KR950019723A - Prediction Method of Molten Steel Temperature and Component Change Using Artificial Neural Network - Google Patents

Prediction Method of Molten Steel Temperature and Component Change Using Artificial Neural Network Download PDF

Info

Publication number
KR950019723A
KR950019723A KR1019930031774A KR930031774A KR950019723A KR 950019723 A KR950019723 A KR 950019723A KR 1019930031774 A KR1019930031774 A KR 1019930031774A KR 930031774 A KR930031774 A KR 930031774A KR 950019723 A KR950019723 A KR 950019723A
Authority
KR
South Korea
Prior art keywords
neural network
artificial neural
molten steel
steel temperature
conditions
Prior art date
Application number
KR1019930031774A
Other languages
Korean (ko)
Other versions
KR970010980B1 (en
Inventor
윤상엽
장근수
이인범
양대륙
변선민
박종진
Original Assignee
조말수
포항종합제철 주식회사
백덕현
재단법인산업과학기술연구소
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 조말수, 포항종합제철 주식회사, 백덕현, 재단법인산업과학기술연구소 filed Critical 조말수
Priority to KR1019930031774A priority Critical patent/KR970010980B1/en
Publication of KR950019723A publication Critical patent/KR950019723A/en
Application granted granted Critical
Publication of KR970010980B1 publication Critical patent/KR970010980B1/en

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
    • G01N27/02—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance
    • G01N27/04—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance
    • G01N27/20—Investigating the presence of flaws

Landscapes

  • Chemical & Material Sciences (AREA)
  • Chemical Kinetics & Catalysis (AREA)
  • Electrochemistry (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Carbon Steel Or Casting Steel Manufacturing (AREA)

Abstract

본 발명은 제철소내 제강공정에서의 전로 조업지원을 위한 기술로써, 양질의 강을 생산하는데 필수적인 강의 온도 및 성분 변화를 인공신경회로망을 이용하여 예측하는 방법에 관한 것으로써, 전로공정에 내재하는 상호 복잡히 얽힌 반응 상황들에 관계없이 센서에 의존하지 않고 전로 조업의 주요 목표인 용강온도 및 탄소성분 함량은 물론 인, 망간 등의 반응시간에 따른 성분변화를 일괄처리 할 수 있고 수식모델시 규명할 수 없는 요인들로 인하여 조업상황에 영향을 끼쳤던 문제들 및 다양한 조업조건들까지도 해결할 수 있는 인공회로망을 이용한 용강온도 및 성분 변화 예측방법을 제공하고자 하는데, 그 목적이 있다.The present invention relates to a technology for supporting converter operation in steelmaking processes in steelworks, and relates to a method for predicting temperature and component changes of steels essential to producing high-quality steels using an artificial neural network. Regardless of the complicated reaction conditions, it is possible to batch process the change of components according to reaction time such as phosphorus and manganese as well as molten steel temperature and carbon content, which are the main targets of converter operation, without relying on sensors. The purpose of the present invention is to provide a method of predicting molten steel temperature and component change using artificial networks that can solve problems and various operating conditions that have affected the operation situation due to numerous factors.

본 발명은 전로조업시 중요한 조업인자 및 조건의 주요항목들을 매 조업시 입력층, 중간층 및 출력층으로 구성되어 통상의 인공신경회로망에 입력하여 통상의 방법으로 가중치(weight)를 조정하므로써, 시간에 따른 용강온도 및 성분값을 구하는 단계: 상기한 조업인자 및 조건과 동일한 조업 인자 및 조건에서 얻어진 과거 연속 채취 데이타와 상기 인공신경회로망에서 구한 시간에 따른 용강온도 및 성분값을 비교하여 허용오차 범위내에 오도록 가중치를 통상의 방법으로 학습시키고, n번째 조업까지 학습시켰을 때의 오차의 합이 허용범위내에 들어왔을때 학습을 종료하는 단계: 상기와 같이 학습되어진 인공신 회로망을 학습시 적용시키지 않는 새로운 조업에 적용하는 단계를 포함하여 구성되는 인공신경회로망을 이용한 용강온도 및 성분변화 예측방법을 그 요지로 한다.The present invention is composed of the input layer, the middle layer and the output layer in the operation of the operation parameters and conditions important in the operation of the operation by inputting to the artificial artificial neural network in the usual way, by adjusting the weight (weight) in the usual way, Obtaining the molten steel temperature and the component value: comparing the continuous continuous data obtained under the same operating factors and conditions with the molten steel temperature and the component value over time obtained from the artificial neural network to be within the tolerance range. Learning the weights in the usual way, and terminating the learning when the sum of the errors when learning up to the nth operation falls within the allowable range: applying to the new operation that does not apply the artificial neural network learned as described above. Prediction of molten steel temperature and component change using artificial neural network That's the point.

Description

인공신경회로망을 이용한 용장온도 및 성분 변화예측방법Prediction of Redundant Temperature and Component Changes Using Artificial Neural Networks

본 내용은 요부공개 건이므로 전문내용을 수록하지 않았음Since this is an open matter, no full text was included.

제1도는 본 발명의 원리를 나타내는 기능 블록도,1 is a functional block diagram illustrating the principles of the present invention;

제2도는 본 발명에 적응되는 인공 신경회로망의 일례를 나타내는 전체 구성도,2 is an overall configuration diagram showing an example of an artificial neural network adapted to the present invention;

제3도는 본 발명에 있어서 신경회로망 적용시의 일례를 나타내는 흐름도,3 is a flowchart showing an example of neural network application in the present invention;

제4도는 전로공정에서 시간에 따른 탄소 농도변화를 나타내는 조업 실적도4 is an operation record showing the change of carbon concentration over time in the converter process.

Claims (2)

전로조업시 중요한 조업인자 및 조건의 주요항목들을 매 조업시 입력층, 중간층 및 출력층으로 구성되는 통상의 인공신경회로망에 입력하여 통상의 방법으로 가중치(weight)를 조정하므로써, 시간에 따른 용강온도 및 성분값을 구하는 단계; 상기한 조업인자 및 조건과 동일한 조업 인자 및 조건에서 얻어진 과거 연속 채취 데이타와 상기 인공신경회로망에서 구한 시간에 따른 용강온도 및 성분값을 비교하여 허용오차 범위내에 오도록 가중치를 통상의 방법으로 학습시키고, n번째 조업가지 학습 시켰을 때의 오차의 합이 허용범위내에 들어 왔을 때 학습을 종료하는 단계; 상기와 같이 학습되어진 인공시녕 회로망을 학습시 적용시키지 않는 새로운 조업에 적용하는 단계를 포함하여 구성됨을 특징으로 하는 인공신경회로망을 이용한 용감온도 및 성분변화 예측방법.Molten steel temperature and time by adjusting the weight in the usual way by inputting the main items of the important operation factors and conditions in the converter operation into a conventional artificial neural network consisting of the input layer, the middle layer and the output layer during each operation Obtaining a component value; The weight is trained in a conventional manner so as to be within an allowable range by comparing the molten steel temperature and the component values with time obtained from the artificial neural network and the historical continuous sampling data obtained under the same operation factors and conditions. terminating the learning when the sum of the errors when the n-th operation branch is within the allowable range is learned; A method for predicting brazing temperature and composition change using an artificial neural network, comprising applying the learned artificial neural network to a new operation not applied during learning. 제1항에 있어서, 인공신경회로망의 출력충에 시간항이 포함되어 있는 것을 특징으로 하는 인공신경 회로망을 이용한 용강온도 및 성분 변화 예측방법.The method of claim 1, wherein the output terminus of the artificial neural network includes a time term. ※ 참고사항 : 최초출원 내용에 의하여 공개하는 것임.※ Note: The disclosure is based on the initial application.
KR1019930031774A 1993-12-31 1993-12-31 Temperature and components in furnace bath using neural network KR970010980B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1019930031774A KR970010980B1 (en) 1993-12-31 1993-12-31 Temperature and components in furnace bath using neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1019930031774A KR970010980B1 (en) 1993-12-31 1993-12-31 Temperature and components in furnace bath using neural network

Publications (2)

Publication Number Publication Date
KR950019723A true KR950019723A (en) 1995-07-24
KR970010980B1 KR970010980B1 (en) 1997-07-05

Family

ID=19374705

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1019930031774A KR970010980B1 (en) 1993-12-31 1993-12-31 Temperature and components in furnace bath using neural network

Country Status (1)

Country Link
KR (1) KR970010980B1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110686409A (en) * 2019-08-19 2020-01-14 珠海格力电器股份有限公司 Water heater and bathing control method and system thereof

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110686409A (en) * 2019-08-19 2020-01-14 珠海格力电器股份有限公司 Water heater and bathing control method and system thereof

Also Published As

Publication number Publication date
KR970010980B1 (en) 1997-07-05

Similar Documents

Publication Publication Date Title
KR930013177A (en) Method of decarburizing molten metal using neural network during steel refining
CN109447346B (en) Converter oxygen consumption prediction method based on gray prediction and neural network combined model
MX9702973A (en) Model predictive control apparatus and method.
WO1994020887A3 (en) Method and apparatus for analyzing a neural network within desired operating parameter constraints
CN102766728B (en) Method and device for real-time prediction of sulfur content of molten steel in refining process of ladle refining furnace
EP0949318A3 (en) Process for determining the nitrogen content of the effluent of the pretreatment reactor in a catalytic hydrocracking plant
PL364460A1 (en) Method for controlling and driving a technical process
KR950019723A (en) Prediction Method of Molten Steel Temperature and Component Change Using Artificial Neural Network
CN111933221B (en) Method for predicting dynamic recrystallization fraction of Nb microalloyed steel
JPH07198588A (en) Creep deformation estimating method and creep life estimating method
Cateni et al. Prediction of steel hardenability and related reliability through neural networks
GB2003929A (en) Controlling converter blow end-point by waste gas analysis
JP2921970B2 (en) Converter end point control method
CN115058555B (en) Intelligent soft measurement method and system for measuring carbon content of converter endpoint
van der Wolk et al. Prediction of the continuous cooling transformation diagram of vanadium containing steels using artificial neural networks
JPS57161016A (en) Refining method of low sulfur high chromium steel
Lund Carburizing Steels: Hardenability prediction and hardenability control in steel-making
CN117153281A (en) Method, equipment and medium for predicting alloy recommendation rate and yield
JPS57131310A (en) Method for controlling content of phosphorus of molten steel
Alex et al. An empirical observer for wastewater treatment plants
Kavic Investigation of Process Influences on the Steel Composition and Inclusion Modification in Secondary Metallurgy using Diverse Statistical Learning and Data Mining Methods
JPH0657319A (en) Method for estimating manganese concentration in tapped steel from converter
Wang et al. Based on BP network terminal quality prediction for BOF steelmaking process
Hintikka Quality control of continuously cast slabs
Lund et al. The Measurement, Prediction and Control of Jominy-Hardenability of Carburizing Steels

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
G160 Decision to publish patent application
E701 Decision to grant or registration of patent right
GRNT Written decision to grant
FPAY Annual fee payment

Payment date: 20031002

Year of fee payment: 7

LAPS Lapse due to unpaid annual fee