Integrated Computational Materials Engineering and Artificial Intelligence for Sustainable Casting Industry

This introduction paper is based on the paper "Integrated Computational Materials Engineering and Artificial Intelligence for Sustainable Casting Industry" published by "The 75th World Foundry Congress".

Figure 2 CALPHAD-based approach for alloy design[2]
Figure 2 CALPHAD-based approach for alloy design[2]

1. Overview:

  • Title: Integrated Computational Materials Engineering and Artificial Intelligence for Sustainable Casting Industry
  • Author: Alan A. Luo, Diran Apelian
  • Year of publication: 2024
  • Journal/academic society of publication: The 75th World Foundry Congress
  • Keywords: metal casting; sustainability; integrated computational materials engineering (ICME); artificial intelligence (AI); machine learning (ML)

2. Abstract:

The global metal casting industry is facing enormous sustainability and regulatory challenges related to carbon reduction and carbon neutrality. Digital design and manufacturing of metal castings, enabled by integrated computational materials engineering (ICME) technologies and the recent boom in artificial intelligence and machine learning (AI/ML), provide great opportunities for the industry to overcome these challenges. This presentation provides some examples of cast alloy design and process innovations using the ICME approach. The talk will also present application cases of AI/ML tools to support casting quality control and property prediction. We will also discuss future opportunities to combine ICME and AI/ML tools to revitalize and revolutionize the metal casting industry for sustainable growth.

3. Introduction:

Materials and manufacturing industries, including metal casting, are fundamental to global economies but also contribute to climate liability, making their decarbonization a key priority. To meet the Paris Climate Agreement goal of climate neutrality by 2050, major economies have set ambitious emission reduction targets. Consequently, the global casting industry faces significant regulatory and sustainability challenges, particularly concerning carbon emission reduction, as its customers, especially in the automotive sector, shift towards clean energy and sustainable production. Currently, iron and steel production accounts for 24% of industrial emissions, and aluminum production for 3%. Recycling rates for these metals are low (45% for steel, 30% for aluminum). Increasing metal circularity and reducing energy consumption in manufacturing are vital for a carbon-neutral society and circular economy. Figure 1 illustrates a vision for material circularity and sustainability, achievable through: 1) reduced/prolonged usage via better material design and manufacturing/energy efficiency; 2) increased repair, reuse, refurbish, remanufacturing, and recycling; 3) limited but clean primary material production; and 4) minimal or no disposal of non-renewable materials.

4. Summary of the study:

Background of the research topic:

The global metal casting industry is confronted with significant sustainability and regulatory pressures due to carbon reduction goals and the push for carbon neutrality. Key customer sectors, like automotive, are transitioning to clean energy, demanding sustainable production practices from suppliers.

Status of previous research:

Integrated computational materials engineering (ICME) has emerged as a methodology integrating materials information via computational tools with engineering product performance analysis and manufacturing-process simulation. This contrasts with traditional CAD/CAE/CAM approaches that often rely on uniform material properties. Artificial intelligence (AI) and machine learning (ML) are rapidly developing fields, with ML defined as systems that generate outputs like predictions or decisions for specific objectives. Casting simulation and digital manufacturing tools have been increasingly adopted.

Purpose of the study:

This presentation aims to:

  • Provide examples of cast alloy design and process innovations using the ICME approach.
  • Present application cases of AI/ML tools for supporting casting quality control and property prediction.
  • Discuss future opportunities for combining ICME and AI/ML tools to revitalize and revolutionize the metal casting industry towards sustainable growth.

Core study:

The core of the study focuses on the application and integration of ICME and AI/ML technologies to address sustainability challenges in the metal casting industry. This includes leveraging ICME for advanced alloy design (e.g., for recycled aluminum alloys as shown in Figure 2) and process development (Figure 3), and utilizing AI/ML for predictive quality control, property prediction (e.g., UTS prediction as shown in Figure 4), and optimizing manufacturing processes. The study explores how these digital tools can enhance material circularity, energy efficiency, and overall sustainability.

5. Research Methodology

Research Design:

This paper is a presentation and review that discusses the application of Integrated Computational Materials Engineering (ICME) and Artificial Intelligence/Machine Learning (AI/ML) methodologies to enhance sustainability within the metal casting industry. It highlights examples of innovations and applications, and outlines future directions.

Data Collection and Analysis Methods:

The paper describes the use of and refers to results from:

  • Integrated Computational Materials Engineering (ICME): This includes CALPHAD (Calculation of Phase Diagrams)-based approaches for alloy design (as exemplified in Figure 2) and frameworks for integrating casting design, manufacturing process models, microstructure models, and property models (Figure 3). These models are based on computational thermodynamics, kinetics, and process simulations.
  • Artificial Intelligence (AI) and Machine Learning (ML): This involves the use of extensive datasets generated from metal casting processes (e.g., high-pressure die casting) to train ML algorithms, such as neural network models, for tasks like predicting the ultimate tensile strength (UTS) of die castings (Figure 4) and identifying good parts versus process scrap.

Research Topics and Scope:

The research topics and scope discussed in the paper include:

  • The application of ICME for designing cast alloys (particularly secondary/recycled alloys) and developing casting processes.
  • The use of AI/ML tools for casting quality control, prediction of material properties (e.g., UTS), defect control, predictive maintenance, and supply chain logistics in metal casting.
  • The potential for combining ICME and AI/ML tools for the co-design of alloys, processes, and component topology to achieve multi-objective optimization for sustainable castings.
  • The overarching goal is to leverage these advanced computational and data-driven approaches to improve efficiency, reduce environmental impact, and promote sustainability in the metal casting industry.

6. Key Results:

Key Results:

  • ICME technologies enable digital design and manufacturing of metal castings by integrating materials information with engineering product performance and manufacturing-process simulation. ICME models, based on computational thermodynamics, kinetics, and process models, provide location-specific microstructure and property predictions, offering improvements over traditional CAD/CAE/CAM approaches that use uniform material properties.
  • The ICME approach, such as CALPHAD-based methods, can be used for designing secondary (recycled) aluminum alloys for structural die casting applications (Figure 2).
  • AI/ML tools, trained on extensive datasets from casting processes, can be applied to support casting quality control and property prediction. For instance, neural network ML models can predict the ultimate tensile strength (UTS) of die castings (Figure 4) and can be used to model the prediction of good parts and process scrap.
  • The recent development of large thin-wall die casting (giga-casting), when combined with ICME tools, can enhance the sustainability of the metal casting industry.
  • Future applications of AI/ML in metal casting include defect control, predictive maintenance, and supply chain logistics.
  • Combining ICME and AI/ML tools presents opportunities for the co-design of alloys, processes, and topology for multi-objective optimization of sustainable castings, contributing to the revitalization and revolutionizing of the metal casting industry.

Figure Name List:

Figure 1 Vision of material circularity and manufacturingsustainability (modified, based on [1])
Figure 1 Vision of material circularity and manufacturingsustainability (modified, based on [1])
Figure 3 ICME framework for casting design and process development
[3]
Figure 3 ICME framework for casting design and process development [3]
  • Figure 1 Vision of material circularity and manufacturing sustainability (modified, based on [1])
  • Figure 2 CALPHAD-based approach for alloy design [2]
  • Figure 3 ICME framework for casting design and process development[3]
  • Figure 4 Actual UTS testing data vs. the values predicted by the Neural Network model of die castings[5]

7. Conclusion:

Metal casting, despite its long history, currently faces significant technical and social challenges related to its carbon footprint. It is imperative for the industry to embrace the use of recycled alloys and the concept of a circular material economy. To overcome these challenges, the casting industry needs to utilize new ICME and AI/ML tools to improve its efficiency and reduce energy consumption. Furthermore, as transportation industries migrate to clean energy technologies, opportunities will arise for lightweight and high-performance castings, where ICME and AI/ML can play a crucial role.

8. References:

  • [1] https://www.brunel.ac.uk/research/Centres/BCAST/About-us.
  • [2] CinkilicE, RidgewayCD, YanX, Luo AA. Metall. Mater. Trans. A, 2019, 50A: 5945-5956.
  • [3] LuoAA, SachdevAK, ApelianDJ. Mater. Process.Technol., 2022, 306: 117606.
  • [4] https://nvlpubs.nist.gov/nistpubs/ai/NIST.AI.100-1.pdf.
  • [5] KopperA, Karkare, R, Paffenroth, RC, Apelian A. Integr. Mater. Manuf. Innov., 2020, 9: 287-300.

9. Copyright:

  • This material is a paper by "Alan A. Luo, Diran Apelian". Based on "Integrated Computational Materials Engineering and Artificial Intelligence for Sustainable Casting Industry".
  • Source of the paper: DOI URL not provided in the source document. Paper presented at The 75th World Foundry Congress, October 25-30, 2024, Deyang, Sichuan, China.

This material is summarized based on the above paper, and unauthorized use for commercial purposes is prohibited.
Copyright © 2025 CASTMAN. All rights reserved.