Cement production line carbon emission data cross coupling verification method

Abstract

The invention relates to the technical field of cement mine digital mining engineering, and discloses a cement production line carbon emission data cross coupling verification method, which comprises a carbon emission online monitoring module A1, a carbon emission online accounting module A2 and a big data cross coupling model A3, the carbon emission on-line monitoring module A1 is a carbon emission on-line monitoring system mounted at a pollution source discharge port at the kiln tail of a cement production line; the carbon emission online monitoring system comprises automatic monitoring of flue gas CO2 concentration, flue gas flow rate, flue gas humidity, flue gas temperature and flue gas pressure parameters and real-time carbon emission obtained through calculation of the automatic monitoring parameters. According to the invention, cross checking and data coupling can be carried out on the carbon emission data obtained by the two methods through a big data model, a universal automatic monitoring technology and the accounting business of a cement production line are fused, the influence of alternative fuel on the monitoring of the carbon emission in the cement industry is solved, and the effective connection of an accounting method and online monitoring is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of digital mining engineering for cement mines, and particularly to a method for cross-coupling verification of carbon emission data in a cement production line. Background Technique

[0002] The carbon emission measurement methods in process industries mainly include the accounting-based method and the continuous monitoring-based method. The carbon emission measurement method in the cement industry mainly uses the accounting method. Most of the data of the accounting method comes from the statement vouchers provided by cement factories, and the data reliability is lacking. The measurement of clinker output has always been a difficult problem to be solved in the cement industry. Moreover, the carbon accounting work belongs to ex-post accounting, with the problem of time lag, and it is impossible to timely discover the problems existing in the current enterprise production operation process. Therefore, if the cement industry starts to be included in the carbon trading market, relying solely on the accounting method, it is difficult to effectively guarantee the monitoring accuracy and credibility, and it is difficult to effectively avoid risks such as data fraud. At the same time, it is also impossible to form a highly executable standardized plan, and the credibility and reliability of the data may not be able to support carbon market transactions.

[0003] Carrying out the direct monitoring of carbon emissions in cement factories to obtain "first-hand" data is particularly important. The carbon emission measurement and monitoring technologies represented by continuous monitoring have many advantages and highlights. However, since the carbon online monitoring technology field in current cement factories has not formed a standardized and replicable cement industry CO2 monitoring technology system, nor has it formed a standardized carbon online monitoring data management method, and there are problems such as a large workload of manual maintenance and calibration for online monitoring instruments, resulting in the ineffective promotion of the method for continuously monitoring carbon emissions in cement factories. It is generally recognized that the uncertainty of the online monitoring method is better than that of the accounting method, but the accounting method also has its advantages. The carbon emission data obtained by the online monitoring method includes the carbon dioxide emissions generated by alternative fuels, and the use of alternative fuels is one of the important ways for the cement industry to save energy and reduce carbon emissions. In the accounting guidelines, the use of alternative fuels adopts carbon emission factor deduction or even does not include it in the total carbon emissions of cement factories to encourage cement factories to use alternative fuels on a large scale. The online monitoring method cannot eliminate the impact of alternative fuels on carbon emissions.

[0004] Therefore, neither the nuclear algorithm nor the online monitoring method can solve the problems existing in the accurate measurement of carbon emissions in cement enterprises. What the cement industry needs is to combine the comprehensiveness of the nuclear algorithm and the accuracy advantages of the online monitoring method, formulate a multi-level carbon emission data management and cross-check mechanism, promote the evolution of carbon emission measurement and supervision to a higher-level and higher-level technical stage, realize the systematic improvement from accounting to monitoring measurement, comprehensively improve the quality of carbon emission data. Patent CN114166998 proposes a carbon emission measurement scheme and system for cement production enterprises. The carbon emissions are calculated based on the summary of the directly monitored carbon emissions and the verified carbon emissions. However, the data summary method used in this method is data summation, which only solves the problem of real-time calculation of carbon emissions and cannot eliminate the impact of alternative fuels on the accurate measurement of carbon emissions. Patent CN114705260 proposes an online measurement system for the gas emissions of a cylindrical chimney based on multi-point monitoring, which can accurately measure the flow rate and carbon dioxide emissions by fully considering the velocity distribution inside the interface, but does not combine the monitoring technology with the carbon emission business of the cement production line and still cannot truly and accurately evaluate the carbon emissions. Summary of the Invention

[0005] (1) Technical problems to be solved

[0006] Aiming at the deficiencies of the existing technology, the present invention provides a cross-coupling verification method for carbon emission data of a cement production line. Through a big data model, the carbon emission data obtained by two methods are cross-verified and data-coupled, and the general automatic monitoring technology is integrated with the accounting business of the cement production line to solve the impact of alternative fuels on the monitoring of carbon emissions in the cement industry and realize the effective connection between the nuclear algorithm and online monitoring.

[0007] (2) Technical solutions

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A cross-coupling verification method for carbon emission data of a cement production line, including an online carbon emission monitoring module A1, an online carbon emission accounting module A2, and a big data cross-coupling model A3. The online carbon emission monitoring module A1 is an online carbon emission monitoring system installed at the exhaust port of the pollution source at the kiln tail of the cement production line. The online carbon emission monitoring system includes the automatic monitoring of parameters such as flue gas CO2 concentration, flue gas flow rate, flue gas humidity, flue gas temperature, and flue gas pressure, and the real-time carbon emissions calculated through the automatic monitoring parameters;

[0010] The online carbon emission accounting module A2 calculates the real-time carbon emissions obtained based on the online measurement of clinker production, the online measurement of fossil fuel consumption, the chemical composition of clinker, and the calorific value of fossil fuels. The real-time carbon emissions include real-time fossil fuel combustion emissions and real-time process emissions. The equipment for online measurement of clinker production is a clinker rail weigher, which is installed under the clinker conveyor of the cement production line after the grate cooler. The clinker rail weigher consists of high-precision sensors, a material balance algorithm, and a dynamic weighing instrument. The material balance algorithm is used to eliminate the influence of the tension between the chain buckets of the grate cooler clinker conveyor and the weight of the clinker conveyor on the weighing accuracy.

[0011] The big data cross-coupling model A3 is a model for data cross-calibration and coupling verification between the carbon online monitoring module based on machine learning and the online accounting module. Through the big data cross-coupling model, the real-time accounting and real-time online monitoring results are fused.

[0012] Furthermore, the online carbon emission monitoring module A1 calculates the carbon dioxide emissions based on the real-time monitoring data of the cement kiln tail emission source.

[0013] Based on the above-mentioned scheme, the online carbon emission accounting module is the carbon dioxide emissions of clinker production obtained by accounting based on the online measurement data of fossil fuel consumption and clinker production.

[0014] As a further scheme of the present invention, the experimental data of the big data cross-coupling model consists of an offline model training process and an online model operation process.

[0015] Furthermore, the steps of the offline model training process are as follows:

[0016] First step, according to the carbon emission data obtained by the online carbon emission monitoring module and the online carbon emission accounting module, several working conditions are divided according to the similarity of the data.

[0017] Second step, for each working condition, a machine learning or deep learning regression model is selected, and it is trained using the automatically monitored carbon emission data, clinker online measurement data, raw coal consumption online measurement data, and non-carbonate alternative fuel consumption data to obtain an online monitoring data calibration value prediction model.

[0018] Third step, by comparing the online monitoring data calibration value and the online accounting data, if the gap error between the two is within K, it can be considered that the calibration calculation method is effective; if the gap error between the two is outside K, the parameters of the calibration calculation method are optimized with the result of the accounting method as the target until the error is less than K.

[0019] Based on the above-mentioned scheme, the steps of the online model operation process are as follows:

[0020] First step, determine which working condition the real-time automatic monitoring data belongs to;

[0021] Second step, according to different working conditions, calculate the monitoring data calibration value based on data such as automatically monitored carbon emission data, online metering data of clinker, online metering data of raw coal consumption, and data of non-carbonate alternative fuel consumption;

[0022] Third step, record and save the calibration value, fill in the abnormal value of the online monitoring data. When new carbon emission accounting data is obtained, compare the average value of the calibration values in the past within the same range with this accounting value. If the error is less than K, it means that the calibration algorithm does not need to be modified; if the error is greater than K, calibrate the parameters of the target optimization calibration calculation method corresponding to this working condition.

[0023] As a further solution of the present invention, the allowable error range of the online monitoring data calibration value and the online accounting data is 1-5%.

[0024] Furthermore, the time granularity for calculating the online monitoring data calibration value is T0, and the time is 1-30 min. The time granularity for obtaining new real-time carbon emission accounting data is T1, and the time is 1-24 h (III) Beneficial effects

[0025] Compared with the prior art, the present invention provides a cross-coupling verification method for carbon emission data of a cement production line, having the following beneficial effects:

[0026] 1. In the present invention, through the comprehensiveness of the comprehensive accounting method and the accuracy advantage of the online monitoring method, a multi-level carbon emission data management and cross-check mechanism is formulated. Combining the online metering of clinker and raw coal, the real-time accounting and real-time online monitoring technologies are integrated, and a calibration optimization model is established to realize the effective connection between the accounting method and the online monitoring. The two check each other, further improving the data credibility of carbon emission measurement and realizing the accurate measurement of carbon dioxide emissions of cement enterprises.

[0027] 2. The present invention uses a clinker rail scale to measure the clinker output in real time, directly measuring the clinker output from the end of the production process. Compared with the material consumption coefficient method and the manual inventory method, it can accurately measure the clinker output, effectively eliminating the human intervention in the clinker output accounting process. Compared with the method of measuring the volume of the clinker by the intelligent inventory method to calculate the clinker output, it effectively reduces the influence of the clinker bulk density on the clinker output measurement, and improves the credibility of the carbon emissions calculated by the accounting method. Description of the drawings

[0028] Figure 1 It is a schematic diagram of the principle of a cross-coupling verification method for carbon emission data of a cement production line proposed by the present invention;

[0029] Figure 2Schematic diagram of using a clinker output weighbridge for a method of cross-coupled verification of carbon emission data in a cement production line proposed by the present invention;

[0030] Figure 3 Schematic diagram of the installation position of the clinker output rail weighing scale for a method of cross-coupled verification of carbon emission data in a cement production line proposed by the present invention;

[0031] Figure 4 Technical roadmap of the big data cross-coupled model system for a method of cross-coupled verification of carbon emission data in a cement production line proposed by the present invention. Detailed implementation manners

[0032] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0033] Refer to Figures 1-4 , an embodiment of the present invention provides a technical solution: a method for cross-coupled verification of carbon emission data in a cement production line, including an on-line carbon emission monitoring module A1, an on-line carbon emission accounting module A2, and a big data cross-coupled model A3.

[0034] As Figure 1 shown, the on-line carbon emission monitoring module A1 is an on-line carbon emission monitoring system installed at the exhaust port of the kiln tail pollution source of the cement production line. For a cement production line with multiple pollution source exhaust ports, the on-line carbon emission monitoring system should include all the exhaust ports of the kiln tail pollution sources. The on-line carbon emission monitoring system should include the automatic monitoring of flue gas CO2 concentration, flue gas flow rate, flue gas humidity, flue gas temperature, and flue gas pressure parameters. Among the above monitoring parameters, the flue gas humidity, flue gas temperature, and flue gas pressure parameters can be a new system or shared with the original pollutant CEMS. The real-time carbon emissions obtained by calculating through the above automatic monitoring parameters;

[0035]

[0036] Among them, Qb is the hourly carbon emission of the cement production line, is the average flue gas flow rate at the measurement section; is the carbon dioxide mass concentration of the kiln tail chimney; F is the cross-sectional area of the measurement section; is the moisture content of the kiln tail flue gas; is the average temperature and average pressure of the kiln tail chimney flue gas; P h is the local atmospheric pressure. The on-line monitored carbon emissions obtained by calculating through the above formula are hourly data.

[0037] Furthermore, by using the same method, the online monitored carbon emission data for each day and month can be calculated and used for cross-comparison with the carbon emission data accounted for at the same time granularity.

[0038] In the carbon emission online monitoring system of the present invention, the flue gas velocity monitoring can adopt online ultrasonic flow monitoring, online single-point Pitot tube flow monitoring, online multi-point Pitot tube flow monitoring, online matrix Pitot tube flow monitoring, and online three-dimensional Pitot tube flow monitoring. The relative error of the flow velocity monitoring should not be lower than 5%. The position of the flue gas velocity monitoring in the carbon emission online monitoring system should be selected at the vertical chimney position 4D in front and 2D behind, where D is the chimney diameter. If the position of the flue gas velocity monitoring in the carbon emission online monitoring system does not meet the vertical chimney position of 4D in front and 2D behind, fluid numerical calculation is required. The flue gas CO2 concentration monitoring in the carbon emission online monitoring system can adopt the cold-dry extraction non-dispersive infrared absorption spectroscopy technology, and the relative error of the concentration monitoring should not be lower than 3%. The flue gas humidity monitoring in the carbon emission online monitoring system can adopt the in-situ or extraction capacitance method, and the relative error of the humidity monitoring should not be lower than 5%.

[0039] In particular, the flue gas humidity monitoring in the carbon emission online monitoring system can also adopt the technology of soft measurement by big data model. By establishing an algorithm model of humidity data with the oxygen content obtained from environmental protection CEMS monitoring, flue gas temperature, and carbon dioxide concentration obtained from the carbon emission online monitoring module, the real-time flue gas humidity data can be calculated through the big data algorithm model and compared with the real-time monitoring system to realize the abnormal warning of the online monitoring system.

[0040] It should be noted that the online carbon emission accounting module calculates the real-time accounted carbon emissions based on the online measurement of clinker production, online measurement of fossil fuel consumption, clinker chemical composition, and fossil fuel calorific value. The online carbon emission accounting includes real-time fossil fuel combustion emissions and real-time process emissions. The clinker production online measurement device is a clinker rail weigher, which directly measures the clinker production from the end of the production process. The clinker rail weigher is installed under the clinker conveyor of the grate cooler in the cement production line. The clinker rail weigher consists of high-precision sensors, material balance algorithms, and dynamic weighing instruments. Among them, the material balance algorithm is used to eliminate the influence of the tension between the chain buckets of the grate cooler clinker conveyor and the weight of the clinker conveyor on the weighing accuracy. The carbon dioxide emissions from clinker production calculated based on the online measurement data of fossil fuel consumption and clinker production are calculated according to the following formula:

[0041] E ck =E ck,燃烧 +E ck,过程

[0042]

[0043] Among them, E ck is the hourly carbon emission calculated online, E ck,燃烧 is the carbon emission from fossil fuel combustion obtained by online accounting, E ck,过程 is the process carbon emission obtained by online accounting, FC ck is the fossil fuel consumption obtained by real-time measurement, NCV ar is the hourly calorific value of fossil fuel obtained by the quality system, CC and OF are the carbon content per unit calorific value and carbon oxidation rate of fossil fuel, Q ck and Q p are the clinker output and non-carbonate raw material consumption obtained by real-time measurement, EF ck is the process emission factor of clinker, EF p is the deduction coefficient of non-carbonate raw materials.

[0044] It should be noted that for the real-time measurement of fossil fuel consumption, a rotor scale or a belt scale can be used. The real-time measurement of fossil fuel consumption can also be carried out by measuring the gas flow rate and pulverized coal concentration in the metering coal delivery pipeline, and the real-time consumption of fossil fuel can be obtained through calculation. The calorific value of fossil fuel can be analyzed by laboratory calorific value analysis or online analyzer calorific value analysis. The metering of non-carbonate raw material consumption can be carried out by a belt scale. The carbon content per unit calorific value of fossil fuel, carbon oxidation rate, process emission factor of clinker, and deduction coefficient of non-carbonate raw materials are replaced by the default values issued by the Ministry of Ecology and Environment. The monthly cumulative result of the real-time measurement of fossil fuel consumption can be compared with the monthly fossil fuel consumption calculated by the nuclear method. The monthly fossil fuel consumption is calculated by "incoming quantity + beginning inventory - ending inventory - external sales volume". The monthly cumulative result of the real-time measurement of clinker output can be compared with the monthly clinker output calculated by the nuclear method. The monthly clinker output is obtained by calculating "consumption + external sales volume + ending inventory - beginning inventory - purchase quantity".

[0045] Furthermore, combined with the production data of the cement plant, a real-time comparison and verification system for clinker output and fossil fuel consumption based on a big data model can be established. Based on the massive production data of the production line, the clinker output and fossil fuel consumption can be predicted. If the measured data deviates from the predicted value generated by the comparison and verification system, a warning will be issued, and the predicted value will be filled into the online measurement data to ensure the continuity of the real-time measurement data.

[0046] Accurate;

[0047] The big data cross-coupling model A3 is a data cross-calibration and coupling test model based on machine learning for the carbon online monitoring module and the online accounting module. The real-time accounting and real-time online monitoring technologies are integrated through the big data cross-coupling model, which consists of the offline model training process and the online model operation process of the cross-coupling verification data model.

[0048] Further, the steps of the offline model training process are as follows:

[0049] In the first step, according to the carbon emission data obtained from the carbon emission online monitoring module and the carbon emission accounting module, several working conditions are divided based on the similarity of the data.

[0050] In the second step, for each working condition, a machine learning or deep learning regression model is selected and trained using the automatically monitored carbon emission data, clinker online metering data, raw coal consumption online metering data, and non-carbonate alternative fuel consumption data to obtain an online monitoring data calibration value prediction model.

[0051] In the third step, by comparing the online monitoring data calibration value and the online accounting data, if the difference error between the two is within K, it can be considered that the calibration calculation method is effective; if the difference error between the two is outside K, the parameters of the calibration calculation method are optimized with the result of the accounting method as the target until the error is less than K.

[0052] Further, the steps of the online model operation process are as follows:

[0053] In the first step, it is judged which working condition the real-time automatic monitoring data belongs to;

[0054] In the second step, according to different working conditions, the calibration value of the monitoring data is calculated based on the automatically monitored carbon emission data, clinker online metering data, raw coal consumption online metering data, non-carbonate alternative fuel consumption data, etc., and the calculation time granularity is T0;

[0055] In the third step, the calibration value is recorded and saved, and the outliers in the online monitoring data are filled. When new carbon emission accounting data is obtained (time granularity is T1), the average value of the calibration values in the past same range is compared with the accounting value. If the error is less than K, it means that the calibration algorithm does not need to be modified; if the error is greater than K, the parameters of the target optimization calibration calculation method corresponding to this working condition are calibrated.

[0056] In particular, the allowable error range between the online monitoring data calibration value and the online accounting data is 1-5%, the time granularity for calculating the online monitoring data calibration value is T0, generally 1-30 minutes, and the time granularity T1 for obtaining new real-time carbon emission accounting data is generally 1-24 hours.

[0057] In the description in this article, it should be noted that relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device.

Claims

1. A method for cross-coupling verification of carbon emission data of a cement production line, characterized in that: It includes a carbon emission online monitoring module A1, a carbon emission online accounting module A2 and a big data cross-coupling model A3. The carbon emission online monitoring module A1 is an online carbon emission monitoring system installed at the outlet of the pollution source at the tail of the cement production line kiln. The carbon emission online monitoring system includes automatic monitoring of flue gas CO2 concentration, flue gas flow rate, flue gas humidity, flue gas temperature and flue gas pressure parameters, and real-time carbon emissions calculated by the automatic monitoring parameters; The carbon emission online accounting module A2 is a real-time accounting carbon emission calculated based on the online measurement of clinker output, the online measurement of fossil fuel consumption, the chemical composition of clinker and the calorific value of fossil fuels. The real-time accounting carbon emission includes real-time fossil fuel combustion emissions and real-time process emissions. The equipment for online measurement of clinker output is a clinker track weighing scale, which is installed below the clinker conveyor of the grate cooler of the cement production line. The clinker track weighing scale is composed of a high-precision sensor, a material weighing algorithm, and a dynamic weighing instrument. The material weighing algorithm is used to eliminate the influence of the tension between the chain buckets of the grate cooler clinker conveyor and the weight of the clinker conveyor on the weighing accuracy; The big data cross-coupling model A3 is a model for data cross-calibration and coupling verification of the carbon online monitoring module and the online accounting module based on machine learning. The real-time accounting and real-time online monitoring results are integrated through the big data cross-coupling model.

2. According to claim 1, a cement production line carbon emission data cross-coupling verification method is characterized in that: The carbon emission online monitoring module A1 calculates carbon dioxide emissions based on real-time monitoring data of cement kiln tail emission sources.

3. According to claim 1, a cement production line carbon emission data cross-coupling verification method is characterized in that: The carbon emission online calculation module is used to calculate the carbon dioxide emissions from clinker production based on the online measurement of fossil fuel consumption and the online measurement of clinker output data.

4. According to claim 1, a cement production line carbon emission data cross-coupling verification method is characterized in that: The experimental data of the big data cross-coupling model consists of an offline model training process and an online model running process.

5. A cement production line carbon emission data cross-coupling verification method according to claim 4, characterized in that: The offline model training process steps are as follows: In the first step, the carbon emission data obtained by the carbon emission online monitoring module and the carbon emission online accounting module are divided into several working conditions according to the similarity of the data. In the second step, for each working condition, a machine learning or deep learning regression model is selected, and the automatic monitoring carbon emission data, clinker online metering data, raw coal consumption online metering data and non-carbonate alternative fuel consumption data are used for training to obtain an online monitoring data calibration value prediction model. The third step is to compare the calibration value of the online monitoring data and the online calculation data. If the difference error between the two is within K, the calibration calculation method can be considered effective. If the difference error between the two is outside K, the parameters of the calibration calculation method are optimized based on the results of the calculation method until the error is less than K.

6. A cement production line carbon emission data cross-coupling verification method according to claim 4, characterized in that: The steps of the online model operation process are as follows: The first step is to determine which working condition the real-time automatic monitoring data belongs to; The second step is to calculate the monitoring data calibration value according to different working conditions based on the automatic monitoring carbon emission data, clinker online measurement data, raw coal consumption online measurement data, non-carbonate alternative fuel consumption data and other data; The third step is to record and save the calibration value and fill in the abnormal value of the online monitoring data. When the new carbon emission accounting data is obtained, the average value of the calibration value in the same range in the past is compared with the accounting value. If the error is less than K, it means that the calibration algorithm does not need to be modified; if the error is greater than K, calibrate the parameters of the target optimization calibration calculation method corresponding to the working condition.

7. A cement production line carbon emission data cross-coupling verification method according to claim 5, characterized in that: The error tolerance range of the online monitoring data calibration value and the online calculation data is 1-5%.

8. A cement production line carbon emission data cross-coupling verification method according to claim 6, characterized in that: The time granularity for calculating the calibration value of the online monitoring data is T0, which is 1-30 minutes. The time granularity for obtaining new carbon emission real-time accounting data is T1, which is 1-24 hours.

Images