单齿轮齿损坏的三个指标耦合：使用高分辨率相机监控的裂纹，使用疲劳机负载传感器测量的残余负载以及通过压电传感器检测到的声发射。该模型具有预测能力，有助于更好地了解疲劳测试过程中材料损坏如何演变。