Application of Improved Local Models of Large Scale Database-based Online Modeling to Prediction of Molten Iron Temperature of Blast Furnace

Jiao

et al. 2022

steel research int.

Table 2. Characteristic parameters selected under different thresholds. re Characteristic parameters 0.5 Coke M10, Theoretical alkalinity, Proportion of lump ore, Blast volume, Sinter particle size <5 mm, Sinter particle size >40 mm, SiO 2 content of sinter, S content of hot metal, Slag ratio, H 2 content, Si content of hot metal, Coke particle size <15 mm, Coke batch, Coke ratio 0.6 Blast temperature, Coke particle size 40-25 mm, Raceway flame temperature, Coke M10, Theoretical alkalinity, S content of hot metal, Slag ratio, H 2 content, Si content of hot metal, S content of hot metal, Sinter tumble strength, Coke particle size <15 mm, Coke batch, Coke ratio 0.7 Blast temperature, Coke particle size 4025 mm, Bottom center temperature, Raceway flame temperature, Coke M10, Theoretical alkalinity, Slag ratio, H 2

Section: Introductionmentioning

confidence: 99%

Prediction of Blast Furnace Fuel Ratio Based on Back‐Propagation Neural Network and K‐Nearest Neighbor Algorithm

Jiao

et al. 2022

steel research int.

“…HMT was predicted by a fuzzy inference system 8) or a data-based online model. 9) Silicon content of hot metal was predicted by a neural network model, 10) switching linear systems, 11) a non-Gaussian local regression, 12) a support vector regression, 13) or a just-in-time model. 14) Predictive control of silicon content with a neural network model 15) or a state-space model 16) was proposed.…”

Section: Introductionmentioning

confidence: 99%

Practical Operation Guidance on Thermal Control of Blast Furnace

Hashimoto¹,

Okamoto²,

Kaise³

et al. 2019

ISIJ Int.

Blast furnaces are still manually operated in the steel industry. To realize an efficient and stable operation, the authors developed an operation guidance system for controlling hot metal temperature and applied it to actual furnaces. The system features a newly developed 2D transient model that describes the sluggish and complicated process dynamics. Based on the transient model, nonlinear model predictive control (NMPC) and moving horizon estimation (MHE) were adopted to provide operators with appropriate control actions while minimizing the undesirable influence of disturbances. The online validation results demonstrated that the developed operation guidance system successfully reduced the variance of HMT by 1.9°C.

“…Based on artificial neural networks (ANN) models belong to this type of model. They have been used extensively in recent years due to its multiple advantages: are easy to program, have a good adaptation for nonlinear systems, its parameters can be calculated online, are robust against noise and are easy to reprogram to adapt the changes of system conditions [3,4,5].…”

Section: Introductionmentioning

confidence: 99%

Artificial Neural Network Model for Predict of Silicon Content in Hot Metal Blast Furnace

David

Machado

2016

MSF

The growing focus on the efficiency of the reduction process in blast furnace generates an alteration in the way they operate. This modifies the conditions of transfer of silicon for the hot metal and can cause problems in the added value of your product. To evaluate the changes of the operational parameters of the reduction on the conditions of transfer of silicon process a mathematical model based on artificial neural networks has been implemented. Through this model it was possible to predict the silicon content to determine the influence of each operational parameter. Artificial neural networks were able to predict the silicon content through parameters of the reduction in blast furnace process, and this was verified by the precision of this model. The ANN showed that Theoretical flame temperature, Pressure blow and Coke rate have a positive influence on the silicon content in hot metal, and the Hot metal rate is inversely proportional to the silicon content of the hot metal.