Artificial Intelligence Tools Reduce Costly Faults in Steel Production BENEFITS TO OUR NATION Implementing operational AI tools for strip break classification will enhance the
Project Objective
The project objective is to demonstrate an automated metal strip break classification system that is capable of predicting the reason for a ‘strip break’ event. P2TB16#y1
Benefits to Industry
By swiftly and accurately identifying breakages in steel strips, this technology reduces production interruptions, enhances efficiency, and ultimately leads to improved productivity and cost savings. The real-time monitoring and predictive capabilities of AI improve overall quality control, ensuring that American steel sets the global standard for quality. This not only boosts profitability but also secures the industry's role as a key player in domestic and global markets. Embracing AI in steel manufacturing paves the way for a more efficient, sustainable, and prosperous future for American steel manufacturers.
Technical Approach
Develop the digital twins to detect and classify strip breaks using the Falkonry Operational AI platform by analyzing historical and real time data. Tune the digital twins using semi-supervised reinforcement learning and validate the solution to monitor and classify in near real-time conditions.
Accomplishments
- Completed digital twin baselining from historical data that was collected to classify normal and strip break behaviors.
- Created cold rolling mill operational data dashboards.
- Operationalized digital twin that continuously monitors and classifies strip breaks in real time.
- Developed complete documentation package for strip break classification solution performance.
- Published Technical Paper: Automated and Scalable Strip Break Classification in Tandem Cold Rolling Mills Using Time Series AI, Mehta & Chukwulebe, AIST Digital Transformation Forum, March 2023
Deliverables
- Delivered Metal Strip / Weld Break SM Profile.
- Delivered Classification Solution Performance Report.
- Published Use Case: Strip Break Classification in a Cold Rolling Steel Mill Using Machine Learning.
Reusable Outcomes
- Metal Strip / Weld Break SM Profile.