CN111651729A - Method for predicting blockage of secondary cooling water nozzle in continuous casting - Google Patents
Method for predicting blockage of secondary cooling water nozzle in continuous casting Download PDFInfo
- Publication number
- CN111651729A CN111651729A CN202010487846.8A CN202010487846A CN111651729A CN 111651729 A CN111651729 A CN 111651729A CN 202010487846 A CN202010487846 A CN 202010487846A CN 111651729 A CN111651729 A CN 111651729A
- Authority
- CN
- China
- Prior art keywords
- data
- nozzle
- secondary cooling
- pressure
- continuous casting
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000009749 continuous casting Methods 0.000 title claims abstract description 26
- 239000000498 cooling water Substances 0.000 title claims abstract description 20
- 230000002159 abnormal effect Effects 0.000 claims abstract description 29
- 238000012417 linear regression Methods 0.000 claims abstract description 20
- 238000001816 cooling Methods 0.000 claims abstract description 15
- 238000007405 data analysis Methods 0.000 claims abstract description 14
- 238000005266 casting Methods 0.000 claims abstract description 12
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 8
- 239000010959 steel Substances 0.000 claims abstract description 8
- 238000003860 storage Methods 0.000 claims abstract description 6
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims description 14
- 230000005856 abnormality Effects 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 6
- 238000010219 correlation analysis Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 6
- 238000005192 partition Methods 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000005507 spraying Methods 0.000 claims description 3
- 230000000903 blocking effect Effects 0.000 abstract description 5
- 230000007547 defect Effects 0.000 abstract description 3
- 238000010438 heat treatment Methods 0.000 abstract description 2
- 238000005096 rolling process Methods 0.000 abstract description 2
- 238000004519 manufacturing process Methods 0.000 description 4
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 2
- 230000032683 aging Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000007619 statistical method Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 238000001134 F-test Methods 0.000 description 1
- 238000011001 backwashing Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000012938 design process Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 229910052742 iron Inorganic materials 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000611 regression analysis Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 239000007921 spray Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- B22D11/22—Controlling or regulating processes or operations for cooling cast stock or mould
- B22D11/225—Controlling or regulating processes or operations for cooling cast stock or mould for secondary cooling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/22—Indexing; Data structures therefor; Storage structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Software Systems (AREA)
- Mathematical Physics (AREA)
- Pure & Applied Mathematics (AREA)
- Computational Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- Mechanical Engineering (AREA)
- Algebra (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Continuous Casting (AREA)
Abstract
The invention relates to a method for predicting blockage of a continuous casting secondary cooling water nozzle, which is characterized in that a sensor is installed on site, a PLC (programmable logic controller) throws data into a Postgre database through a network shutdown machine for storage, data analysis is carried out on original data of the Postgre database after being preprocessed on python, and the relation between flow and pressure of the nozzle in a normal state is quantified through establishing a linear regression model through the data analysis, so that abnormal judgment logic is designed, and an alarm threshold value is continuously optimized. The real-time early warning function of abnormal conditions of nozzles in different flow areas of the two cooling areas is realized, the blocked nozzles are cleaned and replaced in a targeted manner when the machine is stopped for a short time, the blocking conditions are visualized, the workload is reduced, and the replacement accuracy is improved. The casting blank quality defect caused by uneven heating of the secondary cooling area is reduced, and a more stable casting blank is provided for the downstream steel rolling link, so that the yield of steel is improved.
Description
Technical Field
The invention belongs to the technical field of continuous casting, and particularly relates to a method for predicting blockage of a secondary cooling water nozzle in continuous casting.
Background
In the long-term production process of continuous casting, the nozzle blockage of the secondary cooling area is found, and the longer the time interval is, the more the nozzle blockage is, and the more the nozzle blockage occurs in 3, 4 and 5 areas. In the statistical process, the nozzles close to the edge are easy to block, and the blocking probability of the nozzles is far greater than that of the nozzles close to the central position. In the production process, some regional partial nozzles can be observed through the range estimation and can not go out water or the water yield is not enough, and most region can't be judged through the range estimation, and operational environment is high temperature environment moreover, and artifical observation is consuming time and is laboursome and have great potential safety hazard. In the maintenance process, after the nozzles are unscrewed, some nozzles are not blocked by visual observation; some can find the nozzle blockage of the yellow brown color; some nozzles cannot judge the blocking condition visually, and the blocking degree and the aging degree are different, so that an accurate replacement and cleaning standard cannot be established.
In the continuous casting production process, many defects on the surface and inside of a casting blank are caused by improper water spray cooling, and the blockage condition of a continuous casting nozzle is closely related to the secondary cooling effect of a sector section of a continuous casting machine, so that the blockage condition is one of important factors influencing the internal quality of the casting blank. The good working condition of the continuous casting nozzle can not only improve the quality of a casting blank, but also improve the operation rate of a continuous casting machine. Therefore, the early warning of the blockage of the nozzle in the secondary cooling area is a problem with practical significance.
Disclosure of Invention
The invention aims to provide a method for predicting the blockage of a secondary cooling water nozzle in continuous casting.
The technical scheme adopted by the invention for solving the technical problems is as follows: a method for predicting the blockage of a secondary cooling water nozzle in continuous casting comprises the following steps: the method comprises the steps that a sensor is installed on the site, a PLC (programmable logic controller) throws data into a Postgre database through a network shutdown machine for storage, original data of the Postgre database are preprocessed through python and then subjected to data analysis, and the relation between flow and pressure of a nozzle in a normal state is quantified through a linear regression model through the data analysis, so that abnormal judgment logic is designed, and an alarm threshold value is continuously optimized.
Further, the data stored in the Postgre database comprises original data collected by a PLC, ERP recorded data and manual recorded data. The original data collected by the PLC comprise header pipe pressure, header pipe flow, crystallizer water inlet temperature, crystallizer water outlet temperature, pulling speed, pressure on a partition branch pipe, flow and valve opening; the ERP recorded data comprises molten steel components and tundish temperature; the manual recording data comprises nozzle blockage recording, replacement recording and abnormal working condition recording.
Further, the preprocessing comprises unified time frequency and abnormal value processing, data are uniformly processed into an average value in one minute through python and stored according to a time sequence structure, and for part of original data with abnormality, the data are directly removed when the python reads the data.
Further, the data was analyzed as follows: the computer adjusts the set flow of the secondary cooling water in real time according to the change of the pulling speed, the valve corrects the difference value between the set flow and the actual flow by changing the size of the opening, the relation between the pressure and the actual flow is preliminarily determined to well reflect the spraying state of the nozzle, further correlation analysis finds that the linear correlation coefficient between the pressure and the actual flow is greater than 0.9, and an abnormal early warning algorithm is designed by utilizing the pressure, the actual flow and the nozzle blockage recording data for strong correlation.
Further, the linear regression model is y ═ a × x2Wherein y is pressure, x is actual flow value, and the square x of the pressure y and the actual flow value corresponding to the normal condition of the nozzle is obtained according to the corresponding linear regression model2The corresponding relation, i.e. anormalThe value of (c).
Further, the abnormality determination logic is as follows:
calculating the pretreated pressure per minute and actual flow rate value within one hour to obtain
i＝|ai-anormal1, …,60 according toiWhether the threshold t is larger than a specific threshold t or not is judged to be abnormal, and the threshold t is continuously optimized by observing and comparing the nozzle abnormality:
(1)
the nozzle is in a normal state, and the nozzle does not need to be cleaned or replaced;
(2)
the nozzle is in an abnormal state and needs to be replaced as soon as possible;
(3) otherwise, the nozzle state may be abnormal, and a technician is required to further visually judge the quality conditions of the nozzle and the casting blank.
Further, the linear regression model needs to train data under the condition of normal operation of the secondary cooling zone, and then judges that the nozzle is abnormal according to the data, in the online deployment process, data of the previous hour is collected every 15 minutes, data preprocessing is carried out, whether the machine is stopped and the flow is blocked is judged according to the pulling speed, whether the quantity and the quality of the data are qualified is judged, and finally, the state of the nozzle is judged according to the trained model, so that the real-time alarm of the state of the nozzle for 15 minutes is realized.
The method mainly utilizes python to carry out data mining and analysis modeling on data recorded on site and a large amount of data acquired by the sensor. By statistical analysis of the parameters, combined with field experience, 3 important classes of parameters were screened out from about 200 data relating to the caster: flow, pressure and secondary cold zone anomalies were recorded. And according to the correlation analysis result, preliminarily establishing a multi-class linear regression model, and selecting an optimal linear regression model by F-test, AIC test and the like. The relationship between the flow and the pressure under the normal condition is found by training the data of the nozzle under the normal condition and the linear regression model. Correspondingly, establishing an abnormal alarm logic: if the relation between the flow and the pressure in the previous hour deviates from the relation between the normal flow and the pressure, the abnormal condition of the second cold area can be judged. After the model is deployed on line, the real-time early warning of the abnormal conditions of the nozzles in different flow areas of the two cold areas can be realized every 15 minutes.
The model algorithm is designed on python, embedded into Java for deployment, extracted from a database in real time for calculation, and finally displayed to a user through a front-end design interface. The system interface is divided into 5 zones and 8 streams, and the nozzle states are represented by different colors: abnormal (red), early warning (yellow), normal (green), data acquisition (blue) and shut-down flow blockage (grey). And 8 historical alarm records of the expanded blocks and the visualization of the current linear model are clicked, so that the nozzle state can be known more intuitively on site. And at the same time, after the condition is changed or the nozzle combination is changed, the model can be retrained through the interface.
In the algorithm design process, the current situation and problems are analyzed by experts in the early stage, and the requirements of instruments and meters are determined. And installing a sensor on the site, and throwing the data to a Postgre database by the PLC through a network shutdown machine for storage. And then, carrying out preprocessing such as uniform time frequency and abnormal value processing on the database original data in python, and then carrying out data analysis, such as visualization, correlation analysis and the like. Through data analysis, the relation between the flow and the pressure under the normal state of the nozzle is quantified by establishing a linear regression model, so that an abnormal judgment logic is designed, and an alarm threshold value is continuously optimized.
According to the invention, a classical statistical method is used for mining and analyzing data recorded on site and a large amount of data collected by a sensor, important characterization parameters are searched through correlation analysis, and an abnormality detection model is correspondingly established, so that the real-time early warning function of nozzle abnormality conditions in different flow areas of a secondary cooling area is realized, the blocked nozzles are cleaned and replaced in a targeted manner when the machine is stopped for a short time, the nozzle replacement quality is improved, the blocked conditions are visualized, the workload is reduced, and the replacement accuracy is improved. The casting blank quality defect caused by uneven heating of the secondary cooling area is reduced, and a more stable casting blank is provided for the downstream steel rolling link, so that the yield of steel is improved, and considerable economic value is generated.
Detailed Description
The following are specific examples of the present invention and further describe the technical solutions of the present invention, but the scope of the present invention is not limited to these examples. All changes, modifications and equivalents that do not depart from the spirit of the invention are intended to be included within the scope thereof.
Direct cooling water backwater of the continuous casting machine is respectively collected in the iron sheet ditch and then automatically flows to the cyclone well, and then enters the in-workshop casting machine after the procedures of deoiling, cooling, backwashing, filtering and the like. Taking the No. 4 continuous casting machine with the side of the permanent front 165 as an example, the No. 4 continuous casting machine is provided with 5 zones and 8 streams, cooling water is introduced into the two cooling zones through a main pipe, and then atomized and sprayed through nozzles of the flow zones through branch pipes respectively, so that a casting blank in straightening is cooled continuously. The branch pipe is provided with an actual flowmeter, a set flowmeter, a pressure gauge, a valve opening degree and other electronic instruments. And the secondary cooling area dynamically adjusts and sets water flow according to the pulling speed through a cooling mathematical model set by a computer. Meanwhile, according to the actual water flow condition, the opening degree of the valve is subjected to negative feedback adjustment.
A method for predicting the blockage of a secondary cooling water nozzle in continuous casting comprises the following steps: the method comprises the steps that a sensor is installed on the site, a PLC (programmable logic controller) throws data into a Postgre database through a network shutdown machine for storage, original data of the Postgre database are preprocessed through python and then subjected to data analysis, and the relation between flow and pressure of a nozzle in a normal state is quantified through a linear regression model through the data analysis, so that abnormal judgment logic is designed, and an alarm threshold value is continuously optimized.
Data collection and storage: no. 4 continuous casting machine of the permanent steel plant produces 165 square billets, and all production data are transmitted to an energy pipe network and can only be stored for a short time. Therefore, the data is transferred to the Postgre database for storage through the network shutdown. In the collected data, the original data collected by the PLC comprises: the pressure of a main pipe, the flow of the main pipe, the temperature of water entering a crystallizer, the temperature of water exiting the crystallizer, the pulling speed, the pressure on a partition flow dividing branch pipe, the flow, the opening degree of a valve and the like; ERP recorded data has indexes: molten steel components, tundish temperature and the like; the manual recording data mainly comprises: nozzle blockage recording, replacement recording, abnormal condition recording and the like.
Data preprocessing: through the initial search of the data, most of the data come from the sensors, the data are transmitted in a sensor data forwarding mode, the recording frequencies are inconsistent but are all in millisecond level, and therefore the data are uniformly processed into an average value in one minute through python and stored according to a time sequence structure, and the data are conveniently analyzed later. Meanwhile, some original data are found to have abnormity, such as the pulling speed and the pressure are negative, the valve opening exceeds 100, the data value is extremely large, and the like. Because the original data is many and the abnormal data is small in proportion, the data is directly removed when the python reads the data. Meanwhile, the data has the individual condition that the deviation from the sample is too large or too small, and although the data possibly accords with the actual condition, the rationality of the whole result is greatly influenced, so that the special data are removed by using the statistical 3-time standard deviation principle in the data analysis process.
And (3) data analysis: the data parameters are subjected to descriptive analysis, and the correlations of the pulling speed and the crystallizer water quantity with the temperature and the nozzle state are not high. Meanwhile, the information is found in the combing analysis of a secondary cooling water regulating mechanism by field technicians: the computer adjusts the set flow of the secondary cooling water in real time according to the change of the pulling speed, and the valve corrects the difference value between the set flow and the actual flow by changing the size of the opening, so that the relationship between the pressure and the actual flow is preliminarily determined to well reflect the spraying state of the nozzle. Further correlation analysis also found that the linear correlation coefficient between the pressure and the actual flow rate is greater than 0.9, which is a strong correlation. And then, designing an abnormity early warning algorithm by utilizing the pressure, the actual flow and the nozzle blockage record data.
Establishing a linear regression model:
definition of regression model: in statistics, Linear Regression (Linear Regression) is a Regression analysis that models the relationship between one or more independent and dependent variables using a least squares function called a Linear Regression equation. Such a function is a linear combination of one or more model parameters called regression coefficients. The case of only one independent variable, greater than one, is called simple regressionThe mathematical expression is as follows, the least squares function (y-X β) (y-X β) is solved for errors following a normal distribution with a mean value of 0, y ═ X β +, -N (0,)TThe closed solution β ═ XX (XX) can be obtainedT)- 1XTy, in practical application, python often solves an optimal solution β through gradient descent to establish a linear regression model, wherein the Pearson correlation coefficient between pressure and actual flow is more than 0.9 through data analysis, and visual drawing verification shows that the pressure (y) has strong linear correlation with the actual flow value (x) and the pressure should be zero when the actual flow is zero under the condition of normal data, therefore, the following 3 types of linear models are established:
y＝a*x2
y＝a*x3/2
y＝a*x2
finally obtaining the best fitting equation y ═ a x through the fitting test of the model in the last chapter2. The 4-zone 8-flow nozzle combinations (number and type) may be different, i.e. the pressure (y) and the actual flow value (x) do not correspond to the same value. Therefore, the data of each zone flow nozzle in the normal operation state are fitted, and a corresponding linear regression model is trained, so that the square (x) of the pressure (y) and the actual flow value corresponding to the nozzle in the normal condition is obtained2) The corresponding relation, i.e. anormalThe value of (c).
An abnormality judgment logic:
from the preamble section, it follows that in normal nozzle conditions, the pressure (y) is anormalSquare of actual flow value (x)2). Therefore, if an abnormality occurs in the nozzle,
should deviate from anormal. In particular, the more severe the nozzle clogging,
should be much larger than anormal(ii) a And if a water leakage occurs from the nozzle,
should be much less than anormal. Accordingly, the following abnormality determination logic is established:
calculating the pretreated pressure per minute and actual flow rate value within one hour to obtain
i＝|ai-anormal1, …,60 according toiWhether the threshold t is larger than a specific threshold t or not is judged to be abnormal, and the threshold t is continuously optimized by observing and comparing the nozzle abnormality:
(1)
the nozzle is in a normal state, and the nozzle does not need to be cleaned or replaced;
(2)
the nozzle is in an abnormal state and needs to be replaced as soon as possible;
(3) otherwise, the nozzle state may be abnormal, and a technician is required to further visually judge the quality conditions of the nozzle and the casting blank. In summary, the algorithm is based on the following conditions: different pressure versus flow relationships can characterize different nozzle operating conditions (e.g., different nozzle combinations, different nozzle aging levels, different nozzle plugging levels, etc.). Therefore, the model needs to train data under the normal operation condition of the secondary cooling zone, and then the nozzle abnormality is judged according to the data. In the online deployment process, data of the previous hour are collected every 15 minutes, data preprocessing is carried out, whether stopping and flow blocking are carried out or not is judged according to the pulling speed, whether the quantity and the quality of the data are qualified or not is judged, and finally the nozzle state is judged according to a trained model, so that the real-time alarm of the nozzle state for 15 minutes is realized.
The present invention is not limited to the above embodiments, and any structural changes made under the teaching of the present invention shall fall within the scope of the present invention, which is similar or similar to the technical solutions of the present invention.
The techniques, shapes, and configurations not described in detail in the present invention are all known techniques.
Claims (8)
1. A method for predicting the blockage of a secondary cooling water nozzle in continuous casting is characterized by comprising the following steps: the method comprises the steps that a sensor is installed on the site, a PLC (programmable logic controller) throws data into a Postgre database through a network shutdown machine for storage, original data of the Postgre database are preprocessed through python and then subjected to data analysis, and the relation between flow and pressure of a nozzle in a normal state is quantified through a linear regression model through the data analysis, so that abnormal judgment logic is designed, and an alarm threshold value is continuously optimized.
2. The method of predicting clogging of a continuous casting secondary cooling water nozzle as set forth in claim 1, wherein said Postgre database stores data including PLC acquired raw data, ERP recorded data, manual recorded data.
3. The method for predicting the blockage of the continuous casting secondary cooling water nozzle according to claim 2, wherein the raw data collected by the PLC comprises the pressure of a main pipe, the flow rate of the main pipe, the temperature of water entering the crystallizer, the temperature of water exiting the crystallizer, the pulling speed, the pressure on a partition branch pipe, the flow rate and the opening degree of a valve; the ERP recorded data comprises molten steel components and tundish temperature; the manual recording data comprises nozzle blockage recording, replacement recording and abnormal working condition recording.
4. The method for predicting the blockage of the continuous casting secondary cooling water nozzle as claimed in claim 1, wherein the preprocessing comprises the processing of a uniform time frequency and an abnormal value, the data is uniformly processed into an average value within one minute through python and stored according to a time sequence structure, and the data is directly rejected when the python reads the data for a part of original data with the abnormality.
5. A method of predicting clogging of a continuous casting secondary cooling water nozzle as set forth in claim 3, wherein said data is analyzed as follows: the computer adjusts the set flow of the secondary cooling water in real time according to the change of the pulling speed, the valve corrects the difference value between the set flow and the actual flow by changing the size of the opening, the relation between the pressure and the actual flow is preliminarily determined to well reflect the spraying state of the nozzle, further correlation analysis finds that the linear correlation coefficient between the pressure and the actual flow is greater than 0.9, and an abnormal early warning algorithm is designed by utilizing the pressure, the actual flow and the nozzle blockage recording data for strong correlation.
6. The method of predicting continuous casting secondary cold water nozzle clogging according to claim 1, wherein said linear regression model is y ═ a x2Wherein y is pressure, x is actual flow value, and the square x of the pressure y and the actual flow value corresponding to the normal condition of the nozzle is obtained according to the corresponding linear regression model2The corresponding relation, i.e. anormalThe value of (c).
7. The method of predicting clogging of a continuous casting secondary cooling water nozzle as set forth in claim 1, wherein said anomaly decision logic is as follows:
calculating the pretreated pressure per minute and actual flow rate value within one hour to obtain
i＝|ai-anormal1, …,60 according toiWhether the threshold t is larger than a specific threshold t or not is judged to be abnormal, and the threshold t is continuously optimized by observing and comparing the nozzle abnormality:
(1)
the nozzle is in a normal state, and the nozzle does not need to be cleaned or replaced;
(2)
the nozzle is in an abnormal state and needs to be replaced as soon as possible;
(3) otherwise, the nozzle state may be abnormal, and a technician is required to further visually judge the quality conditions of the nozzle and the casting blank.
8. The method for predicting the blockage of the continuous casting secondary cooling water nozzle according to claim 6, wherein the linear regression model needs to train data under the condition of normal operation of a secondary cooling zone, and then the abnormality of the nozzle is judged according to the data, in the online deployment process, the data of the previous hour is collected every 15 minutes, the data is preprocessed, whether the machine is stopped and the flow is blocked is judged according to the pulling speed, whether the quantity and the quality of the data are qualified is judged, and finally the state of the nozzle is judged according to the trained model, so that the real-time alarm of the state of the nozzle for 15 minutes is realized.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010487846.8A CN111651729A (en) | 2020-06-02 | 2020-06-02 | Method for predicting blockage of secondary cooling water nozzle in continuous casting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010487846.8A CN111651729A (en) | 2020-06-02 | 2020-06-02 | Method for predicting blockage of secondary cooling water nozzle in continuous casting |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111651729A true CN111651729A (en) | 2020-09-11 |
Family
ID=72344699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010487846.8A Pending CN111651729A (en) | 2020-06-02 | 2020-06-02 | Method for predicting blockage of secondary cooling water nozzle in continuous casting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111651729A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112958751A (en) * | 2021-01-27 | 2021-06-15 | 唐山不锈钢有限责任公司 | Online prediction and management method for continuous casting secondary cooling state |
| CN113158466A (en) * | 2021-04-23 | 2021-07-23 | 浙江捷创智能技术有限公司 | Method for predicting maintenance of pipe blockage of phosphorus chemical dilute phosphoric acid supply system |
| CN116186936A (en) * | 2023-03-01 | 2023-05-30 | 华院计算技术(上海)股份有限公司 | Method, system, equipment and medium for determining continuous casting process parameters |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104568410A (en) * | 2015-01-05 | 2015-04-29 | 河北钢铁股份有限公司邯郸分公司 | Continuous-casting secondary cooling nozzle working state online judgment method |
| CN105939801A (en) * | 2014-02-07 | 2016-09-14 | 首要金属科技奥地利有限责任公司 | Monitoring of a line system |
| CN106180622A (en) * | 2016-07-18 | 2016-12-07 | 中冶赛迪工程技术股份有限公司 | The dynamic judgment method of a kind of secondary cooling nozzle of continuous casting duty and system |
| CN107052291A (en) * | 2017-04-21 | 2017-08-18 | 抚顺新钢铁有限责任公司 | A kind of continuous casting secondary cooling water dynamic water allocation system and technique |
| CN109723699A (en) * | 2017-10-27 | 2019-05-07 | 北京精密机电控制设备研究所 | A kind of nozzle-flapper servo valve prestage discharge coefficient test device and method |
| CN110625080A (en) * | 2019-09-23 | 2019-12-31 | 中冶南方连铸技术工程有限责任公司 | Intelligent detection and control method for online working state of continuous casting secondary cooling system |
| CN110699683A (en) * | 2019-11-15 | 2020-01-17 | 昆山开信精工机械股份有限公司 | Self-checking type intelligent early warning cold spraying equipment and operation process thereof |
-
2020
- 2020-06-02 CN CN202010487846.8A patent/CN111651729A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105939801A (en) * | 2014-02-07 | 2016-09-14 | 首要金属科技奥地利有限责任公司 | Monitoring of a line system |
| CN104568410A (en) * | 2015-01-05 | 2015-04-29 | 河北钢铁股份有限公司邯郸分公司 | Continuous-casting secondary cooling nozzle working state online judgment method |
| CN106180622A (en) * | 2016-07-18 | 2016-12-07 | 中冶赛迪工程技术股份有限公司 | The dynamic judgment method of a kind of secondary cooling nozzle of continuous casting duty and system |
| CN107052291A (en) * | 2017-04-21 | 2017-08-18 | 抚顺新钢铁有限责任公司 | A kind of continuous casting secondary cooling water dynamic water allocation system and technique |
| CN109723699A (en) * | 2017-10-27 | 2019-05-07 | 北京精密机电控制设备研究所 | A kind of nozzle-flapper servo valve prestage discharge coefficient test device and method |
| CN110625080A (en) * | 2019-09-23 | 2019-12-31 | 中冶南方连铸技术工程有限责任公司 | Intelligent detection and control method for online working state of continuous casting secondary cooling system |
| CN110699683A (en) * | 2019-11-15 | 2020-01-17 | 昆山开信精工机械股份有限公司 | Self-checking type intelligent early warning cold spraying equipment and operation process thereof |
Non-Patent Citations (1)
| Title |
|---|
| 鹿艳: "连铸二冷喷嘴试验装置及自动化控制系统", 现代电子技术, no. 11, pages 84 - 86 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112958751A (en) * | 2021-01-27 | 2021-06-15 | 唐山不锈钢有限责任公司 | Online prediction and management method for continuous casting secondary cooling state |
| CN113158466A (en) * | 2021-04-23 | 2021-07-23 | 浙江捷创智能技术有限公司 | Method for predicting maintenance of pipe blockage of phosphorus chemical dilute phosphoric acid supply system |
| CN116186936A (en) * | 2023-03-01 | 2023-05-30 | 华院计算技术(上海)股份有限公司 | Method, system, equipment and medium for determining continuous casting process parameters |
| CN116186936B (en) * | 2023-03-01 | 2024-03-22 | 华院计算技术(上海)股份有限公司 | Method, system, equipment and medium for determining continuous casting process parameters |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111651729A (en) | Method for predicting blockage of secondary cooling water nozzle in continuous casting | |
| US7827006B2 (en) | Heat exchanger fouling detection | |
| CN110625080B (en) | Intelligent detection and control method for online working state of continuous casting secondary cooling system | |
| US20080082295A1 (en) | Abnormal situation prevention in a coker heater | |
| CN106649755B (en) | Threshold value self-adaptive setting abnormity detection method for multi-dimensional real-time power transformation equipment data | |
| CN102147613B (en) | Model predictive control method and system for moisture and temperature of tobacco redryer | |
| CN107643015B (en) | A kind of micro-channel heat exchanger Fouling Monitoring processing system and method | |
| CN116184948A (en) | Intelligent monitoring disc for water plant and application system and method of early warning diagnosis technology | |
| CN110469496B (en) | Intelligent early warning method and system for water pump | |
| WO2008014344A2 (en) | Model based method for detecting abnormal operation of a level regulatory control loop and associated apparatus | |
| CN114580852B (en) | Water pump breakwater clearance real-time warning system based on industry big data | |
| CN113219910A (en) | Full-flow production self-diagnosis and optimization system | |
| Trunzer et al. | Failure mode classification for control valves for supporting data-driven fault detection | |
| Zhou et al. | Application of time series data anomaly detection based on deep learning in continuous casting process | |
| CN110083797B (en) | Method and system for monitoring abnormity of drip irrigation pipe production line in real time | |
| CN113405741B (en) | Leakage detection system and diagnosis method for dry quenching primary dedusting cooling system | |
| Chen et al. | Design of multivariate alarm systems based on online calculation of variational directions | |
| US20180104624A1 (en) | Device to separate water and solids of spray water in a continuous caster, and method to monitor and control corrosion background | |
| CN115186754A (en) | Unit energy efficiency monitoring and diagnosing method based on information entropy and auto-associative regression model | |
| CN112215493A (en) | Method for judging and evaluating process quality of cigarette shredding workshop section | |
| CN112130542A (en) | Control loop performance evaluation method based on normal operation data and system identification | |
| CN112958751A (en) | Online prediction and management method for continuous casting secondary cooling state | |
| RU2608141C1 (en) | Method for real-time monitoring of water and sand carry-over with extracted product from well in automated process control systems of gas-field objects of oil and gas condensate of deposits of far north | |
| CN115471059B (en) | Water turbine online economic operation system based on cavitation vibration area planning | |
| CN116943329A (en) | Wind turbine generator gearbox filter blockage early warning diagnosis method and system |
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 |