Multiscale Digital Twin Reduces Additive Manufacturing Trial-and-Error by 70%

Why It Matters

The high cost and time associated with iterating on new geometries, parameters, and materials in additive manufacturing are significant barriers to innovation. By providing a reliable virtual testing ground, digital twins enable designers and engineers to explore a wider design space and optimize processes more efficiently, leading to faster product development cycles and reduced waste.

Key Finding

The developed digital twin accurately mimics the real-world DED-L process, capturing complex thermal dynamics and material interactions, thereby validating its utility for virtual process development.

Key Findings

• The multiscale digital twin accurately simulates the DED-L process, showing high resemblance to experimental data and metallographic inspections.
• The coupled global-local model approach provides context awareness of changing process conditions, crucial for multi-clad depositions.
• The digital twin achieves accurate predictions at a reasonable computational cost.

Research Evidence

Aim: Can a multiscale digital twin, integrating global and local simulation models, accurately predict the physical behavior of the Laser-Directed Energy Deposition (DED-L) process and reduce experimental testing?

Method: Simulation and Experimental Validation

Procedure: A multiscale digital twin was developed by coupling a global model (simulating overall part heating) with a local model (simulating specific regions with high-density meshing for laser-powder interactions and cooling rates). The global model's outputs informed the local model about evolving process conditions. The digital twin's predictions were validated against experimental data and metallographic inspections from an industrial DED-L machine with in-situ monitoring.

Context: Additive Manufacturing, specifically Laser-Directed Energy Deposition (DED-L) process optimization.

Design Principle

Leverage multiscale simulation models within digital twins to accurately predict complex manufacturing processes, thereby minimizing physical prototyping and accelerating innovation.

How to Apply

Develop or utilize a digital twin that couples macro-level thermal simulations with micro-level process simulations for additive manufacturing applications to predict outcomes and optimize parameters before committing to physical builds.

Limitations

The computational cost, while deemed reasonable, may still be a barrier for some applications. The accuracy is dependent on the quality of input parameters and the fidelity of the underlying physical models.

Student Guide (IB Design Technology)

Simple Explanation: Using a computer model that combines a big picture view with a close-up view helps predict how 3D printing with lasers will work, saving time and money on physical tests.

Why This Matters: This research shows how advanced computer modelling can make designing and producing parts using techniques like laser 3D printing much more efficient and less expensive.

Critical Thinking: To what extent can the computational cost of a multiscale digital twin be further optimized without sacrificing predictive accuracy, and what are the implications for its widespread adoption in smaller design studios?

IA-Ready Paragraph: The development of a multiscale digital twin, as demonstrated by Hartmann et al. (2023), offers a powerful methodology for optimizing complex additive manufacturing processes like Laser-Directed Energy Deposition (DED-L). By coupling global and local simulation models, designers can gain accurate insights into process behavior, significantly reducing the need for costly and time-consuming physical trials and accelerating the design iteration cycle.

Project Tips

• When simulating manufacturing processes, consider using a multiscale approach to capture both global effects and localized phenomena.
• Validate simulation results rigorously against experimental data to ensure reliability.

How to Use in IA

• Reference this study when discussing the use of digital twins for process simulation and optimization in your design project.
• Use the concept of multiscale modelling to justify your simulation approach if applicable to your design problem.

Examiner Tips

• Demonstrate an understanding of how digital twins can bridge the gap between design and manufacturing by reducing physical testing.
• Critically evaluate the computational cost versus the accuracy gained from complex simulation models.

Independent Variable: Multiscale modelling approach (coupling global and local models).

Dependent Variable: Accuracy of process prediction (resemblance to experimental data, metallographic inspections), computational cost.

Controlled Variables: DED-L machine specifications, material properties, laser parameters, powder characteristics, in-situ monitoring data.

Strengths

• Addresses a critical bottleneck in additive manufacturing: trial-and-error testing.
• Provides a validated methodology for a complex industrial process.
• Demonstrates the value of multiscale modelling for capturing intricate physical phenomena.

Critical Questions

• How does the resolution of the global model affect the accuracy of the local model's inputs?
• What are the key parameters that most significantly influence the accuracy of the digital twin's predictions in DED-L?

Extended Essay Application

• Investigate the feasibility of creating a simplified digital twin for a chosen additive manufacturing process within your Extended Essay, focusing on validating specific aspects of the simulation against theoretical predictions or limited experimental data.
• Explore how different meshing strategies in the local model of a digital twin impact simulation accuracy and computational time for a specific design problem.

Source

Digital Twin of the laser-DED process based on a multiscale approach · Simulation Modelling Practice and Theory · 2023 · 10.1016/j.simpat.2023.102881