Prediction of Residual Stress of Carburized Steel Based on Machine Learning

Zhenlong Zhu,Yilong Liang

doi:10.3390/app10217759

Zhenlong Zhu, Yilong Liang

Open Access

https://doi.org/10.3390/app10217759

Copy DOI

Abstract

In recent years, the number of machine learning applications (especially those involving deep learning) applied to predicting and discovering material properties has been increasing. This paper is based on using microstructure and carbon content to train machine learning models to predict the residual stress of carburized steel. First, a semantic segmentation model of the material organization structure (SegModel-MOS) was constructed based on the AlexNet network and initially trained on the PASCAL VOC2012 dataset. Then, the trained model was fine-tuned on an enhanced homemade dataset consisting of optical microstructures. The experimental results show that SegModel-MOS can distinguish acicular martensite, retained austenite, and lath martensite in microstructures. Finally, we used both support vector machine (SVM) and decision tree (DT) algorithms to establish a mapping relationship between the microstructure, carbon content, and residual stress to predict the residual stress of steel from its microstructure and carbon content. The experiments verified that the prediction model constructed in this study exhibits high accuracy and can directly predict residual stress without requiring any long-term measurements. Thus, the developed model provides a new approach to the study of residual stress in steel.

Highlights

Carburized steel specimens are subject to complex friction, wear, and cyclic stresses during service.The steel surface is most prone to fatigue failure [1,2,3,4]
Involves the stress caused by the temperature difference between surface and theOne coreexplanation of the quenched part, and thermal stress caused byinvolves the temperature difference between surface and the core ofthe the integrated quenched the second explanation the structural stress caused the by the phase change and part, the second explanation involves structural stress caused by the phaseofchange and the value.and
We will seek to establish high requirements images, its computational is relatively high; the relationship between thethe optical images andthe residual stress directly. network. This initial network themapping model can appropriately reduce pixels, increase number of residual layers, and model has high requirements for the input images, and its computational burden is relatively reduce the amount of calculation

Summary

Introduction

Carburized steel specimens are subject to complex friction, wear, and cyclic stresses during service.The steel surface is most prone to fatigue failure [1,2,3,4]. Increasing the strength of the surface of a part and increasing its residual compressive stress both play key roles in increasing its fatigue life. Residual compressive stress can suppress the initiation of surface cracks, thereby improving the fatigue life of manufactured parts [5,6]. Because residual stress field tests are lengthy and damage the component, we instead aimed to predict the residual stress of steel using its microstructure and carbon content based on a data-driven method. This approach can be used to conduct performance analysis for most low-alloy carburized steels. It provides a new method for research into the performance of carburized steels

Objectives

Methods

Results

Conclusion