Reheat furnaces are used in many steel and metallurgical processes to heat large slabs of metal to temperatures suitable for rolling, forging, extruding and further processing.
See Taber's digital twin of this process in action, constructed in the Griffin AI Toolkit® platform!
In a reheat furnace there are often numerous opportunities for optimization and improvement, due to any one or combinations of:
- Highly variable loading of slabs often place different product types and slab sizes adjacent to each other, resulting in very different heating demands at each point in the furnace.
- Inconsistent burner performance and unstable flame location.
- Process unpredictability, often caused by downstream processing or maintenance delays.
- Location of process temperature measurements don’t coincide with where downstream temperature targets must be met (e.g., furnace temperature is measured, but it’s the slab temperature downstream after some processing that determines product acceptance).
- Furnace degradation changes the behavior of the process overtime, making it increasingly difficult to achieve temperature setpoints and product acceptance.
Taber's Reheat Furnace Optimizer uses a powerful and unique combination of Artificial Intelligence, first-principles modeling, expert knowledge from site engineers and operators, and fuzzy logic to identify optimal furnace firing and slab traversal timelines to realize each unique slab's target temperature using the least amount of energy while avoiding the rejection of any product. In practice, this batch process system optimizer has been shown to nearly eliminate reject slabs while harmonizing furnace operation, improving process operability and efficiency.