HP and NVIDIA Collaborate on Open Source Manufacturing Digital Twins

HP 3D Printing and NVIDIA Modulus have announced a collaboration to develop an open-source manufacturing digital twin using physics-based machine learning (physics-ML). According to the NVIDIA Technical Blog, the partnership aims to accelerate innovation in AI engineering applications by embedding the laws of physics into the learning process.

Advances in Physics – ML

Physics-ML is an emerging field that integrates the laws of physics into machine learning models to improve the generalizability and efficiency of neural networks. NVIDIA Modulus, an open-source framework, facilitates the construction, training, and fine-tuning of these models with a simple Python interface. The framework provides reference applications that help domain experts apply Physics-ML to real-world use cases.

HP’s 3D Printing Software organization’s Digital Twin team has leveraged physics-ML models for manufacturing digital twins and contributed this work to Modulus. As a leader in additive manufacturing, HP aims to accelerate the onboarding of new applications and the introduction of this technology into production environments. HP’s Distinguished Technologist, Dr. Jun Zheng, emphasized the importance of a physics simulation engine based on manufacturing process variability, noting the significant speedup achieved with well-trained physics-ML models.

Digital Twins in Additive Manufacturing

HP has a rich history of technological innovation, including the development of thermal inkjet technology. The company’s latest innovation, HP Metal Jet, enables the production of industrial 3D metal parts. HP is developing a digital twin for its Metal Jet technology to optimize design parameters and process control, thereby improving part quality and manufacturing yield.

The HP team created a Virtual Foundry Graphnet model to accelerate the computation of metal powder material phase transitions by applying physics-ML. The model achieved significant speedups, enabling near-real-time, high-fidelity emulation of the metal sintering process. The model also demonstrated applicability to a variety of geometric designs and process parameter configurations.

HP’s Physics-ML Innovation

Physics-ML is still in its early stages, but the HP Digital Twin team believes the open source community is instrumental in accelerating development. By open sourcing Virtual Foundry Graphnet with NVIDIA Modulus, HP has joined the physics-ML open source community. Traditional high-fidelity physics simulations are computationally intensive, often taking hours or days for a single design iteration. Physics-ML surrogate models provide high-fidelity emulation, enabling faster design iterations.

Physics-ML surrogate models now enable immediate feedback on product design manufacturability and automated design reviews. These models also allow product design teams to use previous simulation data as a real-world data source. The integration of product design and manufacturing optimization, which traditionally required multiple iterations across departments, can now be significantly accelerated.

HP’s process physics simulation software, Digital Sintering, has been deployed to HP Metal Jet customers to improve manufacturing results. Running a well-trained metal sintering inference engine can take only seconds to obtain final sintering warpage values, significantly reducing the time required for design iterations.

Empowering researchers

Physics-ML surrogate models are at the forefront of near-real-time simulation workflows. Innovations like Virtual Foundry Graphnet demonstrate the power of AI to accelerate simulation workflows and deliver predictions in seconds. Democratizing AI for manufacturing is essential to empower a broad range of innovators to solve industrial challenges.

AI researchers and HP 3D printing teams collaborate with domain experts using the NVIDIA Modulus open source project. NVIDIA supports the physics-ML research community by providing a platform that fosters collaboration and innovation, ensuring that advanced AI tools are accessible to everyone.

Image source: Shutterstock