Developing an optimisation model for industrial furnace gaseous fuel distribution for energy cost savings

The South African industrial sector is under significant financial pressure and need to reduce production costs to remain competitive in the global and local market. Some steel plants use reheating furnaces to reheat materials before hot rolling. These furnaces use industrial fuel gases as an energy...

Full description

Saved in:
Bibliographic Details
Main Authors: van Niekerk, Sybrand G. J., Breytenbach, Wynand J. J., Marais, Johan H.
Format: Conference Proceeding
Language:English
Subjects:
Data models
Energy
Fuels
furnace
Furnaces
Gases
industry
optimisation
Optimization
Production
Steel
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The South African industrial sector is under significant financial pressure and need to reduce production costs to remain competitive in the global and local market. Some steel plants use reheating furnaces to reheat materials before hot rolling. These furnaces use industrial fuel gases as an energy source. Process by-product and purchased gases can be used as fuel. Inefficiencies in gas distribution can occur in a system where there are multiple furnaces with different energy efficiencies and a shortage of by-product gas. This is a complex system where gas distribution is controlled manually by an operator. In this paper, a methodology is developed to optimise the furnace gas distribution network. Regression models for the different furnaces' consumption of by-product and purchased gases is developed. These models are used in an optimisation process for energy distribution to reduce energy costs. The methodology compiled in this study was validated on a South African steel production facility. In this case study, redistributing the use of by-product and purchased gas between four furnaces indicated energy saving opportunity of R20-million per year. In order to implement an optimisation model of this type, a real-time optimisation solution will need to be adapted.
ISSN:2166-059X
DOI:10.23919/ICUE.2017.8067988