Abstract Pulsed eddy current (PEC) testing was originally reported to sort out heat-treated bearing rings noncontactly and nondestructively in terms of hardness, which is expected to solve the difficulty of online classification and small parts inspection. The signals gathered from PEC system were processed by fast Fourier transform and principal component analysis (PCA) respectively. Then, random forest algorithm was used for smart classification. The results obtained with these processes were presented in this paper and the classification results were discussed. It was observed that the use of PCA produces more robust and better prediction performance on bearing rings classification.

中文翻译：

---

使用脉冲涡流检测对热处理轴承套圈进行非接触和无损评估

摘要 脉冲涡流（PEC）检测最初是在硬度方面对热处理轴承套圈进行非接触、非破坏性分类，有望解决在线分类和小零件检测的困难。从 PEC 系统收集的信号分别通过快速傅立叶变换和主成分分析 (PCA) 进行处理。然后，使用随机森林算法进行智能分类。本文介绍了通过这些过程获得的结果，并对分类结果进行了讨论。据观察，PCA 的使用对轴承套圈分类产生更稳健和更好的预测性能。