Scholarlink

<p>One method for assessing alloy properties is to analyze its microstructure image. Many factors influence the microstructure of alloys, particularly the content of alloying elements and the type of heat treatment. Depending on the alloy type, various alloying elements are used. In assessing the microstructure image, it is crucial to define the elements present. The quantity, shape, and arrangement of microstructural elements indicate the final parameters of the products. In the case of unalloyed cast iron, including vermicular cast iron, the basic matrix components are ferrite and pearlite. Currently, the analysis and assessment of microstructure phases is mostly performed manually. Microstructure assessment requires knowledge and experience in analysis, and its results have a significant impact on the production process control and decision-making, which will consequently allow for avoiding defects during the next process cycle. The lack of clearly defined rules makes the use of classical analysis methods ineffective. The aim of this work is to use machine learning methods to create an algorithm for predicting matrix phases such as pearlite, ferrite, and graphite in microstructure images of vermicular cast iron. This solution would accelerate and automate the process of metal melt inspection.</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

A model for estimating the phase fraction in the microstructure of densified graphite iron (CGI) using machine learning techniques

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年4月28日

查看全文
  • Michał Zmarły,
  • Łukasz Marcjan,
  • Dorota Wilk-Kołodziejczyk,
  • Grzegorz Gumienny,
  • Sandra Gajoch

摘要

One method for assessing alloy properties is to analyze its microstructure image. Many factors influence the microstructure of alloys, particularly the content of alloying elements and the type of heat treatment. Depending on the alloy type, various alloying elements are used. In assessing the microstructure image, it is crucial to define the elements present. The quantity, shape, and arrangement of microstructural elements indicate the final parameters of the products. In the case of unalloyed cast iron, including vermicular cast iron, the basic matrix components are ferrite and pearlite. Currently, the analysis and assessment of microstructure phases is mostly performed manually. Microstructure assessment requires knowledge and experience in analysis, and its results have a significant impact on the production process control and decision-making, which will consequently allow for avoiding defects during the next process cycle. The lack of clearly defined rules makes the use of classical analysis methods ineffective. The aim of this work is to use machine learning methods to create an algorithm for predicting matrix phases such as pearlite, ferrite, and graphite in microstructure images of vermicular cast iron. This solution would accelerate and automate the process of metal melt inspection.