Scholarlink

The degradation of the coating through erosion is a serious problem in erosion prone engineering systems that are utilized in process industries, aerospace, high-technology manufacturing systems, and power generation. Traditional experimental and empirical methods to assess erosion are resource intensive, time consuming and frequently fail to describe the complicated, nonlinear relationship among coating composition, microstructure, processing conditions, and service environments. Following these constraints, this chapter proposes a complete model to predict erosion behavior of coating with the aid of machine learning in accordance with Industry 4.0, Industry 5.0 and the idea of sustainability. The chapter discusses in a systematic manner, erosion mechanisms, data acquisition schemes, feature engineering, and appropriate machine learning algorithms that can predict erosion. The focus is to be put on explainable artificial intelligence to guarantee physical interpretability and human-oriented decision-making. It is emphasized that the combination of machine learning models and digital twin architectures is a revolutionary solution to predictive maintenance and lifecycle management of coated components. The chapter illustrates the allowance of reduced experimental trials, improved coating design, and reliability of a system through representative industrial case studies in which data-driven erosion prediction can allow reducing experimental trials, improving the design of the coating, and increasing the reliability of a system. The major issues, such as the lack of data, quantifying uncertainty, and the barrier to deployment, are critically discussed and then the future research directions in the field of physics-informed models, multiscale data fusion, and sustainability-aware optimization. In general, machine learning-driven erosion forecasting is one of the major facilitators of intelligent, strong, and sustainable surface engineering that the chapter will facilitate the shift to the next generation of smart manufacturing systems.

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

Machine Learning–Driven Prediction of Erosion Performance in Coatings for Smart and Sustainable Manufacturing

  • 章节

  • 发表日期：

    2026年

查看全文
  • Hitesh Vasudev,
  • Bhuvan Unhelkar,
  • Satish Kumar

摘要

The degradation of the coating through erosion is a serious problem in erosion prone engineering systems that are utilized in process industries, aerospace, high-technology manufacturing systems, and power generation. Traditional experimental and empirical methods to assess erosion are resource intensive, time consuming and frequently fail to describe the complicated, nonlinear relationship among coating composition, microstructure, processing conditions, and service environments. Following these constraints, this chapter proposes a complete model to predict erosion behavior of coating with the aid of machine learning in accordance with Industry 4.0, Industry 5.0 and the idea of sustainability. The chapter discusses in a systematic manner, erosion mechanisms, data acquisition schemes, feature engineering, and appropriate machine learning algorithms that can predict erosion. The focus is to be put on explainable artificial intelligence to guarantee physical interpretability and human-oriented decision-making. It is emphasized that the combination of machine learning models and digital twin architectures is a revolutionary solution to predictive maintenance and lifecycle management of coated components. The chapter illustrates the allowance of reduced experimental trials, improved coating design, and reliability of a system through representative industrial case studies in which data-driven erosion prediction can allow reducing experimental trials, improving the design of the coating, and increasing the reliability of a system. The major issues, such as the lack of data, quantifying uncertainty, and the barrier to deployment, are critically discussed and then the future research directions in the field of physics-informed models, multiscale data fusion, and sustainability-aware optimization. In general, machine learning-driven erosion forecasting is one of the major facilitators of intelligent, strong, and sustainable surface engineering that the chapter will facilitate the shift to the next generation of smart manufacturing systems.