CN116340795B - Operation data processing method for pure oxygen combustion heating furnace - Google Patents
Operation data processing method for pure oxygen combustion heating furnace Download PDFInfo
- Publication number
- CN116340795B CN116340795B CN202310609217.1A CN202310609217A CN116340795B CN 116340795 B CN116340795 B CN 116340795B CN 202310609217 A CN202310609217 A CN 202310609217A CN 116340795 B CN116340795 B CN 116340795B
- Authority
- CN
- China
- Prior art keywords
- data
- surface temperature
- abnormal
- associated data
- influence factor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- MYMOFIZGZYHOMD-UHFFFAOYSA-N Dioxygen Chemical compound O=O MYMOFIZGZYHOMD-UHFFFAOYSA-N 0.000 title claims abstract description 37
- 238000002485 combustion reaction Methods 0.000 title claims abstract description 37
- 238000010438 heat treatment Methods 0.000 title claims abstract description 34
- 238000003672 processing method Methods 0.000 title abstract description 6
- 230000002159 abnormal effect Effects 0.000 claims abstract description 133
- 238000000034 method Methods 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 22
- 230000005856 abnormality Effects 0.000 claims abstract description 13
- 239000000446 fuel Substances 0.000 claims description 15
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims description 13
- 239000001301 oxygen Substances 0.000 claims description 13
- 229910052760 oxygen Inorganic materials 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 8
- 230000008859 change Effects 0.000 claims description 7
- 238000005259 measurement Methods 0.000 claims description 6
- 230000001419 dependent effect Effects 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 22
- 238000004458 analytical method Methods 0.000 abstract description 15
- 238000004422 calculation algorithm Methods 0.000 abstract description 10
- 238000012098 association analyses Methods 0.000 abstract description 2
- 239000007789 gas Substances 0.000 description 14
- 230000008569 process Effects 0.000 description 9
- 230000005855 radiation Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000012546 transfer Methods 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000013506 data mapping Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000004880 explosion Methods 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F27—FURNACES; KILNS; OVENS; RETORTS
- F27D—DETAILS OR ACCESSORIES OF FURNACES, KILNS, OVENS, OR RETORTS, IN SO FAR AS THEY ARE OF KINDS OCCURRING IN MORE THAN ONE KIND OF FURNACE
- F27D21/00—Arrangements of monitoring devices; Arrangements of safety devices
- F27D21/0014—Devices for monitoring temperature
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01K—MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR
- G01K13/00—Thermometers specially adapted for specific purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2433—Single-class perspective, e.g. one-against-all classification; Novelty detection; Outlier detection
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/005—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles during manufacturing process
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E20/00—Combustion technologies with mitigation potential
- Y02E20/34—Indirect CO2mitigation, i.e. by acting on non CO2directly related matters of the process, e.g. pre-heating or heat recovery
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Mechanical Engineering (AREA)
- Manufacturing & Machinery (AREA)
- Waste-Gas Treatment And Other Accessory Devices For Furnaces (AREA)
- Testing And Monitoring For Control Systems (AREA)
Abstract
The invention relates to the technical field of data processing, in particular to an operation data processing method for a pure oxygen combustion heating furnace. The method acquires associated data of surface temperature data, and determines abnormal interference factors at corresponding moments by carrying out data discrete analysis and associated analysis on the associated data. And further carrying out data deviation analysis on the surface temperature data, and determining suspected abnormal influence factors of each surface temperature data. Combining the abnormal interference factors and the suspected abnormal influence factors to obtain comprehensive abnormal influence factors, processing the surface temperature data according to the comprehensive abnormal influence factors, and identifying abnormal temperature data. According to the invention, through processing and association analysis of the associated data, the index representing the affected degree of the surface temperature data of the pure oxygen combustion heating furnace is obtained, and then the abnormality detection algorithm is combined, so that accurate abnormal temperature data is identified.
Description
Technical Field
The invention relates to the technical field of data processing, in particular to an operation data processing method for a pure oxygen combustion heating furnace.
Background
The pure oxygen heating furnace is high temperature heating equipment, generates high temperature gas through the combustion of pure oxygen and fuel, and heats by utilizing the radiation and convection heat transfer modes of the high temperature gas. The pure oxygen combustion heating furnace has higher combustion efficiency and lower energy consumption, and is widely used. In the use process, because the furnace wall is thinned and cracks and other abnormal conditions are caused by long-time use, if the furnace wall is not maintained in time, gas leakage occurs to the furnace wall so as to generate explosion. Therefore, the temperature data of the pure oxygen heating furnace needs to be monitored in real time and the abnormality detection is carried out, so that the abnormality is found out in time and the abnormality is treated in time.
In the prior art, common data anomaly detection includes LOF algorithm, COF algorithm, random forest algorithm, etc., and these anomaly detection algorithms can screen out anomaly temperature data through fluctuation and anomaly of temperature presentation in time sequence. However, in the pure oxygen combustion heating furnace, the surface temperature of the furnace wall is interfered by various factors, such as high-temperature exhaust flow, oxygen flow, fuel flow and the like, under the interference of the factors, the furnace wall is abnormal at a certain moment, but normal results are shown under the surface temperature data, the abnormality is not identified in the abnormality detection process, so that the abnormality detection is not timely, and the maintenance and overhaul of the pure oxygen combustion heating furnace are influenced, so that potential safety hazards are formed.
Disclosure of Invention
In order to solve the technical problem that abnormal detection is inaccurate because a plurality of factors influence abnormal detection of the surface temperature of the pure oxygen combustion heating furnace, the invention aims to provide a data processing method for the operation of the pure oxygen combustion heating furnace, which adopts the following specific technical scheme:
the invention provides a method for processing operation data of a pure oxygen combustion heating furnace, which comprises the following steps:
acquiring an associated data sequence and a surface temperature sequence of a region to be treated of the pure oxygen combustion heating furnace according to a preset adoption frequency and a sampling period; the associated data includes at least exhaust flow, fuel flow, and oxygen flow;
acquiring a first influence factor of the associated data at each moment according to the data discrete degree of the elements in the associated data sequence; fitting the historical data to obtain a surface temperature prediction model; the independent variable of the surface temperature prediction model is all associated data, the dependent variable is a surface temperature prediction value, and the surface temperature prediction model comprises the associated coefficient of all the associated data; obtaining abnormal interference factors at corresponding moments according to each association coefficient and the corresponding first influence factor;
obtaining an average surface temperature predicted value according to the associated data sequence and the surface temperature predicted model; obtaining suspected abnormal influence factors of the corresponding surface temperatures according to the deviation degree of each surface temperature in the surface temperature sequence relative to the average surface temperature predicted value;
obtaining a comprehensive abnormal influence factor of the surface temperature at a corresponding moment according to the suspected abnormal influence factor and the abnormal interference factor; and identifying abnormal temperature data according to the comprehensive abnormal influence factor and the data value of the surface temperature.
Further, the obtaining the first influence factor of the associated data at each moment according to the change characteristics of the elements in the associated data sequence includes:
obtaining a data average value and a data standard deviation of the associated data sequence, obtaining a data difference between associated data and the data average value at each moment, and taking the product of the data difference and the data standard deviation as the first influence factor of the associated data at the corresponding moment.
Further, the method for obtaining the surface temperature prediction model comprises the following steps:
and (3) counting historical data, taking each associated data as an independent variable, taking the surface temperature predicted value as an independent variable, and fitting a multi-element primary model to obtain the surface temperature predicted model.
Further, the method for acquiring the association coefficient comprises the following steps:
and the association coefficient is the absolute value of the coefficient corresponding to each association data in the multi-element primary model.
Further, the method for acquiring the abnormal interference factor comprises the following steps:
multiplying the association coefficient corresponding to each associated data at each moment by the first influence factor, and accumulating the products to obtain the abnormal interference factor at the corresponding moment.
Further, the method for obtaining the average surface temperature predicted value according to the surface temperature predicted model comprises the following steps:
acquiring an associated data mean value in each associated data sequence; and inputting the associated data mean value into the surface temperature prediction model, and outputting the average surface temperature predicted value.
Further, the obtaining the suspected abnormal influence factor of the corresponding surface temperature according to the deviation degree of each surface temperature in the surface temperature sequence comprises the following steps:
obtaining a temperature difference between each surface temperature and the average surface temperature prediction value; taking the average surface temperature predicted value as a reference value, and obtaining a temperature standard deviation of the surface temperature sequence relative to the reference value; taking the product of the temperature difference and the temperature standard deviation as a first deviation degree of corresponding surface temperature data; normalizing the temperature difference to obtain a second deviation degree of corresponding surface temperature data; and adding the first deviation degree and the second deviation degree to obtain the suspected abnormal influence factor.
Further, the obtaining the comprehensive abnormal influence factor according to the suspected abnormal influence factor and the abnormal interference factor includes:
and taking the absolute value of the difference between the abnormal interference factor and the suspected abnormal influence factor as the comprehensive abnormal influence factor.
Further, the identifying abnormal temperature data from the integrated abnormal impact factor and the data value of the surface temperature includes:
in the DBSCAN clustering method, taking the sum of the comprehensive abnormal influence factors of each surface temperature and a constant 1 as a weight, and adjusting the distance measurement between each surface temperature and other surface temperatures according to the weight to obtain a clustering result; and screening out abnormal temperature data according to the clustering result.
The invention has the following beneficial effects:
in order to avoid abnormal analysis errors caused by independently analyzing the surface temperature of the pure oxygen combustion heating furnace, the invention simultaneously collects the related data and the surface temperature data. Firstly, analyzing the discrete degree of the associated data, namely showing a first influence factor through abnormal transformation of the associated data, and further combining historical data to obtain an abnormal interference factor with stronger reference through the associated coefficient. Further processing the surface temperature data to obtain suspected abnormal factors corresponding to the surface temperature, and further obtaining comprehensive abnormal factors. The comprehensive anomaly factors simultaneously consider the anomaly condition of the surface temperature and the anomaly condition represented by the associated data, so that the accurate anomaly information can be detected by combining the anomaly detection of the surface temperature by the comprehensive anomaly factors, the anomaly detection error caused by processing the surface temperature data only is avoided, the anomaly detection precision is improved, and the maintenance and overhaul of the pure oxygen combustion heating furnace by the anomaly data can be timely identified.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a method for processing operating data of a pure oxygen combustion furnace according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of a specific implementation, structure, characteristics and effects of a method for processing operation data of a pure oxygen combustion heating furnace according to the invention, which is provided by the invention, with reference to the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the operation data processing method for the pure oxygen combustion heating furnace provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for processing operation data of a pure oxygen combustion heating furnace according to an embodiment of the present invention is shown, where the method includes:
step S1: and acquiring a related data sequence and a surface temperature sequence of a region to be treated of the pure oxygen combustion heating furnace according to the preset adoption frequency and the sampling period.
Because the analysis of the surface temperature of the pure oxygen combustion heating furnace alone can cause errors in anomaly detection, in order to ensure the accuracy of anomaly detection, the surface temperature of the pure oxygen combustion heating furnace needs to be acquired, and meanwhile, the related data of the influence temperature in the furnace needs to be acquired. Because the pure oxygen combustion heating furnace is heated based on the radiation and convection principles of high-temperature gas generated by combustion of pure oxygen and fuel, the related data influencing the surface temperature at least comprises exhaust gas flow, fuel flow and oxygen flow, namely the three basic data are key information of the influenced and determined furnace wall surface temperature, and analysis is needed based on the three data when the related characteristics are analyzed later.
It should be noted that, in other embodiments, on the basis of the three data of the exhaust flow, the fuel flow and the oxygen flow, multiple factors such as the air pressure and the pure oxygen concentration in the furnace may be added to perform the correlation analysis, the analysis method of each correlation data is consistent, and the type of the sensor required for acquiring the correlation data and the surface temperature data is well known to those skilled in the art, which is not repeated and limited herein.
In the embodiment of the invention, the working areas of the pure oxygen heating furnace comprise four working areas of a combustion chamber, a feed and discharge port, an emission chimney and a furnace tube, so that independent data anomaly analysis is required to be carried out on all the four working areas. In addition, because the temperature of the local area cannot represent the temperature information of the whole working area, even sampling points are required to be set in each working area, and data under different areas are acquired. In the embodiment of the invention, 5s is set as a sampling time interval for each sampling point, and 50 is the sampling length to obtain the associated data sequence and the surface temperature sequence. I.e. each data sequence has a length of 50 and the time interval between adjacent elements in the sequence is 5s.
It should be noted that, in order to facilitate the subsequent processing of the data, after the data of the area to be processed is obtained, the data needs to be normalized to achieve the purpose of dimension removal, that is, the subsequent data are all dimensionless data with a value range between 0 and 1.
Step S2: acquiring a first influence factor of the associated data at each moment according to the data discrete degree of the elements in the associated data sequence; fitting the historical data to obtain a surface temperature prediction model; independent variables of the surface temperature prediction model are all associated data, the dependent variables are surface temperature prediction values, and the surface temperature prediction model comprises associated coefficients of all the associated data; and obtaining abnormal interference factors at corresponding moments according to each association coefficient and the corresponding first influence factors.
In the pure oxygen combustion heating furnace, when the exhaust flow is increased, the flow speed of the high-temperature gas in the furnace is increased, so that the contact time of the high-temperature gas and the surface of the furnace wall is shorter, the energy density of other heat transfer radiation at high temperature is increased, and the temperature of the surface of the furnace wall is correspondingly increased; when the fuel flow rate is increased, the combustion reaction in the hearth is stronger, more heat is generated, and the heat is transferred to the surface of the furnace wall along with gas convection and radiation, so that the temperature of the surface of the furnace wall is increased; when the oxygen flow increases, the combustion reaction will be more complete, the temperature of the other and burning flames at the high temperature emitted will also increase, and heat is transferred to the furnace wall by radiation, so that the surface of the furnace wall receives more heat, resulting in an increase in the surface temperature. Therefore, if a micro crack is formed in the furnace wall of a certain working area or the furnace wall becomes thin, the air pressure is abnormal, high-temperature other fuel and oxygen in the furnace can generate abnormal violent flow due to the action of the air pressure, so that the flow is increased, the surface temperature of the furnace wall is increased, and abnormal data cannot be accurately identified through surface temperature data in the abnormal detection process. It is therefore necessary to analyze the correlation data, analyzing its data changes and correlation of the correlation data to the surface temperature data.
Firstly, the data change in the associated data needs to be analyzed, and in a time sequence process, if abnormal fluctuation and deviation of the data occur in the corresponding associated data, the fact that the associated data corresponding to the time period is abnormal is indicated, namely, the data reflected by the surface temperature data at the time is possibly inaccurate. Therefore, the first influence factor of the associated data at each moment is obtained according to the data discrete degree of the elements in the associated data sequence, namely, the larger the first influence factor is, the more abnormal the associated data at the corresponding moment is indicated, and the greater the influence of the abnormal condition on the surface temperature data is.
Preferably, in one embodiment of the present invention, obtaining the first influence factor of the associated data at each time instant according to the change characteristics of the elements in the associated data sequence includes:
and obtaining a data average value and a data standard deviation of the associated data sequence, obtaining a data difference between the associated data and the data average value at each moment, and taking the product of the data difference and the data standard deviation as a first influence factor of the associated data at the corresponding moment. The first influence factor is formulated as:
wherein,,is the firstThe first of the seed association dataA first influencing factor at a time instant,is the firstThe first of the associated data sequences of the associated dataThe value of the element at the moment in time,is the firstThe data mean in the associated data sequence of the associated data,is the firstData standard deviation in an associated data sequence of associated data.
As can be seen from the first influence factor formula, the larger the data standard deviation is, the larger the data difference is, which means that the larger the change of the whole elements in the sequence is, namely the more abnormal the corresponding associated data change is, the easier the influence on the surface temperature data is caused.
It should be noted that, the method for obtaining the first influence factors of the associated data sequences corresponding to each associated data is the same, that is, three first influence factors can be obtained after analysis, in the embodiment of the present invention, the first influence factor is the firstThe three first influencing factors at the moment are respectively recorded as: first influencing factor of exhaust gas flowFirst influencing factor of fuel flowAnd a first influencing factor of oxygen flow。
The first influence factor characterizes abnormality of the associated data by utilizing the discrete degree of the data through analyzing the data in the associated data sequence, so that the influence on the surface temperature is reflected. However, the degree of influence of different associated data on the surface temperature is different, so that the degree of association of each associated data on the surface temperature needs to be analyzed, and the obtained first influence factor is adjusted.
And fitting historical data, wherein the historical data comprises associated data and surface temperature data at one moment, so that a surface temperature prediction model can be obtained, and in the surface temperature prediction model, independent variables are associated data, and dependent variables are surface temperature prediction values. Therefore, the surface temperature prediction model comprises correlation coefficients of each correlation data to the surface temperature. And combining the association coefficient and the corresponding first influence factor to obtain the abnormal interference factor at the corresponding moment.
Preferably, in one embodiment of the present invention, the method for obtaining the surface temperature prediction model includes:
and (3) counting historical data, taking each associated data as an independent variable, taking a surface temperature predicted value as an independent variable, and fitting a multi-element primary model to obtain a surface temperature predicted model. Because the associated data in one embodiment of the invention comprises three data, the fitted surface temperature prediction model is a ternary primary model, and the specific expression is:
wherein,,is the firstPredicted values of the surface temperatures of the individual areas to be treated,as the data of the flow rate of the exhaust gas,as the coefficient corresponding to the exhaust gas flow rate data,as a function of the fuel flow rate data,for the coefficient corresponding to the fuel flow rate data,as the data of the flow rate of oxygen,is the coefficient corresponding to the oxygen flow data,is the model offset.
In the surface temperature prediction model, the coefficient represents the degree of change of the surface temperature predicted value due to the associated data, the model offset represents an error term between the predicted value and the true value, and the larger the absolute value of the associated coefficient is, the larger the influence of the corresponding associated data on the surface temperature data is, so that the absolute value of the coefficient corresponding to each associated data in the multi-element primary model can be used as the corresponding associated coefficient.
Preferably, in one embodiment of the present invention, the method for acquiring the abnormal interference factor includes:
multiplying the association coefficient corresponding to each associated data at each moment by the first influence factor, and accumulating the products to obtain the abnormal interference factor at the corresponding moment. The abnormal interference factor is formulated as:
wherein,,is the firstAbnormal interference factors at each moment in time,as a correlation coefficient of the flow rate of the exhaust gas,as a first influencing factor for the exhaust gas flow,as a correlation coefficient for the fuel flow rate,as a first influencing factor for the fuel flow,is the correlation coefficient of the oxygen flow,is the first influencing factor of the oxygen flow. And taking the association coefficient as the weight corresponding to the first influence factor, and carrying out weighted summation on the first influence factors of each associated data to obtain the abnormal interference factor. The obtained abnormal interference factor represents the current momentThe overall effect of each associated data within the area to be treated on the surface temperature data.
Step S3: obtaining an average surface temperature predicted value according to the associated data sequence and the surface temperature predicted model; and obtaining suspected abnormal influence factors of the corresponding surface temperatures according to the deviation degree of the surface temperatures relative to the average surface temperature predicted value in the surface temperature sequence.
After the analysis in the step S2, the overall influence of the associated data in the current area to be processed on the surface temperature data can be obtained. It is further necessary to analyze the surface temperature data itself for abnormal characteristics using the surface temperature data.
Firstly, obtaining an average surface temperature predicted value according to a surface temperature predicted model, wherein the surface temperature predicted value is obtained based on historical data, according to priori knowledge, abnormal data is extremely small compared with normal data, so that the surface temperature predicted model obtained according to the historical data integrally shows a relationship between normal surface temperature and various associated data, namely, the data obtained according to the surface temperature predicted model is also normal data, and therefore the average surface temperature predicted value obtained according to an associated data sequence and the surface temperature predicted model can represent the normal surface temperature data, and therefore, the suspected abnormal influence factors corresponding to the surface temperature can be determined based on the deviation degree of each surface temperature in the average surface temperature predicted value obtained surface temperature sequence.
It should be noted that, in an embodiment of the present invention, it is considered that some areas to be processed are larger as the working area, and different areas in the working area cannot be combined and analyzed, so for such areas to be processed, data of one sampling area may be constructed into one surface temperature sequence, that is, one area to be processed corresponds to a plurality of surface temperature sequences, and then the data in the surface temperature sequence is analyzed by the deviation degree of the same method, so that the subsequent anomaly detection is more accurate, and the moment and the position of the generation of the anomaly data can be accurately located. For convenience of description, only one surface temperature sequence is described as an example in the subsequent process.
Preferably, in one embodiment of the present invention, in consideration of the need to reduce the influence of abnormal changes of the exhaust gas flow rate, the fuel flow rate, the oxygen flow rate on the surface temperature when acquiring the reference data used in the surface temperature data deviation degree analysis process, the associated data average value in each associated data sequence is obtained; and inputting the average value of the associated data into a surface temperature prediction model, and outputting an average surface temperature predicted value.
Preferably, the obtaining the suspected abnormal influence factor of the corresponding surface temperature according to the deviation degree of each surface temperature in the surface temperature sequence in one embodiment of the invention includes:
obtaining a temperature difference between each surface temperature and the average surface temperature prediction value; taking the average surface temperature predicted value as a reference value to obtain a temperature standard deviation of the surface temperature sequence relative to the reference value; taking the product of the temperature difference and the temperature standard deviation as a first deviation degree of corresponding surface temperature data; normalizing the temperature difference to obtain a second deviation degree of the corresponding surface temperature data; and adding the first deviation degree and the second deviation degree to obtain a suspected abnormal influence factor. The formulation of the suspected anomaly impact factor is:
wherein,,is the firstA suspected anomaly impact factor for the corresponding surface temperature at each time instant,for the first degree of deviation,for the second degree of deviation the first degree of deviation,is the firstData values for the corresponding surface temperatures at each time instant,is the firstAverage surface temperature predictions under the individual areas to be treated,is a normalization function.
According to the formula of the suspected abnormal factor, the greater the temperature difference is, the greater the deviation degree of the corresponding data is; the larger the standard deviation is, the larger the fluctuation degree of the data in the corresponding sequence is, namely the larger the deviation degree is, so that the larger the suspected abnormal influence factor is, and the surface temperature at the corresponding time is the abnormal temperature.
Step S4: obtaining a comprehensive abnormal influence factor of the surface temperature at a corresponding moment according to the suspected abnormal influence factor and the abnormal interference factor; and identifying abnormal temperature data according to the data value of the comprehensive abnormal influence factor and the surface temperature.
In order to detect whether the primary operation of the pure oxygen combustion heating furnace is abnormal or not through the data characteristics of the surface temperature, the interference of the related data on the surface temperature needs to be reduced, so that the suspected abnormal influence factor and the abnormal interference factor can be combined to obtain the comprehensive abnormal influence factor of the surface temperature at the corresponding moment, the influenced degree of the surface temperature at the moment is judged according to the comprehensive influence factor, and in the subsequent abnormal detection process, the abnormal temperature data can be identified by combining the data values of the comprehensive abnormal influence factor and the surface temperature.
Preferably, in one embodiment of the present invention, obtaining the integrated anomaly impact factor from the suspected anomaly impact factor and the anomaly interference factor includes:
and taking the absolute value of the difference between the abnormal interference factor and the suspected abnormal influence factor as the comprehensive abnormal influence factor. If the abnormal interference factor is larger and the suspected abnormal influence factor is smaller, the surface data feature displayed by the surface temperature data at the moment is not abnormal data, but the surface temperature data at the moment is influenced by the related data, so that the obtained absolute value of the difference is larger, namely the comprehensive abnormal influence factor is larger; if the abnormal interference factor is smaller and the suspected abnormal factor is larger, the fact that the associated data does not affect the surface data at the moment is indicated, but the surface data already presents obvious abnormal characteristics, so that the absolute value of the obtained difference is larger, namely the comprehensive abnormal influence factor is larger, and the abnormal referential of the corresponding data can be further amplified; if the abnormal interference factor and the suspected abnormal influence factor are both smaller values, the surface temperature data at the moment is normal data, and the processing is not needed, so that the comprehensive abnormal influence factor is smaller; if the abnormal interference factor and the suspected abnormal influence factor are both larger values, the fact that larger abnormal characteristics are generated in the furnace at the moment is indicated, problems are caused on the associated data and the surface temperature, the abnormal characteristics represented by the surface temperature data at the moment are enough to be detected by an abnormal detection algorithm, and therefore the abnormal characteristics do not need to be processed, and the comprehensive abnormal influence factor is smaller.
Preferably, in one embodiment of the present invention, identifying the abnormal temperature data based on the data value of the integrated abnormal impact factor and the surface temperature includes:
because the DBSCAN cluster anomaly analysis has smaller calculation amount and faster processing speed compared with other anomaly detection algorithms, the DBSCAN cluster is used for identifying the anomaly temperature data. In the DBSCAN clustering method, taking the sum of the comprehensive abnormal influence factors of each surface temperature and a constant 1 as a weight, and adjusting the distance measurement between each surface temperature and other surface temperatures according to the weight to obtain a clustering result; and screening out abnormal temperature data according to the clustering result. I.e. the adjusted distance metric expression is:
wherein,,is the surface temperatureAnd other surface temperatures b,is the surface temperatureAnd other surface temperatures b,is the surface temperatureIs a comprehensive abnormality influencing factor of (1). It should be noted that, the distance measurement before adjustment may be obtained by using the euclidean distance obtaining method, and in other embodiments, other distance measurement algorithms may also be used, and such operations are technical means well known to those skilled in the art, and are not described herein.
After the distance measurement is adjusted, each data point can be clustered in the two-dimensional data space of each surface data mapping value through DBSCAN clustering, and in the finally obtained clustering result, the isolated data point and the sample with fewer samples in the cluster are abnormal temperature data. In the embodiment of the invention, the surface temperature data corresponding to the cluster with the least isolated data point and the sample in the cluster is used as the abnormal temperature data.
It should be noted that, in other embodiments, other anomaly detection algorithms may be used, and in the anomaly detection process, the corresponding difference metric is adjusted by combining with the comprehensive anomaly impact factor, so that the accurate detection of the anomaly temperature data can be achieved.
The staff can determine the abnormal condition of the area to be processed based on the time and the position corresponding to the obtained abnormal temperature data, and perform abnormal positioning in time sequence and space, so that the staff can conveniently perform subsequent overhaul and maintenance processing.
In summary, the embodiment of the invention acquires the associated data of the surface temperature data, and determines the abnormal interference factor at the corresponding time by performing data discrete analysis and associated analysis on the associated data. And further carrying out data deviation analysis on the surface temperature data, and determining suspected abnormal influence factors of each surface temperature data. Combining the abnormal interference factors and the suspected abnormal influence factors to obtain comprehensive abnormal influence factors, processing the surface temperature data according to the comprehensive abnormal influence factors, and identifying abnormal temperature data. According to the embodiment of the invention, the index representing the affected degree of the surface temperature data of the pure oxygen combustion heating furnace is obtained through processing and association analysis of the associated data, so that an abnormality detection algorithm is combined, and accurate abnormal temperature data is identified.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (5)
1. A method for processing operating data of a pure oxygen combustion heating furnace, the method comprising:
acquiring an associated data sequence and a surface temperature sequence of a region to be treated of the pure oxygen combustion heating furnace according to a preset adoption frequency and a sampling period; the associated data includes at least exhaust flow, fuel flow, and oxygen flow;
acquiring a first influence factor of the associated data at each moment according to the data discrete degree of the elements in the associated data sequence; fitting the historical data to obtain a surface temperature prediction model; the independent variable of the surface temperature prediction model is all associated data, the dependent variable is a surface temperature prediction value, and the surface temperature prediction model comprises the associated coefficient of all the associated data; obtaining abnormal interference factors at corresponding moments according to each association coefficient and the corresponding first influence factor;
obtaining an average surface temperature predicted value according to the associated data sequence and the surface temperature predicted model; obtaining suspected abnormal influence factors of the corresponding surface temperatures according to the deviation degree of each surface temperature in the surface temperature sequence relative to the average surface temperature predicted value;
obtaining a comprehensive abnormal influence factor of the surface temperature at a corresponding moment according to the suspected abnormal influence factor and the abnormal interference factor; identifying abnormal temperature data according to the comprehensive abnormal influence factors and the data value of the surface temperature;
the obtaining the first influence factor of the associated data at each moment according to the change characteristics of the elements in the associated data sequence comprises the following steps:
obtaining a data average value and a data standard deviation of the associated data sequence, obtaining a data difference between associated data and the data average value at each moment, and taking the product of the data difference and the data standard deviation as the first influence factor of the associated data at the corresponding moment;
the method for acquiring the surface temperature prediction model comprises the following steps:
the statistical historical data, each associated data is used as an independent variable, the surface temperature predicted value is used as an independent variable, and a multi-element primary model is fitted to obtain the surface temperature predicted model;
the obtaining the suspected abnormal influence factor of the corresponding surface temperature according to the deviation degree of each surface temperature in the surface temperature sequence comprises the following steps:
obtaining a temperature difference between each surface temperature and the average surface temperature prediction value; taking the average surface temperature predicted value as a reference value, and obtaining a temperature standard deviation of the surface temperature sequence relative to the reference value; taking the product of the temperature difference and the temperature standard deviation as a first deviation degree of corresponding surface temperature data; normalizing the temperature difference to obtain a second deviation degree of corresponding surface temperature data; adding the first deviation degree and the second deviation degree to obtain the suspected abnormal influence factor;
the identifying abnormal temperature data according to the integrated abnormal influence factor and the data value of the surface temperature comprises:
in the DBSCAN clustering method, taking the sum of the comprehensive abnormal influence factors of each surface temperature and a constant 1 as a weight, and adjusting the distance measurement between each surface temperature and other surface temperatures according to the weight to obtain a clustering result; and screening out abnormal temperature data according to the clustering result.
2. The method for processing operation data of a pure oxygen combustion heating furnace according to claim 1, wherein the method for acquiring the correlation coefficient comprises:
and the association coefficient is the absolute value of the coefficient corresponding to each association data in the multi-element primary model.
3. The method for processing operation data of a pure oxygen combustion heating furnace according to claim 1, wherein the method for acquiring the abnormal interference factor comprises:
multiplying the association coefficient corresponding to each associated data at each moment by the first influence factor, and accumulating the products to obtain the abnormal interference factor at the corresponding moment.
4. The method for processing operation data of a pure oxygen combustion heating furnace according to claim 1, wherein the obtaining method for obtaining an average surface temperature predicted value according to the surface temperature prediction model comprises:
acquiring an associated data mean value in each associated data sequence; and inputting the associated data mean value into the surface temperature prediction model, and outputting the average surface temperature predicted value.
5. The method for processing operation data of a pure oxygen combustion heating furnace according to claim 1, wherein the obtaining a comprehensive abnormality influence factor from the suspected abnormality influence factor and the abnormality influence factor comprises:
and taking the absolute value of the difference between the abnormal interference factor and the suspected abnormal influence factor as the comprehensive abnormal influence factor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310609217.1A CN116340795B (en) | 2023-05-29 | 2023-05-29 | Operation data processing method for pure oxygen combustion heating furnace |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310609217.1A CN116340795B (en) | 2023-05-29 | 2023-05-29 | Operation data processing method for pure oxygen combustion heating furnace |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116340795A CN116340795A (en) | 2023-06-27 |
| CN116340795B true CN116340795B (en) | 2023-07-28 |
Family
ID=86888040
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310609217.1A Active CN116340795B (en) | 2023-05-29 | 2023-05-29 | Operation data processing method for pure oxygen combustion heating furnace |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116340795B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116611017B (en) * | 2023-07-17 | 2023-09-19 | 山东一然环保科技有限公司 | Nitrogen oxide emission detection method of low-nitrogen combustion heating furnace |
| CN116680446B (en) * | 2023-08-03 | 2023-09-29 | 山东青上化工有限公司 | Intelligent acquisition method and system for fermentation production data of ecological fertilizer containing bacteria |
| CN116805065B (en) * | 2023-08-25 | 2023-11-14 | 山东荣信集团有限公司 | Intelligent management method for monitoring data of coal powder heating furnace burner |
| CN116821833B (en) * | 2023-08-29 | 2023-11-14 | 好林(威海)新材料有限公司 | Data anomaly detection method for machine filter cloth adhesion |
| CN117251744B (en) * | 2023-11-16 | 2024-03-15 | 苏州科恩新能科技股份有限公司 | Kiln operation abnormal state detection method based on artificial intelligence |
| CN117666365B (en) * | 2024-02-01 | 2024-04-16 | 云扬八方工程有限公司 | Intelligent dust removal system of coke dry quenching furnace |
| CN118724581A (en) * | 2024-09-02 | 2024-10-01 | 营口仁威矿产有限公司 | Calcining preparation process of high-quality spinel |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130023886A (en) * | 2011-08-30 | 2013-03-08 | 현대제철 주식회사 | Method for predicting variation of furnace heat for blast furnace |
| CN112285287A (en) * | 2020-10-19 | 2021-01-29 | 广东长天思源环保科技股份有限公司 | Flue gas online monitoring data preprocessing method |
| CN116108355A (en) * | 2023-04-12 | 2023-05-12 | 广东顺德禾庄能源科技有限公司 | Management method for cloud hot water monitoring platform data |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5640647B2 (en) * | 2010-10-26 | 2014-12-17 | 新日鐵住金株式会社 | Method and apparatus for measuring surface temperature of steel in furnace |
| CN106777652B (en) * | 2016-12-09 | 2019-12-17 | 中冶赛迪工程技术股份有限公司 | method for predicting air permeability of blast furnace |
| CN113570145A (en) * | 2021-07-30 | 2021-10-29 | 安徽工业大学 | Prediction method for temperature of dead material column of iron-making blast furnace core |
| WO2023036427A1 (en) * | 2021-09-09 | 2023-03-16 | Sumitomo SHI FW Energia Oy | Method for determining a local temperature anomaly in a fluidized bed of a combustion boiler, method for calibrating a numerical model of a fluidized bed of a combustion boiler, method for estimating risk of fluidized bed combustion boiler bed sintering, method of controlling a fluidized bed boiler, as well as a combustion boiler |
| CN115858630B (en) * | 2023-02-21 | 2023-05-09 | 新风光电子科技股份有限公司 | Abnormality detection method for energy storage data of energy storage power station |
-
2023
- 2023-05-29 CN CN202310609217.1A patent/CN116340795B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20130023886A (en) * | 2011-08-30 | 2013-03-08 | 현대제철 주식회사 | Method for predicting variation of furnace heat for blast furnace |
| CN112285287A (en) * | 2020-10-19 | 2021-01-29 | 广东长天思源环保科技股份有限公司 | Flue gas online monitoring data preprocessing method |
| CN116108355A (en) * | 2023-04-12 | 2023-05-12 | 广东顺德禾庄能源科技有限公司 | Management method for cloud hot water monitoring platform data |
Non-Patent Citations (1)
| Title |
|---|
| 钢铁企业高炉煤气发生量异常数据检测;王红君;白鹏;赵辉;岳有军;;化工自动化及仪表(06);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116340795A (en) | 2023-06-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116340795B (en) | Operation data processing method for pure oxygen combustion heating furnace | |
| CN116659589B (en) | Donkey-hide gelatin cake preservation environment monitoring method based on data analysis | |
| CN111352408B (en) | Multi-working-condition process industrial process fault detection method based on evidence K nearest neighbor | |
| TW201037779A (en) | Etching apparatus, analysis apparatus, etching treatment method, and etching treatment program | |
| CN116611017B (en) | Nitrogen oxide emission detection method of low-nitrogen combustion heating furnace | |
| CN107702905B (en) | Rubber ring reliable life prediction method based on Weibull distribution | |
| CN108897354B (en) | Aluminum smelting process hearth temperature prediction method based on deep belief network | |
| CN116825243B (en) | Multi-source data-based thermal barrier coating service life prediction method and system | |
| CN118194026B (en) | Gas power generation data analysis system | |
| CN117573667A (en) | Optimized cleaning method and system for coal mine inter-well data | |
| CN112949735A (en) | Liquid hazardous chemical substance volatile concentration abnormity discovery method based on outlier data mining | |
| CN110097175B (en) | Embedded DCNN compression method for heavy pipe identification of ethylene cracking furnace pipe | |
| Wohlmeister et al. | Differentiation of Candida albicans, Candida glabrata, and Candida krusei by FT-IR and chemometrics by CHROMagar™ Candida | |
| US20230333023A1 (en) | Method of non-contact material identification, method of non-contact temperature identification, and method for identifying progress of heat treatment processing | |
| JP3732768B2 (en) | Semiconductor processing equipment | |
| CN117078023B (en) | Kiln fault risk assessment method based on big data analysis | |
| CN104460653B (en) | A kind of heating furnace monitoring system evaluation method | |
| CN109214268B (en) | Packed tower flooding state online monitoring method based on integrated manifold learning | |
| Harrington et al. | Validation using sensitivity and target transform factor analyses of neural network models for classifying bacteria from mass spectra | |
| JP6643202B2 (en) | Plasma processing apparatus and analysis method for analyzing plasma processing data | |
| CN113688895A (en) | Method and system for detecting abnormal firing zone of ceramic roller kiln based on simplified KECA | |
| WO2020253485A1 (en) | Dbscan-based outer surface temperature measurement method for cracking furnace tube | |
| Hasegawa et al. | Towards automated gas leak detection through cluster analysis of mass spectrometer data | |
| CN117556274B (en) | Temperature data anomaly analysis method for heat pipe backboard | |
| CN112229212A (en) | Roller kiln energy consumption abnormity detection method based on dynamic principal component analysis |
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 |