A neural network model for predicting the silicon content of the hot metal at No. 2 blast furnace of SSAB Luleaa

Book · Tue Dec 31 04:00:00 EST 1996

OSTI ID:460626

Guangqing, Zuo; Jitang, Ma; Bo, B ^[1]

---

1. Luleaa Univ. of Technology (Sweden). Div. of Process Metallurgy

To predict the silicon content of hot metal at No. 2 blast furnace, SSAB, Luleaa Works, a three-layer Back-Propagation network model has been established. The network consists of twenty-eight inputs, six middle nodes and one output and uses a generalized delta rule for training. Different network structures and different training strategies have been tested. A well-functioning network with dynamic updating has been designed. The off-line test and the on-line application results showed that more than 80% of the predictions can match the actual silicon content in hot metal in a normal operation, if the allowable prediction error was set to {+-}0.05% Si, while the actual fluctuation of the silicon content was larger than {+-}0.10% Si.

• OSTI ID: 460626
• Report Number(s): CONF-960317--; ISBN 1-886362-12-2
• Country of Publication: United States
• Language: English

Related Subjects

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATION
99 GENERAL AND MISCELLANEOUS
BLAST FURNACES
CHEMICAL COMPOSITION
CONCENTRATION RATIO
METAL INDUSTRY
NEURAL NETWORKS
SILICON
STABILIZATION
STEELS
THEORETICAL DATA
TRAINING