WO2023070691A1 - Procédé de construction de modèle de trajectoire de charge de haut fourneau - Google Patents

Procédé de construction de modèle de trajectoire de charge de haut fourneau Download PDF

Info

Publication number
WO2023070691A1
WO2023070691A1 PCT/CN2021/128114 CN2021128114W WO2023070691A1 WO 2023070691 A1 WO2023070691 A1 WO 2023070691A1 CN 2021128114 W CN2021128114 W CN 2021128114W WO 2023070691 A1 WO2023070691 A1 WO 2023070691A1
Authority
WO
WIPO (PCT)
Prior art keywords
charge
population
trajectory model
shape data
surface shape
Prior art date
Application number
PCT/CN2021/128114
Other languages
English (en)
Chinese (zh)
Inventor
叶理德
高田翔
李清忠
余成明
刘书文
梁小兵
梅超凡
Original Assignee
中冶南方工程技术有限公司
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 中冶南方工程技术有限公司 filed Critical 中冶南方工程技术有限公司
Publication of WO2023070691A1 publication Critical patent/WO2023070691A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/10—Geometric CAD
    • G06F30/17—Mechanical parametric or variational design
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G06F30/28—Design optimisation, verification or simulation using fluid dynamics, e.g. using Navier-Stokes equations or computational fluid dynamics [CFD]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/12—Computing arrangements based on biological models using genetic models
    • G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T3/00—Geometric image transformations in the plane of the image
    • G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
    • G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/06—Multi-objective optimisation, e.g. Pareto optimisation using simulated annealing [SA], ant colony algorithms or genetic algorithms [GA]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2113/00—Details relating to the application field
    • G06F2113/08—Fluids
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces

Definitions

  • the invention relates to the technical field of blast furnace ironmaking, in particular to a method for constructing a blast furnace charge trajectory model.
  • the charge enters from the bunker at the top of the blast furnace, passes through the material flow valve, fork pipe, throat, and chute, and finally falls to the throat of the furnace. Then, as the charge is consumed, the charge at the throat of the furnace gradually falls To the furnace shaft to complete the conversion process.
  • the shape of the charge surface is crucial to the normal operation of the blast furnace. A good charge surface can make the gas flow distribution more reasonable and effectively fill the consumption of charge in different parts.
  • the traditional manual method is to measure the depth of several parts with a probe, and then make a rough estimate of the material level in combination with the gas composition and furnace top temperature, which requires the operator to have rich experience in material distribution, and it is difficult to accurately describe the continuous distribution of the material level .
  • the material level distribution of the furnace throat can be directly measured by technical means such as radar, but in practical applications, the service life of the radar is limited, and it is inconvenient to replace and maintain, and most radars can only measure one at a time. Point, to complete the scanning of the entire material surface needs to be measured tens of hundreds of times, if you want to scan both before and after the cloth, it will take a long time.
  • the present invention provides a method for building a blast furnace charge trajectory model, which includes the following steps,
  • the initial charge trajectory model formula is as follows:
  • v is the speed when the charge slides out of the chute
  • ⁇ is the inclination angle of the chute
  • r is the hydraulic radius of the material flow valve
  • H is the distance from the outlet of the chute to the material surface
  • t is the time for the charge to move in the air
  • R is The lateral distance between the charging point and the outlet of the chute
  • a, b, c, d, e are model parameters
  • acquiring the real material surface shape data in S3 specifically includes detecting the real material surface shape data of the furnace top material surface through radar scanning, and the real material surface shape data is a discrete point vector S with a fixed length.
  • model parameters in the initial charge trajectory model formula are optimized through a genetic algorithm, and the initial charge trajectory model formula
  • the model parameters in are iteratively updated as follows:
  • the construction of the initial population by the chromosome coding method in S301 specifically includes: the chromosome coding method adopts floating-point number coding, and the population individuals of the initial population include model parameters a, b, c, d, and e.
  • the cross mutation operation is performed on the reserved population individuals to obtain the next generation population operation, specifically: performing genetic crossover processing on the retained population individuals through a crossover operator, and performing genetic crossover processing on the two paired population individuals
  • the next generation of population individuals is obtained by exchanging genes with each other through single-point crossover; the reserved population individuals are subjected to gene mutation processing through a mutation operator, and after determining the gene mutation position in the population individuals, the mutation is performed by means of basic bit mutation Operation, inverting the original gene at the gene mutation position according to the preset probability.
  • Another object of the present invention is to provide a computer system, which is configured to perform the following steps: first, model the process of blast furnace charge descent to obtain the initial charge trajectory model formula; then obtain a charge level data set, and convert the The material surface data set is input to the initial furnace charge trajectory model formula to calculate the set of the drop position of the charge at the furnace throat, and the simulated charge is obtained according to the set of the charge surface data set and the drop position of the charge at the furnace throat. surface shape data; finally obtain the real material surface shape data, according to the real material surface shape data and the simulated material surface shape data, the model parameters in the initial charge trajectory model formula are optimized by genetic algorithm, and the The model parameters in the formula of the initial charge trajectory model described above are iteratively updated to obtain the charge trajectory model.
  • Another object of the present invention is to provide a computer-readable storage medium, which includes a memory, and a computer program is stored in the memory.
  • the computer program is executed by a processor, the above-mentioned method for building a blast furnace charge trajectory model is realized.
  • the beneficial effects of the present invention are: continuously adjust and optimize multiple parameters of the furnace charge trajectory model according to the radar scanning data, and finally obtain the most suitable furnace charge trajectory model for the current blast furnace, so that without the need for radar, through the distribution matrix directly
  • the shape of the material surface after the material distribution is calculated, and by simulating the genetic evolution of nature, the optimal model parameters of the furnace material trajectory model can be efficiently found in the solution space, which not only makes the modeling process simple but also improves the accuracy of the model.
  • Fig. 1 is a method block diagram of a method for constructing a blast furnace charge trajectory model of the present invention
  • Fig. 2 is the structural representation of blast furnace throat place of the present invention
  • Fig. 3 is a flow chart of the method for optimizing the model parameters of the blast furnace charge trajectory model of the present invention.
  • the present invention provides a method for building a blast furnace charge trajectory model, comprising the following steps,
  • the present invention continuously adjusts and optimizes multiple parameters of the furnace material trajectory model, and finally obtains the most suitable furnace material trajectory model for the current blast furnace, so that the material distribution matrix can be directly calculated through the material distribution matrix without the need for radar.
  • the shape of the charge surface by simulating the genetic evolution of nature, can efficiently find the optimal model parameters of the charge trajectory model in the solution space, which not only makes the modeling process simple but also improves the accuracy of the model.
  • the charge descending process is modeled using classical mechanical formulas to obtain a charge trajectory model formula.
  • the structure at the throat of the blast furnace is shown in Figure 2. Modeling is performed on the blast furnace charging process to obtain the initial charging trajectory model formula.
  • the initial charging trajectory model formula is as follows:
  • v is the speed when the charge slides out of the chute
  • ⁇ is the inclination angle of the chute
  • r is the hydraulic radius of the material flow valve
  • H is the distance from the outlet of the chute to the material surface
  • t is the time for the charge to move in the air
  • R is The lateral distance between the charging point and the outlet of the chute
  • a, b, c, d, e are model parameters.
  • the position of the drop point of the charge on the furnace throat can be calculated.
  • the material level data set includes the hydraulic radius of the material flow valve, the angle of the chute and the drop height, the total weight, density, and distribution matrix of the material batch.
  • the temperature of each gear of the blast furnace can be calculated. Cloth volume.
  • the shape of the material surface of the gear can be simulated. Repeat the above method to calculate the shape of each gear of the current fabric, and then obtain the simulated material surface shape data of the material surface after the current material batch is finished.
  • the real material surface shape data of the furnace top material surface is detected by radar scanning, and the real material surface shape data is a discrete point vector S with a fixed length.
  • the preset number of generations is generally set to 100-500 generations.
  • the chromosome encoding method adopts floating-point number encoding
  • the population individuals of the initial population include model parameters a, b, c, d, and e.
  • Set the lower limit of the solution space and the upper limit of the solution space of the model parameters set the lower limit of the solution space of the five parameters to ⁇ -50,-50,-50,-50,-50 ⁇ , and the upper limit of the solution space of the parameters to ⁇ 50 ,50,50,50,50 ⁇ , generate the first generation population ⁇ a1,b1,c1,d1,e1 ⁇ , ⁇ a2,b2,c2,d2,e2 ⁇ ,... ⁇ according to the chromosome coding method, the population size is 100.
  • the initial population is constructed by the chromosome coding method, and the population individuals in the initial population are input into the formula of the initial charge trajectory model to obtain the simulated material surface shape data of the population individual, wherein the simulated material surface shape data is a fixed-length discrete point vector C.
  • Carry out interpolation or elimination processing to described discrete point vector S obtain the applicability of described simulated material level shape data according to applicability function, make the discrete point vector S length of described real material level shape data and described simulated material level shape
  • the discrete point vector C of the data has the same length, and the applicability function is the opposite number of the two norm of the discrete point vector, then the applicability function f is specifically:
  • the first N population individuals are retained; the cross-mutation operation is performed on the retained population individuals to obtain the next generation Population, after inputting the population individuals in the current population into the formula of the initial charge trajectory model to obtain the simulated material surface shape data of the population individuals, carry out genetic crossover processing on the reserved population individuals through the crossover operator, and perform genetic crossover processing on the two mutual
  • the paired population individuals exchange genes with each other through single-point crossover, thereby forming the next generation of population individuals; the reserved population individuals are subjected to gene mutation processing through mutation operators, and after determining the gene mutation position in the population individuals, through the basic position
  • the mutation operation is carried out in the way of mutation, and the original gene at the mutation position of the gene is reversed according to the preset probability.
  • the population uses the fitness ratio method of the selection operation to select individuals with high applicability and keep them.
  • the optimized individuals in the population are directly inherited to the next generation through the selection operator selection, or new individuals are generated through pairing and crossover, and then passed on to the next generation.
  • using the crossover operator of single point crossover by selecting two chromosomes, splitting at a randomly selected position point and exchanging the right part, so as to obtain two different daughter chromosomes.
  • mutation and crossover operations are performed on the basis of selected individuals to generate the next generation of populations. Through crossover operations, two paired chromosomes exchange genes in a certain way to form a new population. individual.
  • the crossover operation plays a key role in the genetic algorithm, and changing the gene values at some loci of the individual strings in the population is the main method to generate new individuals.
  • Mutation operation is to change the gene value on a certain locus according to a small probability, and it is also an operation method to generate new individuals.
  • the present invention adopts the method of basic bit mutation to carry out the variation operation, and its specific operation process is: firstly determine the gene variation position of each individual, and then reverse the original gene of the variation point according to a certain probability. After the population P(t) undergoes selection, crossover and mutation operations, the next generation population P(t+1) is obtained. Genetic algorithm has local random search ability.
  • the local random search ability of the mutation operator can be used to accelerate the convergence to the parameters in the formula of the optimal charge trajectory model, and it can also make the genetic algorithm Population diversity can be maintained to prevent premature convergence. Select the population individual with the greatest applicability to iteratively update the model parameters in the initial charge trajectory model formula to obtain the charge trajectory model.
  • Another object of the present invention is to provide a computer system, which is configured to perform the following steps: first, model the process of blast furnace charge descent to obtain the initial charge trajectory model formula; then obtain a charge level data set, and convert the The material surface data set is input to the initial furnace charge trajectory model formula to calculate the set of the drop position of the charge at the furnace throat, and the simulated charge is obtained according to the set of the charge surface data set and the drop position of the charge at the furnace throat. surface shape data; finally obtain the real material surface shape data, according to the real material surface shape data and the simulated material surface shape data, the model parameters in the initial charge trajectory model formula are optimized by genetic algorithm, and the The model parameters in the formula of the initial charge trajectory model described above are iteratively updated to obtain the charge trajectory model.
  • Another object of the present invention is to provide a computer-readable storage medium, which includes a memory, and a computer program is stored in the memory.
  • the computer device mainly includes a processor, a memory 302 and a bus, the memory stores at least one program, and the program is executed by the processor to realize the configuration of the computer as described above.
  • the processor includes one or more processing cores.
  • the processor is connected to the memory through the bus.
  • the memory is used to store program instructions.
  • the memory can be realized by any type of volatile or non-volatile storage device or their combination, such as Static Anytime Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), Magnetic Memory, Flash Memory, Magnetic Disk or Optical Disk.
  • SRAM Static Anytime Access Memory
  • EEPROM Electrically Erasable Programmable Read Only Memory
  • EPROM Erasable Programmable Read Only Memory
  • PROM Programmable Read Only Memory
  • ROM Read Only Memory
  • Magnetic Memory Flash Memory
  • Flash Memory Magnetic Disk or Optical Disk.
  • the present invention proposes a computer-readable storage medium, at least one program is stored in the storage medium, and the at least one program is executed during operation to realize the configuration of the above-mentioned computer system and/or execute the above-mentioned construction method of the blast furnace charge trajectory model .

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Computer Hardware Design (AREA)
  • Software Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computing Systems (AREA)
  • Evolutionary Biology (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Physics (AREA)
  • Pure & Applied Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Genetics & Genomics (AREA)
  • Algebra (AREA)
  • Medical Informatics (AREA)
  • Computational Mathematics (AREA)
  • Fluid Mechanics (AREA)
  • Physiology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Biomedical Technology (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

Procédé de construction de modèle de trajectoire de charge de haut fourneau, qui consiste à : modéliser un processus de descente de charge de haut fourneau pour obtenir une formule de modèle de trajectoire de charge de four initial ; acquérir des données de forme de surface de matériau réel ; en fonction des données de forme de surface de matériau réel et des données de forme de surface de matériau simulé, optimiser les paramètres de modèle dans la formule de modèle de trajectoire de charge de four initiale au moyen d'un algorithme génétique pour comparer la valeur d'erreur des individus optimaux ; et si la valeur d'erreur est inférieure à l'erreur minimale précédente, mettre à jour les paramètres dans la formule de modèle de trajectoire de charge de four à l'aide des paramètres de chromosomes de l'individu. Selon le procédé, le processus de modélisation est simple, et la précision du modèle est améliorée.
PCT/CN2021/128114 2021-10-26 2021-11-02 Procédé de construction de modèle de trajectoire de charge de haut fourneau WO2023070691A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111244739.3A CN114139438A (zh) 2021-10-26 2021-10-26 一种高炉炉料轨迹模型构建方法
CN202111244739.3 2021-10-26

Publications (1)

Publication Number Publication Date
WO2023070691A1 true WO2023070691A1 (fr) 2023-05-04

Family

ID=80395158

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2021/128114 WO2023070691A1 (fr) 2021-10-26 2021-11-02 Procédé de construction de modèle de trajectoire de charge de haut fourneau

Country Status (2)

Country Link
CN (1) CN114139438A (fr)
WO (1) WO2023070691A1 (fr)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114921598B (zh) * 2022-04-27 2023-05-09 中南大学 一种高炉炉顶炉料运动轨迹建模方法及***

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4400206A (en) * 1981-05-28 1983-08-23 Kawasaki Steel Corporation Process for estimating particle size segregation of burden layer in blast furnace top
CN102629286A (zh) * 2012-02-24 2012-08-08 北京首钢自动化信息技术有限公司 一种基于智能算法的高炉布料数值模拟方法
CN102653801A (zh) * 2012-04-25 2012-09-05 北京科技大学 基于遗传算法建立的无料钟高炉炉顶布料控制方法
CN112176136A (zh) * 2020-09-24 2021-01-05 中南大学 一种高炉u型溜槽上炉料运动轨迹建模方法及***

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4400206A (en) * 1981-05-28 1983-08-23 Kawasaki Steel Corporation Process for estimating particle size segregation of burden layer in blast furnace top
CN102629286A (zh) * 2012-02-24 2012-08-08 北京首钢自动化信息技术有限公司 一种基于智能算法的高炉布料数值模拟方法
CN102653801A (zh) * 2012-04-25 2012-09-05 北京科技大学 基于遗传算法建立的无料钟高炉炉顶布料控制方法
CN112176136A (zh) * 2020-09-24 2021-01-05 中南大学 一种高炉u型溜槽上炉料运动轨迹建模方法及***

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
ZHU, QINGTIAN ET AL.: "Mathematical Model of Burden Trajectory in a Blast Furnace", JOURNAL OF UNIVERSITY OF SCIENCE AND TECHNOLOGY BEIJING, vol. 29, no. 9, 30 September 2007 (2007-09-30), pages 932 - 936, XP009545905 *

Also Published As

Publication number Publication date
CN114139438A (zh) 2022-03-04

Similar Documents

Publication Publication Date Title
CN111191316B (zh) 一种基于响应面的建筑自然通风性能优化模型及优化方法
JP3404532B2 (ja) 最適フィッティングパラメータ決定方法および装置、並びに最適フィッティングパラメータ決定プログラム
CN111832101B (zh) 一种水泥强度预测模型的构建方法及水泥强度预测方法
CN114692265B (zh) 一种基于深度强化学习的零碳建筑优化设计方法
CN106022473A (zh) 一种融合粒子群和遗传算法的基因调控网络构建方法
WO2023070691A1 (fr) Procédé de construction de modèle de trajectoire de charge de haut fourneau
CN114969953B (zh) 基于CatBoost-NSGA-Ⅲ的盾构下穿隧道优化设计方法及设备
CN110399928A (zh) 固体氧化物燃料电池电压预测方法、终端设备及存储介质
CN111833970B (zh) 一种水泥熟料质量表征参数预测模型构建方法及其应用
CN105786759A (zh) 一种改进标准混合蛙跳算法的方法
CN108805346A (zh) 一种基于多隐层极限学习机的热连轧轧制力预报方法
CN112069656A (zh) 基于lssvm-nsgaii耐久性混凝土配合比多目标优化的方法
CN110046434A (zh) 基于高斯过程回归和遗传算法的机械零件可靠性设计方法
JP2003288579A (ja) 最適フィッティングパラメータ決定方法および装置、並びに最適フィッティングパラメータ決定プログラム
CN116542001B (zh) 基于改进谱聚类及遗传算法的供水管网独立计量分区方法
CN111144569A (zh) 一种基于遗传算法的良率提升适用模型的优化方法
CN110210072A (zh) 基于近似模型及差分进化算法求解高维优化问题的方法
CN115481510B (zh) 基于改进nsga-ⅱ算法的径向压气机叶片多工况优化方法及装置
CN113935556B (zh) 一种基于dna遗传算法的温度传感器优化布置方法
CN116663660A (zh) 一种基于基因评价遗传算法的一维下料方法
CN113378276B (zh) 一种基于遗传算法及基因表达式编程的复合地基智能设计方法
CN112861276B (zh) 基于数据与知识双驱动的高炉料面优化方法
CN114742173A (zh) 一种基于神经网络的变压器故障诊断方法及***
Showkatbakhsh et al. Multi-objective optimization of robotically bent in-situ reinforcement system
Vitingerova Evolutionary algorithms for multi-objective parameter estimation

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21962041

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE