Computer vision-enabled digital twins for granular flow optimization
This solution leverages computer vision and machine learning to create digital twins that optimize granular flow in manufacturing, improving product quality and reducing costs. Applicable in food, pharmaceutical, and 3D printing sectors.
Overview
The proposed technology integrates computer vision with machine learning to develop digital twins for granular flow processes in manufacturing. By capturing real-time imaging of granular flow patterns, the system extracts descriptors like avalanching energy and Zernike moments, which serve as digital fingerprints. These metrics enable predictive modeling, allowing manufacturers to refine simulations and optimize workflows, thereby reducing trial-and-error costs and enhancing quality control.
Technical specifications
- Real-time imaging captures granular flow patterns for analysis
- Descriptors extracted: Avalanching energy, surface fractal number, Zernike moments
- Machine learning models correlate descriptors with particle properties and processing conditions
- Digital twin framework supports real-time process adjustments and workflow optimization
- Applicable in industries such as food production, pharmaceuticals, and 3D printing
Technology readiness level
Currently at TRL 6, the framework has been validated through preliminary tests and is preparing for further validation in industrial settings. The next steps include building experimental devices and developing simulations to refine the technology for broader application.
About Lawrence Technical University
Lawrence Technological University is a private STEM- and design-focused university of a few thousand students with a hands-on, industry-centric culture. Based in Southfield within the Detroit metro manufacturing and mobility hub, LTU connects companies to faculty expertise, student talent, and shared prototyping spaces for rapid development. A structured co-op and internship model, plus professional studios and capstone collaborations, streamlines applied engagements and recruiting. Research is supported by competitive federal and state funding, including National Science Foundation awards and industry contracts. A dedicated technology transfer office supports IP strategy, prototyping, supplier introductions, and commercialization.
delivered to your inbox.