Furnace control system, furnace control method, and furnace provided with same control system

Patent metadata

Field	Value
Publication number	US-2021033341-A1
Application number	US-202016944215-A
Country	US
Kind code	A1
Filing date	Jul 31, 2020
Priority date	Aug 2, 2019
Publication date	Feb 4, 2021
Grant date	—

Abstract

To provide a furnace control system which can predict production of flammable gases produced inside a furnace during melting, incineration, and fusion batch processes and effectively carry out furnace combustion control on the basis of said prediction results in order to reduce the conventional problem of time lag. A furnace control system has a flammable gas quantity of state calculation unit 12 which calculates a flammable gas quantity of state corresponding to prediction factor data using a quantity of state estimation model for flammable gas originating in volatile organic compounds produced using intelligent information processing technology using as learning data past data relating to furnaces, data relating to materials, and data relating to exhaust gases, and a combustion control unit 11 which controls furnace combustion on the basis of the flammable gas quantity of state calculated by the flammable gas quantity of state calculation unit 12.

First claim

What is claimed is:

1 . A furnace control system, having a flammable gas quantity of state calculation unit which calculates a flammable gas quantity of state corresponding to prediction factor data using a quantity of state estimation model for flammable gas originating in volatile organic compounds produced using intelligent information processing technology using as learning data past data relating to furnaces, data relating to materials, and data relating to exhaust gases, and
a combustion control unit which controls furnace combustion on the basis of the flammable gas quantity of state calculated by the flammable gas quantity of state calculation unit.

2 . The furnace control system as claimed in claim 1 , wherein the furnace control unit
(1) educes or increases fuel supplied to a burner depending on a reference burner combustion state which is pre-set; (2) increases or decreases an oxidizing agent supplied to the burner depending on a reference burner combustion state which is pre-set; (3) decreases fuel supplied to the burner on the basis of the reference burner combustion state which is pre-set and/or increases the oxidizing agent supplied to the burner; (4) supplies the oxidizing agent to an oxidizing agent supply means which is separate from the burner; or (5) reduces the fuel and oxidizing agent supplied to the burner and supplies the oxidizing agent to the oxidizing agent supply means in accordance with the calculated flammable gas quantity of state.

3 . The furnace control system as claimed in claim 1 , having a display control unit which displays to a display unit the calculated flammable gas quantity of state as a time series graph.

4 . The furnace control system as claimed in claim 1 , having
a learning data generation unit which generates learning data, a training data generation unit which generates training data, and an estimation model generation unit which generates a flammable gas quantity of state estimation model using the learning data and the training data.

5 . The furnace as claimed in claim 1 , comprising
a main furnace body, a burner which is provided to the main furnace body, an exhaust port for exhaust gas emitted from the main furnace body, an exhaust duct which is provided having a certain gap away from the exhaust port, a combustion region in which uncombusted gas included in exhaust gas and air flowing in through the gap cause a combustion reaction inside the exhaust duct, and a temperature measurement unit and/or a pressure measurement unit which measures the temperature and/or pressure of the combustion region.

6 . A furnace combustion control method, comprising
a flammable gas quantity of state calculation step of calculating a flammable gas quantity of state corresponding to prediction factor data using a quantity of state estimation model for flammable gas originating in volatile organic compounds produced using intelligent information processing technology using as learning data past data relating to furnaces, data relating to materials, and data relating to exhaust gases, and a combustion control step of controlling furnace combustion on the basis of the flammable gas quantity of state calculated in the flammable gas quantity of state calculation step.

7 . The furnace control method as claimed in claim 5 , wherein the furnace control step
(1) reduces or increases fuel supplied to a burner depending on a reference burner combustion state which is pre-set; (2) increases or decreases an oxidizing agent supplied to the burner depending on a reference burner combustion state which is pre-set; (3) decreases fuel supplied to the burner on the basis of the reference burner combustion state which is pre-set and/or increases the oxidizing agent supplied to the burner; (4) supplies the oxidizing agent to an oxidizing agent supply means which is separate from the burner; or (5) reduces the fuel and oxidizing agent supplied to the burner and supplies the oxidizing agent to the oxidizing agent supply means in accordance with the calculated flammable gas quantity of state.

8 . The furnace combustion control method as claimed in claim 5 , having a display step of displaying to a display unit the calculated flammable gas quantity of state as a time series graph.

9 . The furnace control method as claimed in claim 5 , having
a learning data generation step of generating learning data, a training data generation step of generating training data, and an estimation model generation step of generating a flammable gas quantity of state estimation model using the learning data and the training data.

Assignees

• Air Liquide

Inventors

• Fujishima Shoichi

Classifications

• F27D2019/004

  Fuel quantity · CPC title

• F27D19/00 Primary

  Arrangements of controlling devices · CPC title

• F27D2019/0043

  Amount of air or O2 to the burner · CPC title

• F23N5/102

  using electronic means · CPC title

• G06N20/00

  Machine learning · CPC title