Above Burden Temperature Data Probes Interpretation to Prevent Malfunction of Blast Furnaces - Part 2: Factory Applications

  1. R. Barea 1
  2. R. Martín D. 1
  3. J. Mochón 1
  4. L.F. Verdeja 2
  5. P. Rusek 3
  6. J. Jiménez 1
  1. 1 Centro Nacional de Investigaciones Metalúrgicas
    info

    Centro Nacional de Investigaciones Metalúrgicas

    Madrid, España

    ROR https://ror.org/04m7z8d34

  2. 2 Siderurgical Dpt. ETSIMO.UNiovi
  3. 3 Cracow University
Montrer des affiliations +
Revue:
Steel Research International

ISSN: 1611-3683,  1869-344X

Année de publication: 2009

Volumen: 80

Número: 3

Pages: 194-201

Type: Article

GOOGLE SCHOLAR lock_openAccès ouvert editor

D'autres publications dans: Steel Research International

Résumé

The application of neural networks to the interpretation of a large amount of data from blast furnaces is still very innovative in the steel and metallurgical industry. Contrary to the deterministic research which is based on mass and energy balances, as well as on chemical kinetics, the development of simulation in “black box” processes has strongly appeared as a consequence of the stochastic origin of the variables used. Specifically, this paper shows the application of neural networks to the processing of thermal information provided by the temperature measuring probes located at the furnace top, above the level of the ferric and reductant charge. As a result of this work, a computer tool as a user-friendly aid to the person in charge of the process was developed with the following information: (i) A tool that supplies a real time thermal distribution of the blast furnace gases which are properly classified (Operational Thermal Standards). (ii) It provides the system with alarms which prevent potential incidents (collapses/slippages) over an hour in advance of any incident. (iii) It guides the person in charge as to how to regulate the blast parameters in order to control the situation.