CN114854929B - Converter blowing CO 2 Method for dynamically predicting molten steel components and temperature in real time - Google Patents
Converter blowing CO 2 Method for dynamically predicting molten steel components and temperature in real time Download PDFInfo
- Publication number
- CN114854929B CN114854929B CN202210493643.9A CN202210493643A CN114854929B CN 114854929 B CN114854929 B CN 114854929B CN 202210493643 A CN202210493643 A CN 202210493643A CN 114854929 B CN114854929 B CN 114854929B
- Authority
- CN
- China
- Prior art keywords
- element content
- converter
- steel
- target
- temperature
- 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.)
- Active
Links
Images
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
- C21C5/35—Blowing from above and through the bath
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01D—MEASURING NOT SPECIALLY ADAPTED FOR A SPECIFIC VARIABLE; ARRANGEMENTS FOR MEASURING TWO OR MORE VARIABLES NOT COVERED IN A SINGLE OTHER SUBCLASS; TARIFF METERING APPARATUS; MEASURING OR TESTING NOT OTHERWISE PROVIDED FOR
- G01D21/00—Measuring or testing not otherwise provided for
- G01D21/02—Measuring two or more variables by means not covered by a single other subclass
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/20—Metals
- G01N33/205—Metals in liquid state, e.g. molten metals
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P10/00—Technologies related to metal processing
- Y02P10/25—Process efficiency
Abstract
Converter blowing CO 2 The real-time dynamic prediction method of molten steel components and temperature comprises establishing converter blowing CO 2 A molten steel component and temperature prediction model database; establishing converter blowing CO 2 The molten steel component and temperature real-time prediction model is used for carrying out classification according to the types, proportions and flow rates of bottom blowing gas and top blowing gas in different time periods, and real-time prediction is carried out on each class by using different submodels; according to the blowing of CO into the converter 2 The real-time dynamic prediction model of the components and the temperature of the molten steel predicts the components and the temperature of the molten steel in real time; and (4) updating the database in real time after eliminating abnormal heat data, and correcting the correction item under certain conditions. The invention blows CO to the converter 2 The real-time dynamic prediction of the molten steel components and the temperature in the smelting process solves the problem of O blowing of the original converter 2 The inapplicability of the model causes the prolongation of the smelting period, the inaccurate predication of the end point components, the overoxidation of the end point of the converter and the waste of raw materials, shortens the smelting time, reduces the production cost, and avoids the blowing of CO by the converter 2 The black box and experience operation of the method improve the hit rate of the end point.
Description
Technical Field
The invention belongs to the technical field of prediction of components and temperature of molten steel in converter steelmaking and provides a method for predicting the composition and the temperature of molten steel in converter steelmakingRelates to a converter blowing CO 2 A real-time dynamic prediction method for molten steel components and temperature.
Background
Converter steelmaking is one of the most important links in steel production, and the tasks of dephosphorization, decarburization and bath temperature rise in the converter steelmaking process are mainly completed by oxygen supply. Along with the acceleration of smelting rhythm and the continuous improvement of oxygen supply intensity, the problems of unstable dephosphorization, excessive smoke dust, serious molten steel peroxidation and the like are easily caused. Based on this, some researchers began to mix a certain proportion of CO by top and bottom blowing in converter steelmaking 2 The product is used as a diluting and weak oxidizing agent and successfully completes the steel-making task. In order to achieve better smelting effect, CO is required to be combined with different blowing time periods 2 Reaction characteristics of (2) and converter metallurgy task, and pertinently formulating CO 2 And (4) blowing process. The smelting process of the converter mainly completes the tasks of slagging, desiliconization, demanganization, dephosphorization, decarburization and the like according to the time sequence, and CO is used 2 Compared with oxygen, the oxygen mainly shows endothermic effect in converter smelting, in order to better regulate and control the temperature of a molten pool and complete the metallurgical task of corresponding time periods, the whole blowing time is generally divided into a plurality of time periods, and CO injection is regulated and controlled in different time periods 2 Flow rate and ratio. At present, part of the steel plant converters blow CO in China 2 And using segmented control of CO 2 The flow rate and the proportion of the mixture obtain good metallurgical effect.
Since the converter does not blow CO before 2 The original molten steel component and temperature prediction model are not considered and can not predict the blowing CO 2 The change of material balance caused by the input of carbon atoms and oxygen atoms and energy balance caused by the endothermic effect of the reaction is caused, and when the heat supplement agent is added, the waste of the heat supplement agent is often caused because of no reference; in addition, the original model does not consider blowing CO 2 The oxygen atoms introduced are CO 2 Replacing part of O 2 After that, the model considers that O is reduced 2 The flow rate and the predicted blowing time are increased, the end point of the converter is seriously oxidized, and the smelting time is prolonged; and blowing CO into the converter 2 The process parameters used by the process in different converting stages are different, so that the prediction life of the original prediction model is causedThe medium rate is obviously reduced, and CO is blown in the converter 2 The metallurgical effect obtained is not stable.
If CO can be injected into the converter 2 Real-time dynamic prediction of molten steel components and temperature, and local setting and adjustment of CO 2 The blowing parameters in the molten steel use different CO at different periods 2 The proportion and the flow rate can pertinently complete the smelting task of the converter, enhance the stirring, control the end point peroxidation, reduce the smoke generation amount, accurately predict the real-time components and the temperature of the molten steel, and can lead CO to be mixed with the molten steel 2 The effect of the converter is more excellent, and the production efficiency of a steel plant is improved.
Disclosure of Invention
Aiming at the problems, the invention provides a converter for blowing CO 2 The real-time dynamic prediction method of molten steel components and temperature considers the blowing of CO in a converter 2 CO resulting from differences in reaction characteristics and metallurgical tasks at different converting periods 2 Blowing process variations and converting CO from time periods to time periods 2 And classifying the blowing process, and establishing a corresponding classification prediction submodel.
The invention is realized by the following technical scheme:
converter blowing CO 2 The method for dynamically predicting the components and the temperature of the molten steel in real time comprises the following steps of: considering the blowing of CO into the converter 2 CO resulting from differences in reaction characteristics and metallurgical tasks at different converting periods 2 Variation of blowing Process to remove CO from time periods 2 Classifying the blowing process, and using a prediction model containing a balance calculation term and a correction term of the corresponding classification to realize blowing CO to the converter 2 And (4) real-time dynamic prediction of the components and the temperature of the process molten steel.
Blowing CO in a converter as described above 2 The real-time dynamic prediction method of the components and the temperature of the molten steel specifically comprises the following steps:
step 1: establishing converter blowing CO 2 A molten steel component and temperature prediction model database;
step 1.1: acquiring historical data of the whole process from molten iron charging to molten steel discharging in converter smelting and real-time data eliminated by applying the model;
step 1.1.1: obtaining the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of the end-point tapping of each steel type converter;
step 1.1.2: obtaining molten iron components, molten iron temperature, molten iron weight, scrap steel components, scrap steel temperature, scrap steel adding weight, steel tapping components, steel tapping temperature and steel tapping weight in the smelting process of each steel type;
step 1.1.3: acquiring the components of various auxiliary materials, the temperature of the auxiliary materials and the adding weight and adding time of the auxiliary materials in the smelting process of various steel grades;
step 1.1.4: acquiring the type of bottom blowing gas, the proportion of the bottom blowing gas, the total flow of the bottom blowing gas, the temperature of the bottom blowing gas and the total amount of the bottom blowing gas at each moment;
step 1.1.5: acquiring top-blown gas types, top-blown gas proportion, total top-blown gas flow, top-blown gas temperature and total top-blown gas amount at each moment;
step 1.1.6: acquiring the height of the liquid level in the furnace and the height of the oxygen lance at each moment;
step 1.1.7: acquiring furnace gas flow, furnace gas temperature and component proportions in the furnace gas at each moment;
step 1.1.8: and obtaining the blowing time and smelting time of each heat.
Step 1.2: classifying the acquired data of each heat according to the combination of the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content;
step 1.2.1: the standard target carbon element content, silicon element content, manganese element content, phosphorus element content and sulfur element content of the tapping at the end point of the converter are respectively recorded as mC0, mSi0, mMn, mP0 and mS0;
step 1.2.2: if (1 + xa) mC0 is less than or equal to the target carbon element content < (1 + (x + 1) a) mC0, the target carbon element content of the steel is marked as mC (x + 1); if (1- (x + 1) a) mC 0. Ltoreq. The target carbon content < (1-xa) mC0, the target carbon content of the steel is denoted mC- (x + 1).
Step 1.2.3: if (1 + × b) mSi0 is less than or equal to target silicon element content < (1 + (x + 1) b) mSi0, the target silicon element content of the steel is recorded as mSi (x + 1); if (1- (x + 1) b) mSi0 is less than the target silicon element content < (1-xb) mSi0, the target silicon element content of the steel is denoted as mSi- (x + 1).
Step 1.2.4: if (1 + xc) mMn ≦ target manganese content < (1 + (x + 1) c) mMn, the target manganese content of the steel type is noted as mMn (x + 1); if (1- (x + 1) c) mMn ≦ target manganese content < (1-xc) mMn, the target manganese content of this steel grade is noted as mMn- (x + 1).
Step 1.2.5: if (1 + xd) mP0 is less than or equal to the target phosphorus element content < (1 + (x + 1) d) mP0, the target phosphorus element content of the steel is marked as mP (x + 1); if (1- (x + 1) d) mP0 is less than or equal to the target phosphorus element content < (1-xd) mP0, the target phosphorus element content of the steel is marked mP- (x + 1).
Step 1.2.6: if (1 + xe) mS0 is less than or equal to the target sulfur element content < (1 + (x + 1) e) mS0, the target sulfur element content of the steel is recorded as mS (x + 1); if (1- (x + 1) e) mS0 is less than or equal to the target sulfur element content < (1-xe) mS0, the target sulfur element content of the steel is recorded as mS- (x + 1);
x is a non-negative integer; a. b, c, d and e are positive numbers.
Step 1.2.7: recording the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of each steel type according to the steps 1.2.1-1.2.6, and classifying the steel types with the same marks of the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content, the target sulfur element content and the like.
And 2, step: establishing converter blowing CO 2 A real-time prediction model of molten steel components and temperature;
step 2.1: establishing converter blowing CO 2 A molten steel component real-time prediction model;
step 2.1.1: dividing the converter blowing period into N time periods according to time, recording the time periods as N1-Nn, wherein each time period can be unequal; n is a positive integer.
Step 2.1.2: classifying each time period according to the type of bottom blowing gas, the range of bottom blowing gas proportion, the range of bottom blowing gas flow, the type of top blowing gas, the range of top blowing gas proportion and the range of top blowing gas flow, recording the categories as Ci respectively, and calling the data in the step 1.2 for matching;
i is a positive integer.
Step 2.1.3: according to material balance, the converter predicts real-time carbon element content rtC = converter income carbon element content item-converter expenditure carbon element content item + correction carbon element content item, and is matched with each category Ci to establish a sub-model.
Step 2.1.4: according to the material balance, the converter predicts the real-time silicon element content rtSi = converter income silicon element content item-converter expenditure silicon element content item + correction silicon element content item, and matches with each class Ci to establish a sub-model.
Step 2.1.5: according to material balance, the converter predicts real-time manganese content rtMn = converter incoming manganese content item-converter outgoing manganese content item + corrected manganese content item, and matches with each category Ci to establish a sub-model.
Step 2.1.6: according to the material balance, the converter predicts the real-time phosphorus content rtP = the converter income phosphorus content item-the converter expenditure phosphorus content item + the corrected phosphorus content item, and matches with each class Ci to establish a sub-model.
Step 2.1.7: according to material balance, the converter predicts real-time sulfur element content rtS = converter income sulfur element content item-converter expenditure sulfur element content item + correction sulfur element content item, and matches with each category Ci to establish a sub-model;
the corrected carbon element content item, the corrected silicon element content item, the corrected manganese element content item, the corrected phosphorus element content item and the corrected sulfur element content item are all correction items in material balance calculation of molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, the height of the liquid level in the furnace, the height of an oxygen lance and other items of all the types Ci after classification in the step 2.1.2, and the initial correction item after the model is started is obtained by automatically fitting the embedded initial correction item after the model is assigned with the to-be-assigned item and according to the established database.
Step 2.2: establishing converter blowing CO 2 A molten steel temperature real-time prediction model;
step 2.2.1: according to energy balance, the converter predicts real-time molten steel temperature rtT = converter income energy item-converter expenditure energy item + correction energy item, and matches with each category Ci to establish a sub-model;
the corrected energy items are correction items in energy balance calculation of molten iron, steel scraps, auxiliary materials, bottom blowing gas, top blowing gas, furnace liquid level height, oxygen lance height and other items of each category Ci after classification in the step 2.1.2, and the initial correction items after the model is started are obtained by performing automatic fitting according to the established database after the value of the to-be-assigned items of the embedded initial correction items are assigned to the model.
And step 3: according to the blowing of CO into the converter 2 Blowing CO into converter by using molten steel component and temperature real-time prediction model 2 Predicting the components and the temperature of the molten steel in real time, and comparing the components and the temperature with an actual detection data result;
step 3.1: acquiring the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of the steel type converter end-point tapping, and performing automatic matching of a database; .
Step 3.2: collecting molten iron components, molten iron temperature, molten iron weight, steel scrap components, steel scrap temperature, steel scrap adding weight, auxiliary material components, auxiliary material temperature and auxiliary material adding weight which are added at each moment of the current smelting furnace, and data information such as bottom blowing gas type, bottom blowing gas proportion, total bottom blowing gas flow, bottom blowing gas temperature, top blowing gas type, top blowing gas proportion, total top blowing gas flow, top blowing gas temperature, furnace liquid level height, oxygen lance height, furnace gas flow, furnace gas temperature and each component proportion in the furnace gas, and the like as input items of the N1 time period.
Step 3.3: automatically selecting a corrected carbon element content item, a corrected silicon element content item, a corrected manganese element content item, a corrected phosphorus element content item, a corrected sulfur element content item and a corrected energy item corresponding to the stage in the database selected in the step 3.1 according to the classification of the step 2.1.2, and carrying out CO blowing of the converter 2 And (4) predicting the components and the temperature of the molten steel in real time.
Step 3.4: and using data information of the predicted real-time carbon element content, the predicted real-time silicon element content, the predicted real-time manganese element content, the predicted real-time phosphorus element content, the predicted real-time sulfur element content and the predicted real-time molten steel temperature at the end of the N1 time period, as well as molten iron components, molten iron temperatures, molten iron weights, scrap steel components, scrap steel temperatures, scrap steel adding weights, auxiliary material components, auxiliary material temperatures and auxiliary material adding weights which are added at various moments, and data information of bottom blowing gas types, bottom blowing gas proportions, bottom blowing gas total flow, bottom blowing gas temperatures, top blowing gas types, top blowing gas proportions, top blowing gas total flow, top blowing gas temperatures, furnace liquid level heights, oxygen lance heights, furnace gas flow, furnace gas temperatures and component proportions in furnace gas and the like as input items of the N2 time period.
Step 3.5: automatically selecting a corrected carbon element content item, a corrected silicon element content item, a corrected manganese element content item, a corrected phosphorus element content item, a corrected sulfur element content item and a corrected energy item corresponding to the stage in the database selected in the step 3.1 according to the classification of the step 2.1.2, and performing converter CO injection in the N2 time period 2 And (4) predicting the components and the temperature of the molten steel in real time.
Step 3.6: converting Ny in the steps 3.4-3.5 into N (y + 1), and repeating the steps until (y + 1) = N;
wherein y is not less than 1 and not more than n-1;
step 3.7: and comparing rtC, rtSi, rtMn, rtP, rtS and rtT obtained by the prediction model through real-time prediction with rC, rSi, rMn, rP, rS and rT obtained by actual detection data, and recording.
And 4, step 4: after the abnormal heat data are removed, updating the database in real time, and correcting the correction term under a certain condition;
step 4.1: rejecting abnormal heat with [ (1+M%) × average converting time less than or equal to converting time ] < U [ (1+M%) × average smelting time less than or equal to smelting time ], and importing other heat data into a database in real time;
step 4.2: selecting the data of the nearest L furnaces in each database every Z furnaces to correct the correction items, wherein L meets the condition that the data group matched with each category Ci in the step 2.1.2 is not less than L;
step 4.3: when the absolute value of the difference value between the predicted value and the actual detection value of the continuous F furnace is larger than the maximum allowable error value of each molten steel component and temperature, namely | rtC-rC ≧ MaxC or | rtSi-rSi | ≧ MaxSi or | rtMn-rMn | ≧ MaxMn or | rtP-rP | ≧ MaxP or | rtS-rS | ≧ MaxP or | rtT-rT | ≧ MaxT, correction of the correction term is immediately carried out, and L and L are changed.
Further, the molten iron components in the step 1.1.2 comprise molten iron carbon content, molten iron silicon content, molten iron manganese content, molten iron phosphorus content and molten iron sulfur content;
the scrap steel comprises scrap steel carbon element content, scrap steel silicon element content, scrap steel manganese element content, scrap steel phosphorus element content and scrap steel sulfur element content;
the tapping components comprise tapping carbon element content, tapping silicon element content, tapping manganese element content, tapping phosphorus element content and tapping sulfur element content.
Further, the auxiliary material components in the step 1.1.3 comprise auxiliary material carbon element content, auxiliary material silicon element content, auxiliary material manganese element content, auxiliary material phosphorus element content and auxiliary material sulfur element content;
the auxiliary materials comprise sintered ore, active lime, quicklime, light-burned dolomite, sintered dolomite, ferrosilicon, fluorite, coke, ferromanganese and a heat-compensating agent.
Further, step 1.1.4 bottom blowing gas comprises N 2 Ar and CO 2 (ii) a Step 1.1.5 the top-blown gas comprises O 2 And CO 2 。
6. The process of claim 2, wherein the blowing of CO in the converter is carried out in a continuous mode 2 The method for dynamically predicting the components and the temperature of the molten steel in real time is characterized in that in the step 1.1.7, the furnace gas components comprise CO and CO 2 、SO 2 、O 2 And N 2 。
Further, the actual detection data in the steps 3 and 3.7 and the actual detection value in the step 4.3 refer to molten steel components and temperature data obtained by sublance at the later stage of blowing after the abnormal heat is removed, molten steel end point sample detection after blowing and sampling detection in the smelting process.
The prediction model established by the method comprises the balance calculation term and the correction term, and compared with a method for fitting only by using historical data, the method reduces the phenomenon that the fitting error is overlargeThe resulting likelihood of an excessively low end-point hit rate. CO is considered in the balance calculation term 2 The reaction characteristic of the converter is used for causing the change of the balance of materials and energy, and a correction term capable of automatically correcting is used for integrating various information to correct the calculation, so that the CO is injected into the converter 2 The real-time dynamic prediction of the components and the temperature of the process molten steel solves the problem of O blowing of the original converter 2 The problems of prolonged smelting time, inaccurate terminal component prediction, overoxidation of the converter terminal and waste of raw materials caused by the inapplicability of the model.
The method can be used for blowing CO into the converter 2 The real-time dynamic prediction of the components and the temperature of the molten steel shortens the smelting time, reduces the production cost, provides favorable reference information for the operation adjustment of operators according to the actual conditions, and avoids the CO injection of the converter 2 The black box and experience operation of the method improve the hit rate of the end point.
Drawings
FIG. 1 shows the CO injection in a converter according to the invention 2 A flow chart of a concrete implementation mode of the real-time dynamic prediction method of the components and the temperature of the molten steel;
FIG. 2 shows CO injection in a converter according to the invention 2 A real-time dynamic prediction flow chart of a specific implementation mode of the method for real-time dynamic prediction of the components and the temperature of the molten steel;
FIG. 3 is a graph of hit rate results for carbon content predictions using the prediction method of the present invention;
FIG. 4 is a graph of the hit rate prediction result of manganese content predicted by the prediction method of the present invention;
FIG. 5 is a graph showing the hit rate of phosphorus content prediction predicted by the prediction method of the present invention;
FIG. 6 is a graph of the hit rate results of sulfur content predictions using the prediction method of the present invention;
FIG. 7 is a graph of hit rate results for temperature predictions using the prediction method of the present invention.
Detailed Description
In order to more clearly illustrate the embodiments of the present invention, the present invention will be further described with reference to a 300t converter example in a certain plant. It should be understood that the specific embodiments described herein are merely illustrative of the invention and do not delimit the invention.
The invention aims to inject CO into the converter 2 The real-time dynamic prediction of the components and the temperature of the molten steel is carried out, and the main prediction process comprises the following steps: a database is established by collecting a large amount of historical data information, the difference values between the carbon element content, the silicon element content, the manganese element content, the phosphorus element content and the sulfur element content required by each steel type endpoint and the reference content are grouped, after a model is started and the value of a model to be assigned is given, the model is classified according to the top and bottom blowing gas types, the proportion range and the flow range of each time period, and then an embedded initial correction term is calculated and fitted according to the established database group and classification to obtain the initial correction term. When smelting starts, the model automatically obtains required data, the required data is classified according to the content requirement and parameters of smelting steel species target elements, correction items in corresponding database groups are called, dynamic prediction is carried out, when a certain stage is finished, information such as molten steel components and energy at the end of the previous stage is used as calculation income and input items of the next stage of the model, the corresponding correction items are called according to the parameter classification of the next stage, prediction is continuously carried out until blowing is finished, the predicted data and the detected data are compared and recorded, and after abnormal heat data are eliminated, other effective heat data are led into the database after smelting is finished. After a certain condition is met, the model automatically corrects the correction term.
As shown in FIG. 1, the present invention relates to a converter blowing CO 2 The real-time dynamic prediction method of the components and the temperature of the molten steel comprises the following steps:
step 1: establishing converter blowing CO 2 A molten steel component and temperature prediction model database;
step 1.1: acquiring historical data of the whole process from molten iron charging to molten steel discharging in converter smelting and real-time data eliminated by applying the model;
step 1.1.1: obtaining the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of the end-point tapping of each steel type converter;
step 1.1.2: obtaining molten iron components, molten iron temperature, molten iron weight, scrap steel components, scrap steel temperature, scrap steel adding weight, steel tapping components, steel tapping temperature and steel tapping weight in the smelting process of each steel type;
the molten iron comprises molten iron carbon content, molten iron silicon content, molten iron manganese content, molten iron phosphorus content and molten iron sulfur content;
the steel scrap components comprise steel scrap carbon element content, steel scrap silicon element content, steel scrap manganese element content, steel scrap phosphorus element content and steel scrap sulfur element content;
the tapping components comprise tapping carbon element content, tapping silicon element content, tapping manganese element content, tapping phosphorus element content and tapping sulfur element content;
step 1.1.3: acquiring the components of various auxiliary materials, the temperature of the auxiliary materials and the adding weight and adding time of the auxiliary materials in the smelting process of various steel grades;
the auxiliary material components comprise auxiliary material carbon element content, auxiliary material silicon element content, auxiliary material manganese element content, auxiliary material phosphorus element content and auxiliary material sulfur element content;
the auxiliary materials comprise sintered ore, active lime, quicklime, light-burned dolomite, sintered dolomite, ferrosilicon, fluorite, coke, ferromanganese and a heat-supplementing agent;
step 1.1.4: acquiring the type of bottom blowing gas, the proportion of the bottom blowing gas, the total flow of the bottom blowing gas, the temperature of the bottom blowing gas and the total amount of the bottom blowing gas at each moment;
the bottom-blown gas comprises N 2 Ar and CO 2 ；
Step 1.1.5: acquiring top-blown gas types, top-blown gas proportion, total top-blown gas flow, top-blown gas temperature and total top-blown gas amount at each moment;
the top-blown gas comprises O 2 And CO 2 ；
Step 1.1.6: acquiring the height of the liquid level in the furnace and the height of the oxygen lance at each moment;
step 1.1.7: obtaining the furnace gas flow, the furnace gas temperature and the proportion of each component in the furnace gas at each moment;
furnace gas component packageComprises CO and CO 2 、SO 2 、O 2 And N 2 ；
Step 1.1.8: obtaining the blowing time and smelting time of each heat;
step 1.2: classifying the obtained data of each heat according to the combination of the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content;
step 1.2.1: the standard target carbon element content, silicon element content, manganese element content, phosphorus element content and sulfur element content of the steel tapped at the end point of the converter are respectively recorded as mC0, mSi0, mMn, mP0 and mS0, and in the embodiment, mC0, mSi0, mMn, mP0 and mS0 are respectively assigned with values of 0.04%, 0.005%, 0.045%, 0.006% and 0.006%;
step 1.2.2: if (1 + xa) mC0 is less than or equal to the target carbon element content < (1 + (x + 1) a) mC0, the target carbon element content of the steel is marked as mC (x + 1); if (1- (x + 1) a) mC 0. Ltoreq. The target carbon content < (1-xa) mC0, the target carbon content of the steel is denoted mC- (x + 1), in this example a is assigned 1;
step 1.2.3: if (1 + × b) mSi0 is less than or equal to target silicon element content < (1 + (x + 1) b) mSi0, the target silicon element content of the steel is recorded as mSi (x + 1); if (1- (x + 1) b) mSi 0. Ltoreq. Target elemental silicon content < (1-xb) mSi0, the target elemental silicon content of the steel is denoted mSi- (x + 1), in this example b is assigned 1;
step 1.2.4: if (1 + xc) mMn ≦ target manganese content < (1 + (x + 1) c) mMn, the target manganese content of the steel type is noted as mMn (x + 1); if (1- (x + 1) c) mMn ≦ target manganese content < (1-xc) mMn, the target manganese content of the steel grade is noted as mMn- (x + 1), in this example c is assigned 1;
step 1.2.5: if (1 + xd) mP0 is less than or equal to the target phosphorus element content < (1 + (x + 1) d) mP0, the target phosphorus element content of the steel is marked as mP (x + 1); if (1- (x + 1) d) mP0 is less than or equal to the target phosphorus element content < (1-xd) mP0, the target phosphorus element content of the steel is marked mP- (x + 1), and d is assigned 0.5 in the embodiment;
step 1.2.6: if (1 + xe) mS0 is less than or equal to the target sulfur element content < (1 + (x + 1) e) mS0, the target sulfur element content of the steel is recorded as mS (x + 1); if (1- (x + 1) e) mS0 is less than or equal to the target elemental sulfur content < (1-xe) mS0, the target elemental sulfur content of the steel is recorded as mS- (x + 1), and e is assigned 0.5 in the present example;
x is a non-negative integer; a. b, c, d and e are positive numbers;
step 1.2.7: recording the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of each steel type according to the steps 1.2.1 to 1.2.6, and classifying the steel types with the same marks of the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content into a class, such as a group with more data amount [ mC1, mSi1, mMn1, mP1, mS1] and [ mC1, mSi1, mMn, mP2, mS1] in the embodiment, wherein each group contains a set of a large amount of data;
step 2: establishing converter blowing CO 2 A real-time prediction model of molten steel components and temperature;
step 2.1: establishing converter blowing CO 2 A molten steel component real-time prediction model;
step 2.1.1: dividing the converter blowing period into N time periods according to time, recording the time periods as N1-Nn, wherein each time period can be unequal, and N is a positive integer; in this embodiment, it is mainly considered that different CO is injected in different blowing time 2 And taking N as 4 according to the proportion and the flow, wherein N1: blowing for 0-3 min, and N2: blowing for 3-6 min, N3: converting for 9 min-TSC, N4: blowing TSC till the end of blowing;
step 2.1.2: classifying each time period according to the type of bottom-blown gas, the proportion range of bottom-blown gas, the flow range of bottom-blown gas, the type of top-blown gas, the proportion range of top-blown gas and the flow range of top-blown gas, respectively recording the types as Ci, and calling the data in the step 1.2 for matching, wherein i is a positive integer; in this example, the bottom-blown gas was CO in the N1-N3 periods 2 And the flow rate is kept unchanged, the bottom blowing gas is Ar in the N4 time period, and the flow rate is greatly increased; the total flow rate in the top-blown gas is kept constant in the N1-N3 time period, and CO is added 2 The proportion is changed from 3.1 percent to 7 percent and 4.7 percent in sequence, the total flow of top-blown gas is greatly increased in the N4 period, and the top-blown CO is 2 The proportion is 7.5%; the flow rate of the top and bottom blowing gas in each time period is compared withCombining the proportions, and respectively marking the corresponding categories in the time periods from N1 to N4 as C1, C2, C3 and C4;
step 2.1.3: according to material balance, the converter predicts the real-time carbon element content rtC = the converter income carbon element content item-the converter expenditure carbon element content item + the corrected carbon element content item, and matches with each category Ci to establish a sub-model; in this example, the functional relationship between the term of the carbon content and the blowing time t is modified
f C (t j ) A time-varying correction term sub-function is obtained by data fitting in the calculation of material balance for molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, the height of the liquid level in the furnace, the height of an oxygen lance and other data;
step 2.1.4: according to material balance, the converter predicts real-time silicon element content rtSi = converter income silicon element content item-converter expenditure silicon element content item + correction silicon element content item, and matches with each category Ci to establish a sub-model; in this example, the functional relation between the silicon element content term and the blowing time t is corrected
f Si (t j ) The method comprises the following steps of obtaining a time-varying correction term subfunction by data fitting in material balance calculation for molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, liquid level height in a furnace, oxygen lance height and other data;
step 2.1.5: according to material balance, the converter predicts real-time manganese content rtMn = converter incoming manganese content item-converter outgoing manganese content item + corrected manganese content item, and matches with each category Ci to establish a sub-model; in this example, the functional relation between the manganese content term and the blowing time t is corrected
f Mn (t j ) Is molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, the height of the liquid level in the furnace and the height of an oxygen lanceA time-varying correction term subfunction obtained by data fitting in material balance calculation with other data items;
step 2.1.6: according to material balance, the converter predicts real-time phosphorus content rtP = converter income phosphorus content item-converter expenditure phosphorus content item + correction phosphorus content item, and matches with each class Ci to establish a sub-model; in this example, the functional relation between the term of the phosphorus content and the blowing time t is modified
f P (t j ) The method comprises the following steps of obtaining a time-varying correction term subfunction by data fitting in material balance calculation for molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, liquid level height in a furnace, oxygen lance height and other data;
step 2.1.7: according to material balance, the converter predicts real-time sulfur element content rtS = converter income sulfur element content item-converter expenditure sulfur element content item + correction sulfur element content item, and matches with each category Ci to establish a sub-model; in this example, the functional relationship between the sulfur content term and the blowing time t is modified
f S (t j ) The method comprises the following steps of obtaining a time-varying correction term subfunction by data fitting in material balance calculation for molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, liquid level height in a furnace, oxygen lance height and other data;
step 2.2: establishing converter blowing CO 2 A molten steel temperature real-time prediction model;
step 2.2.1: according to energy balance, the converter predicts real-time molten steel temperature rtT = converter income energy item-converter expenditure energy item + correction energy item, and matches with each category Ci to establish a sub-model; in the present example, the functional relationship between the energy term and the blowing time t is modified
f T (t j ) The method comprises the following steps of obtaining a time-varying correction term subfunction by data fitting in energy balance calculation for molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, the height of the liquid level in a furnace, the height of an oxygen lance and other data;
and step 3: according to the blowing of CO into the converter 2 Molten steel composition and temperature real-time prediction model for blowing CO into converter 2 The components of the molten steel and the temperature are predicted in real time and compared with the actual detection data result, as shown in figure 2;
step 3.1: acquiring the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of the steel type converter end-point tapping, and performing automatic matching of a database;
step 3.2: collecting molten iron components, molten iron temperature, molten iron weight, steel scrap components, steel scrap temperature, steel scrap adding weight, auxiliary material components, auxiliary material temperature and auxiliary material adding weight which are added at each moment of the current smelting furnace, and data information of bottom blowing gas type, bottom blowing gas proportion, total bottom blowing gas flow, bottom blowing gas temperature, top blowing gas type, top blowing gas proportion, total top blowing gas flow, top blowing gas temperature, furnace liquid level height, oxygen lance height, furnace gas flow, furnace gas temperature and proportion of each component in the furnace gas as input items of N1 time period;
step 3.3: automatically selecting a corrected carbon element content item, a corrected silicon element content item, a corrected manganese element content item, a corrected phosphorus element content item, a corrected sulfur element content item and a corrected energy item corresponding to the stage in the database selected in the step 3.1 according to the classification of the step 2.1.2, and carrying out CO blowing of the converter 2 Real-time prediction of molten steel components and temperature;
step 3.4: predicting real-time carbon element content, predicting real-time silicon element content, predicting real-time manganese element content, predicting real-time phosphorus element content, predicting real-time sulfur element content and predicting real-time molten steel temperature at the end of the N1 time period, and using the molten iron component, the molten iron temperature, the molten iron weight, the scrap steel component, the scrap steel temperature, the scrap steel adding weight, the auxiliary material component, the auxiliary material temperature and the auxiliary material adding weight which are added at each moment, and the type of bottom blowing gas, the bottom blowing gas proportion, the total bottom blowing gas flow, the bottom blowing gas temperature, the top blowing gas type, the top blowing gas proportion, the total top blowing gas flow, the top blowing gas temperature, the liquid level in the furnace, the height of an oxygen lance, the furnace gas flow, the furnace gas temperature and the proportion data information of each component in the furnace gas as input items of the N2 time period;
step 3.5: automatically selecting a corrected carbon element content item, a corrected silicon element content item, a corrected manganese element content item, a corrected phosphorus element content item, a corrected sulfur element content item and a corrected energy item corresponding to the stage in the database selected in the step 3.1 according to the classification of the step 2.1.2, and performing converter CO injection in the N2 time period 2 Real-time prediction of molten steel components and temperature;
step 3.6: converting Ny in the steps 3.4-3.5 into N (y + 1), and then repeating the steps until the N4 time period is finished;
step 3.7: comparing rtC, rtSi, rtMn, rtP, rtS and rtT obtained by real-time prediction of the prediction model with rC, rSi, rMn, rP, rS and rT obtained by actual detection data, and recording;
and 4, step 4: after the abnormal heat data are removed, updating the database in real time, and correcting the correction term under a certain condition;
step 4.1: eliminating abnormal heats of [ (1+M%) × average converting time less than or equal to converting time ] < U [ (1+M%) × average smelting time less than or equal to smelting time ], and importing other heat data into a database in real time; in this example, M is assigned a value of 20;
step 4.2: selecting the data of the nearest L furnaces in each database every Z furnaces to correct the correction items, wherein L meets the condition that the data group matched with each category Ci in the step 2.1.2 is not less than L; in this embodiment, Z and l are assigned values of 10 and 100, respectively;
step 4.3: when the absolute value of the difference value between the predicted value and the actual detection value of the continuous F furnace is larger than the maximum allowable error value of each molten steel component and temperature, namely | rtC-rC | ≧ MaxC or | rtSi-rSi | ≧ MaxSi or | rtMn-rMn | ≧ MaxMn or | rtP-rP | ≧ MaxP or | rtS-rS | ≧ MaxP or | rtT-rT | ≧ MaxT, immediately correcting the correction term, and changing L and L; in the present example, F is assigned with values of 5, and MaxC, maxSi, maxMn, maxP, maxS and MaxT are assigned with values of 0.01%, 0.003%, 0.01%, 0.002% and 7 ℃.
The actual detection data and the actual detection values refer to molten steel components and temperature data obtained by sublance at the later stage of blowing after the abnormal heat is eliminated, the detection of a molten steel terminal sample after the blowing is finished and the sampling detection in the smelting process.
Converter CO based on certain factory blowing 2 Actual production data, the process of the embodiment is repeated for many times, and finally, the hit rate of the predicted single index carbon element content of the molten steel at the end point of the converter (plus or minus 0.01%) is 94.32%, the hit rate of the predicted manganese element content (plus or minus 0.01%) is 91.35%, the hit rate of the predicted phosphorus element content (plus or minus 0.002%) is 92.93%, the hit rate of the predicted sulfur element content (plus or minus 0.002%) is 90.57%, and the hit rate of the predicted molten steel temperature at the TSO (plus or minus 7 ℃) is 90.14%, which is shown in FIGS. 3, 4, 5, 6 and 7, because the plant does not detect the silicon element content, the hit rate of the predicted silicon element content cannot be obtained; the double hit rate of the predicted carbon content and the predicted molten steel temperature is 83.21%. Proves that the method can be used for blowing CO into the converter 2 The real-time dynamic prediction of the components and the temperature of the molten steel shortens the smelting time, reduces the production cost, provides favorable reference information for the operation adjustment of operators according to the actual conditions, and avoids the CO injection of the converter 2 The black box and experience operation of the method improve the hit rate of the end point.
Claims (6)
1. Converter blowing CO 2 The method for dynamically predicting the components and the temperature of the molten steel in real time is characterized in that the CO blowing of a converter is considered 2 CO resulting from differences in reaction characteristics and metallurgical tasks at different converting periods 2 Variation of blowing Process to remove CO from time periods 2 Classifying the blowing process, and using a prediction model containing a balance calculation term and a correction term of the corresponding classification to realize blowing CO to the converter 2 Real-time dynamic prediction of process steel liquid components and temperature;
the method comprises the following steps:
step 1: establishing converter to blow CO 2 A molten steel component and temperature prediction model database;
step 1.1: acquiring historical data of the whole process from molten iron charging to molten steel discharging in converter smelting and real-time data eliminated by applying the model;
step 1.1.1: obtaining the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of the end-point tapping of each steel type converter;
step 1.1.2: obtaining molten iron components, molten iron temperature, molten iron weight, scrap steel components, scrap steel temperature, scrap steel adding weight, steel tapping components, steel tapping temperature and steel tapping weight in the smelting process of each steel type;
step 1.1.3: acquiring the components of various auxiliary materials, the temperature of the auxiliary materials and the adding weight and adding time of the auxiliary materials in the smelting process of various steel grades;
step 1.1.4: acquiring the type of bottom blowing gas, the proportion of the bottom blowing gas, the total flow of the bottom blowing gas, the temperature of the bottom blowing gas and the total amount of the bottom blowing gas at each moment;
step 1.1.5: acquiring top-blown gas types, top-blown gas proportions, total top-blown gas flow, top-blown gas temperature and total top-blown gas amount at each moment;
step 1.1.6: acquiring the height of the liquid level in the furnace and the height of the oxygen lance at each moment;
step 1.1.7: obtaining the furnace gas flow, the furnace gas temperature and the proportion of each component in the furnace gas at each moment;
step 1.1.8: obtaining the blowing time and smelting time of each heat;
step 1.2: classifying the obtained data of each heat according to the combination of the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content;
step 1.2.1: the standard target carbon element content, silicon element content, manganese element content, phosphorus element content and sulfur element content of the tapping at the end point of the converter are respectively recorded as mC0, mSi0, mMn, mP0 and mS0;
step 1.2.2: if (1 + xa) mC0 is less than or equal to the target carbon element content < (1 + (x + 1) a) mC0, the target carbon element content of the steel is recorded as mC (x + 1); if (1- (x + 1) a) mC 0. Ltoreq. The target carbon content < (1-xa) mC0, the target carbon content of the steel is denoted mC- (x + 1);
step 1.2.3: if (1 + xb) mSi0 is less than or equal to the target silicon element content < (1 + (x + 1) b) mSi0, the target silicon element content of the steel is marked as mSi (x + 1); if (1- (x + 1) b) mSi0 is less than or equal to the target silicon element content < (1-xb) mSi0, the target silicon element content of the steel is recorded as mSi- (x + 1);
step 1.2.4: if (1 + xc) mMn ≦ target manganese content < (1 + (x + 1) c) mMn, the target manganese content of the steel type is noted as mMn (x + 1); if (1- (x + 1) c) mMn ≦ target manganese content < (1-xc) mMn, the target manganese content of the steel grade is noted as mMn- (x + 1);
step 1.2.5: if (1 + xd) mP0 is less than or equal to the target phosphorus element content < (1 + (x + 1) d) mP0, the target phosphorus element content of the steel is marked as mP (x + 1); if the target phosphorus element content of (1- (x + 1) d) mP0 is less than (1-xd) mP0, the target phosphorus element content of the steel is marked as mP- (x + 1);
step 1.2.6: if (1 + xe) mS0 is less than or equal to the target sulfur element content < (1 + (x + 1) e) mS0, the target sulfur element content of the steel is recorded as mS (x + 1); if (1- (x + 1) e) mS0 is less than or equal to the target sulfur element content < (1-xe) mS0, the target sulfur element content of the steel is recorded as mS- (x + 1);
x is a non-negative integer; a. b, c, d and e are positive numbers;
step 1.2.7: recording the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of each steel type according to the steps 1.2.1-1.2.6, and classifying the steel types with the same marks of the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content;
step 2: establishing converter blowing CO 2 A real-time prediction model of molten steel components and temperature;
step 2.1: establishing converter blowing CO 2 A molten steel component real-time prediction model;
step 2.1.1: dividing the converter blowing period into N time periods according to time, recording the time periods as N1-Nn, wherein each time period can be unequal; n is a positive integer;
step 2.1.2: classifying each time period according to the type of bottom blowing gas, the range of bottom blowing gas proportion, the range of bottom blowing gas flow, the type of top blowing gas, the range of top blowing gas proportion and the range of top blowing gas flow, recording the categories as Ci respectively, and calling the data in the step 1.2 for matching;
i is a positive integer;
step 2.1.3: according to material balance, the converter predicts the real-time carbon element content rtC = the converter income carbon element content item-the converter expenditure carbon element content item + the corrected carbon element content item, and matches with each category Ci to establish a sub-model;
step 2.1.4: according to the material balance, the converter predicts the real-time silicon element content rtSi = the silicon element content item of the converter income-the silicon element content item of the converter expenditure + the silicon element content item of the correction, and is matched with each category Ci to establish a sub-model;
step 2.1.5: according to material balance, the converter predicts real-time manganese content rtMn = converter incoming manganese content item-converter outgoing manganese content item + corrected manganese content item, and matches with each category Ci to establish a sub-model;
step 2.1.6: according to material balance, the converter predicts real-time phosphorus content rtP = converter income phosphorus content item-converter expenditure phosphorus content item + correction phosphorus content item, and matches with each class Ci to establish a sub-model;
step 2.1.7: according to the material balance, the converter predicted real-time sulfur element content rtS = converter income sulfur element content item-converter expenditure sulfur element content item + correction sulfur element content item, and is matched with each category Ci to establish a sub-model;
the corrected carbon element content item, the corrected silicon element content item, the corrected manganese element content item, the corrected phosphorus element content item and the corrected sulfur element content item are all correction items in material balance calculation of molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, the height of the liquid level in the furnace, the height of an oxygen lance and other items of all the types Ci after classification in the step 2.1.2, and the initial correction item after the model is started is obtained by automatically fitting the embedded initial correction item after the model is assigned with the to-be-assigned item and according to the established database;
step 2.2: establishing converter blowing CO 2 A molten steel temperature real-time prediction model;
step 2.2.1: according to energy balance, the converter predicts real-time molten steel temperature rtT = converter income energy item-converter expenditure energy item + correction energy item, and matches with each category Ci to establish a sub-model;
the corrected energy items are correction items in energy balance calculation of the various types Ci of molten iron, scrap steel, auxiliary materials, bottom blowing gas, top blowing gas, liquid level height in the furnace, oxygen lance height and other items after classification in the step 2.1.2, and the initial correction items after the model is started are obtained by automatically fitting the embedded initial correction items according to the established database after the evaluation of the model to-be-evaluated items is carried out;
and 3, step 3: according to the blowing of CO into the converter 2 Molten steel composition and temperature real-time prediction model for blowing CO into converter 2 Predicting the components and the temperature of the molten steel in real time, and comparing the components and the temperature with an actual detection data result;
step 3.1: acquiring the target carbon element content, the target silicon element content, the target manganese element content, the target phosphorus element content and the target sulfur element content of the steel type converter end-point tapping, and performing automatic matching of a database;
step 3.2: collecting molten iron components, molten iron temperature, molten iron weight, steel scrap components, steel scrap temperature, steel scrap adding weight, auxiliary material components, auxiliary material temperature and auxiliary material adding weight which are added at each moment of the current smelting furnace, and data information of bottom blowing gas type, bottom blowing gas proportion, total bottom blowing gas flow, bottom blowing gas temperature, top blowing gas type, top blowing gas proportion, total top blowing gas flow, top blowing gas temperature, furnace liquid level height, oxygen lance height, furnace gas flow, furnace gas temperature and proportion of each component in the furnace gas as input items of N1 time period;
step 3.3: automatically selecting a corrected carbon element content item, a corrected silicon element content item, a corrected manganese element content item, a corrected phosphorus element content item, a corrected sulfur element content item and a corrected energy item corresponding to the stage in the database selected in the step 3.1 according to the classification of the step 2.1.2, and carrying out CO blowing of the converter 2 Real-time prediction of molten steel components and temperature;
step 3.4: predicting real-time carbon element content, predicting real-time silicon element content, predicting real-time manganese element content, predicting real-time phosphorus element content, predicting real-time sulfur element content and predicting real-time molten steel temperature at the end of the N1 time period, and using the molten iron component, the molten iron temperature, the molten iron weight, the scrap steel component, the scrap steel temperature, the scrap steel adding weight, the auxiliary material component, the auxiliary material temperature and the auxiliary material adding weight which are added at each moment, and the type of bottom blowing gas, the bottom blowing gas proportion, the total bottom blowing gas flow, the bottom blowing gas temperature, the top blowing gas type, the top blowing gas proportion, the total top blowing gas flow, the top blowing gas temperature, the liquid level in the furnace, the height of an oxygen lance, the furnace gas flow, the furnace gas temperature and the proportion data information of each component in the furnace gas as input items of the N2 time period;
step 3.5: automatically selecting a corrected carbon element content item, a corrected silicon element content item, a corrected manganese element content item, a corrected phosphorus element content item, a corrected sulfur element content item and a corrected energy item corresponding to the stage in the database selected in the step 3.1 according to the classification of the step 2.1.2, and carrying out CO blowing in the converter in the N2 time period 2 Real-time prediction of molten steel components and temperature;
step 3.6: converting Ny in the steps 3.4-3.5 into N (y + 1), and repeating the steps until (y + 1) = N;
wherein y is not less than 1 and not more than n-1;
step 3.7: comparing rtC, rtSi, rtMn, rtP, rtS and rtT obtained by real-time prediction of the prediction model with rC, rSi, rMn, rP, rS and rT obtained by actual detection data, and recording;
and 4, step 4: after the abnormal heat data are removed, updating the database in real time, and correcting the correction term under a certain condition;
step 4.1: rejecting abnormal heat with [ (1+M%) × average converting time less than or equal to converting time ] < U [ (1+M%) × average smelting time less than or equal to smelting time ], and importing other heat data into a database in real time;
step 4.2: selecting the data of the nearest L furnaces in each database every Z furnaces to correct the correction items, wherein L meets the condition that the data group matched with each category Ci in the step 2.1.2 is not less than L;
step 4.3: when the absolute value of the difference value between the predicted value and the actual detection value of the continuous F furnace is larger than the maximum allowable error value of each molten steel component and temperature, namely | rtC-rC ≧ MaxC or | rtSi-rSi | ≧ MaxSi or | rtMn-rMn | ≧ MaxMn or | rtP-rP | ≧ MaxP or | rtS-rS | ≧ MaxP or | rtT-rT | ≧ MaxT, correction of the correction term is immediately carried out, and L and L are changed.
2. The process of claim 1 wherein said blowing of CO in said converter 2 The method for dynamically predicting the components and the temperature of the molten steel in real time is characterized in that the molten iron components in the step 1.1.2 comprise the carbon content of the molten iron, the silicon content of the molten iron, the manganese content of the molten iron, the phosphorus content of the molten iron and the sulfur content of the molten iron;
the steel scrap components comprise steel scrap carbon element content, steel scrap silicon element content, steel scrap manganese element content, steel scrap phosphorus element content and steel scrap sulfur element content;
the tapping components comprise tapping carbon element content, tapping silicon element content, tapping manganese element content, tapping phosphorus element content and tapping sulfur element content.
3. The process of claim 1, wherein the blowing of CO in the converter is carried out in a single pass 2 The method for dynamically predicting the components and the temperature of the molten steel in real time is characterized in that in the step 1.1.3, the auxiliary material components comprise the carbon element content of an auxiliary material, the silicon element content of the auxiliary material, the manganese element content of the auxiliary material, the phosphorus element content of the auxiliary material and the sulfur element content of the auxiliary material;
the auxiliary materials comprise sintered ore, active lime, quicklime, light-burned dolomite, sintered dolomite, ferrosilicon, fluorite, coke, ferromanganese and a heat-compensating agent.
4. The process of claim 1, wherein the blowing of CO in the converter is carried out in a single pass 2 The method for dynamically predicting the components and the temperature of the molten steel in real time is characterized in that in the step 1.1.4, the bottom blowing gas comprises N 2 Ar and CO 2 (ii) a Step 1.1.5 the top-blown gas comprises O 2 And CO 2 。
5. The process of claim 1, wherein the blowing of CO in the converter is carried out in a single pass 2 The method for dynamically predicting the components and the temperature of the molten steel in real time is characterized in that in the step 1.1.7, the furnace gas components comprise CO and CO 2 、SO 2 、O 2 And N 2 。
6. The process of claim 1 wherein said blowing of CO in said converter 2 The method for dynamically predicting the components and the temperature of the molten steel in real time is characterized in that the actual detection data in the steps 3 and 3.7 and the actual detection value in the step 4.3 refer to the components and the temperature data of the molten steel obtained by removing a sub lance at the later stage of blowing after an abnormal heat, detecting a molten steel terminal sample after blowing and sampling and detecting in the smelting process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210493643.9A CN114854929B (en) | 2022-05-07 | 2022-05-07 | Converter blowing CO 2 Method for dynamically predicting molten steel components and temperature in real time |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210493643.9A CN114854929B (en) | 2022-05-07 | 2022-05-07 | Converter blowing CO 2 Method for dynamically predicting molten steel components and temperature in real time |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114854929A CN114854929A (en) | 2022-08-05 |
| CN114854929B true CN114854929B (en) | 2023-03-17 |
Family
ID=82634573
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210493643.9A Active CN114854929B (en) | 2022-05-07 | 2022-05-07 | Converter blowing CO 2 Method for dynamically predicting molten steel components and temperature in real time |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114854929B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115927784B (en) * | 2022-11-30 | 2023-07-14 | 北京科技大学 | Based on CO 2 Converter steelmaking end point control method by dynamic mixed blowing |
| CN116640906B (en) * | 2023-07-27 | 2023-10-20 | 江苏永钢集团有限公司 | Ladle bottom blowing carbon dioxide smelting method and system based on 5G technology |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101608250A (en) * | 2009-07-20 | 2009-12-23 | 北京科技大学 | A kind of winding-up CO 2The method of gas control temperature of steelmaking melting pool of converter |
| CN101845531B (en) * | 2010-05-07 | 2011-11-02 | 河北省首钢迁安钢铁有限责任公司 | Converter smelting endpoint molten steel carbon and temperature control system and method thereof |
| CN103882176B (en) * | 2014-03-25 | 2015-09-30 | 东北大学 | The online dynamic control method of a kind of convertor steelmaking process based on data-driven |
| CN108359766B (en) * | 2018-03-02 | 2019-03-15 | 北京科技大学 | A kind of oxygen bottom blowing converter smelting process control method based on continuous measurement of molten steel temperature |
| CN108676955B (en) * | 2018-05-02 | 2019-07-12 | 中南大学 | A kind of BOF Steelmaking Endpoint carbon content and temprature control method |
| CN108588324B (en) * | 2018-06-26 | 2019-04-26 | 北京科技大学 | Pneumatic steelmaking passes through CO2The method that (FeO) and dust generate in high-intensitive input control slag |
| CN111500815B (en) * | 2020-05-28 | 2021-06-11 | 北京科技大学 | Bottom blowing O2-CO2Dynamic control method for steelmaking process of CaO converter |
-
2022
- 2022-05-07 CN CN202210493643.9A patent/CN114854929B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN114854929A (en) | 2022-08-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114854929B (en) | Converter blowing CO 2 Method for dynamically predicting molten steel components and temperature in real time | |
| CN108647407B (en) | Method for analyzing and determining carbon in converter steelmaking flue gas | |
| CN103361461B (en) | A kind of converter smelting soft steel phosphorus content on-line prediction control method | |
| CN113919206A (en) | Narrow composition control method for steelmaking alloying | |
| US20150337404A1 (en) | Method and device for predicting, controlling and/or regulating steelworks processes | |
| JP6314484B2 (en) | Hot metal dephosphorization method | |
| CN115522012B (en) | Method for controlling TSC stage oxygen blowing amount of large converter | |
| CN111624324A (en) | Method for testing quality of steel scrap for steelmaking by using converter | |
| CN113981167B (en) | Multi-mode smelting method based on molten iron grading system | |
| CN113961865A (en) | Method for accurately controlling addition amount of temperature regulator in TSC (thyristor switched capacitor) stage of large converter | |
| CN113930575A (en) | Converter double-slag smelting method for high-silicon high-phosphorus molten iron | |
| CN115927784B (en) | Based on CO 2 Converter steelmaking end point control method by dynamic mixed blowing | |
| CN113362903B (en) | Method for intelligently adding lime in TSC (thyristor switched capacitor) stage of large converter | |
| JP2012136767A (en) | Method for estimating phosphorus concentration in converter | |
| US4474361A (en) | Oxygen-blown steelmaking furnace | |
| CN107502698A (en) | A kind of automatic Steelmaking method suitable for less-slag melting | |
| CN115906538A (en) | Method for predicting molten steel components in ladle refining furnace | |
| JP2013144819A (en) | Method for recovering exhaust gas of converter | |
| EP3989013A1 (en) | Method for controlling process, operation guidance method, method for operating blast furnace, method for manufacturing molten iron, and device for controlling process | |
| CN110991772B (en) | Efficient furnace protection method for predicting final slag viscosity model of converter | |
| Terpák et al. | The Mathematical Model for Indirect Measurement of Carbon Concentration in the Steelmaking Process and its Utilization in Process Control | |
| KR100470066B1 (en) | Method of depressing slop by calculating and controlling the retained slag in the oxygen converter after tapping | |
| CN112863609A (en) | Method and device for calculating oxygen blowing amount and coolant amount in converter steelmaking | |
| RU2817694C1 (en) | Refining process control device and refining process control method | |
| JP2000309817A (en) | Converter blowing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |