CN107096899A - The crystallizer bleedout prediction electric thermo system that a kind of logic-based judges - Google Patents
The crystallizer bleedout prediction electric thermo system that a kind of logic-based judges Download PDFInfo
- Publication number
- CN107096899A CN107096899A CN201710346785.1A CN201710346785A CN107096899A CN 107096899 A CN107096899 A CN 107096899A CN 201710346785 A CN201710346785 A CN 201710346785A CN 107096899 A CN107096899 A CN 107096899A
- Authority
- CN
- China
- Prior art keywords
- temperature
- thermocouple
- electric thermo
- data
- alarm
- 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.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B22—CASTING; POWDER METALLURGY
- B22D—CASTING OF METALS; CASTING OF OTHER SUBSTANCES BY THE SAME PROCESSES OR DEVICES
- B22D11/00—Continuous casting of metals, i.e. casting in indefinite lengths
- B22D11/16—Controlling or regulating processes or operations
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Continuous Casting (AREA)
Abstract
The invention belongs to metallurgical continuous casting controlling of production process field, the crystallizer bleedout prediction electric thermo system judged more specifically to a kind of logic-based.The hardware composition part of the present invention includes fourplatemold and thermocouple, steel leakage prediction system computer, PLC, interchanger, thermocouple compensation cable, audible-visual annunciator, relay, distant station, in addition to data acquisition unit, bleed-out data compress memory cell, self study and off-line analysis unit, alarm and execution unit automatically;Wherein, steel leakage prediction system computer includes client and server computer, and client computer is used for the monitoring display of bleed-out data;Wherein, thermocouple bury columns and rows spacing design to bond tearing port it is identical the time required to one thermocouple spacing of horizontal and vertical propagation.The present invention realizes the promptly and accurately forecast to bonding steel leakage, it is ensured that continuous casting direct motion and raising slab quality.
Description
Technical field
The invention belongs to metallurgical continuous casting controlling of production process field, judge more specifically to a kind of logic-based
Crystallizer bleedout prediction electric thermo system.
Background technology
The fast development of high-efficient continuous casting technology is modernized, high-quality continuous casting billet is not required nothing more than, and requires high pulling rate, is bonded
Bleed-out is always an important problem for restricting high casting speed continuous casting, solves bonding steel leakage to ensureing that continuous casting direct motion and production are high
Quality strand is significant, at present in addition to improving and inducing the process conditions that green shell is bonded, and its Main Means is that exploitation is high
The fast crystallizer bleedout prediction electric thermo system of forecast accuracy, low rate of false alarm, response.The crystallizer bleedout prediction electric thermo system commonly used both at home and abroad
The change of certain amount thermocouple measurement copper plate temperature is buried mainly in copper plate of crystallizer, is propagated based on V-type tearing port is bonded
When electric thermo-couple temperature performance the spatial-temporal characteristics such as " time lag " and " temperature inversions ", using logic judgment algorithm or intelligently calculation
Method recognizes these features, realizes the forecast of bonding steel leakage.Wherein, although intelligent algorithm is solving actual nonlinear problem tool
There is preferable performance, but realize that simple, workable, real-time is high, reliability is high, easy care in view of logic judgment algorithm
The advantages of, by broad practice in major steel mill's steel leakage prediction system.
On the various logic evaluation algorithm of breakout prediction, due to the difference that thermocouple is arranged and logic rules are designed, lead
Cause algorithm performance and time of fire alarming to have very big difference, particularly in practical application, due to thermocouple failure or shell growth not
It is uniform wait caused by electric thermo-couple temperature fluctuation, easily cause strike-alarm not in time, bond false alarm.Once strike-alarm is too late
When, then it can not make to bond reparation so as to avoid bleed-out by dropping pulling rate measure;And false alarm can cause casting machine to drop suddenly each time
Speed is shut down, and has a strong impact on the high-efficiency reform of slab quality and casting machine, huge economy can be all caused as bleed-out accident
Loss.So in the logic judgment algorithm of design bonding, cohesive behavior is not only quoted in time, it is considered to alarm ageing, contracted
Short alarm time, is that the reparation of follow-up bonding steel leakage creates favorable conditions, while must reduce false alarm as far as possible again.
Patent document CN101332499A discloses a kind of slab continuous-casting bleedout forecast control method, the forecast control method
It is to be based on Thermocouple temperature change, realizes what bonding steel leakage judged by logic judgment algorithm.The forecast control method
It is representative temperature waveform when judging that single thermocouple is bonded using temperature amplitude inspection and duration, and passes through neighboring thermocouple
The time lag of temperature anomaly change considers cohesive vertical and horizontal dissemination, finally determines that the bleed-out of every thermocouple is several
Rate, when the bleed-out probability for certain branch thermocouple occur reaches 95%, reduction pulling rate stops cast.Although the forecast control method one
Determine the accuracy that degree improves breakout prediction, reduce error-alert degree, still, the forecast control method does not account for viscous
Alarm time is tied and ageing, particularly when being checked using temperature amplitude, to single thermocouple sticking temperature waveform
Recognize that time of fire alarming is later, cause not necessarily quote cohesive behavior in time, if later alarm, even if directly shutdown
The reparation for bonding breach can not be ensured without bleed-out, if alarmed earlier, keep relatively low green shell to heal pulling rate with regard to that can make to glue
Tie breach reparation, and and need not shut down, stop cast, and increase the unnecessary operating time, reduce production efficiency and increase
Equipment loss.
Patent document CN102825234A discloses a kind of judgement alarm method of bonding steel leakage, and the judgement alarm method is
(i.e. temperature is inverted) feature is intersected in the time lag and temperature curve of spatial variations based on multiple rows of electric thermo-couple temperature, is for bonding
When representative temperature change waveform pattern or theoretical temperatures change waveform pattern and design, and in actual production process, and
Not all sticking temperature change waveform pattern can all show " temperature inversion " phenomenon, and due to liquid steel level during actual production
The influence of the factors such as fluctuation, covering slag performance, MOLTEN STEEL FLOW, thermocouple failure, often occurs thermoelectricity above and below copper plate of crystallizer
Even temperature is inverted, or in the case of the non-adherent such as unsteady statecalculation, thermocouple failure, also often has larger thermocouple
Temperature fluctuation, so also easily produces temperature phenomena of inversion, and institute's design in this way undoubtedly adds report by mistake and fail to report several
Rate.
It is comprehensive that current steel leakage prediction system both at home and abroad and the investigation of logic judgment forecasting procedure are found, it is primarily present and asks as follows
Topic:(1) thermocouple failure itself causes measured temperature anomalous variation, the judgement of influence breakout prediction algorithm and system.Such as:By
In thermocouple quality problems itself, performance is unstable or thermocouple loses pyroelecthc properties, and temperature curve fluctuation is excessive;Thermocouple is filled
It is bad with mode mistake or sealing, cause greasy dirt or cooling water to enter rear measurement temperature and actual temperature deviation greatly, make temperature
Curve is in undulating;Occur distortion phenomenon in thermocouple signal transmitting procedure, fluctuate temperature curve.(2) open pour, tundish
Change, pour eventually etc. in the case of unsteady statecalculation, larger is influenceed on electric thermo-couple temperature, can also increase the false alarm of algorithm;In theory
As long as differentiating under various unsteady state conditions and bond the electric thermo-couple temperature waveform of process just can reduce wrong report, but real process,
Sometimes electric thermo-couple temperature change closely bonds process temperature waveform under unsteady state condition, and this adds increased breakout prediction model
The difficulty of design.(3) some important parameters of logic judgment algorithm depend on specific technique and appointed condition, as equipment is damaged
The casting condition change such as consumption, material replacing, steel grade and covering slag, algorithm is more and more inaccurate.(4) although bond when single thermoelectricity
Even temperature change over time " time lag " and " temperature inversion " feature of the multiple rows of multiple row electric thermo-couple temperature change in feature, space into
Common recognition, is the foundation of breakout prediction logical algorithm design, although had nothing in common with each other according to the logic judgment algorithm of these characteristic Designs,
But there is deficiency to forecast accuracy and considering for time of fire alarming.
In summary, how to overcome continuous casting breakout prediction method in the prior art and device forecast accuracy and alarm and
It is not enough present on when property, it is the technical barrier of urgent need to resolve in the prior art.
The content of the invention
1. the invention technical problem to be solved
It is an object of the invention to overcome continuous casting breakout prediction method in the prior art and device in forecast accuracy and report
Deficiency is realized to bonding there is provided the crystallizer bleedout prediction electric thermo system that a kind of logic-based judges present on alert promptness
The promptly and accurately forecast of bleed-out, it is ensured that continuous casting direct motion and raising slab quality.
2. technical scheme
To reach above-mentioned purpose, the technical scheme that the present invention is provided is:
The crystallizer bleedout prediction electric thermo system that the logic-based of the present invention judges, the steel leakage prediction system hardware composition part bag
Include fourplatemold and thermocouple, steel leakage prediction system computer, PLC, interchanger, thermocouple compensation cable, sound and light alarm
Device, relay, distant station, wherein, steel leakage prediction system computer includes client and server computer, client computer
For the monitoring display of bleed-out data, server computer is used for breakout prediction model calculation, breakout prediction model and client
Information between breakout prediction graphical interfaces is exchanged, and the compression of bleed-out data is stored;
Wherein, the design of the embedded columns and rows spacing of thermocouple meets following condition:
In formula, txAnd tyTo bond tearing port the time required to one thermocouple spacing of horizontal and vertical propagation;DxAnd DyPoint
Wei not the embedded row and column spacing of copper plate of crystallizer thermocouple;VxAnd VyRespectively bond the horizontal and vertical spread speed of tearing port;β
To bond the tear line and horizontal angle of breach, 20 °~50 ° are taken;
The steel leakage prediction system carries out crystallizer bleedout prediction electric thermo using following steps:
Step 1: from parameter configuration files read breakout prediction method in each logic judgment parameter of regularity;
Step 2: the collection of copper plate of crystallizer electric thermo-couple temperature time series data, storage and pretreatment;
Step 3: entering trip temperature respectively to all thermocouples on copper plate of crystallizer simultaneously rises inspection;
Step 4: the thermocouple for rising Rule of judgment to meeting temperature in step 3 carries out temperature rate-of-rise inspection;
Step 5:To 3 thermocouple TC (i, j-1), TC (i, j+1) and the TC around temperaturing lifting abnormality thermocouple TC (i, j)
(i+1, j) progress temperature change delay checks, and judges whether that making bonding gently alarms according to result;
Step 6:Meet thermocouple TC (i, j) constant temperature and decline inspection or TC (m, n) surrounding's electric thermo-couple temperature change
Change after delay inspection, then into step 7;
Step 7:Electric thermo-couple temperature, which is inverted, to be checked;
Step 8:According to alarm shielding condition, judge whether that make bonding steel leakage alarms again.
As further improvement of the present invention, each logic judgment parameter of regularity includes data prediction logic in step one
Judge that parameter, electric thermo-couple temperature rise and check that logic judgment parameter, electric thermo-couple temperature climbing speed check logic judgment parameter, heat
The delay of galvanic couple temperature change checks that logic judgment parameter, electric thermo-couple temperature decline and checks that logic judgment parameter, electric thermo-couple temperature are fallen
Inspection logic judgment parameter, bonding steel leakage is put gently to alarm and weigh alarming logic and judge parameter.
As further improvement of the present invention, in step 2, all electric thermo-couple temperatures in copper plate of crystallizer are gathered first
Time series data simultaneously stores numbering, the temperature of thermocouple of collection per second according to the predetermined row and column arrangement mode of thermocouple;
Then following data filtering and smooth pretreatment are carried out:
A) electric thermo-couple temperature is less than 50 DEG C or more than 200 DEG C, is not involved in algorithm calculating;
B) temperature fluctuation per second is more than 20 DEG C, and every temperature fluctuation in 2 seconds is more than 38 DEG C, is not involved in algorithm calculating;
C) temperature data for gathering thermocouple current time is made comparisons with the temperature data that the thermocouple previous second gathers,
Temperature data by temperature change more than 20% replaces with the temperature data of previous second collection;
D) time series data of each electric thermo-couple temperature is smoothed using the method for moving average, above-mentioned movement is put down
Equal method be use before and after current time one it is intersegmental every temperature data calculate average value substitution current time temperature value, if mobile interval
For 2k, k is positive integer, then the moving average at p momentFor:
In formula, T (p-k) is the electric thermo-couple temperature value at p-k moment.
As further improvement of the present invention, in step 3, electric thermo-couple temperature, which rises, checks as follows:When thermocouple is normal
During work, take rise inspection cycle period for a temperature within every 4 seconds, if the temperature of the 1st second is a, the temperature of the 2nd second is b, the 3rd
Second temperature is c, and the 4th second temperature is d;Every moment just checks first 3 seconds on a timeline and whether current time meets temperature
Degree rises Rule of judgment:d>c>b>a;If certain moment the i-th row j row thermocouple TC (i, j) is met, then by the thermocouple, this is followed
τ is labeled as in the ring cycle at the time of 1st second0(i, j), if can not meet, the temperature continued next second, which rises, to be checked;
Wherein:Said temperature rising Rule of judgment is met when some temperature rises proof cycle, then checks next temperature
Rise whether proof cycle still meets said temperature rising Rule of judgment, still meet the τ before then maintaining0(i, j), if
It is unsatisfactory for, the temperature continued next second, which rises, to be checked;
Temperature rising Rule of judgment is continuously unsatisfactory for if the temperature within 4 rises proof cycle, with latter
Temperature rises proof cycle and meets temperature rising Rule of judgment, the then τ before still maintaining again0(i, j);If 5 or 5
Individual temperatures above rises proof cycle and is continuously unsatisfactory for temperature rising Rule of judgment, the then τ before removing0(i, j), Zhi Houzai
When meeting temperature rising Rule of judgment, then a τ is re-established0(i, j), follow-up algorithm is with new τ0On the basis of (i, j).
As further improvement of the present invention, temperature rate-of-rise checks as follows in step 4:When there is τ0(i, j)
When, thermocouple TC (i, j) temperature rate-of-rise u=(d-a)/3 value is calculated, compares u and design temperature climbing speed threshold values uc
Size;If u is more than or equal to uc, then it is assumed that the thermocouple TC (i, j) temperature rise abnormity, while by τ0(i, j) saves as this
Electric thermo-couple temperature begins to ramp up moment t0(i, j), moment acquiescence is temperature change point at first, into step 5.
As further improvement of the present invention, in step 5, in t0Temperature change time delay threshold values after (i, j)
In the range of, to 3 thermocouple TC (i, j-1) of TC (i, j) surroundings, TC (i, j+1) and TC, (i+1, j) repeat the above steps three and step
Numeration is marked in rapid four logic judgment inspection, the thermocouple TC (m, n) for being satisfied by condition, and stores TC (m, n) temperature
Begin to ramp up moment t0(m, n), when mark count is more than 2, sends bonding steel leakage and gently alarms, enter back into step 6.
As further improvement of the present invention, in step 6, thermocouple TC (i, j) constant temperature, which declines, checks as follows:
In t0In 30 seconds after (i, j), temperature drop inspection is carried out to thermocouple TC (i, j), it every 4 seconds is a temperature drop inspection to take
The cycle is looked into, if the 1st second temperature is a', the 2nd second temperature is b', and the 3rd second temperature is c', and temperature is d' within the 4th second, such as meets d' ﹤ c' ﹤
B' ﹤ a', then judge that temperature drop occurs in thermocouple TC (i, j), if persistently meeting 3 temperature drop proof cycles, then it is assumed that
There is constant temperature decline in thermocouple TC (i, j), is directly entered step 7;
TC (m, n) surrounding's electric thermo-couple temperature change delay checks as follows:In t0To institute in 3~15 seconds after (m, n)
Having around TC (m, n) three thermocouple TC (m, n-1), TC (m, n+1) and TC, (m-1, n) repeat step three and step 4 are patrolled
Collect and judge to check, thermocouple TC (i, j) temperature change is postponed to check wherein excluding, the equally thermocouple TC to meeting condition
Numeration is marked in (x, y), and when marking count to be more than or equal to 3, thermocouple TC (i, j) is labeled as into doubtful bonding
Point, is directly entered step 7.
As further improvement of the present invention, in step 7, in t0In 30s after (i, j), by where TC (i, j)
Two row thermocouples enter trip temperature and fallen under three thermocouple TC (i, j-1) of row, TC (i, j) and TC (i, j+1) corresponding column
Logic judgment inspection is put, is specially:
Situation (1):First compare t0Before (i, j) in 10s, whether the mean temperature of i-th row some thermocouple is higher than same
The mean temperature of one of lower two rows, 2 thermocouples of row;
If it is, observation is in t0After (i, j) in 30s, the time-temperature curve of some thermocouple of i-th row
Intersect with whether the time-temperature curve of one of two rows, 2 thermocouples under the same column has on image, if intersecting
Then think temperature inversion occur, be directly entered step 8;
If be less than, compare t0The temperature for whether occurring i-th row some thermocouple after (i, j) in 30s is first higher than
It is less than one of two rows, 2 thermocouples under same column again afterwards, if there is, then it is assumed that occur in that temperature is inverted, be directly entered step
Rapid eight;
Situation (2):When the situation that is unsatisfactory for (1), first compare t010s before (i, j), certain row i+1 row thermocouple it is flat
Whether equal temperature is higher than the mean temperature of the row thermocouple of row i-th+2;
If it is, in t0After (i, j) in 30s, the time-temperature curve of certain row i+1 row thermocouple is observed with being somebody's turn to do
Whether the time-temperature curve of the row thermocouple of row i-th+2 has intersection on image, then thinks that temperature occur falls if intersecting
Put, be directly entered step 8;
If be less than, compare t0The temperature for certain row i+1 row thermocouple whether occur in 30s after (i, j) is first higher than
, when the situation described in appearance, appearance is still believed that after the row thermocouple of row i-th+2 less than the row thermocouple of row i-th+2 again
Temperature inversion, is directly entered step 8.
As further improvement of the present invention, following alarm shielding condition is provided with step 8, if be discontented with
Alarm shielding condition under being enough just is made bonding steel leakage and alarmed again:
1. 120s interior crystallizers liquid level exceeds predetermined minimum and maximum range before alarming, and system is not alarmed;
2. before alarming in 120s pulling rate less than it is default do not alarm pulling rate when, system is not alarmed;
3. before alarming in 120s pulling rate rate of change exceed it is default do not alarm pulling rate rate of change when, system is not alarmed;
4. when cast length be less than it is default do not alarm cast length when, system is not alarmed;
5. bonding steel leakage is alarmed again after in 60s, system is not alarmed;
6. when failure thermocouple number exceedes the threshold values of setting, system is not alarmed.
As further improvement of the present invention, in step 2, k is set to 2;In step 4, temperature rate-of-rise threshold values
ucIt is set to 0.15 DEG C/s;In step 5, t0Temperature change time delay threshold range after (i, j) is set to 3~15 seconds;
In step 8, mould liquid level predetermined minimum, maximum are respectively 80mm, 120mm;Default pulling rate of not alarming is 0.5m/
min;Default pulling rate rate of change of not alarming is 0.006m/min;Default cast length of not alarming is 3.0m;Failure thermoelectricity even number
The threshold values that mesh is set as thermocouple sum 1/3;
The steel leakage prediction system also includes:Data acquisition unit, bleed-out data compress automatically memory cell, self study and from
Line analysis unit, alarm and execution unit;
The data acquisition unit is used to obtain electric thermo-couple temperature real time data and casting cycle product practice, by counting
According to measurement, data transfer and the part of host computer three composition;
Wherein:DATA REASONING part includes sensor and A/D change-over circuits, and A/D change-over circuits are used to measure sensor
To analog signal be converted to data signal;TCP data segment uses field bus technique, for sending data to distal end
Host computer, wherein having from oracle database and from two kinds of acquisition data modes of PLC, is retrieved as by building from oracle database
Vertical continuous casting second databases are connected with the DB links of steel leakage prediction system database, are accessed using synonym;Number is obtained from PLC
According to directly to gather Siemens S7-400 PLC using the OPC of Siemens's host computer Configuration software WinCC communications, gather what is come up
Data are directly stored on the industrial computer for installing WinCC;
The bleed-out data compress memory cell automatically is used for the owning during real time execution by steel leakage prediction system
Data, are compressed according to certain file format by way of data flow, and each casting cycle terminates to automatically save one
File, each casting cycle starts with regard to being recorded automatically;
The self study and off-line analysis unit, can be carried out offline by off-line analysis module to breakout prediction model algorithm
Emulation, specially changes logic judgment parameter of regularity, bleed-out data is played back, pass through forecast accuracy, rate of false alarm, alarm
The performance indications of frequency carry out study optimization to breakout prediction model on-line parameter;Correspondence steel leakage prediction system is in on-line operation
When, each bonding steel leakage is alarmed again send after, site operation personnel is confirmed whether it is correct bleed-out result and is confirmed whether it is just
Really alarm, then system record automatically the correct alarm times of breakout prediction algorithm, error-alert degree, bonding steel leakage quote rate,
Forecast accuracy, rate of false alarm, the index of wrong report frequency, and quote rate, forecast accuracy and false alarm with the bonding steel leakage of setting
Frequency performance index threshold values is compared, once less than setting threshold values, system sends Parameter Self-learning request;Wherein, leakage is bonded
Steel quotes rate=correct alarm times/(failing to report number of times+correct alarm times), and forecast accuracy=correct alarm times/(is failed to report
Number of times+correct alarm times+error-alert degree), rate of false alarm=error-alert degree/(correct alarm times+error-alert degree), by mistake
Report frequency=error-alert degree/casting furnace number;
The alarm and execution unit, bleed-out alarm signal is sent by audible-visual annunciator, and light alarm is bright for amber light, reports again
Alert is when red, if attaching most importance to alarm, warning message uploads to the execution unit of casting speed control, is drawn by sending reduction to PLC
Speed or the instruction directly shut down;
The parameter configuration unit, for setting system operational parameters, including various logic judgment rule parameter, system
Username and password, and system communication connecting interface relevant parameter, and these parameters are stored in parameter configuration files,
Can change parameter in administrator right, and automatically update parameter configuration files, during steel leakage prediction system computing directly from
Put file call parameters.
3. beneficial effect
The technical scheme provided using the present invention, compared with prior art, with following remarkable result:
(1) the crystallizer bleedout prediction electric thermo method and system that logic-based proposed by the present invention judges is as base with bleed-out theory
Plinth, relies on actual bleed-out data and produce reality to be designed, including:1. the design combination of various logic rule, utilizes reality
Various temperature change features when border is bonded, take into full account balance, that is, when a certain Rule of judgment is more strict, other
Part needs suitably to relax, and finds equalization point between forecast accuracy and reduction rate of false alarm improving, and further contemplate alarm response
Time and ageing;2. the determination of breakout prediction algorithm and system core parameter, first by bleed-out data to raising speed in temperature
Rate threshold values, bonding V-type breach horizontal and vertical propagation time, V-type tear line angle, the judgement parameter of data filtering, model screen
The statistical analyses such as conditional parameter are covered, is continued to optimize then in conjunction with steel leakage prediction system operation result, makes overall system performance
Reach optimal.
(2) present invention is on the premise of ensureing that bonding steel leakage is all quoted, in order to reduce false alarm, carries out from many aspects
Consider, such as:1. when logic judgment algorithm is designed, take data filtering and smoothing processing, temperature rise inspection, temperature and risen
Speed is checked, temperature change delay twice is checked, temperature drop inspection and temperature are inverted and checked;2. to unsteady statecalculation, thermoelectricity
Situations such as even number of defects is more adds alarm shielding condition.
(3) present invention is in order to reduce the change of casting process and appointed condition etc. to breakout prediction algorithm and the shadow of system
Ring, devise the function of systematic parameter self study, changed with the production law for adapting to actual complex, so as to improve breakout prediction system
The adaptivity and stability of system.
(4) the crystallizer bleedout prediction electric thermo method and system that logic-based proposed by the present invention judges, meets produce reality, reports
Alert fast response time, can send alarm bonding in V-type tearing port propagation 30s, not fail to report, forecast accuracy is high, false alarm
Number of times is substantially reduced, and with preferable adaptivity and stability.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be attached to what is used required in embodiment
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, therefore is not construed as pair
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, can also be according to this
A little accompanying drawings obtain other related accompanying drawings.
Fig. 1 is the flow chart of the crystallizer bleedout prediction electric thermo method of logic-based judgement in embodiment 1;
The thermocouple schematic diagram that Fig. 2 respectively walks for breakout prediction logical check on copper plate of crystallizer in embodiment 1;
Fig. 3 is the structural representation of crystallizer bleedout prediction electric thermo system in embodiment 1;
Fig. 4 is affixed side E row electric thermo-couple temperature change curves in embodiment 1;
Fig. 5 is affixed side F row electric thermo-couple temperature change curves in embodiment 1;
Fig. 6 is affixed side G row electric thermo-couple temperature change curves in embodiment 1;
Fig. 7 is affixed side H row electric thermo-couple temperature change curves in embodiment 1;
Fig. 8 is affixed side I row electric thermo-couple temperature change curves in embodiment 1.
Embodiment
To make the purpose, technical scheme and advantage of the embodiment of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that described embodiment is
A part of embodiment of the present invention, rather than whole embodiments.Therefore, embodiments of the invention below to providing in the accompanying drawings
Detailed description be not intended to limit the scope of claimed invention, but be merely representative of the present invention selected embodiment.
Based on the embodiment in the present invention, it is all that those of ordinary skill in the art are obtained under the premise of creative work is not made
Other embodiment, belongs to the scope of protection of the invention.
To further appreciate that present disclosure, the present invention is described in detail in conjunction with the accompanying drawings and embodiments.
Embodiment 1
As shown in figure 1, the crystallizer bleedout prediction electric thermo method that the logic-based of the present embodiment judges, comprises the following steps:
Step 1: from parameter configuration files read breakout prediction method in each logic judgment parameter of regularity, including data are pre-
Processing logic judgment parameter, electric thermo-couple temperature, which rise, checks that logic judgment parameter, electric thermo-couple temperature climbing speed check that logic is sentenced
Disconnected parameter, electric thermo-couple temperature change delay check that logic judgment parameter, electric thermo-couple temperature decline and check logic judgment parameter, thermoelectricity
Even temperature, which is inverted, checks that logic judgment parameter, bonding steel leakage are gently alarmed and weighed alarming logic and judge parameter.
Step 2: the collection of copper plate of crystallizer electric thermo-couple temperature time series data, storage and pretreatment.
All electric thermo-couple temperature time series datas and arranged first in collection copper plate of crystallizer according to the predetermined row and column of thermocouple
Row mode is stored in numbering, the present embodiment, the thermocouple line number buried in copper plate of crystallizer is designated as 1 successively from top to bottom, 2,
3rd ..., row number by it is left-to-right be designated as successively A, B, C ..., such as, 3B represents the 3rd row B row thermocouples.
The temperature of thermocouple of collection per second, then carries out following data filtering and smooth pretreatment, to reduce mistake
The influence of data and data fluctuations to breakout prediction method, specific as follows by mistake:
A) electric thermo-couple temperature is less than 50 DEG C or more than 200 DEG C, is not involved in algorithm calculating.
B) temperature fluctuation per second is more than 20 DEG C, and every temperature fluctuation in 2 seconds is more than 38 DEG C, is not involved in algorithm calculating.
C) temperature data for gathering thermocouple current time is made comparisons with the temperature data that the thermocouple previous second gathers,
Temperature data by temperature change more than 20% replaces with the temperature data of previous second collection.
D) time series data of each electric thermo-couple temperature is smoothed using the method for moving average, is ensureing not lose
While removing former electric thermo-couple temperature variation characteristic, reduce influence of the thermo-electric couple temperature data fluctuation to model.
Here the method for moving average be use before and after current time one it is intersegmental every temperature data to calculate average value substitution current
Moment temperature value, if mobile at intervals of 2k, k is positive integer, then the moving average at p momentFor:
In formula, T (p-k) is the electric thermo-couple temperature value at p-k moment.K is set to 2 in the present invention.
Step 3: entering trip temperature respectively to all thermocouples on copper plate of crystallizer simultaneously rises inspection.All single heat
Galvanic couple temperature, which rises, checks as follows:
When thermocouple normal work, take rise inspection cycle period for a temperature within every 4 seconds, if the temperature of the 1st second is
A, the temperature of the 2nd second is b, and the 3rd second temperature is c, and the 4th second temperature is d;Every moment just checks first 3 seconds on a timeline
Temperature, which whether is met, with current time rises Rule of judgment:d>c>b>a;If certain moment the i-th row j row thermocouple TC (i, j) is full
Foot, then be labeled as τ at the time of by the 1st second in the thermocouple this cycle period0(i, j), if can not meet, continues next second
Temperature rise check.
Said temperature rising Rule of judgment is met when some temperature rises proof cycle, then checks next temperature and rises inspection
Look into whether the cycle still meets said temperature rising Rule of judgment, still meet the τ before then maintaining0(i, j), if discontented
Foot, the temperature continued next second, which rises, to be checked.
(1) temperature rising Rule of judgment is continuously unsatisfactory for if the temperature within 4 rises proof cycle, with latter
Individual temperature rises proof cycle and meets temperature rising Rule of judgment, the then τ before still maintaining again0(i, j);(2) if 5
Individual or 5 temperatures above rise proof cycle and are continuously unsatisfactory for temperature rising Rule of judgment, the then τ before removing0(i, j),
When meeting temperature rising Rule of judgment again afterwards, then a τ is re-established0(i, j), follow-up algorithm is with new τ0(i, j) is base
It is accurate.
Step 4: the thermocouple for rising Rule of judgment to meeting temperature in step 3 carries out temperature rate-of-rise inspection.Tool
Body is as follows:
When there is τ0When (i, j), thermocouple TC (i, j) temperature rate-of-rise u=(d-a)/3 value is calculated, compares u with setting
Determine temperature rate-of-rise threshold values ucSize.If u is more than or equal to uc, then it is assumed that the thermocouple TC (i, j) temperature rise abnormity,
Simultaneously by τ0(i, j) saves as the electric thermo-couple temperature and begins to ramp up moment t0(i, j), moment acquiescence is temperature change point at first,
Into step 5.Wherein, temperature rate-of-rise threshold values u in the present embodimentcIt is set to 0.15 DEG C/s.
Step 5:As shown in Fig. 2 to 3 thermocouple TC (i, j-1), TC around temperaturing lifting abnormality thermocouple TC (i, j)
(i+1, j) progress temperature change delay checks, and judges whether that making bonding gently alarms according to result by (i, j+1) and TC.Specifically
For:
In t0(it is set in temperature change time delay threshold range after (i, j) in the present embodiment in 3~15 seconds),
To 3 thermocouple TC (i, j-1) of TC (i, j) surroundings, TC (i, j+1) and TC (i+1, j) repeat step three and step 4 logics
When judging to check, numeration is marked in the thermocouple TC (m, n) for being satisfied by condition, and storing TC (m, n) temperature and begin to ramp up
Carve t0(m, n), when mark count is more than 2, sends bonding steel leakage and gently alarms, into step 6.
Step 6:Meet thermocouple TC (i, j) constant temperature and decline inspection or TC (m, n) surrounding's electric thermo-couple temperature change
Change after delay inspection, then into step 7;
(1) thermocouple TC (i, j) constant temperature, which declines, checks as follows:
In t0In 30 seconds after (i, j), temperature drop inspection is carried out to thermocouple TC (i, j), it every 4 seconds is a temperature to take
Degree declines proof cycle, if the 1st second temperature is a', and the 2nd second temperature is b', and the 3rd second temperature is c', and the 4th second temperature is d', such as full
Sufficient d' ﹤ c' ﹤ b' ﹤ a', then judge that temperature drop occurs in thermocouple TC (i, j), if persistently meeting 3 temperature drops checks week
Phase, then it is assumed that constant temperature decline occurs in thermocouple TC (i, j), is directly entered step 7.
(2) as shown in Fig. 2 TC (m, n) surrounding's electric thermo-couple temperature change delay checks following (wherein, exclude to thermoelectricity
The temperature change delay of even TC (i, j) is checked):
In t0To three thermocouple TC (m, n-1), TC (m, n around all TC (m, n) in 3~15 seconds after (m, n)
+ 1) and TC (m-1, n) repeat step three and step 4 logic judgment inspection, equally the thermocouple TC (x, y) to meeting condition
Numeration is marked, when marking count to be more than or equal to 3, thermocouple TC (i, j) doubtful bonding point is labeled as, directly
Into step 7.
Step 7:Electric thermo-couple temperature, which is inverted, to be checked.
In t0In 30s after (i, j), TC (i, j) is expert at three thermocouple TC (i, j-1), TC (i, j) and TC
Two row thermocouples enter trip temperature inversion logic judgment inspection under (i, j+1) corresponding column, i.e., for example for TC (i, j),
Then by two row thermocouple TC under the corresponding columns of TC (i, j), (i+1, j) (i+2 j) enters trip temperature and is inverted logic judgment with TC
Check;Specially:
Situation (1):First compare t0Before (i, j) in 10s, the i-th row some thermocouple (i.e. thermocouple TC (i, j-1), TC
Any one in (i, j) and TC (i, j+1)) mean temperature whether be higher than one of two rows, 2 thermocouples under same column
Mean temperature.
If it is, just comparing t again0Whether occurs temperature inversion therewith in 30s after (i, j), i.e., in t0After (i, j)
In 30s, one of two rows, 2 thermocouples under the time-temperature curve and the same column of some thermocouple of i-th row
(" one of them " selection herein is in t0Before (i, j) in 10s, mean temperature is less than described some thermocouple of i-th row
That) time-temperature curve whether have intersection on image, if intersect then think temperature inversion occur, directly enter
Enter step 8;
If be less than, compare t0Whether there is some thermocouple of the i-th row (i.e. thermocouple TC in 30s after (i, j)
Any one in (i, j-1), TC (i, j) and TC (i, j+1)) temperature first higher than rear again less than two rows, 2 thermoelectricity under same column
It is one of even, i.e., for example for TC (i, j), its t0Temperature after (i, j) in 30s is first higher than rear again less than two rows under same column
2 thermocouple TC (i+1, j) and TC (i+2, one of j), if there is, then it is assumed that occur in that temperature is inverted, be directly entered
Step 8.
Situation (2):When the situation that is unsatisfactory for (1), just compare the thermocouple of certain row i+1 row and the heat of the row of row i-th+2
Whether galvanic couple there is temperature inversion, i.e., first compare t0Whether 10s before (i, j), the mean temperature of certain row i+1 row thermocouple is high
In the mean temperature of the row thermocouple of row i-th+2.
If it is, just comparing t again0Whether occurs temperature inversion therewith in 30s after (i, j), i.e., in t0After (i, j)
In 30s, the time-temperature curve of the time-temperature curve of certain row i+1 row thermocouple and the row thermocouple of row i-th+2 is being schemed
Whether there is intersection as upper, then think temperature inversion occur if intersecting, be directly entered step 8;
If be less than, compare t0The temperature for certain row i+1 row thermocouple whether occur in 30s after (i, j) is first higher than
, when the situation described in appearance, appearance is still believed that after the row thermocouple of row i-th+2 less than the row thermocouple of row i-th+2 again
Temperature inversion, is directly entered step 8.
As long as said temperature, which is inverted in logical check, 1 temperature inversion occurs, just it is directly entered step 8 and prepares to send viscous
Junction leakage steel is alarmed again.Temperature inversion is that the important change of cohesive tearing port communication process electric thermo-couple temperature is special in theory
Levy, whether occur so there are many breakout prediction algorithms to be inverted temperature, and whether the inverted difference of temperature reaches setting value
As the necessary condition for judging bonding steel leakage, but influenceed by the condition of continuous casting actual complex, this temperature phenomena of inversion is simultaneously
It is not regular appearance, the present invention has found for substantial amounts of actual bleed-out data analysis, is bonding in the 30s occurred, certain row
Other row thermocouples are difficult temperature inversion occur with the first row thermocouple, and in actual continuous casting, it may appear that first
The temperature of row thermocouple is lower than the phenomenon of other row thermocouples of the row before generation is bonded.In order to reduce these as far as possible
Interference, and reduce dependence of the temperature inversion for breakout prediction algorithm, temperature is inverted by the breakout prediction method of the present embodiment
As one of final condition, and condition is very loose, and this is also the place that this method is different from other breakout prediction methods.
Step 8:According to alarm shielding condition, judge whether that make bonding steel leakage alarms again.
In order to reduce wrong report, there is provided following alarm shielding condition, if being unsatisfactory for following alarm shielding condition
Bonding steel leakage is just made to alarm again:
1. 120s interior crystallizers liquid level exceeds predetermined minimum (present embodiment is set to 80mm) and maximum before alarming
(present embodiment is set to 120mm) scope, system is not alarmed;
2. before alarming in 120s pulling rate less than it is default do not alarm pulling rate (present embodiment is set to 0.5m/min) when, system
Do not alarm;
3. pulling rate rate of change exceedes and default does not alarm that (present embodiment is set to pulling rate rate of change in 120s before alarming
When 0.006m/min), system is not alarmed;
4. open and pour stage judgement:When cast length is less than default cast length of not alarming (present embodiment is set to 3.0m)
When, system is not alarmed;
5. bonding steel leakage is alarmed again after in 60s, system is not alarmed.
6. when failure thermocouple number exceedes threshold values (present embodiment is set to the 1/3 of thermocouple sum) of setting, it is
System judges without bleed-out, does not alarm.
When sending bonding gently alarm, it is necessary to which site operation personnel carries out benefit slag, drags for the operation such as slag in time, and observe bleed-out
Thermocouple temperature variation curve in forecast system operation interface, when send bond alarm again when, take drop pulling rate arrange
Apply and avoid bleed-out.
Corresponding with the above method, the present embodiment also provides the crystallizer bleedout prediction electric thermo system that a kind of logic-based judges,
As shown in figure 3, specifically including:
Crystallizer bleedout prediction electric thermo system hardware part includes:Modular panel billet crystallizer, totally four pieces of copper coins, conticaster
The two pieces wide face copper coin of outer arc affixed side, two pieces of narrow copper plates of inner arc active side, crystallizer copper plate thickness is 41mm, copper coin surface
Comprising 1mm nickel plating layer thicks, copper coin material is CuCrZr alloys.High accuracy K-type armouring heat is buried in the system copper plate of crystallizer
Galvanic couple, thermocouple is mounted the copper coin surface temperature that positioning is contacted to detect with molten steel, and measuring accuracy is up to 0.1 DEG C, measurement
The form of thermocouple is using measurement point and compensating wire distributing installation, and thermocouple is using initial tension of spring crimping and clip fixation side
Formula is connected, outside also to use rubber seal, prevents thermocouple from seeping water.Thermocouple insertion depth is 20.5mm on wide face copper coin, narrow
Thermocouple insertion depth is 20mm on the copper coin of face.Thermocouple temperature survey and its accurate stable are steel leakage prediction systems
The premise and key of work, especially high casting speed continuous casting process.
Meanwhile, the thermocouple arrangement of the steel leakage prediction system is using high density thermocouple arrangement, wide face copper coin cloth
Put m rows n1Row, narrow copper plate arrangement m rows n2Row, leptoprosopy and wide face are embedded highly identical per a line thermocouple, and the first row thermoelectricity
Occasionally it is arranged in away from below meniscus 50mm positions, to reduce liquid fluctuating and slag circle etc. to the first row thermocouple temperature measurement accuracy
With the influence of temperature fluctuation.Wherein, m >=3, n1Take the maximum integer less than (strand width/thermocouple buries column pitch+1), n2
Take the maximum integer less than (slab thickness/thermocouple buries column pitch+1).The present invention on the embedded row of thermocouple and column and row with
The design of capable spacing, in order to improve the accuracy rate of breakout prediction, and recognizes that the behavior of bonding is alarmed, is that this should make as soon as possible
Cohesive V-type tearing port matches in the columns and rows spacing that vertically and horizontally spread speed is buried with thermocouple, that is, ensures to bond V
Type breach is approximately equivalent in the time of one thermometric spacing of horizontal and vertical propagation.And from substantial amounts of bonding steel leakage green shell and thermoelectricity
Knowable to even temperature data, bond breach and laterally propagate substantially faster than longitudinal propagation, so the embedded column pitch of thermocouple must be than row
Spacing is big.The design that the present invention buries columns and rows spacing particularly with regard to thermocouple must is fulfilled for following condition:
In formula, txAnd tyTo bond tearing port the time required to one thermocouple spacing of horizontal and vertical propagation;DxAnd DyPoint
Wei not the embedded row and column spacing of copper plate of crystallizer thermocouple;VxAnd VyRespectively bond the horizontal and vertical spread speed of tearing port;β
To bond the tear line and horizontal angle of breach, typically in 20 °~50 ° scopes.
In the present embodiment, on Thermocouple arrangement, the first row thermocouple is arranged in away from meniscus
70mm positions, while determine that β is averagely about 35 ° according to the measurement of bleed-out green shell and related bleed-out data statistics, then in order to ensure to glue
Tie V-type tearing port equal the time required to one thermocouple spacing of horizontal and vertical propagation, thermocouple line space is designed as
126mm, column pitch is 180mm, and two wide face copper coin is respectively arranged 5 rows 12 and arranged, and two narrow copper plates are respectively arranged 5 rows 2 and arranged, altogether
140 thermocouples.
The crystallizer bleedout prediction electric thermo system hardware part that the present invention is provided also include fourplatemold and thermocouple,
Steel leakage prediction system computer, PLC, interchanger, thermocouple compensation cable, audible-visual annunciator, relay, distant station etc..Wherein,
Steel leakage prediction system computer includes client and server computer, and the monitoring that client computer is used for bleed-out data shows
Show, any data are not preserved, and server computer is used for core breakout prediction model calculation, breakout prediction model and client
Enter the exchange of row information and data between steel leakage prediction system graphical interfaces, and the compression of bleed-out data is stored.PLC is on the one hand
Real-time data collection, passes to steel leakage prediction system, and another aspect logic judgment algorithm sends alarm signal to PLC, and PLC is sent
Pulling rate execute instruction is dropped, bleed-out is prevented.
The crystallizer bleedout prediction electric thermo system architecture that the present invention is provided includes:
Data acquisition unit, for obtaining electric thermo-couple temperature real time data and casting cycle product practice, by data
Measurement, data transfer and the part of host computer three composition.DATA REASONING part is mainly made up of sensor and A/D change-over circuits etc.,
Effect is that the analog signal for measuring sensor is converted to data signal, the temperature of such as thermocouple measurement.Data transfer part
Divide and use field bus technique, effect is to send data to distal end host computer.Epigynous computer section is mainly software data interface
Realization, it is contemplated that the existing data for coming from PLC of the system fetched data, also have the data for coming from continuous casting database,
So having from oracle database and from two kinds of acquisition data modes of PLC.Obtained from oracle database, by setting up continuous casting two
DBMS storehouse is connected with the DB links of steel leakage prediction system database, is accessed using synonym;Data are obtained from PLC, are utilized
The OPC communications of the advanced host computer Configuration software WinCC of Siemens directly gather Siemens S7-400 PLC, gather the number come up
On the industrial computer for installing WinCC according to being directly stored in, steel leakage prediction system can just be directly read.In data acquisition,
In order to eliminate interference of the site environment to signal, the measures such as compensating cable and digital filtering are taken, reduce the influence of noise.
Bleed-out data compress memory cell automatically, for all numbers by steel leakage prediction system during real time execution
According to being compressed according to certain file format by way of data flow, each casting cycle terminates to automatically save a text
Part, each casting cycle starts with regard to being recorded automatically, and the data volume of system operation over a year is left in 50G after compression
The right side, so 200G hard drive spaces can just preserve 4 annual datas without cleaning.
The breakout prediction model algorithm unit that logic-based judges, according to the electric thermo-couple temperature, pulling rate and crystallization received
The casting cycle real time data such as device liquid level, and read each logic judgment parameter of regularity in systematic parameter configuration file, Ran Houyun
Light alarm and the weight alarm decision of bonding steel leakage are carried out with model algorithm.
Steel leakage prediction system self study and off-line analysis unit, on the one hand can be used for the weight of history heat process data
Put, history electric thermo-couple temperature, pulling rate and mould liquid level etc. can be checked, continuous casting process personnel can be to casting by history playback
Journey crash analysis, finds out cause of accident, optimizes casting process parameter.Prior one side also may be used by off-line analysis module
Off-line simulation is carried out to the breakout prediction model algorithm of core, specially changes logic judgment parameter of regularity, bleed-out data is entered
Row is played back, and now steel leakage prediction system passes through forecast accuracy, rate of false alarm, alert frequency etc. using the parameter computing just set
Performance indications compare just constantly can carry out study optimization to breakout prediction model on-line parameter.Meanwhile, correspondence breakout prediction system
System in on-line operation, each bonding steel leakage is alarmed again send after, site operation personnel is according to whether bleed-out result confirms in time
Whether it is correct alarm, then system records the correct alarm times of breakout prediction algorithm, error-alert degree, bonding steel leakage automatically
The performance indications such as rate, forecast accuracy, rate of false alarm, wrong report frequency are quoted, and rate (this embodiment party is quoted with the bonding steel leakage of setting
Taken in formula 100%), forecast accuracy (85% is taken in present embodiment) and false alarm frequency (0.15% is taken in present embodiment)
Important performance indexes threshold values is compared, once less than setting threshold values, system can send Parameter Self-learning request.Glued in the present invention
Junction leakage steel quotes rate=correct alarm times/(failing to report number of times+correct alarm times), and forecast accuracy=correct alarm times/
(failing to report number of times+correct alarm times+error-alert degree), rate of false alarm=error-alert degree/(correct alarm times+false alarm
Number), report frequency=error-alert degree/casting furnace number by mistake.
Steel leakage prediction system is alarmed and execution unit, and bleed-out alarm signal is sent by audible-visual annunciator, and light alarm is Huang
Lamp is bright, and alarm again is when red, if attaching most importance to alarm, warning message can upload to the execution unit of casting speed control, by PLC
Send reduction pulling rate or the instruction directly shut down.Simultaneously all warning messages can all be transferred to system man-machine interface and show, and
Preserve that steel leakage prediction system sends and these warning messages can be also saved in log file, in alarm logging file all goes through
The details such as alarm moment, type of alarm, the alert locations of history, are easy to follow-up accident to search and statistical analysis.
Steel leakage prediction system parameter configuration unit, for setting system operational parameters, including various logic judgment rule ginseng
Number, username and password of system etc., and system communication connecting interface relevant parameter, and these parameters are stored in parameter
In configuration file, parameter can be changed in administrator right, and automatically update parameter configuration files, steel leakage prediction system computing
When directly from configuration file call parameters.
The 3rd stove molten steel pours in slab caster in the present embodiment, company, and casting steel grade is SS400, casting section for 1550 ×
230mm2, work pulling rate is 1.2m/min, peaceful by data filtering to the Thermocouple temperature data collected first
Sliding pretreatment, pretreated typical row electric thermo-couple temperature change curve is as shown in Fig. 4-Fig. 8, then, during according to these temperature
Ordinal number is according to the logic judgment inspection for carrying out bonding steel leakage.
(1) electric thermo-couple temperature, which rises, checks.
Affixed side 1G thermocouples are 05:03:53 moment began to ramp up, and rose in this temperature and checked in circulation, a=
99.48 DEG C, b=99.68 DEG C, c=99.92 DEG C, d=100.24 DEG C, meet d>c>b>A, then record τ0(1, G)=05:03:
53。
(2) electric thermo-couple temperature climbing speed is checked
Calculate affixed side 1G electric thermo-couple temperature climbing speed u=(100.24-99.48)/3=0.2533 DEG C/s>0.15℃/
S, then it is assumed that affixed side 1G electric thermo-couple temperature rise abnormities, while recording the electric thermo-couple temperature begins to ramp up moment t0(1, G)=
05:03:53。
(3) to 05:03:1F, 1H and 2G thermocouple after 53 in 3~15s around affixed side 1G thermocouples enters trip temperature
Rise abnormity is checked.
05:04:04 moment affixed side 1F thermocouple is begun to ramp up, and is risen in this temperature and is checked in circulation, a=
102.58 DEG C, b=102.76 DEG C, c=103.00 DEG C, d=103.26 DEG C, meet d>c>b>A, then record τ0(1, F)=05:
04:04.Moreover, affixed side 1F electric thermo-couple temperature climbing speed u=(103.26-102.58)/3=0.2267 DEG C/s>0.15℃/
S, then it is assumed that affixed side 1F electric thermo-couple temperature rise abnormities, while recording the electric thermo-couple temperature begins to ramp up moment t0(1, F)=
05:04:04。
05:03:56 moment affixed side 1H thermocouples are begun to ramp up, and are risen in this temperature and are checked in circulation, a=
94.58 DEG C, b=94.86 DEG C, c=95.66 DEG C, d=96.54 DEG C, meet d>c>b>A, then record τ0(1, H)=05:03:56.
Moreover, affixed side 1H electric thermo-couple temperature climbing speed u=(96.54-94.58)/3=0.6533 DEG C/s>0.15 DEG C/s, then it is assumed that
Affixed side 1H electric thermo-couple temperature rise abnormities, while recording the electric thermo-couple temperature begins to ramp up moment t0(1, H)=05:03:56.
05:04:05 moment affixed side 2G thermocouple is begun to ramp up, and is risen in this temperature and is checked in circulation, a=
92.80 DEG C, b=93.02 DEG C, c=93.28 DEG C, d=93.58 DEG C, meet d>c>b>A, then record τ0(2, G)=05:04:05.
Moreover, affixed side 2G electric thermo-couple temperature climbing speed u=(93.58-92.80)/3=0.26 DEG C/s>0.15 DEG C/s, then it is assumed that Gu
Side 2G electric thermo-couple temperature rise abnormities are determined, while recording the electric thermo-couple temperature begins to ramp up moment t0(2, G)=05:04:05.
Gently alarmed from the foregoing, it will be observed that temperature rise abnormity thermocouple quantity sends bonding steel leakage for 3.
(4) surrounding's thermoelectricity of the inspection of affixed side 1G electric thermo-couple temperatures continuous decrease or affixed side 1F, 1H and 2G thermocouple
Even temperature change delay is checked.
1) affixed side 1G electric thermo-couple temperatures continuous decrease is checked:
From 05:03:53 moment start after 30 seconds in, if the 1st second temperature is a', the 2nd second temperature is b', the 3rd second temperature
For c', temperature is d' within the 4th second, occurs in that a'=102.24 DEG C, b'=102.20 DEG C, c'=102.12 DEG C, d'=101.98 DEG C,
D' ﹤ c' ﹤ b' ﹤ a' are met, and 05:04:12 moment continued 3 circulations, then 05:04:12 moment judged affixed side 1G heat
There is continuous decrease in galvanic couple temperature, is directly entered " (5) electric thermo-couple temperature is inverted logical check ".
2) surrounding's electric thermo-couple temperature change delay of affixed side 1F, 1H and 2G thermocouples is checked, including:
Wherein, with reference to Fig. 2, although surrounding's thermocouple of affixed side 1F, 1H and 2G thermocouple has 9 in theory, but excludes
Affixed side 1G thermocouples, and (i.e. affixed side 2F, 2H thermocouple is only carried out affixed side 2F, the 2H thermocouple of repeat count respectively
Temperature change delay is checked), the thermocouple that actual progress temperature change delay is checked only has 5, and cross is used in fig. 2
Lines have carried out filling mark):
1. to 05:04:1E and 2F thermocouples after 04 in 3~15s around affixed side 1F thermocouples enter trip temperature rising
Abnormal examination.
05:04:11 moment affixed side 1E thermocouples are begun to ramp up, and are risen in this temperature and are checked in circulation, a=
92.52 DEG C, b=92.68 DEG C, c=92.84 DEG C, d=93.02 DEG C, meet d>c>b>A, then record τ0(1, E)=05:04:11.
Moreover, affixed side 1E electric thermo-couple temperature climbing speed u=(93.02-92.52)/3=0.1667 DEG C/s>0.15 DEG C/s, then it is assumed that
Affixed side 1E electric thermo-couple temperature rise abnormities, while recording the electric thermo-couple temperature begins to ramp up moment t0(1, E)=05:04:11.
05:04:18 moment affixed side 2F thermocouples are begun to ramp up, and are risen in this temperature and are checked in circulation, a=
91.66 DEG C, b=91.80 DEG C, c=91.96 DEG C, d=92.14 DEG C, meet d>c>b>A, then record τ0(1, E)=05:04:18.
Moreover, affixed side 2F electric thermo-couple temperature climbing speed u=(92.14-91.66)/3=0.16 DEG C/s>0.15 DEG C/s, then it is assumed that Gu
Side 2F electric thermo-couple temperature rise abnormities are determined, while recording the electric thermo-couple temperature begins to ramp up moment t0(2, F)=05:04:18.
2. to 05:03:1I and 2H thermocouples after 56 in 3~15s around affixed side 1H thermocouples enter trip temperature rising
Abnormal examination.
05:04:06 moment affixed side 1I thermocouple is begun to ramp up, and is risen in this temperature and is checked in circulation, a=
99.62 DEG C, b=99.74 DEG C, c=99.94 DEG C, d=100.26 DEG C, meet d>c>b>A, then record τ0(1, I)=05:04:
06.Moreover, affixed side 1I electric thermo-couple temperature climbing speed u=(100.26-99.62)/3=0.2133 DEG C/s>0.15 DEG C/s, then
Affixed side 1I electric thermo-couple temperature rise abnormities are thought, while recording the electric thermo-couple temperature begins to ramp up moment t0(1, I)=05:
04:06。
05:04:04 moment affixed side 2H thermocouple is begun to ramp up, and is risen in this temperature and is checked in circulation, a=
93.20 DEG C, b=93.32 DEG C, c=93.48 DEG C, d=93.68 DEG C, meet d>c>b>A, then record τ0(2, H)=05:04:04.
Moreover, affixed side 2H electric thermo-couple temperature climbing speed u=(93.68-93.20)/3=0.16 DEG C/s>0.15 DEG C/s, then it is assumed that Gu
Side 2H electric thermo-couple temperature rise abnormities are determined, while recording the electric thermo-couple temperature begins to ramp up moment t0(2, H)=05:04:04.
3. to 05:04:3G thermocouples after 05 in 3~15s around affixed side 2G thermocouples enter trip temperature rise abnormity
Check.
05:04:05 moment affixed side 3G thermocouple is begun to ramp up, and is risen in this temperature and is checked in circulation, a=
94.14 DEG C, b=94.28 DEG C, c=94.48 DEG C, d=94.70 DEG C, meet d>c>b>A, moreover, affixed side 3G electric thermo-couple temperatures
Climbing speed u=(94.70-94.14)/3=0.28 DEG C/s>0.15 DEG C/s, but the electric thermo-couple temperature begins to ramp up the moment not
05:04:After 05 in 3~15s, so affixed side 3G thermocouples are not counted in the abnormal thermocouple of sticking temperature.
It can thus be concluded that, there are 4 thermocouples to meet temperature change delay around affixed side 1F, 1H and 2G thermocouple and check bar
Affixed side thermocouple 1G is then labeled as doubtful bonding point by part, quantity more than 3, into " (5) electric thermo-couple temperature is inverted logic
Check ".
It should be noted that above-mentioned " 1) affixed side 1G electric thermo-couple temperatures continuous decrease is checked ", " 2) affixed side 1F, 1H and
The surrounding electric thermo-couple temperature change delay of 2G thermocouples is checked " if meet wherein any one be then directly entered " (5) thermocouple
Temperature is inverted logical check ", in the present embodiment in order to become apparent from, comprehensively introduce whole breakout prediction method, be described.
(5) electric thermo-couple temperature is inverted logical check
With reference to Fig. 7,05:03:10s before 53 moment, (92.78 DEG C) of affixed side 2H thermocouples mean temperature, which is less than, fixes
Side 1H thermocouples mean temperature (94.37 DEG C).05:03:After 53 moment in 30s, 05:04:13 moment affixed side 2H thermoelectricity
It is even temperature inversion first occur with 1H thermocouples, send bonding and alarm again, used time 20s, than the actual bleed-out moment (05:04:40)
It is advanced by about 27s.
In order to further verify the value of forecasting of the present invention, the 1569 furnace electric thermo-couple temperature of one month of collection in worksite are utilized
Deng casting cycle data, the crystallizer bleedout prediction electric thermo method and system that the logic-based that the present embodiment is proposed judges are surveyed
Examination, and contrasted with live steel leakage prediction system.Test obtained result as shown in table 1.
The present invention of table 1 and the comparison of live steel leakage prediction system
It was found from the test result of table 1, it is pre- that the every evaluation index of the steel leakage prediction system of the present embodiment is better than live bleed-out
Reporting system.Live steel leakage prediction system, failing to report once to bond causes bleed-out, and wrong report number of times reaches 20 times, the alarm used time 50s with
Interior, alarm response speed is slower, adds the risk that follow-up drop pulling rate repairs bleed-out.The breakout prediction method of the present embodiment and it is
System is reported by mistake 1 time without failing to report, and wrong report number of times is substantially reduced, and bonding steel leakage quotes rate and reaches 100%, and forecast accuracy reaches
88.89%, rate of false alarm is 11.11%, and false alarm frequency is only 0.06% time/stove, and the alarm used time of all bonding steel leakages
All within 30s, alarm response speed is fast, and repairing bleed-out for follow-up drop pulling rate provides time enough.
Schematical above that the present invention and embodiments thereof are described, the description does not have restricted, institute in accompanying drawing
What is shown is also one of embodiments of the present invention, and actual structure is not limited thereto.So, if the common skill of this area
Art personnel are enlightened by it, without departing from the spirit of the invention, are designed and the technical scheme without creative
Similar frame mode and embodiment, all should belong to protection scope of the present invention.
Claims (10)
1. the crystallizer bleedout prediction electric thermo system that a kind of logic-based judges, it is characterised in that the steel leakage prediction system hardware is constituted
Part includes fourplatemold and thermocouple, steel leakage prediction system computer, PLC, interchanger, thermocouple compensation cable, sound
Light crossing-signal, relay, distant station, wherein, steel leakage prediction system computer includes client and server computer, client
Computer be used for bleed-out data monitoring display, server computer be used for breakout prediction model calculation, breakout prediction model and
Information between client breakout prediction graphical interfaces is exchanged, and the compression of bleed-out data is stored;
Wherein, the design of the embedded columns and rows spacing of thermocouple meets following condition:
<mrow>
<mfrac>
<msub>
<mi>t</mi>
<mi>x</mi>
</msub>
<msub>
<mi>t</mi>
<mi>y</mi>
</msub>
</mfrac>
<mo>=</mo>
<mfrac>
<mrow>
<msub>
<mi>D</mi>
<mi>x</mi>
</msub>
<mo>/</mo>
<msub>
<mi>V</mi>
<mi>x</mi>
</msub>
</mrow>
<mrow>
<msub>
<mi>D</mi>
<mi>y</mi>
</msub>
<mo>/</mo>
<msub>
<mi>V</mi>
<mi>y</mi>
</msub>
</mrow>
</mfrac>
<mo>=</mo>
<mfrac>
<msub>
<mi>D</mi>
<mi>x</mi>
</msub>
<msub>
<mi>D</mi>
<mi>y</mi>
</msub>
</mfrac>
<mi>t</mi>
<mi>a</mi>
<mi>n</mi>
<mi>&beta;</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
In formula, txAnd tyTo bond tearing port the time required to one thermocouple spacing of horizontal and vertical propagation;DxAnd DyRespectively
Copper plate of crystallizer thermocouple buries row and column spacing;VxAnd VyRespectively bond the horizontal and vertical spread speed of tearing port;β is viscous
The tear line and horizontal angle of breach are tied, 20 °~50 ° are taken;
The steel leakage prediction system carries out crystallizer bleedout prediction electric thermo using following steps:
Step 1: from parameter configuration files read breakout prediction method in each logic judgment parameter of regularity;
Step 2: the collection of copper plate of crystallizer electric thermo-couple temperature time series data, storage and pretreatment;
Step 3: entering trip temperature respectively to all thermocouples on copper plate of crystallizer simultaneously rises inspection;
Step 4: the thermocouple for rising Rule of judgment to meeting temperature in step 3 carries out temperature rate-of-rise inspection;
Step 5:To 3 thermocouple TC (i, j-1) around temperaturing lifting abnormality thermocouple TC (i, j), TC (i, j+1) and TC (i+1,
J) delay of progress temperature change is checked, and judges whether that making bonding gently alarms according to result;
Step 6:The surrounding's electric thermo-couple temperature change for meeting thermocouple TC (i, j) constant temperature decline inspection or TC (m, n) is prolonged
After checking late, then into step 7;
Step 7:Electric thermo-couple temperature, which is inverted, to be checked;
Step 8:According to alarm shielding condition, judge whether that make bonding steel leakage alarms again.
2. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 1 judges, it is characterised in that:In step one
Each logic judgment parameter of regularity include data prediction logic judgment parameter, electric thermo-couple temperature rise check logic judgment parameter,
Electric thermo-couple temperature climbing speed checks that logic judgment parameter, electric thermo-couple temperature change delay check logic judgment parameter, thermocouple
Temperature drop checks that logic judgment parameter, electric thermo-couple temperature are inverted and checks that logic judgment parameter, bonding steel leakage are gently alarmed and reported again
Alert logic judgment parameter.
3. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 2 judges, it is characterised in that:Step 2
In, all electric thermo-couple temperature time series datas and according to the predetermined row and column arrangement mode of thermocouple in copper plate of crystallizer are gathered first
Storage numbering, the temperature of thermocouple of collection per second;
Then following data filtering and smooth pretreatment are carried out:
A) electric thermo-couple temperature is less than 50 DEG C or more than 200 DEG C, is not involved in algorithm calculating;
B) temperature fluctuation per second is more than 20 DEG C, and every temperature fluctuation in 2 seconds is more than 38 DEG C, is not involved in algorithm calculating;
C) temperature data for gathering thermocouple current time is made comparisons with the temperature data that the thermocouple previous second gathers, by temperature
Temperature data of the degree change more than 20% replaces with the temperature data of previous second collection;
D) time series data of each electric thermo-couple temperature is smoothed using the method for moving average, the above-mentioned method of moving average
Be use before and after current time one it is intersegmental every temperature data calculate average value substitution current time temperature value, if it is mobile at intervals of
2k, k are positive integer, then the moving average at p momentFor:
<mrow>
<mover>
<mrow>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>)</mo>
</mrow>
</mrow>
<mo>&OverBar;</mo>
</mover>
<mo>=</mo>
<mfrac>
<mrow>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>-</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>-</mo>
<mi>k</mi>
<mo>+</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mo>...</mo>
<mo>+</mo>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>+</mo>
<mi>k</mi>
<mo>-</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
<mo>+</mo>
<mi>T</mi>
<mrow>
<mo>(</mo>
<mi>p</mi>
<mo>+</mo>
<mi>k</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mn>2</mn>
<mi>k</mi>
</mrow>
</mfrac>
</mrow>
In formula, T (p-k) is the electric thermo-couple temperature value at p-k moment.
4. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 3 judges, it is characterised in that in step 3
Electric thermo-couple temperature, which rises, checks as follows:When thermocouple normal work, take rise inspection cycle period for a temperature within every 4 seconds,
If the temperature of the 1st second is a, the temperature of the 2nd second is b, and the 3rd second temperature is c, and the 4th second temperature is d;Every one on a timeline
Moment just checks first 3 seconds and whether current time meets temperature and rise Rule of judgment:d>c>b>a;If certain moment the i-th row j is arranged
Thermocouple TC (i, j) is met, then τ is labeled as at the time of by the 1st second in the thermocouple this cycle period0(i, j), if can not
Meet, the temperature continued next second, which rises, to be checked;
Wherein:Said temperature rising Rule of judgment is met when some temperature rises proof cycle, then checks next temperature and rises
Whether proof cycle, which still meets said temperature, rises Rule of judgment, still meets the τ before then maintaining0(i, j), if discontented
Foot, the temperature continued next second, which rises, to be checked;
Temperature rising Rule of judgment is continuously unsatisfactory for if the temperature within 4 rises proof cycle, with latter temperature
Rise proof cycle and meet temperature rising Rule of judgment, the then τ before still maintaining again0(i, j);If 5 or 5 with
Upper temperature rises proof cycle and is continuously unsatisfactory for temperature rising Rule of judgment, the then τ before removing0(i, j), meets again afterwards
When temperature rises Rule of judgment, then a τ is re-established0(i, j), follow-up algorithm is with new τ0On the basis of (i, j).
5. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 4 judges, it is characterised in that in step 4
Temperature rate-of-rise checks as follows:When there is τ0When (i, j), thermocouple TC (i, j) temperature rate-of-rise u=(d-a)/3 is calculated
Value, compare u and design temperature climbing speed threshold values ucSize;If u is more than or equal to uc, then it is assumed that thermocouple TC (i, j)
Temperature rise abnormity, while by τ0(i, j) saves as the electric thermo-couple temperature and begins to ramp up moment t0(i, j), moment acquiescence is temperature
Change point at first is spent, into step 5.
6. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 5 judges, it is characterised in that:Step 5
In, in t0In temperature change time delay threshold range after (i, j), to 3 thermocouple TC (i, j-1) around TC (i, j),
(i+1, the logic judgment inspection of j) repeat the above steps three and step 4, is satisfied by the thermocouple TC of condition by TC (i, j+1) and TC
Numeration is marked in (m, n), and stores TC (m, n) temperature and begin to ramp up moment t0(m, n), mark count is more than 2
When, send bonding steel leakage and gently alarm, enter back into step 6.
7. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 6 judges, it is characterised in that step 6
In, thermocouple TC (i, j) constant temperature, which declines, checks as follows:In t0In 30 seconds after (i, j), thermocouple TC (i, j) is entered
Trip temperature, which declines, to be checked, it every 4 seconds is a temperature drop proof cycle to take, if the 1st second temperature is a', the 2nd second temperature is b', the
3 seconds temperature are c', and temperature is d' within the 4th second, such as meets d' ﹤ c' ﹤ b' ﹤ a', then judge that temperature drop occurs in thermocouple TC (i, j),
If persistently meeting 3 temperature drop proof cycles, then it is assumed that constant temperature decline occurs in thermocouple TC (i, j), is directly entered
Step 7;
TC (m, n) surrounding's electric thermo-couple temperature change delay checks as follows:In t0To all TC in 3~15 seconds after (m, n)
Three thermocouple TC (m, n-1) around (m, n), TC (m, n+1) and TC (m-1, n) logic of repeat step three and step 4 sentence
It is disconnected to check, thermocouple TC (i, j) temperature change is postponed to check wherein excluding, equally to meet condition thermocouple TC (x,
Y) numeration is marked, when marking count to be more than or equal to 3, thermocouple TC (i, j) doubtful bonding point is labeled as, directly
Tap into step 7.
8. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 7 judges, it is characterised in that step 7
In, in t0In 30s after (i, j), TC (i, j) is expert at three thermocouple TC (i, j-1), TC (i, j) and TC (i, j+1)
Two row thermocouples enter trip temperature inversion logic judgment inspection under corresponding column, are specially:
Situation (1):First compare t0Before (i, j) in 10s, whether the mean temperature of i-th row some thermocouple is higher than two under same column
The mean temperature of one of 2 thermocouples of row;
If it is, observation is in t0After (i, j) in 30s, the time-temperature curve of some thermocouple of i-th row with it is described
Whether the time-temperature curve of one of two rows, 2 thermocouples has intersection on image under same column, then thinks if intersecting
There is temperature to be inverted, be directly entered step 8;
If be less than, compare t0Whether the temperature of i-th row some thermocouple is occurred after (i, j) in 30s first higher than rear and low
In one of two rows, 2 thermocouples under same column, if there is, then it is assumed that occur in that temperature is inverted, be directly entered step 8;
Situation (2):When the situation that is unsatisfactory for (1), first compare t010s, the mean temperature of certain row i+1 row thermocouple before (i, j)
Whether the mean temperature of row i-th+2 row thermocouple is higher than;
If it is, in t0After (i, j) in 30s, observe the time-temperature curve of certain row i+1 row thermocouple and the row i-th+
Whether the time-temperature curve of 2 row thermocouples has intersection on image, then thinks temperature inversion occur if intersecting, directly
Into step 8;
If be less than, compare t0Whether occur the temperature of certain row i+1 row thermocouple after (i, j) in 30s and first arrange the higher than this
, when the situation described in appearance, still believe that less than the row thermocouple of row i-th+2 again after i+2 row thermocouples and occur in that temperature
It is inverted, is directly entered step 8.
9. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 8 judges, it is characterised in that in step 8
There is provided following alarm shielding condition, alarms if being unsatisfactory for following alarm shielding condition and just making bonding steel leakage again:
1. 120s interior crystallizers liquid level exceeds predetermined minimum and maximum range before alarming, and system is not alarmed;
2. before alarming in 120s pulling rate less than it is default do not alarm pulling rate when, system is not alarmed;
3. before alarming in 120s pulling rate rate of change exceed it is default do not alarm pulling rate rate of change when, system is not alarmed;
4. when cast length be less than it is default do not alarm cast length when, system is not alarmed;
5. bonding steel leakage is alarmed again after in 60s, system is not alarmed;
6. when failure thermocouple number exceedes the threshold values of setting, system is not alarmed.
10. the crystallizer bleedout prediction electric thermo system that logic-based according to claim 9 judges, it is characterised in that step 2
In, k is set to 2;In step 4, temperature rate-of-rise threshold values ucIt is set to 0.15 DEG C/s;In step 5, t0Temperature after (i, j)
Degree change delay time threshold range is set to 3~15 seconds;In step 8, mould liquid level predetermined minimum, maximum difference
For 80mm, 120mm;Default pulling rate of not alarming is 0.5m/min;Default pulling rate rate of change of not alarming is 0.006m/min;It is default
Cast length of not alarming is 3.0m;The threshold values that failure thermocouple number is set as thermocouple sum 1/3;
The steel leakage prediction system also includes:Data acquisition unit, bleed-out data are compressed memory cell, self study and divided offline automatically
Analyse unit, alarm and execution unit;
The data acquisition unit is used to obtain electric thermo-couple temperature real time data and casting cycle product practice, is surveyed by data
Amount, data transfer and the part of host computer three composition;
Wherein:DATA REASONING part includes sensor and A/D change-over circuits, and A/D change-over circuits are used for measure sensor
Analog signal is converted to data signal;TCP data segment uses field bus technique, upper for sending data to distal end
Machine, wherein having from oracle database and from two kinds of acquisition data modes of PLC, is retrieved as by the company of foundation from oracle database
Casting second databases are connected with the DBlinks of steel leakage prediction system database, are accessed using synonym;Obtaining data from PLC is
Siemens S7-400 PLC is directly gathered using the OPC communications of Siemens's host computer Configuration software WinCC, the data come up are gathered
Directly it is stored on the industrial computer for installing WinCC;
The bleed-out data compress memory cell automatically is used for all data by steel leakage prediction system during real time execution,
It is compressed according to certain file format by way of data flow, each casting cycle terminates to automatically save a file,
Each casting cycle starts with regard to being recorded automatically;
The self study and off-line analysis unit, can offline be imitated breakout prediction model algorithm by off-line analysis module
Very, specially change logic judgment parameter of regularity, bleed-out data are played back, pass through forecast accuracy, rate of false alarm, alarm frequency
The performance indications of rate carry out study optimization to breakout prediction model on-line parameter;Steel leakage prediction system is corresponded in on-line operation,
Each bonding steel leakage is alarmed again send after, site operation personnel is confirmed whether it is correct bleed-out result and is confirmed whether it is correct report
Alert, then system records the correct alarm times of breakout prediction algorithm, error-alert degree, bonding steel leakage and quotes rate, forecast automatically
Accuracy rate, rate of false alarm, the index of wrong report frequency, and quote rate, forecast accuracy and false alarm frequency with the bonding steel leakage of setting
Performance indications threshold values is compared, once less than setting threshold values, system sends Parameter Self-learning request;Wherein, bonding steel leakage report
Extracting rate=correct alarm times/(failing to report number of times+correct alarm times), forecast accuracy=correct alarm times/(fails to report number of times
+ correct alarm times+error-alert degree), rate of false alarm=error-alert degree/(correct alarm times+error-alert degree), wrong report frequency
Rate=error-alert degree/casting furnace number;
The alarm and execution unit, bleed-out alarm signal is sent by audible-visual annunciator, and light alarm is bright for amber light, and alarm again is
When red, if attaching most importance to alarm, warning message uploads to the execution unit of casting speed control, by PLC send reduction pulling rate or
The instruction directly shut down;
The parameter configuration unit, for setting system operational parameters, including various logic judgment rule parameter, the user of system
Name and password, and system communication connecting interface relevant parameter, and these parameters are stored in parameter configuration files, in pipe
Reason person's authority can change parameter, and automatically update parameter configuration files, directly from configuration text during steel leakage prediction system computing
Part call parameters.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710346785.1A CN107096899B (en) | 2017-05-17 | 2017-05-17 | A kind of crystallizer bleedout prediction electric thermo system that logic-based judges |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710346785.1A CN107096899B (en) | 2017-05-17 | 2017-05-17 | A kind of crystallizer bleedout prediction electric thermo system that logic-based judges |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107096899A true CN107096899A (en) | 2017-08-29 |
CN107096899B CN107096899B (en) | 2018-09-11 |
Family
ID=59670237
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710346785.1A Active CN107096899B (en) | 2017-05-17 | 2017-05-17 | A kind of crystallizer bleedout prediction electric thermo system that logic-based judges |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107096899B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109029830A (en) * | 2018-09-25 | 2018-12-18 | 首钢集团有限公司 | Steel leakage detection system and method |
CN109396375A (en) * | 2018-12-11 | 2019-03-01 | 大连理工大学 | A kind of crystallizer bleedout prediction electric thermo method based on feature vector and hierarchical clustering |
CN110523941A (en) * | 2019-08-06 | 2019-12-03 | 北京数钰科技发展有限公司 | Continuous casting bonding steel leakage multistage risk control method and control device |
CN111496211A (en) * | 2020-04-28 | 2020-08-07 | 中冶南方连铸技术工程有限责任公司 | Method for tracking and identifying bonding point on surface of casting blank |
CN112149823A (en) * | 2020-08-20 | 2020-12-29 | 汉威科技集团股份有限公司 | Combined implementation method for filtering alarm information |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
BR8600427A (en) * | 1985-02-01 | 1986-10-14 | Nippon Steel Corp | PROCESS AND APPARATUS TO AVOID A FOUNDATION DEFECT IN A CONTINUOUS FOUNDATION |
JP2001239353A (en) * | 2000-02-28 | 2001-09-04 | Nippon Steel Corp | Detecting method of abnormal casting condition inside mold in continuous casting |
JP3501054B2 (en) * | 1999-12-09 | 2004-02-23 | Jfeスチール株式会社 | Method for preventing constraining breakout in continuous casting |
JP4214818B2 (en) * | 2003-01-31 | 2009-01-28 | Jfeスチール株式会社 | Abnormality detection method for temperature sensor for predicting constrained breakout |
JP2009178746A (en) * | 2008-01-31 | 2009-08-13 | Jfe Steel Corp | Method and apparatus for changing continuous casting mold width |
CN101934353A (en) * | 2009-06-30 | 2011-01-05 | 上海宝信软件股份有限公司 | Device and method for breakout prediction during slab continuous casting |
CN101985166A (en) * | 2009-07-29 | 2011-03-16 | 宝山钢铁股份有限公司 | Continuous casting slab longitudinal-crack steel leakage prediction method |
CN105522133A (en) * | 2015-12-21 | 2016-04-27 | 安徽工业大学 | Judgment method of bonding behavior between continuous casting slab and crystallizer copper plate |
-
2017
- 2017-05-17 CN CN201710346785.1A patent/CN107096899B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
BR8600427A (en) * | 1985-02-01 | 1986-10-14 | Nippon Steel Corp | PROCESS AND APPARATUS TO AVOID A FOUNDATION DEFECT IN A CONTINUOUS FOUNDATION |
JP3501054B2 (en) * | 1999-12-09 | 2004-02-23 | Jfeスチール株式会社 | Method for preventing constraining breakout in continuous casting |
JP2001239353A (en) * | 2000-02-28 | 2001-09-04 | Nippon Steel Corp | Detecting method of abnormal casting condition inside mold in continuous casting |
JP4214818B2 (en) * | 2003-01-31 | 2009-01-28 | Jfeスチール株式会社 | Abnormality detection method for temperature sensor for predicting constrained breakout |
JP2009178746A (en) * | 2008-01-31 | 2009-08-13 | Jfe Steel Corp | Method and apparatus for changing continuous casting mold width |
CN101934353A (en) * | 2009-06-30 | 2011-01-05 | 上海宝信软件股份有限公司 | Device and method for breakout prediction during slab continuous casting |
CN101985166A (en) * | 2009-07-29 | 2011-03-16 | 宝山钢铁股份有限公司 | Continuous casting slab longitudinal-crack steel leakage prediction method |
CN105522133A (en) * | 2015-12-21 | 2016-04-27 | 安徽工业大学 | Judgment method of bonding behavior between continuous casting slab and crystallizer copper plate |
Non-Patent Citations (2)
Title |
---|
刘瑞邦: "板坯连铸机自动控制系统设计", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
郝苏伟: "连铸结晶器漏钢预报系统设计与实现", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109029830A (en) * | 2018-09-25 | 2018-12-18 | 首钢集团有限公司 | Steel leakage detection system and method |
CN109029830B (en) * | 2018-09-25 | 2020-12-15 | 首钢集团有限公司 | Steel leakage detection system and method |
CN109396375A (en) * | 2018-12-11 | 2019-03-01 | 大连理工大学 | A kind of crystallizer bleedout prediction electric thermo method based on feature vector and hierarchical clustering |
CN109396375B (en) * | 2018-12-11 | 2019-09-27 | 大连理工大学 | A kind of crystallizer bleedout prediction electric thermo method based on feature vector and hierarchical clustering |
WO2020119156A1 (en) * | 2018-12-11 | 2020-06-18 | 大连理工大学 | Casting mold breakout prediction method based on feature vectors and hierarchical clustering |
US11105758B2 (en) | 2018-12-11 | 2021-08-31 | Dalian University Of Technology | Prediction method for mold breakout based on feature vectors and hierarchical clustering |
CN110523941A (en) * | 2019-08-06 | 2019-12-03 | 北京数钰科技发展有限公司 | Continuous casting bonding steel leakage multistage risk control method and control device |
CN110523941B (en) * | 2019-08-06 | 2023-03-17 | 中冶南方连铸技术工程有限责任公司 | Continuous casting bonding bleed-out multistage risk control method and control device |
CN111496211A (en) * | 2020-04-28 | 2020-08-07 | 中冶南方连铸技术工程有限责任公司 | Method for tracking and identifying bonding point on surface of casting blank |
CN111496211B (en) * | 2020-04-28 | 2021-08-06 | 中冶南方连铸技术工程有限责任公司 | Method for tracking and identifying bonding point on surface of casting blank |
CN112149823A (en) * | 2020-08-20 | 2020-12-29 | 汉威科技集团股份有限公司 | Combined implementation method for filtering alarm information |
Also Published As
Publication number | Publication date |
---|---|
CN107096899B (en) | 2018-09-11 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107159862B (en) | A kind of crystallizer bleedout prediction electric thermo method that logic-based judges | |
CN107096899B (en) | A kind of crystallizer bleedout prediction electric thermo system that logic-based judges | |
He et al. | Mold breakout prediction in slab continuous casting based on combined method of GA-BP neural network and logic rules | |
CN207552827U (en) | A kind of rail system safe condition comprehensive monitoring and intelligent analysis system | |
CN105522133A (en) | Judgment method of bonding behavior between continuous casting slab and crystallizer copper plate | |
CN108024207A (en) | Flow of the people monitoring method based on three layers of prevention and control circle | |
CN107858883A (en) | A kind of rail system safe condition comprehensive monitoring and intelligent analysis method | |
CN102324066A (en) | Radar chart representation method for early warning and assessment index of power system | |
CN102495549A (en) | Remote maintenance decision system and method for engineering machinery | |
CN210924883U (en) | Bridge structure health monitoring system | |
CN101667339B (en) | Modeling system and method based on measured traffic service level of urban road area | |
CN103617561A (en) | System and method for evaluating state of secondary equipment of power grid intelligent substation | |
CN105758661B (en) | Boiler heating surface service life evaluation system and method | |
CN106980729A (en) | A kind of continuous casting breakout prediction method based on mixed model | |
CN109365769A (en) | A kind of crystallizer bleedout prediction electric thermo method based on mixed model judgement | |
CN116595327B (en) | Sluice deformation monitoring data preprocessing system and method | |
CN101393680A (en) | Plant alarm apparatus and plant alarm method | |
CN110069551A (en) | Medical Devices O&M information excavating analysis system and its application method based on Spark | |
CN116739384A (en) | Mining equipment operation management system based on 5G wireless communication | |
CN107609989A (en) | A kind of bridge health monitoring intelligence CS architecture systems of road network level | |
CN102636624A (en) | Method for soft measurement of alumina concentration in electrolyzer during aluminum electrolysis process | |
CN109902265A (en) | A kind of underground method for early warning based on hidden Markov model | |
CN117114240B (en) | Safety evaluation method, system, equipment and medium for heightened gravity dam structure | |
CN117973131A (en) | Finite element model-based power transmission tower geological disaster failure prediction method and system | |
He et al. | Development and application of mould breakout prediction system with online thermal map for steel continuous casting |
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 |