DocumentCode
296052
Title
Neuro-aided guidance system for optimum desulphurization treatment in torpedo cars
Author
Nakamura, Ken ; Sagara, Kaichi
Author_Institution
Neuro Eng. Sect., Kobe Steel Ltd., Japan
Volume
1
fYear
1995
fDate
Nov/Dec 1995
Firstpage
575
Abstract
Kobe Steel Ltd. has developed an automatic guidance system for controlling the amount of desulphurization reagents to be used in the torpedo cars at Kakogawa Works. The automatic guidance system consists of a neuro-model to simulate the desulphurization process and an optimization system using the neuro-model, which minimizes the cost of desulphurization reagents to be consumed in the process. The conventional guidance system uses three linear multiregression models for predicting the amount of desulphurization reagents needed in three different regions of the hot metal sulphur content. However, the conventional system requires manual adjustment to achieve the required sulphur content in the hot metal. The neuro-desulphurization model developed features: 1) prediction of the sulphur content after the treatment; 2) development of the automatic guidance system; 3) a considerable annual cost savings; and 4) engineering studies to achieve more efficient operational conditions in the future
Keywords
chemical variables control; neurocontrollers; optimisation; process control; steel industry; Kakogawa Works; Kobe Steel; automatic guidance system; desulphurization treatment; neural networks; neuro-model; optimization system; process control; reagents; steel industry; sulphur content; torpedo cars; Automatic control; Calcium; Content addressable storage; Control systems; Cost function; Impurities; Iron; Predictive models; Real time systems; Refining; Silicon; Steel;
fLanguage
English
Publisher
ieee
Conference_Titel
Neural Networks, 1995. Proceedings., IEEE International Conference on
Conference_Location
Perth, WA
Print_ISBN
0-7803-2768-3
Type
conf
DOI
10.1109/ICNN.1995.488243
Filename
488243
Link To Document