CN105925750A - Steelmaking end point prediction method based on neural networks - Google Patents
Steelmaking end point prediction method based on neural networks Download PDFInfo
- Publication number
- CN105925750A CN105925750A CN201610314715.3A CN201610314715A CN105925750A CN 105925750 A CN105925750 A CN 105925750A CN 201610314715 A CN201610314715 A CN 201610314715A CN 105925750 A CN105925750 A CN 105925750A
- Authority
- CN
- China
- Prior art keywords
- converter
- information
- neutral net
- flame
- steelmaking
- Prior art date
- Legal status (The legal status 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 status listed.)
- Pending
Links
Classifications
-
- C—CHEMISTRY; METALLURGY
- C21—METALLURGY OF IRON
- C21C—PROCESSING OF PIG-IRON, e.g. REFINING, MANUFACTURE OF WROUGHT-IRON OR STEEL; TREATMENT IN MOLTEN STATE OF FERROUS ALLOYS
- C21C5/00—Manufacture of carbon-steel, e.g. plain mild steel, medium carbon steel or cast steel or stainless steel
- C21C5/28—Manufacture of steel in the converter
- C21C5/30—Regulating or controlling the blowing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Chemical & Material Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Metallurgy (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Materials Engineering (AREA)
- Organic Chemistry (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Manufacturing & Machinery (AREA)
- Carbon Steel Or Casting Steel Manufacturing (AREA)
Abstract
The invention relates to a steelmaking end point prediction method based on neural networks. The neural networks are used for replacing manual experience to predict the end point of steelmaking of a medium and small sized converter, a plurality of sets of production parameters are collected to serve as independent variables, flame temperature information and flame spectral information in the converter are collected through optical fibers to serve as independent variables, the multiple sets are used for training three layers of BP neural networks, MIV is adopted to screen the independent variables, the independent variable most affecting the end point is selected out to serve as an input parameter, an input parameter of the converter to be measured is selected to be inputted to the trained neural networks, and predicted converter end point temperature and the end point carbon content are obtained; and the shortcoming of manual experience prediction of the end point of steelmaking of the medium and small sized converter is overcome, optical fiber conduction is used for precisely measuring flame information inside the converter, and the accurate neutral networks are obtained.
Description
Technical field
The present invention relates to steel-making automation control area, more particularly, it relates to a kind of BOF Steelmaking Endpoint Forecasting Methodology.
Background technology
Terminal point control of making steel in pneumatic steelmaking is one of key technology of pneumatic steelmaking, and pneumatic steelmaking steel output accounts for more than the 80% of total steel output.Occupy an leading position in big-and-middle-sized Key Iron And Steel converter steel yield, therefore improve and improve the production capacity of pneumatic steelmaking and control level and be constantly subjected to the attention of people.Pneumatic steelmaking is sufficiently complex metallurgical reaction process, and influence factor is a lot.In order to realize automatically controlling of converter steelmaking process, developing many detection techniques both at home and abroad, conventional method mainly has artificial experience method, chemical analysis, static terminal point control, sublance method, analysis of fumes method etc..
Pneumatic steelmaking is an extremely complex process, the terminal point control bessemerized is the important operation bessemerizing latter stage, generally allowing phosphorus, removal of sulphur in the scope required by terminal the most in advance, therefore terminal point control is briefly the control of carbon mass fraction and temperature.In China, primary converter uses at present artificial experience control and traditional static models are extremely difficult to the control accuracy required.And along with people's raising to the prescription of steel, certainly will require to use effective control technology to improve converter smelting level.At present, Japan and more American-European integrated mills, on the basis of static models, are aided with the means such as sublance, analysis of fumes, optic probe and slag on-line checking, successfully achieve the omnidistance dynamically auto-control steel-making of converter.But, the measurement time of chemical analysis can not meet far away the requirement of real-time that smelting process controls, and there is the accident of splash when sampling.Sublance method Target hit rate is high, but expensive, and probe belongs to consumable goods simultaneously, it is impossible to continuously acquire blowing information, heat size is required height, is typically suitable only for more than 100t converter.At present, the device such as combustion gas analyzer and sublance is only used in some large-scale steel mills.The most domestic molten iron that there are about more than 60% is blown by mini-medium BOF plants, due to the restriction of fire door size, it is difficult to use the automation equipments such as sublance, still uses the experience control mode manually seeing fire, causes control accuracy relatively low.
Modeling method conventional in process control can be divided into three major types substantially: white-box model (mechanism model), black-box model (statistical model), grey-box model (model that mechanism and statistics combine).Owing to the mechanism model of complex process is difficult to set up, traditional Optimized-control Technique based on mathematical models is often difficult to be applied in actual production.Along with the maximization of modern industry process units, synthesization, complication, there is non-linear, uncertain, large dead time, parameter distribution and time variation in industrial object, process model building difficulty increases, and needs integrated use control theory, information processing, Modern statistics theory and optimisation technique to realize process flow industry process modeling.The artificial intelligence approach successful Application in a lot of practical problems in recent years makes researchers be introduced into steelmaking process one after another, and steelmaking process can be predicted and control by alternative mechanism model by expectation accurately.The ability to non-linear process matching having due to neural net method and easy implementation, many scholars have used it for the modeling to convertor steelmaking process, and have achieved certain success.
Artificial neural network (Artificial Neural Networks) also referred to as neutral net (NNs) or referred to as link model (Connection Model), refer to the neutral net constructed during biology learns in the field such as information and computer science.It is a kind of imitation human brain neural network's behavior characteristics, a large amount of neurons extensively interconnect and become a kind of complex networks system, carry out the algorithm mathematics model of distributed parallel information processing.This network relies on the complexity of system, by adjusting interconnective relation between internal great deal of nodes, thus reaches the purpose of process information.
Patent documentation 2004100568923 discloses employing artificial neural network to predict the technical scheme of BOF Steelmaking Endpoint temperature and phosphorus content, but there is problems of, the parameter of training neutral net is more miscellaneous accurate not, causes the neutral net obtained can there is bigger forecast error.Affect a lot of because have of endpoint carbon content of converter and temperature, using fire door flame information disclosed in patent documentation 2011103240380 is major parameter, neutral net is jointly trained in conjunction with other parameters, predict carbon content and the temperature of converter terminal, the document exists the accurate not problem of flame acquisition of information, causes forecast error bigger than normal.
Summary of the invention
In view of this, the invention provides a kind of steel-making end-point prediction method based on neutral net, the flame information of employing precise acquisition and other manufacturing parameter information, as variable, to realize improving control accuracy and hit rate, improve converter producing efficiency, the purpose of product quality.Technical solution of the present invention is as follows:
A kind of steel-making end-point prediction method based on neutral net, including: gathering the parameter information in the pneumatic steelmaking of many groups by producing equipment, described parameter information includes the weight of molten iron, the temperature of molten iron, carbon content, waste steel quality, oxygen gun blowing time, oxygen rifle position and blowing oxygen quantity；Arranging passage at converter sidewall, obtain the flame information within converter by optical fiber, described flame information includes spectral information and temperature information；
Using the weight of described molten iron, the temperature of molten iron, carbon content, oxygen gun blowing time, blowing oxygen quantity, flame spectrum information and flame temperature information as independent variable, utilize described independent variable composing training sample；
Set up three layers of BP neutral net, utilize described training sample that the BP neutral net set up is trained, described training sample is input to three layers of BP neutral net, it is calculated outlet temperature and endpoint carbon content, outlet temperature and endpoint carbon content and the actual outlet temperature measured and endpoint carbon content contrast will be calculated, until the differing sufficiently small or be zero of value of calculation and actual value, so that it is determined that the intermediate layer weights of three layers of BP neutral net and threshold value；
Use MIV algorithm that all of described independent variable is screened, filter out the independent variable that the influence degree predicted the outcome is reached preset standard；By described neutral net using the independent variable that the described influence degree predicted the outcome is reached preset standard filtered out as inputting parameter；
Choosing the input parameter of steelmaking converter to be predicted, be input in the BP neutral net trained, neutral net is given and predicts the outcome BOF Steelmaking Endpoint；Described BOF Steelmaking Endpoint includes terminal time, carbon content and liquid steel temperature.
Further, three layers of BP neutral net include input layer, intermediate layer and output layer.
Further, optical fiber sends the flame information within converter to spectral analysis module and image processing module.
Further, spectral analysis module includes that spectroanalysis instrument, spectral analysis module analysis provide the flame spectrum information data within converter.
Further, image processing module includes spectral module and CCD, by separating treatment optical fiber converter internal flame information everywhere, isolated infrared light is imported CCD, the photo collected utilizes two-color thermometry record the flame temperature information data in converter.
Further, the data of 300-500 heat are chosen as training sample.
Further, the influence degree that carbon content predicts the outcome reaches the independent variable of preset standard and includes: flame temperature, flame spectrum information, blowing oxygen quantity, oxygen gun blowing time.
Further, every three months periodic detection, when converter terminal predict the outcome with actual deviation more than predetermined value time, gather steelmaking converter parameter information as training sample, three layers of BP neutral net of re-training.
The invention has the beneficial effects as follows: use neural network BOF Steelmaking Endpoint forecast model, overcome the deficiency of medium and small BOF Steelmaking Endpoint artificial experience prediction, predict the terminal of pneumatic steelmaking more accurately；Utilize optical fiber to pass to accurately and determine the flame information within converter, the parameter information accurately recorded is used for the training of neutral net, make BP neural network prediction error little, it is possible to predict outlet temperature and the terminal phosphorus content information of pneumatic steelmaking the most accurately.
Accompanying drawing explanation
Fig. 1 is the flow chart of embodiment of the present invention BOF Steelmaking Endpoint prediction.
Detailed description of the invention
Describe the present invention the most by way of example in detail.
As it is shown in figure 1, be the flow chart of embodiment of the present invention BOF Steelmaking Endpoint based on neutral net prediction.The method includes:
S1, the parameter information gathered in the pneumatic steelmaking of many groups, parameter information includes the weight of molten iron, the temperature of molten iron, carbon content, waste steel quality, oxygen gun blowing time, oxygen rifle position and blowing oxygen quantity；These parameters are directly obtained by production equipment.Owing in converter, molten steel shows that reaction is acutely, and there is slag to cover, the conventional direct flame information collected from fire door is inaccurate, therefore by arranging a vent at converter sidewall in the present embodiment, go out to be provided with light at vent and passage is set at converter sidewall, obtaining the flame information within converter by optical fiber, flame information includes spectral information and temperature information；Optical fiber is with the material of printing opacity high temperature resistant, high, such as quartz glass, crystalline ceramics etc..
S2, using the weight of molten iron, the temperature of molten iron, carbon content, oxygen gun blowing time, blowing oxygen quantity, flame spectrum information and flame temperature information as independent variable, utilize described independent variable composing training sample；
Setting up three layers of BP neutral net, the BP neutral net set up is trained by the supplemental characteristic choosing 300-500 stove as training sample.In the present embodiment, three layers of BP neutral net are trained by preferably 400 groups converter supplemental characteristics.There is error in the parameter detecting of converter, the data chosen are very few, there is certain randomness, and the neutral net error that so training obtains will be bigger, also can be the biggest to the error of end-point prediction；The converter overabundance of data chosen can introduce unnecessary data deviation, because have chosen too much data, occurs that the probability of the biggest data point of deviation will improve, and easily causes the deviation of training neutral net, and chooses too much data, can increase workload.Prove through test of many times, use the supplemental characteristic training of 300-500 group to obtain neutral net error minimum.Less initial value is composed to intermediate layer weights and threshold value.Training sample is input to three layers of BP neutral net, it is calculated outlet temperature and endpoint carbon content, outlet temperature and endpoint carbon content and the actual outlet temperature measured and endpoint carbon content contrast will be calculated, difference according to value of calculation with actual value adjusts intermediate layer weights and the size of threshold value, repetition training step, until the differing sufficiently small or be zero of value of calculation and actual value, so that it is determined that the intermediate layer weights of three layers of BP neutral net and threshold value；Obtain the three layers of BP artificial neural network trained.
S3, each independent variable in above-mentioned training sample is carried out MIV value calculate；MIV value calculates and includes again:
1, by initial value S corresponding for the independent variable of MIV value to be calculated plus/minus a%S respectively, new training sample P1 and P2 is constituted；That is, training sample P1 includes the initial value of other independent variables, and value S-a%S that the independent variable of MIV value to be calculated is corresponding, and training sample P2 includes the initial value that other independent variables are corresponding, and value S+a%S that the independent variable of MIV value to be calculated is corresponding.
2, utilize the neutral net obtained after above-mentioned training to carry out simulation and prediction using P1 and P2 as simulation sample, obtain two simulation and prediction results A1 and A2；
3, obtain the difference of A1 and A2, affect changing value IV as after variation independent variable to what output produced；
4, IV is averagely drawn, by observation number of cases, the MIV value that the independent variable of MIV value to be calculated is corresponding.
5, according to the size of the absolute value of MIV value, independent variable is ranked up (absolute value of MIV value is the biggest, represents that its influence degree is the biggest), chooses absolute value and reach the independent variable of preset value.Above-mentioned preset value is corresponding with preset standard.
By described neutral net using the independent variable that the described influence degree predicted the outcome is reached preset standard filtered out as inputting parameter；
The independent variable filtered out can be used for instructing acquisition step, simplifies the information category gathered.Saving time, resource and human cost.
S4, choosing the input parameter of steelmaking converter to be predicted, be input in the BP neutral net trained, neutral net is given and predicts the outcome BOF Steelmaking Endpoint；BOF Steelmaking Endpoint includes terminal time, carbon content and liquid steel temperature.
S5, every three months periodic detection, when predicting the outcome of converter terminal and actual deviation are bigger, gather the parameter information of steelmaking converter as training sample, three layers of BP neutral net of re-training.After steelmaking converter uses a period of time, the various manufacturing parameters of converter all can occur a certain degree of change, so that the prediction of BP neutral net produces deviation, therefore periodic detection, inspection neural network prediction end point values and the error of actual value, if error is excessive, more than predetermined value, then Resurvey data, three layers of BP neutral net of re-training, it can be ensured that the accuracy that converter terminal is predicted by neutral net.
The present invention uses neural network BOF Steelmaking Endpoint forecast model, overcomes the deficiency of medium and small BOF Steelmaking Endpoint artificial experience prediction, predicts the terminal of pneumatic steelmaking more accurately；Utilize optical fiber to pass to accurately and determine the flame information within converter, the parameter information accurately recorded is used for the training of neutral net, make BP neural network prediction error little, it is possible to predict outlet temperature and the terminal phosphorus content information of pneumatic steelmaking the most accurately；The converter data message using 300-500 group trains three layers of BP neutral net as training sample, it is possible to obtain accurate neutral net；Use MIV algorithm to screen out the independent variable less to end-point prediction effect, improve efficiency, decrease the workload of data acquisition；By the error of periodic inspection BP neutral net end-point prediction value Yu actual measured value, adjust in time, re-training neutral net, it is ensured that the accuracy of neural network prediction.
Claims (8)
1. a steel-making end-point prediction method based on neutral net, it is characterized in that, including: gathering the parameter information in the pneumatic steelmaking of many groups by producing equipment, described parameter information includes the weight of molten iron, the temperature of molten iron, carbon content, waste steel quality, oxygen gun blowing time, oxygen rifle position and blowing oxygen quantity；Arranging passage at converter sidewall, obtain the flame information within converter by optical fiber, described flame information includes spectral information and temperature information；
Using the weight of described molten iron, the temperature of molten iron, carbon content, oxygen gun blowing time, blowing oxygen quantity, flame spectrum information and flame temperature information as independent variable, utilize described independent variable composing training sample；
Set up three layers of BP neutral net, utilize described training sample that the BP neutral net set up is trained, described training sample is input to three layers of BP neutral net, it is calculated outlet temperature and endpoint carbon content, outlet temperature and endpoint carbon content and the actual outlet temperature measured and endpoint carbon content contrast will be calculated, until the differing sufficiently small or be zero of value of calculation and actual value, so that it is determined that the intermediate layer weights of three layers of BP neutral net and threshold value；
Use MIV algorithm that all of described independent variable is screened, filter out the independent variable that the influence degree predicted the outcome is reached preset standard；By described neutral net using the independent variable that the described influence degree predicted the outcome is reached preset standard filtered out as inputting parameter；
Choosing the input parameter of steelmaking converter to be predicted, be input in the BP neutral net trained, neutral net is given and predicts the outcome BOF Steelmaking Endpoint；Described BOF Steelmaking Endpoint includes carbon content and liquid steel temperature.
Forecasting Methodology the most according to claim 1, it is characterised in that described three layers of BP neutral net include input layer, intermediate layer and output layer.
Forecasting Methodology the most according to claim 2, it is characterised in that described optical fiber sends the flame information within converter to spectral analysis module and image processing module.
Forecasting Methodology the most according to claim 3, it is characterised in that described spectral analysis module includes that spectroanalysis instrument, spectral analysis module analysis provide the flame spectrum information data within converter.
Forecasting Methodology the most according to claim 4, it is characterized in that, described image processing module includes spectral module and CCD, by separating treatment optical fiber converter internal flame information everywhere, isolated infrared light is imported CCD, the photo collected utilizes two-color thermometry record the flame temperature information data in converter.
6. according to the Forecasting Methodology described in any one of claim 1-5, it is characterised in that choose the data of 300-500 heat as training sample.
Forecasting Methodology the most according to claim 6, it is characterised in that the influence degree that described carbon content predicts the outcome reaches the independent variable of preset standard and includes: flame temperature, flame spectrum information, blowing oxygen quantity, oxygen gun blowing time.
Forecasting Methodology the most according to claim 7, it is characterised in that every three months periodic detection, when converter terminal predict the outcome with actual deviation more than predetermined value time, gather steelmaking converter parameter information as training sample, three layers of BP neutral net of re-training.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610314715.3A CN105925750A (en) | 2016-05-13 | 2016-05-13 | Steelmaking end point prediction method based on neural networks |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610314715.3A CN105925750A (en) | 2016-05-13 | 2016-05-13 | Steelmaking end point prediction method based on neural networks |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105925750A true CN105925750A (en) | 2016-09-07 |
Family
ID=56835792
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610314715.3A Pending CN105925750A (en) | 2016-05-13 | 2016-05-13 | Steelmaking end point prediction method based on neural networks |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105925750A (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446405A (en) * | 2016-09-23 | 2017-02-22 | 北京大学深圳研究生院 | Integrated circuit device neural network modeling sample selecting method and device |
| CN108251591A (en) * | 2018-01-15 | 2018-07-06 | 上海大学 | Utilize the top bottom blowing converter producing process control method of LSTM systems |
| CN108676955A (en) * | 2018-05-02 | 2018-10-19 | 中南大学 | A kind of BOF Steelmaking Endpoint carbon content and temprature control method |
| CN108803705A (en) * | 2018-08-21 | 2018-11-13 | 成渝钒钛科技有限公司 | Temperature optimization control method, control device and the application of a kind of steelmaking system and computer readable storage medium |
| CN109186967A (en) * | 2018-07-18 | 2019-01-11 | 西安交通大学 | A kind of rubber shock absorber performance prediction and selection method based on BP neural network |
| CN109234491A (en) * | 2018-11-20 | 2019-01-18 | 北京科技大学 | A kind of BOF Steelmaking Endpoint manganese content prediction technique based on extreme learning machine |
| CN109252009A (en) * | 2018-11-20 | 2019-01-22 | 北京科技大学 | BOF Steelmaking Endpoint manganese content prediction technique based on regularization extreme learning machine |
| CN109975507A (en) * | 2019-04-28 | 2019-07-05 | 华北理工大学 | A kind of real-time determining method and system for making steel later period carbon content of molten steel and temperature value |
| CN110533032A (en) * | 2019-08-22 | 2019-12-03 | 江苏联峰实业有限公司 | A kind of method and apparatus obtaining high-purity cryogenic steel |
| CN111079537A (en) * | 2019-11-18 | 2020-04-28 | 中冶赛迪技术研究中心有限公司 | Method, system, machine readable medium and equipment for identifying smelting working condition of converter |
| CN112195302A (en) * | 2020-10-16 | 2021-01-08 | 中冶赛迪技术研究中心有限公司 | Method and device for predicting electric precipitation explosion risk of primary flue gas of converter |
| CN112907584A (en) * | 2021-01-08 | 2021-06-04 | 昆明理工大学 | Converter steelmaking end point carbon content prediction method for improving MTBCD flame image feature extraction |
| CN113033705A (en) * | 2021-04-22 | 2021-06-25 | 江西理工大学 | Intelligent judgment and verification method for copper converter blowing slagging period end point based on pattern recognition |
| CN113255102A (en) * | 2021-04-23 | 2021-08-13 | 北京科技大学 | Method and device for predicting carbon content and temperature of molten steel at converter end point |
| TWI739364B (en) * | 2019-04-02 | 2021-09-11 | 日商Ｊｆｅ鋼鐵股份有限公司 | The splash prediction method of the converter, the operation method of the converter and the splash prediction system of the converter |
| CN113487520A (en) * | 2021-09-07 | 2021-10-08 | 南通宏耀锅炉辅机有限公司 | High dynamic range image generation method and system based on converter temperature measurement |
| CN114414648A (en) * | 2022-03-29 | 2022-04-29 | 联泰集群（北京）科技有限责任公司 | Automatic potentiometric titration method and system based on machine learning |
| CN115715331A (en) * | 2020-07-01 | 2023-02-24 | 杰富意钢铁株式会社 | Converter blowing control method and converter blowing control system |
| CN116083676A (en) * | 2021-11-08 | 2023-05-09 | 清华大学 | Method, device, equipment and system for monitoring converter steelmaking process |
| CN116128165A (en) * | 2023-04-13 | 2023-05-16 | 深圳大学 | MIV-BP-based building element quality prediction method and system |
| CN117553921A (en) * | 2024-01-12 | 2024-02-13 | 山东钢铁股份有限公司 | Converter molten steel temperature prediction method, system, terminal and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06264129A (en) * | 1992-06-12 | 1994-09-20 | Kawasaki Steel Corp | Method for controlling end point of steelmaking in converter |
| CN1588346A (en) * | 2004-08-30 | 2005-03-02 | 邢台钢铁有限责任公司 | Method for predicting converter terminal point using artificial nurve network technology |
| CN102392095A (en) * | 2011-10-21 | 2012-03-28 | 湖南镭目科技有限公司 | Termination point prediction method and system for converter steelmaking |
| CN102876838A (en) * | 2012-10-30 | 2013-01-16 | 湖南镭目科技有限公司 | System for detecting carbon content and temperature in converter |
-
2016
- 2016-05-13 CN CN201610314715.3A patent/CN105925750A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06264129A (en) * | 1992-06-12 | 1994-09-20 | Kawasaki Steel Corp | Method for controlling end point of steelmaking in converter |
| CN1588346A (en) * | 2004-08-30 | 2005-03-02 | 邢台钢铁有限责任公司 | Method for predicting converter terminal point using artificial nurve network technology |
| CN102392095A (en) * | 2011-10-21 | 2012-03-28 | 湖南镭目科技有限公司 | Termination point prediction method and system for converter steelmaking |
| CN102876838A (en) * | 2012-10-30 | 2013-01-16 | 湖南镭目科技有限公司 | System for detecting carbon content and temperature in converter |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106446405A (en) * | 2016-09-23 | 2017-02-22 | 北京大学深圳研究生院 | Integrated circuit device neural network modeling sample selecting method and device |
| CN106446405B (en) * | 2016-09-23 | 2018-12-18 | 北京大学深圳研究生院 | A kind of integrated circuit device neural net model establishing Method of Sample Selection and device |
| CN108251591A (en) * | 2018-01-15 | 2018-07-06 | 上海大学 | Utilize the top bottom blowing converter producing process control method of LSTM systems |
| CN108676955A (en) * | 2018-05-02 | 2018-10-19 | 中南大学 | A kind of BOF Steelmaking Endpoint carbon content and temprature control method |
| CN108676955B (en) * | 2018-05-02 | 2019-07-12 | 中南大学 | A kind of BOF Steelmaking Endpoint carbon content and temprature control method |
| CN109186967A (en) * | 2018-07-18 | 2019-01-11 | 西安交通大学 | A kind of rubber shock absorber performance prediction and selection method based on BP neural network |
| CN109186967B (en) * | 2018-07-18 | 2019-10-11 | 西安交通大学 | A kind of rubber shock absorber performance prediction and selection method based on BP neural network |
| CN108803705A (en) * | 2018-08-21 | 2018-11-13 | 成渝钒钛科技有限公司 | Temperature optimization control method, control device and the application of a kind of steelmaking system and computer readable storage medium |
| CN109234491A (en) * | 2018-11-20 | 2019-01-18 | 北京科技大学 | A kind of BOF Steelmaking Endpoint manganese content prediction technique based on extreme learning machine |
| CN109252009A (en) * | 2018-11-20 | 2019-01-22 | 北京科技大学 | BOF Steelmaking Endpoint manganese content prediction technique based on regularization extreme learning machine |
| TWI739364B (en) * | 2019-04-02 | 2021-09-11 | 日商Ｊｆｅ鋼鐵股份有限公司 | The splash prediction method of the converter, the operation method of the converter and the splash prediction system of the converter |
| CN109975507A (en) * | 2019-04-28 | 2019-07-05 | 华北理工大学 | A kind of real-time determining method and system for making steel later period carbon content of molten steel and temperature value |
| CN110533032A (en) * | 2019-08-22 | 2019-12-03 | 江苏联峰实业有限公司 | A kind of method and apparatus obtaining high-purity cryogenic steel |
| CN111079537A (en) * | 2019-11-18 | 2020-04-28 | 中冶赛迪技术研究中心有限公司 | Method, system, machine readable medium and equipment for identifying smelting working condition of converter |
| CN111079537B (en) * | 2019-11-18 | 2023-09-26 | 中冶赛迪技术研究中心有限公司 | Method, system, machine-readable medium and equipment for identifying smelting working conditions of converter |
| CN115715331A (en) * | 2020-07-01 | 2023-02-24 | 杰富意钢铁株式会社 | Converter blowing control method and converter blowing control system |
| CN112195302A (en) * | 2020-10-16 | 2021-01-08 | 中冶赛迪技术研究中心有限公司 | Method and device for predicting electric precipitation explosion risk of primary flue gas of converter |
| CN112195302B (en) * | 2020-10-16 | 2023-05-12 | 中冶赛迪技术研究中心有限公司 | Method and device for predicting primary flue gas electric dust removal explosion risk of converter |
| CN112907584A (en) * | 2021-01-08 | 2021-06-04 | 昆明理工大学 | Converter steelmaking end point carbon content prediction method for improving MTBCD flame image feature extraction |
| CN113033705B (en) * | 2021-04-22 | 2022-12-02 | 江西理工大学 | Intelligent judgment and verification method for copper converter blowing slagging period end point based on pattern recognition |
| CN113033705A (en) * | 2021-04-22 | 2021-06-25 | 江西理工大学 | Intelligent judgment and verification method for copper converter blowing slagging period end point based on pattern recognition |
| CN113255102B (en) * | 2021-04-23 | 2022-02-08 | 北京科技大学 | Method and device for predicting carbon content and temperature of molten steel at converter end point |
| CN113255102A (en) * | 2021-04-23 | 2021-08-13 | 北京科技大学 | Method and device for predicting carbon content and temperature of molten steel at converter end point |
| CN113487520B (en) * | 2021-09-07 | 2021-11-05 | 南通宏耀锅炉辅机有限公司 | High dynamic range image generation method and system based on converter temperature measurement |
| CN113487520A (en) * | 2021-09-07 | 2021-10-08 | 南通宏耀锅炉辅机有限公司 | High dynamic range image generation method and system based on converter temperature measurement |
| CN116083676A (en) * | 2021-11-08 | 2023-05-09 | 清华大学 | Method, device, equipment and system for monitoring converter steelmaking process |
| CN114414648A (en) * | 2022-03-29 | 2022-04-29 | 联泰集群（北京）科技有限责任公司 | Automatic potentiometric titration method and system based on machine learning |
| CN116128165A (en) * | 2023-04-13 | 2023-05-16 | 深圳大学 | MIV-BP-based building element quality prediction method and system |
| CN117553921A (en) * | 2024-01-12 | 2024-02-13 | 山东钢铁股份有限公司 | Converter molten steel temperature prediction method, system, terminal and storage medium |
| CN117553921B (en) * | 2024-01-12 | 2024-04-19 | 山东钢铁股份有限公司 | Converter molten steel temperature prediction method, system, terminal and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105925750A (en) | Steelmaking end point prediction method based on neural networks | |
| CN102392095B (en) | Termination point prediction method and system for converter steelmaking | |
| CN108764517B (en) | Method, equipment and storage medium for predicting change trend of silicon content in molten iron of blast furnace | |
| CN1224720C (en) | Blast furnace smelt controlling method with intelligent control system | |
| CN106909705B (en) | Blast furnace molten iron quality forecasting method and system | |
| CN102876838B (en) | Carbon content and system for detecting temperature in a kind of converter | |
| CN111949700A (en) | Intelligent safety guarantee real-time optimization method and system for petrochemical device | |
| CN101542464B (en) | Fuzzy logic control for process with large dead time | |
| CN111651931A (en) | Blast furnace fault diagnosis rule derivation method based on deep neural network | |
| CN102540879A (en) | Multi-target evaluation optimization method based on group decision making retrieval strategy | |
| CN109884892A (en) | Process industry system prediction model based on crosscorrelation time lag grey correlation analysis | |
| CN1207399C (en) | Intelligent blast furnace smelt controlling system | |
| CN104778361B (en) | The method of modified EMD Elman neural network prediction molten iron silicon contents | |
| CN111831719A (en) | Intelligent control method and system for blast furnace ironmaking production process | |
| Martín et al. | Hot metal temperature prediction in blast furnace using advanced model based on fuzzy logic tools | |
| DE102012224184A1 (en) | Method for the prediction, control and / or regulation of steelworks processes | |
| CN105807741A (en) | Industrial production flow prediction method | |
| CN117312816B (en) | Special steel smelting effect evaluation method and system | |
| Wu et al. | Integrated soft sensing of coke-oven temperature | |
| CN110097929A (en) | A kind of blast furnace molten iron silicon content on-line prediction method | |
| US20050137995A1 (en) | Method for regulating a thermodynamic process by means of neural networks | |
| Takalo-Mattila et al. | Explainable Steel Quality Prediction System Based on Gradient Boosting Decision Trees | |
| CN113420500A (en) | Intelligent atmospheric and vacuum system | |
| CN116306272A (en) | Converter heat loss rate prediction method based on big data | |
| CN117612651A (en) | Method for predicting manganese content of converter endpoint |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160907 |