Abnormality detection method and blast furnace operation method

Patent metadata

Field	Value
Publication number	US-9799110-B2
Application number	US-201414899486-A
Country	US
Kind code	B2
Filing date	Jun 18, 2014
Priority date	Jul 29, 2013
Publication date	Oct 24, 2017
Grant date	Oct 24, 2017

Abstract

An abnormality detection method of detecting abnormality of a blast furnace from tuyere images shot by cameras installed in vicinities of a plurality of tuyeres of the blast furnace includes: collecting, in a time-series manner, representative brightness vectors defined by representative brightnesses determined based on brightness values of respective pixels for each of the tuyeres image previously shot by the cameras at a same time; extracting a principal component vector by performing principal component analysis on the representative brightness vectors collected in the time-series manner; calculating, as an evaluation value, a length of a normal line drawn in a direction of the principal component vector from the representative brightness vector collected from the tuyere images shot by the cameras at the same time during an operation; and detecting the abnormality of the blast furnace by comparing the evaluation value with a predetermined threshold.

First claim

The invention claimed is:

1. An abnormality detection method of detecting abnormality of a blast furnace from tuyere images shot by cameras installed in vicinities of a plurality of tuyeres of the blast furnace, the abnormality detection method comprising:
collecting, in a time-series manner, representative brightness vectors defined by representative brightnesses determined based on brightness values of respective pixels for each of the tuyeres image previously shot by the cameras at a same time;
extracting a principal component vector by performing principal component analysis on the representative brightness vectors collected in the time-series manner;
calculating, as an evaluation value, a length of a normal line drawn in a direction of the principal component vector from the representative brightness vector collected from the tuyere images shot by the cameras at the same time during an operation; and
detecting the abnormality of the blast furnace by comparing the evaluation value with a predetermined threshold.

2. The abnormality detection method according to claim 1 , wherein the representative brightness vector is collected while maximum values of the brightness values in the tuyere images are set to the representative brightnesses.

3. The abnormality detection method according to claim 1 , wherein the representative brightness vector is collected while average values of the brightness values in the tuyere images are set to the representative brightnesses.

4. The abnormality detection method according to claim 1 , wherein the representative brightness vector is collected while minimum values of the brightness values in the tuyere images are set to the representative brightnesses.

5. An abnormality detection method of detecting abnormality of a blast furnace from a tuyere image shot by a camera installed in a vicinity of a tuyere of the blast furnace, the abnormality detection method comprising:
collecting, in a time-series manner, representative brightness vectors defined by representative brightnesses determined based on brightness values of respective pixels for a plurality of areas formed by dividing a region of the tuyere image previously shot by the camera into the areas;
extracting a principal component vector by performing principal component analysis on the representative brightness vectors collected in the time-series manner;
calculating, as an evaluation value, a length of a normal line drawn in a direction of the principal component vector from the representative brightness vector collected from the tuyere image shot by the camera during an operation; and
detecting the abnormality of the blast furnace by comparing the evaluation value with a predetermined threshold.

6. The abnormality detection method according to claim 5 , wherein the representative brightness vector is collected while maximum values of the brightness values in the respective areas are set to the representative brightnesses.

7. The abnormality detection method according to claim 5 , wherein the representative brightness vector is collected while average values of the brightness values in the respective areas are set to the representative brightnesses.

8. The abnormality detection method according to claim 5 , wherein the representative brightness vector is collected while minimum values of the brightness values in the respective areas are set to the representative brightnesses.

9. A blast furnace operation method comprising:
detecting abnormality of a blast furnace using the abnormality detection method, the abnormality detection method detecting abnormality of a blast furace from tuyere image shot by cameras installed in vicinities of a plurality of tuyeres the blast furnace and including:
collecting, in a time-series manner,representative brightness vectors defined by representative brightness determined based on brightness value of respective pixels for each of the tuyeres image preyiously shot by the cameras at same time;
extracting a principal component vector by performing principal component analysis on the representative brightness vectors collected in the time series manner;
calculating as an evaluation vaule, a length of a normal line drawn in a direction of the principal component vector from the representative brightness vector collected from the tuyere images shot by by the cameras at the same time during an operation; and
dectecting the abnomality of the blast furnace by comparing the evaluation value with predetermined threshold; and
controlling an operation condition of the blast furnace based on whether the abnormality of the blast furnace has been detected.

10. The blast furnace operation method according to claim 9 , wherein the representative brightness vector is collected while maximum values of the brightness values in the tuyere images are set to the representative brightnesses.

11. The blast furnace operation method according to claim 9 , wherein the representative brightness vector is collected while average values of the brightness values in the tuyere images are set to the representative brightnesses.

12. The blast furnace operation method according to claim 9 , wherein the representative brightness vector is collected while minimum values of the brightness values in the tuyere images are set to the representative brightnesses.

13. A blast furnace operation method comprising:
detecting abnormality of a blast furnace using the an abnormality detection method, the abnormality detection method detecting abnormality of a blast furnace from a tuyere image shot by a camera installed in a vicinity of a tuyere of the blast furnace and including:
collecting, in a time-series manner, representative brightness vectors defined by representative brightnesses determined based on brightness values of respective pixels for a plurality of areas formed by dividing a region of the tuyere image previously shot by the camera into the areas;
extracting a principal component vector by performing principal component analysis on the representative brightness vectors collected in the time-series manner;
calculating, as an evaluation value, a length of a normal line drawn in a direction of the principal component vector from the representative brightness vector collected from the tuyere image shot by the camera during an operation; and
detecting the abnormality of the blast furnace by comparing the evaluation value with a predetermined threshold; and

controlling an operation condition of the blast furnace based on whether the abnormality of the blast furnace has been detected.

14. The blast furnace operation method according to claim 13 , wherein the representative brightness vector is collected while maximum values of the brightness values in the respective areas are set to the representative brightnesses.

15. The blast furnace operation method according to claim 13 , wherein the representative brightness vector is collected while average values of the brightness values in the respective areas are set to the representative brightnesses.

16. The blast furnace operation method according to claim 13 , wherein the representative brightness vector is collected while minimum values of the brightness values in the respective areas are set to the representative brightnesses.

Assignees

• Jfe Steel Corp

Inventors

• Tsuda Kazuro
• Hirata Takehide
• Yamahira Naoshi
• Abe Hidekazu

Classifications

• G06F18/2411

  based on the proximity to a decision surface, e.g. support vector machines · CPC title

• F27D1/16

  Making or repairing linings {; Increasing the durability of linings; Breaking away linings} · CPC title

• G06T2207/10016

  Video; Image sequence · CPC title

• G06T2207/20076

  Probabilistic image processing · CPC title

• G06T7/0004 Primary

  Industrial image inspection · CPC title