CN103361454B - Blast furnace hanging material judgment method based on data filtering - Google Patents
Blast furnace hanging material judgment method based on data filtering Download PDFInfo
- Publication number
- CN103361454B CN103361454B CN201210089565.2A CN201210089565A CN103361454B CN 103361454 B CN103361454 B CN 103361454B CN 201210089565 A CN201210089565 A CN 201210089565A CN 103361454 B CN103361454 B CN 103361454B
- Authority
- CN
- China
- Prior art keywords
- data
- blast furnace
- rule base
- parameter
- rule
- 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
Landscapes
- Testing And Monitoring For Control Systems (AREA)
- Manufacture Of Iron (AREA)
Abstract
The invention discloses a blast furnace hanging material judging method based on data filtering, which adopts a grading processing method to filter data in the actual operation of a blast furnace; establishing a primary rule base module according to the process condition of the blast furnace, and monitoring the change and fluctuation of single parameters of the blast furnace; establishing a secondary rule base module, and monitoring the common change and fluctuation of multiple parameters of the blast furnace; according to the monitoring and evaluation result of the data, early warning or correcting the abnormal condition of the blast furnace in time; and flexibly adjusting the primary rule base and the secondary rule base according to requirements. The data of the primary rule base are a material ruler, a material ruler descending speed, a ratio of hot air pressure to air quantity after normalization calculation, an air permeability index, a ratio of an average value of edge temperatures of four crossed temperature measuring points to a furnace top temperature, a ratio of the average value of the edge temperatures of the four crossed temperature measuring points to a furnace throat temperature and a furnace heat index; the second-level rule base data is alarm data output by the first-level rule base. The invention solves the problems of inconvenient modification, difficult maintenance, complex structure, fuzzy conclusion and the like of the conventional rule base, pointedly monitors the condition of the blast furnace hanging, provides basis for the production operation of the blast furnace, and simultaneously achieves the aims of reducing the fluctuation of the furnace condition, ensuring the stable and smooth operation of the furnace condition, improving the production efficiency of the blast furnace and reducing the energy consumption.
Description
Technical field
The present invention relates to blast furnace ironmaking monitoring field, especially a kind of blast furnace hanging determination methods based on data filter.
Background technology
Blast furnace ironmaking is that a collection physics, chemical reaction are also with high temperature, continuous, complicated commercial run that the solid-liquid-gas three phase of heat transfer, mass transfer is changed mutually, therefore, the various parameter of high furnace interior using traditional detection means to obtain and feature are all isolated, single, completely truly can not reflect the working order of blast furnace reality, also just cannot operate the cloth of blast furnace, the operation such as pushing wind manipulation, breeze blowing effectively instructs.Blast furnace hanging to refer in blast furnace production process because Gas Flow motion occurs abnormal, cause furnace charge air permeability to decline causing furnace charge to stop declining there is a kind of severely subnormal working of a furnace.In blast furnace actual mechanical process, determination methods mainly expects according to 1 minutes, footage occurs that rising suddenly appears in stagnation and hot-blast pressure at present.The harm of blast furnace hanging is very large, if can not early discovery, is easy to cause more serious furnace condition disorder.Therefore if judge that the hanging sign of blast furnace will contribute to the stable smooth operation of blast furnace according to some observable data of blast furnace in advance.Generally, Computer System for Some Blast Furnace directly gathers measuring instrument data in blast furnace foundation automatization, then carries out the process such as normalizing, noise reduction, then delivers in the Computer Database of blast furnace.Blast Furnace Expert System or intelligent control system, the parameter according to monitoring judges blast furnace working order indirectly, and utilize expert knowledge library and inference engine to evaluate conditions of blast furnace, the result of evaluation uses the interface display such as Eight-Diagram, histogram to blast furnace operating person usually.Blast furnace operating person, according to the condition of production of the prediction of result next stage evaluated, adjusts operation of blast furnace production strategy.
[patent No. 95118361.3 blast furnace comprhensive deterministic system] relates to a kind of system of carrying out the judgement of blast furnace index according to blast furnace detection data, and com-parison and analysis judgement is carried out in Case-based Reasoning storehouse and Conventional wisdom storehouse.
[patent No. 200710012215.5 blast furnace production process control information intelligence system] utilizes Design of Inference Engine to make evaluation to blast furnace working condition, database technology is utilized to design event table, in table, rule derives from parameter and proposes eigenwert, use case flip-flop toggle inference engine.
Above two patents, all based on expert knowledge library, due to current blast furnace complicacy, polytropy, are difficult to set up unalterable general expert knowledge library, are also difficult to adapt to other blast furnaces even if having.Expertise is that to be based upon blast furnace Data Detection point few, under the prerequisite that data are unreliable and mechanism is unclear, be unsuitable for modern blast furnace check point many, the reliable present situation of data.This is exactly introduce and reason that the sorts of systems result of use of domestic-developed is undesirable in fact.
[patent No. CN1403593 mono-kind utilizes intelligent control system to control the method for blast-furnace smelting] utilizes intelligent control system to control blast furnace process, local area network is built at iron work, realize iron manufacturing process real time information sampling and automatic transmitting function, utilize intelligent control software module, follow the tracks of operation of blast furnace data variation, then export diagnostic result.The method relates generally to the technology such as Controlling model and control strategy, does not relate to effective filtration of blast furnace data and usage data filter method to judge blast furnace hanging.
Summary of the invention
The object of the present invention is to provide a kind of blast furnace hanging determination methods based on data filter, adopt hierarchical processing method, implement to filter to the data in blast furnace actual motion, effective monitoring blast furnace data variation and fluctuation, judge that blast furnace occurs that hanging possibility provides foundation.
Another object of the present invention is to safeguard by rule problems such as solving the inconvenience of rule base amendment in the past (needing expert to participate in amendment), difficult in maintenance, complex structure, conclusion are fuzzy.And then reach the fluctuation of the minimizing working of a furnace, ensure working of a furnace stable smooth operation, improve blast fumance efficiency, reduce the object of energy consumption.
Another object of the present invention is to set first-order rule library module, secondary rule base module parameter in conjunction with the feature of blast fumance technique, overcome and only judge blast furnace hanging to expect that chi is stagnated, and rise suddenly with hot-blast pressure, and lack the defect that associated real data judges existing for blast furnace " hanging ".
The object of the invention is by following process implementation, based on a blast furnace hanging determination methods for data filter, it is characterized in that comprising and set up digital sampling and processing, first-order rule library module, secondary rule base module, result output module and regular maintenance module.The data in first-order rule storehouse are with material chi and expect chi lowering speed correlation parameter and correlation parameter thresholding, comprise hot-blast pressure, air quantity, hot-blast pressure and the air quantity ratio after normalization method calculates, permeability index, throat temperature, furnace throat temperature, cross temperature four point edge temperature, cross temperature four point edge temperature averages and throat temperature ratio, cross temperature four point edge temperature averages and furnace throat temperature ratio, molten iron [Si] content, hearth thermal state, ore deposit batch weight and coke (per) charge anharmonic ratio value, additional coke quantity in a smelting cycle; Secondary rule base data are that each parameter that first-order rule storehouse exports exceedes thresholding alert data.Implement to filter to the data in the blast furnace actual motion of data collecting module collected by first-order rule library module, the monitoring one-parameter change of blast furnace and fluctuation.The change common by secondary rule base module monitors blast furnace multiparameter and fluctuation.When one-parameter data do not meet the alarm limit that first-order rule specifies, its data are filtered out by first-order rule storehouse automatically, when one-parameter data meet the alarm limit that first-order rule specifies, according to first-order rule through result output module trigger alarm or direct intervention blast furnace operating, the data meeting first-order rule alarm limit trigger alarm are sent into secondary rule base simultaneously and process.When sending into the alarm limit of the non-meeting tier 2 rule predetermining of Multi-parameter data of secondary rule base, its data are filtered out by secondary rule base automatically, when sending into the alarm limit of Multi-parameter data meeting tier 2 rule predetermining of secondary rule base, according to secondary rule through result output module trigger alarm.
1. digital sampling and processing
Raw data is provided by on-the-spot two-level computer system server, is the sequence data of one group of fixed cycle.According to the sampling interval that first-order rule storehouse provides, field data is stored by minimum sampling interval.In order to not miss maximum value in sampling interval and minimum value, every point data records average (as master data), maximum value and minimum value simultaneously.Can use one of them according to processing requirement or combinationally use, raw data is stored in database for subsequent use.
2. first-order rule library module
First-order rule storehouse is set up in requirement according to technique needs or secondary rule base.As according to technique needs, record certain one-parameter sampling interval, retardation time, high alarm setting 1, high alarm setting 2, low alarm limit 2, low alarm limit 1 etc. relative to current molten iron.Regular record is kept at database backup, can be safeguarded according to blast furnace operating mode by furnace superintendent.For another example secondary rule base input parameter is the Output rusults in first-order rule storehouse, sets up first-order rule storehouse with regard to needing according to the requirement of secondary rule base.Adopt first-order rule module monitors one-parameter data, when one-parameter data meet the alarm limit alarm that first-order rule specifies.Blast furnace operating personnel by adjustment one-parameter correction conditions of blast furnace, also can carry out closed-loop control to one-parameter, direct intervention blast furnace operating.
3. secondary rule base module
On the basis of first-order rule, set up secondary rule base (operational stage rule base) according to the technique contact between certain several one-parameter.When certain several one-parameter data meets respective alert if, they meet again secondary rule base defined terms simultaneously, can according to secondary rule base trigger alarm or blast furnace operating.If only there is several one-parameter to meet respective alert if, not during meeting tier 2 rule base rule, screen-picture and audible alarm are also failure to actuate, but go to first-order rule storehouse and carry out respective handling.
4. result output module
According to the result of determination to different pieces of information and feature, classification is sent on the screen of the terminal of operation department or master control room.As result of determination only need to remind blast furnace operating personnel time, only according to judgement situation, alarming result need be shown to computer screen with the form of picture.Time as very urgent in result of determination, according to practical situation, alarming result can be shown to computer screen simultaneously with the form of picture, take measures rapidly to intervene blast furnace operating, avoid blast furnace to fluctuate or abnormal.
5. regular maintenance module
First first-order rule and secondary rule set up basic rule base by iron-smelting process expert according to production technology principle.In use, furnace superintendent can according to State of Blast Furnace labour and raw material condition be modified and parameter increase and decrease.
The present invention is to the method for blast furnace data acquisition stage treatment, implement to filter to the data in blast furnace actual motion, extract the delta data that the steady direct motion of blast furnace is impacted, according to blast furnace technology situation, formulate first-order rule library module, the monitoring one-parameter change of blast furnace and fluctuation.Formulate secondary rule base module, the change of monitoring blast furnace multiparameter and fluctuation, according to the monitoring evaluation result to data, the unusual condition of timely early warning or correction blast furnace.Furnace superintendent can adjust first-order rule storehouse and secondary rule base as required flexibly, solves the problems such as rule base in the past revises inconvenience (needing expert to participate in amendment), difficult in maintenance, complex structure, conclusion are fuzzy.Also reach simultaneously and reduce working of a furnace fluctuation and energy consumption, improve blast fumance efficiency, ensure the object of working of a furnace stable smooth operation.
The present invention makes parameter or the index comprehensive descision blast furnace generation hanging facts such as materials footage certificate, material chi lowering speed, hot-blast pressure and the ratio data of air quantity after normalization method calculates, hearth thermal state, permeability index, cross temperature four point edge temperature averages and throat temperature ratio, cross temperature four point edge temperature averages and furnace throat temperature ratio and has the trend that hanging occurs.Can implement to filter to the data in blast furnace actual motion, facilitate alteration ruler storehouse, effective monitoring blast furnace data variation and fluctuation, particularly monitor the situation of blast furnace hanging, for operation of blast furnace production provides foundation.Thus reach the fluctuation of the minimizing working of a furnace and energy consumption, improve blast fumance efficiency, ensure the object of working of a furnace stable smooth operation.
Accompanying drawing explanation
Accompanying drawing 1 blast furnace data monitoring module schematic diagram;
Accompanying drawing 2 first-order rule data declaration schematic diagram;
Accompanying drawing 3 one-level and secondary rule relation schematic diagram.
Embodiment
Below in conjunction with accompanying drawing, the present invention is described in further detail.
As shown in Figure 1, the invention provides a kind of blast furnace hanging determination methods based on data filter, the system of employing comprises sets up digital sampling and processing, first-order rule library module, secondary rule base module, result output module and regular maintenance module.
1. digital sampling and processing
Raw data is provided by on-the-spot second computer, is the data sequence of one group of fixed cycle.According to the sampling interval that first-order rule storehouse provides, field data is stored by minimum sampling interval.In order to not miss maximum value in sampling interval and minimum value, filter out and influential data are not had to smooth operation of furnace, in data cell, every point data only retains average (as master data), maximum value and minimum value, can use one of them or combinationally use according to processing requirement.Raw data is stored in database for subsequent use, to expect chi lowering speed and material chi data instance: be 5 seconds according to the sampling interval that first-order rule storehouse provides, and records the material chi lowering speed of every 5 seconds and material footage according to for subsequent use as a point data.
2. first-order rule library module
First-order rule storehouse is set up in requirement according to technique needs or secondary rule base.As according to technique needs, record certain one-parameter sampling interval, retardation time, parameter high alarm setting 1, parameter high alarm setting 2, parameter low alarm limit 2, parameter low alarm limit 1 etc. that parameter has an impact relative to current molten iron [Si] content.Regular record is kept at database backup, can be safeguarded according to blast furnace operating mode by furnace superintendent.For another example secondary rule base input parameter is the Output rusults in first-order rule storehouse, sets up first-order rule storehouse with regard to needing according to the requirement of secondary rule base.
Refer to that a certain data are delayed relative to the physics of other data points action time to the time point of blast furnace effect retardation time.Such as: relative to the impact of air quantity on [Si] content in current molten iron, current material loading data are on the approximately delayed smelting cycle (7 ~ 8 hours) of the impact of [Si] content in current molten iron, the i.e. time of 4 sampling interval, according to the combustion intensity of section sometime, smelting cycle is relatively-stationary (see Fig. 2).
Hot-blast pressure and air quantity to blast furnace action time within 10 minutes, hot-blast pressure and air quantity are sync direction change under normal circumstances, if occur that within continuous two sampling periods hot-blast pressure is that increase mode changes, and air quantity is on a declining curve, think to possess and occur hanging probability, but because hot-blast pressure and air quantity are not same quantitative series certificates, have to pass through normalization method to calculate, (i) hot-blast pressure is become with air quantity normalizing, 1] data between, then calculate hot-blast pressure and air quantity ratio, calculating hot-blast pressure and air quantity ratio as single parameter, the each sampling period hot-blast pressure of Continuous plus and air quantity ratio in 1 minute, get its moving average, setup parameter high alarm setting 1, parameter high alarm setting 2, regular record is kept at database backup, alarming value can be safeguarded by furnace superintendent according to blast furnace operating mode.If continuous 2 cycles meet alert if, then in parameters input secondary rule base.
Permeability index to blast furnace action time within 1 minute, if occur declining within continuous two sampling periods, think to possess and occur hanging probability, setup parameter low alarm limit 2, parameter low alarm limit 1 etc., regular record is kept at database backup, can safeguard alarming value according to blast furnace operating mode by furnace superintendent.If continuous 2 cycles meet alert if, then in parameters input secondary rule base.
Throat temperature, furnace throat temperature, cross temperature four point edge temperature to blast furnace action time within 1 minute, three parameter changes in sync direction all the time, but rate of change is variant, wherein cross temperature four point edge temperature variation the fastest, secondly be furnace throat temperature, be thirdly throat temperature.If all there is ascendant trend in continuous two sampling periods in three parameters, Continuous plus cross temperature four point edge temperature averages and throat temperature ratio and cross temperature four point edge temperature averages and furnace throat temperature ratio in 1 minute, get its moving average, when cross temperature four point edge temperature averages and throat temperature ratio are greater than 0.8, simultaneously cross temperature four point edge temperature averages and furnace throat temperature ratio are greater than 1.5 and think to possess and occur hanging probability, the concrete high alarm setting 1 of setup parameter, parameter high alarm setting 2 etc., regular record is kept at database backup, alarming value can be safeguarded by furnace superintendent according to blast furnace operating mode.If continuous 2 cycles meet alert if, then in parameters input secondary rule base.
Hearth thermal state is by local heat balance calculation result data, within every 1 hour, calculate 1 time, if this index is less than 0.75, think to possess and occur hanging probability, the concrete low alarm limit 2 of setup parameter, parameter low alarm limit 1 etc., regular record is kept at database backup, can safeguard alarming value according to blast furnace operating mode by furnace superintendent.If continuous 2 cycles meet alert if, then in parameters input secondary rule base.
Ore deposit batch weight and coke (per) charge tuple are according to being a smelting cycle (7 ~ 8 hours) to blast furnace action time, there is no fixed sample period, data gathering is as triggering according to actual cloth " switch " signal, then normal cloth order is 1 batch of ore is 1 coke charge charcoal, if there is 2 coke charge charcoals, then the 2nd coke charge charcoal then regards as additional coke.Additional coke quantity does cumulative calculation by early stage (0 ~ 2.8 hour), mid-term (2.9 ~ 5.6 hours) and later stage (5.7 ~ 8.0 hours), if interior ore deposit batch weight and the coke (per) charge anharmonic ratio value of normal cloth order are less than 4.5, and have additional coke, think to possess and occur hanging probability.The concrete high alarm setting 1 of setup parameter, parameter high alarm setting 2, parameter high alarm setting 3, parameter low alarm limit 3, parameter low alarm limit 2, parameter low alarm limit 1 etc. are done by early stage, mid-term and later stage and additional coke quantity, regular record is kept at database backup, can safeguard alarming value according to blast furnace operating mode by furnace superintendent.If there is alarming value, then in parameters input secondary rule base.
Adopt first-order rule module monitors one-parameter data, implement to filter to one-parameter data, when one-parameter data do not meet the alarm limit that first-order rule specifies, its data are filtered out by first-order rule storehouse automatically, be not fed in secondary rule base, when one-parameter data meet the alarm limit that first-order rule specifies, can report to the police or send into secondary rule base and process.
To expect chi data instance, sampling interval (5 seconds), retardation time (0 second), high alarm setting I (0.5m), high alarm setting II (0.3m), low alarm limit II (0m), low alarm limit I (0m).To expect chi lowering speed, sampling interval (5 seconds), retardation time (0 second), high alarm setting I (0.004m/s), high alarm setting II (0.002m/s), low alarm limit II (0m/s), low alarm limit I (0m/s).Adopt the rule in first-order rule storehouse to monitor one-parameter data, exceed respective alarm limit and report to the police.Blast furnace operating personnel can carry out closed loop management, direct intervention blast furnace operating by adjustment one-parameter data, revise conditions of blast furnace.In this example, first-order rule can be formulated as follows: when material footage is according to meeting following condition for the moment: higher than 0.3m or higher than 0.5m, result output module produces screen-picture and warning of sounding; When material chi lowering speed meets following condition for the moment: lower than 0.004m/s or lower than 0.002m/s, result output module produces screen-picture and warning of sounding, and then alert data value is transported to secondary rule base module.Assuming that the actual value that the actual value of material chi is 0.25m or material chi lowering speed is 0.01m/s, its data are in rational scope, can not meet first-order rule storehouse, and therefore these data are filtered out by first-order rule storehouse, do not process it.
The ratio data after normalization method calculates for hot-blast pressure and air quantity, when ratio data is more than 1.1 or 1.2, result output module produces screen-picture and warning of sounding, and then alert data value is transported to secondary rule base module.Assumed calculation value is 1.0, and its data are in rational scope, can not meet first-order rule storehouse, and therefore these data are filtered out by first-order rule storehouse, do not process it.
For hearth thermal state, when exponent data is lower than 0.75 or lower than 0.65, result output module produces screen-picture and warning of sounding, and then alert data value is transported to secondary rule base module.If exponent data is 0.76, its data are in rational scope, can not meet first-order rule storehouse, and therefore these data are filtered out by first-order rule storehouse, do not process it.
For permeability index, when index is lower than 28 or lower than 25, result output module produces screen-picture and warning of sounding, and then alert data value is transported to secondary rule base module.If exponent data is 30, its data are in rational scope, can not meet first-order rule storehouse, and therefore these data are filtered out by first-order rule storehouse, do not process it.
For cross temperature four point edge temperature averages and throat temperature ratio and cross temperature four point edge temperature averages and furnace throat temperature ratio, if cross temperature four point edge temperature averages and throat temperature ratio are higher than 0.80 or higher than 0.85, cross temperature four point edge temperature averages and furnace throat temperature ratio are higher than 1.50 or higher than 1.65 simultaneously, result output module produces screen-picture and warning of sounding, and then alert data value is transported to secondary rule base module.If cross temperature four point edge temperature averages and throat temperature ratio are 0.82, but cross temperature four point edge temperature averages and furnace throat temperature ratio are 1.45, its data are still thought and are in rational scope, first-order rule storehouse can not be met, therefore these data are filtered out by first-order rule storehouse, do not process it.
3. secondary rule base module
On the basis of first-order rule, set up secondary rule base (operational stage rule base) according to the technique contact between certain several one-parameter.In other words, when only having blast furnace data to meet alert if, the working of a furnace of blast furnace can not be judged completely, need one group of blast furnace data to satisfy condition just to think blast furnace generation unusual fluctuations simultaneously.Therefore, secondary rule base can be set up according to blast furnace practical situation, when certain several one-parameter data meets respective alert if, they meet again secondary rule base defined terms simultaneously, can according to secondary rule base trigger alarm or blast furnace operating (see Fig. 3).Such as blast furnace " hanging " state secondary rule needs supplemental characteristic to have material chi, material chi lowering speed.When expecting that footage is according to when meeting high alarm setting 2, and when material chi lowering speed also meets high alarm setting 2 simultaneously, " hanging " state of secondary rule can be formed.In this example, when expecting that footage meets following condition according to according to secondary rule: material footage according to higher than 0.3m (high alarm setting 2), material chi lowering speed lower than 0.002m/s (high alarm setting 2) time, result output module produces screen-picture and audible alarm.If the multiple one-parameters in the rule defined in secondary rule base all do not meet respective alert if, namely during non-meeting tier 2 rule base rule, secondary rule base filters out its information, think that it does not affect blast furnace, screen-picture and audible alarm are also failure to actuate, but go back to first-order rule storehouse and carry out respective handling.In this example, if material footage according to according to secondary rule definition is: expect footage certificate higher than 0.3m (high alarm setting 2), material chi lowering speed lower than 0.002m/s (high alarm setting 2).If expect footage certificate higher than 0.3m (high alarm setting 2), material chi lowering speed higher than 0.002m/s (high alarm setting 1), secondary rule base filters out its information, think that it does not affect blast furnace, screen-picture and audible alarm are also failure to actuate, but go back to first-order rule storehouse and carry out respective handling.
4. result output module
According to the result of determination to different pieces of information and feature, classification is sent on the screen of the terminal of operation department or master control room.As result of determination only need to remind blast furnace operating personnel time, only according to judgement situation, alarming result need be shown to computer screen with the form of picture; Time as very urgent in result of determination, can according to practical situation, while alarming result being shown to computer screen with the form of picture and sounding and report to the police, take measures rapidly to intervene blast furnace operating, avoid blast furnace fluctuation or abnormal, such as when expecting that the actual value of chi or continuous two batch material chi data difference reaches the bound of setting, burden distribution system or properly distributed Gas Flow can be changed by closed loop topworks, stablizing conditions of blast furnace.
5. rules administration module
First first-order rule and secondary rule set up basic rule base by the expert of iron-smelting process according to production technology principle, and in use, furnace superintendent can according to State of Blast Furnace labour and raw material condition be modified and parameter increase and decrease.The recent coke quality of such as blast furnace is bad, both can using M40 as parameter of measurement: sampling interval (30 minutes), retardation time (120 ~ 180 minutes), low alarm limit 2 (80%), low alarm limit 1 (78%); Using M10 also as parameter of measurement: sampling interval (30 points), retardation time (120 ~ 180 minutes), high alarm setting 1 (7.0%), high alarm setting 2 (7.2%).By that analogy, just can constantly to adjust and substantial rule base.
Claims (7)
1., based on a blast furnace hanging determination methods for data filter, it is characterized in that comprising and set up digital sampling and processing, first-order rule library module, secondary rule base module, result output module and regular maintenance module; The data in first-order rule storehouse are with material chi and expect chi lowering speed correlation parameter and correlation parameter thresholding, comprise hot-blast pressure, air quantity, hot-blast pressure and the air quantity ratio after normalization method calculates, permeability index, throat temperature, furnace throat temperature, cross temperature four point edge temperature, cross temperature four point edge temperature averages and throat temperature ratio, cross temperature four point edge temperature averages and furnace throat temperature ratio, molten iron [Si] content, hearth thermal state, ore deposit batch weight and coke (per) charge anharmonic ratio value, additional coke quantity in a smelting cycle simultaneously; Secondary rule base data are that each parameter that first-order rule storehouse exports exceedes thresholding alert data; Implement to filter to the data in the blast furnace actual motion of data collecting module collected by first-order rule library module, the monitoring one-parameter change of blast furnace and fluctuation; The change common by secondary rule base module monitors blast furnace multiparameter and fluctuation; When one-parameter data do not meet the alarm limit that first-order rule specifies, its data are filtered out by first-order rule storehouse automatically, when one-parameter data meet the alarm limit that first-order rule specifies, according to first-order rule through result output module trigger alarm or direct intervention blast furnace operating, the data meeting first-order rule alarm limit trigger alarm are sent into secondary rule base simultaneously and process; When sending into the alarm limit of the non-meeting tier 2 rule predetermining of Multi-parameter data of secondary rule base, its data are filtered out by secondary rule base automatically, when sending into the alarm limit of Multi-parameter data meeting tier 2 rule predetermining of secondary rule base, according to secondary rule through result output module trigger alarm.
2. the blast furnace hanging determination methods based on data filter according to claim 1, it is characterized in that the data in blast furnace actual motion are provided by on-the-spot second computer, it is the sequence data of one group of fixed cycle, according to the sampling interval that first-order rule storehouse provides, above-mentioned data are stored by minimum sampling interval, every point data records average, maximum value and minimum value simultaneously, use one of them according to processing requirement or combinationally use, the data of collection are stored in database.
3. the blast furnace hanging determination methods based on data filter according to claim 1, it is characterized in that described first-order rule storehouse sets up according to the requirement of technique needs or secondary rule base, according to technique needs, record parameter sampling interval, the retardation time relative to current molten iron, alarm limit, wherein alarm limit divides one, two two-stages, every grade is divided upper or lower limit according to its effect, and regular record is preserved in a database; According to the requirement of secondary rule base, using the input parameter of secondary rule base needs as first-order rule storehouse Output rusults, be back-calculated to obtain the first-order rule storehouse rule corresponding according to secondary rule base required input parameter.
4. the blast furnace hanging determination methods based on data filter according to claim 1, it is characterized in that, described secondary rule base is on the basis of first-order rule, sets up according to the technique contact between two or more one-parameters; When two or more one-parameter data meet respective alert if, they meet again secondary rule base defined terms simultaneously, can according to secondary rule base trigger alarm or blast furnace operating; If only there is several one-parameter to meet respective alert if, not during meeting tier 2 rule base rule, screen-picture and audible alarm are also failure to actuate, but carry out respective handling according to the rule in first-order rule storehouse to it.
5. the blast furnace hanging determination methods based on data filter according to claim 1, it is characterized in that according to the result of determination to different pieces of information and feature, comprise the result drawn according to first-order rule storehouse and secondary rule base, classification is sent on the screen of the terminal of operation department or master control room.
6. the blast furnace hanging determination methods based on data filter according to claim 1, it is characterized in that first first-order rule storehouse and secondary rule base are set up according to production technology principle, in use, modify and parameter increase and decrease according to State of Blast Furnace labour and raw material condition.
7. the blast furnace hanging determination methods based on data filter according to claim 1, it is characterized in that material footage according to comprise sampling interval 5 seconds, 0 second retardation time, high alarm setting I 0.5m, high alarm setting II 0.3m, low alarm limit II 0m, low alarm limit I 0m; Material chi lowering speed comprise sampling interval 5 seconds, 0 second retardation time, high alarm setting I 0.004m/s, high alarm setting II 0.002m/s, low alarm limit II 0m/s, low alarm limit I 0m/s; Hot-blast pressure and the air quantity ratio data computation period after normalization method calculates 1 minute, high alarm setting I 1.1, high alarm setting II 1.2; Hearth thermal state computation period 1 hour, low alarm limit II 0.65, low alarm limit I 0.75; Permeability index computation period 1 minute, low alarm limit II 25, low alarm limit I 28; Cross temperature four point edge temperature averages and 1 minute throat temperature ratio calculation cycle, high alarm setting I 0.80, high alarm setting II 0.85; Cross temperature four point edge temperature averages and 1 minute furnace throat temperature ratio calculation cycle, high alarm setting I 1.50, high alarm setting II 1.65; When material footage certificate is higher than 0.3m or higher than 0.5m, result output module produces screen-picture and warning of sounding; When material chi lowering speed is lower than 0.004m/s or lower than 0.002m/s, result output module produces screen-picture and warning of sounding; When hot-blast pressure and air quantity, the ratio after normalization method calculates is higher than 1.1 or higher than 1.2, and result output module produces screen-picture and warning of sounding; When hearth thermal state is lower than 0.75 or lower than 0.65, result output module produces screen-picture and warning of sounding; When permeability index is lower than 25 or lower than 28, result output module produces screen-picture and warning of sounding; When cross temperature four point edge temperature averages and throat temperature ratio are higher than 0.80 or higher than 0.85, cross temperature four point edge temperature averages and furnace throat temperature ratio are higher than 1.50 or higher than 1.65 simultaneously, result output module produces screen-picture and warning of sounding, then, alert data value is transported to secondary rule base module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210089565.2A CN103361454B (en) | 2012-03-30 | 2012-03-30 | Blast furnace hanging material judgment method based on data filtering |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210089565.2A CN103361454B (en) | 2012-03-30 | 2012-03-30 | Blast furnace hanging material judgment method based on data filtering |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103361454A CN103361454A (en) | 2013-10-23 |
| CN103361454B true CN103361454B (en) | 2015-03-11 |
Family
ID=49363709
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210089565.2A Active CN103361454B (en) | 2012-03-30 | 2012-03-30 | Blast furnace hanging material judgment method based on data filtering |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103361454B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109306385B (en) * | 2018-10-29 | 2023-04-07 | 甘肃酒钢集团宏兴钢铁股份有限公司 | Blast furnace top pressure stability control system and control method thereof |
| CN111210599A (en) * | 2019-12-11 | 2020-05-29 | 南京连易智能科技有限公司 | Chemical industry index early warning method and system, electronic equipment and storage medium |
| CN110991772B (en) * | 2019-12-27 | 2023-04-18 | 安徽工业大学 | Efficient furnace protection method for predicting final slag viscosity model of converter |
| CN112111618B (en) * | 2020-09-22 | 2022-06-03 | 攀钢集团研究院有限公司 | Blast furnace burden descending uniformity judgment and early warning method and system |
| CN114971015A (en) * | 2022-05-25 | 2022-08-30 | 张家港宏昌钢板有限公司 | Prediction and treatment method of a 2000-level blast furnace overhang accident |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1097804A (en) * | 1993-07-21 | 1995-01-25 | 首钢总公司 | Computerized blast furnace smelting expert system method |
| CN1173542A (en) * | 1997-06-13 | 1998-02-18 | 冶金工业部自动化研究院 | Blast furnace operating consulting system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0726127B2 (en) * | 1987-11-20 | 1995-03-22 | 日本鋼管株式会社 | Blast furnace furnace automatic heat control system |
-
2012
- 2012-03-30 CN CN201210089565.2A patent/CN103361454B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1097804A (en) * | 1993-07-21 | 1995-01-25 | 首钢总公司 | Computerized blast furnace smelting expert system method |
| CN1173542A (en) * | 1997-06-13 | 1998-02-18 | 冶金工业部自动化研究院 | Blast furnace operating consulting system |
Non-Patent Citations (1)
| Title |
|---|
| 高炉专家系统的数据采集及处理;国宏伟 等;《冶金自动化》;20080531;第32卷(第3期);第18-22页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103361454A (en) | 2013-10-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103361454B (en) | Blast furnace hanging material judgment method based on data filtering | |
| CN111652414B (en) | Sliding Window PCA Blast Furnace Abnormality Monitoring Method Based on Gaussian Mixture Model | |
| CN103544273A (en) | Method for assessing integral states of furnace conditions by aid of pattern recognition technology | |
| CN101423348B (en) | Integrated recognition method for sintering conditions of cement rotary kiln | |
| WO2024093533A1 (en) | Hydroelectric signal abnormality determination method based on frequent item set reasoning | |
| CN116805065B (en) | Intelligent management method for monitoring data of coal powder heating furnace burner | |
| US20220390928A1 (en) | Production facilities monitoring method, production facilities monitoring device, and operation method for production facilities | |
| CN105022355A (en) | Cement rotary kiln intelligent optimization control system | |
| CN116567127B (en) | Smart phone with fault monitoring function | |
| CN113177353A (en) | Data model construction method applied to industrial early warning system | |
| CN102830659B (en) | Blast furnace data filtering method | |
| CN108018393A (en) | Intelligent automatic steelmaking system and method | |
| CN110580964A (en) | Online monitoring method and system for thermal power improvement margin of nuclear power unit | |
| CN106483946A (en) | A kind of metallurgical production process energy consumption of unit product on-line detecting system and method | |
| CN111639800B (en) | Method, device and storage medium for setting blast furnace process parameter range | |
| CN118569540B (en) | A negative electrode material production equipment operation control system based on data analysis | |
| CN202755005U (en) | Iron making blast furnace production monitoring alarm system | |
| CN114893332A (en) | Fuel filter state diagnosis method based on Internet of things | |
| CN207867317U (en) | A kind of wire feeder operating state information acquisition device | |
| CN114971015A (en) | Prediction and treatment method of a 2000-level blast furnace overhang accident | |
| CN119200709B (en) | Return gas temperature control system suitable for coke production control | |
| CN212610927U (en) | Monitoring alarm device for electrolytic tank crust breaking | |
| CN115106189B (en) | Implementation method and implementation system for maximizing benefits of coal preparation plant | |
| CN102347635A (en) | Refined monitoring system for power supply reliability | |
| CN115857441B (en) | Coal type control information transmission system in coking coal washing and selecting process |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |