TaiwanManufacturing2020 – ongoingChina Steel Corporation
AI in blast furnace, sintering, coking and reheating furnace operations
China Steel reported yearly savings from AI on its blast furnace, coal blending and reheating furnaces
Problem
Steelmaking depends on operator experience and on many interacting process variables. Sinter machine speed, coal blends and furnace temperatures were set by hand, which cost fuel, raw material and yield.
Approach
China Steel turned its No. 2 blast furnace into a smart blast furnace with 27 AI modules in 2020 and extended the approach to Nos. 1, 3 and 4. It built a platform that predicts coke strength to adjust coal blends. A digital temperature control system covers 16 reheating furnaces across the group. A sintering speed control system, trained with automated machine learning on production data and operator knowledge, started on the No. 2 sintering machine.
Results
Takeaway
The savings come from many narrow models, each tied to one process step with a cost it can be measured against. The figures are China Steel's own.
Categories
- Size
- Large enterprise
- Industry
- Manufacturing
- Business function
- Production and quality
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and MLOptimization
- Adoption
- Built in-house
- Evidence
- Company's own claim
Sources
Source summaries
Commercial Times reported on China Steel's AI work. The No. 2 blast furnace became Taiwan's first smart blast furnace in 2020, with savings of about NT$70.53 million a year. AI coal blending saved about NT$75 million a year in raw material and reheating furnace control saved about NT$154 million in fuel. The article also says China Steel had completed 325 AI projects worth close to NT$1.6 billion a year.
NOWnews reported that China Steel built an AutoML speed control system for its sintering machine, first installed on the No. 2 machine. Sinter yield rose about 4%. A carbon cut of about 22,000 tonnes a year is an estimate.
Full details are in the original sources.