A method and system for real-time monitoring of carbon emissions from coal-fired power plants

Abstract

This invention discloses a method and system for real-time monitoring of carbon emissions from coal-fired power plants, relating to the field of carbon emission measurement. The invention first optimizes the accuracy of carbon content detection in coal before it enters the furnace by adjusting the number of LIBS detection probes on the coal feed belt. Then, based on the detected carbon content data and corresponding power consumption scenarios, the coal is replaced before entering the furnace, which helps save fuel and improve power generation efficiency. Finally, the combustion parameters of the corresponding burners are adjusted based on the carbon content data of the replaced coal, ensuring complete combustion of the coal. This maximizes fuel utilization and reduces the proportion of harmful gases in the final carbon emissions, solving the problem of traditional fixed-system coal combustion, which cannot dynamically adjust the coal quality based on power consumption scenarios, leading to fuel waste and low power generation efficiency.

Description

Technical Field

[0001] This invention belongs to the field of carbon emission measurement, and specifically relates to a method and system for real-time monitoring of carbon emission indicators of coal-fired power plants. Background Technology

[0002] Currently, when coal-fired power plants use burners to burn coal, they cannot dynamically adjust the quality of the coal to be burned according to the electricity consumption scenario (such as peak electricity consumption period and off-peak electricity consumption period), resulting in fuel waste and low power generation efficiency. Summary of the Invention

[0003] To address the problems in related technologies, this invention proposes a method and system for real-time monitoring of carbon emission indicators from coal-fired power plants, in order to overcome the aforementioned technical problems existing in the current related technologies.

[0004] To solve the above-mentioned technical problems, the present invention is achieved through the following technical solution:

[0005] This invention relates to a method for real-time monitoring of carbon emissions from coal-fired power plants, comprising the following steps:

[0006] S1. Collect data from multiple historical studies on the use of the LIBS system to detect the carbon content in coal entering the furnace;

[0007] S2. Construct an error mapping model for coal carbon content detection in the coal feed zone, based on the data collected in S1.

[0008] S3. Collect parameter data of the coal feed zone during the current coal quality burning process of the coal-fired power plant and the corresponding number of LIBS detection probes, and input them into the coal carbon content detection error mapping model of the coal feed zone for mapping. Then, adjust the current initial number of LIBS detection probes in the coal feed zone according to the mapping results to obtain the current final number of LIBS detection probes in the coal feed zone.

[0009] S4. Based on the data of the number of LIBS detection probes in the current coal feed belt, the carbon content of the coal in the current coal feed belt is detected, and the corresponding power consumption scenario type is used to replace the coal in the current coal feed belt, so as to obtain the current coal quality after replacement.

[0010] S5. Adjust the current coal combustion parameter data of various types in conjunction with the carbon content data of the current replaced coal to obtain the current final coal combustion parameter dataset;

[0011] The carbon emissions after combustion are measured in conjunction with the current final coal quality combustion parameter dataset and the current changed coal quality.

[0012] Preferably, step S1 includes the following steps:

[0013] S11. Set several parameter types related to the number of detection probes used to detect the carbon content in the coal feed zone of a coal-fired power plant using the LIBS system, and obtain a set of parameter types affecting the detection of carbon content in the coal feed zone.

[0014] S12. In conjunction with the aforementioned set of parameters affecting carbon content detection in the coal feed zone, collect data on the number of LIBS probes, carbon content detection data, actual carbon content data, and corresponding parameters affecting carbon content detection in the coal feed zone from multiple historical instances of using the LIBS system to detect carbon content in coal feed. This yields a historical set of LIBS probe counts, a historical dataset of coal carbon content detection, a historical dataset of actual coal carbon content, and a matrix of historical parameters affecting carbon content detection in the coal feed zone.

[0015] By defining a set of parameter types affecting the detection of carbon content in the coal feed stream, several parameters that influence the data settings of the LIBS system's detection probes are determined when detecting carbon content. This provides a basis for subsequent data collection of the corresponding influencing parameters. By dynamically linking parameters such as the coal feed stream bandwidth and speed with the number of LIBS probes, the cross-sectional area and dynamic distribution characteristics of the coal flow can be accurately matched, achieving uniform sampling of carbon content across the entire cross-section and addressing complex coal source scenarios. For example, increasing the number of transverse probes when the bandwidth increases can prevent missed detection of carbon content at the edges of the coal flow. Based on the dynamic matching of the feed rate and the number of probes, the data update cycle can be shortened to within 12 minutes (compared to more than 4 hours for traditional laboratory analysis). In high feed rate scenarios (≥2000t / h), parallel acquisition by multiple probes can achieve a full cross-sectional carbon content scan every minute. By parameterizing the number of probes, redundant deployment of fixed detection systems can be avoided.

[0016] Preferably, step S2 includes the following steps:

[0017] S21. Calculate the relative error between the historical coal carbon content detection dataset and the corresponding data in the historical coal carbon content actual dataset to obtain the historical coal carbon content detection error dataset.

[0018] S22. Construct a coal carbon content detection error mapping model for coal feed zone using the historical LIBS detection probe quantity set, the historical coal carbon content detection error dataset, and the historical coal feed zone carbon content detection influence parameter data matrix.

[0019] By constructing a mapping model for the detection error of coal carbon content in the coal feed zone, a quantitative mapping relationship was realized between the number of LIBS detection probes, the data on the influencing parameters of carbon content detection in various types of coal feed zones, and the detection error of coal carbon content in the coal feed zone. This provides a mapping tool for subsequent quantitative adjustment of the number of LIBS detection probes, making it easier and more accurate to obtain the optimal number of LIBS detection probes and minimize the detection error data of coal carbon content in the coal feed zone.

[0020] Preferably, step S22 includes the following steps:

[0021] S221. Construct a first initial SVM model and set a first training data ratio; according to the first training data ratio, divide the historical coal carbon content detection error dataset, the historical LIBS detection probe number set, and the historical coal feed carbon content detection influence parameter data matrix to obtain the historical coal carbon content detection error training dataset, the historical LIBS detection probe number training set, the historical coal feed carbon content detection influence parameter training data matrix, the historical coal carbon content detection error test dataset, the historical LIBS detection probe number test set, and the historical coal feed carbon content detection influence parameter test data matrix.

[0022] S222. Set a first training error threshold; input the training set of the number of historical LIBS detection probes, the training data matrix of historical coal feed carbon content detection influence parameters as training data, and the training dataset of historical coal carbon content detection errors as training labels into the first initial SVM model for training; during the training process, when the training error data is less than the first training error threshold, stop training and obtain the first trained SVM model; otherwise, continue training until the training error data is less than the first training error threshold.

[0023] S223. Set a first test accuracy threshold; input the historical LIBS detection probe quantity test set, the historical coal feed carbon content detection influence parameter test data matrix as test data, and the historical coal carbon content detection error test dataset as test labels into the first trained SVM model for testing; after the test is completed, obtain the first test accuracy data; when the first test accuracy data is greater than or equal to the first test accuracy threshold, use the first trained SVM model as the coal feed carbon content detection error mapping model; otherwise, return to S222 to continue training the first trained SVM model until the first test accuracy data is greater than or equal to the first test accuracy threshold.

[0024] By inputting the collected historical data into the first initial SVM model for training and testing, the final coal feed carbon content detection error mapping model achieves optimal mapping accuracy between the number of LIBS detection probes, the influencing parameters of coal feed carbon content detection, and the coal carbon content detection error test data. This ensures the accuracy of the data relied upon when subsequently adjusting the number of LIBS detection probes, thereby guaranteeing the accuracy of the adjustment of the number of LIBS detection probes.

[0025] Preferably, step S3 includes the following steps:

[0026] S31. In conjunction with the set of parameters affecting carbon content detection in the coal feed zone, collect data on the parameters affecting carbon content detection in the coal feed zone during the current coal quality burning process in the coal-fired power plant, as well as the number of LIBS detection probes set on the corresponding coal feed zone, to obtain the current dataset of parameters affecting carbon content detection in the coal feed zone and the current initial data on the number of LIBS detection probes in the coal feed zone.

[0027] S32. Set the current coal carbon content detection error threshold; input the current coal feed carbon content detection influence parameter dataset and the current initial coal feed LIBS detection probe quantity data into the coal feed carbon content detection error mapping model for mapping to obtain the current coal carbon content detection error data.

[0028] If the current coal carbon content detection error data is greater than or equal to the current coal carbon content detection error threshold, the current initial coal feed with LIBS detection probe count data is adjusted until the current coal carbon content detection error data is less than the current coal carbon content detection error threshold, thus obtaining the current final coal feed with LIBS detection probe count data; otherwise, there is no need to adjust the current initial coal feed with LIBS detection probe count data, and the current initial coal feed with LIBS detection probe count data is used as the current final coal feed with LIBS detection probe count data.

[0029] By adjusting the number of LIBS detection probes in the initial coal feed belt, the corresponding coal carbon content detection error data can meet the requirements. Furthermore, by collecting real-time data on the coal carbon content in the feed belt, carbon content estimation errors caused by coal quality fluctuations can be avoided. Real-time carbon content data can be linked to the boiler control system to dynamically adjust parameters such as the air-coal ratio and combustion temperature. For example, when a sudden increase in carbon content is detected, the secondary air volume is automatically increased to promote complete combustion, reduce unburned carbon loss, and reduce the emission of harmful gases (carbon monoxide). This also provides data for optimizing the blending ratio of coal in the subsequent combustion process and reducing the carbon emission intensity per unit of power generation.

[0030] Preferably, the Pelican optimization algorithm is used to adjust the number of LIBS detection probes in the current initial coal feed into the furnace in step S32.

[0031] Preferably, step S4 includes the following steps:

[0032] S41. In conjunction with the current final coal feed belt LIBS detection probe quantity data, detect the carbon content of the coal in the current coal feed belt to obtain the current coal carbon content data; set low carbon content coal, medium carbon content coal, and high carbon content coal and the corresponding carbon content value range to obtain the low carbon content value range, medium carbon content value range, and high carbon content value range.

[0033] Several electricity consumption scenarios were set for low-carbon coal, medium-carbon coal, and high-carbon coal, respectively, to obtain sets of low-carbon coal consumption scenarios, medium-carbon coal consumption scenarios, and high-carbon coal consumption scenarios.

[0034] S42. Obtain the current electricity consumption scenario and set a set of coal quality replacement judgment conditions; when there is a coal quality replacement judgment condition in the set of coal quality replacement judgment conditions, replace the coal quality on the current coal feed belt until there is no coal quality replacement judgment condition in the set of coal quality replacement judgment conditions, obtain the current replaced coal quality and proceed to S5; otherwise, it is not necessary to replace the coal quality on the current coal feed belt.

[0035] S43. Set parameters for several types of coal combustion processes to obtain a coal combustion parameter type set; collect historical coal combustion parameter data, coal carbon content data, and corresponding coal combustion completeness data corresponding to the coal combustion parameters set to obtain a historical coal combustion parameter data matrix, a historical coal carbon content dataset, and a historical coal combustion completeness dataset; construct a coal combustion completeness mapping model using the historical coal combustion parameter data matrix, the historical coal carbon content dataset, and the historical coal combustion completeness dataset.

[0036] By dynamically updating the coal with different carbon contents fed into the furnace according to different power consumption scenarios, precise control of carbon intensity is achieved, which helps to reduce total carbon emissions, making it more environmentally friendly. It also reduces the unit fuel consumption, thereby reducing power generation costs. In addition, by constructing a historical coal combustion completeness mapping model, a quantitative mapping between coal combustion parameter data, coal carbon content data and coal combustion completeness data is achieved.

[0037] Preferably, step S5 includes the following steps:

[0038] S51. Obtain the carbon content data of the current replaced coal as described in S42, and get the carbon content data of the current replaced coal.

[0039] S52. Adjust the current coal combustion parameter data of various types in conjunction with the current changed coal carbon content data, coal combustion completeness mapping model and coal combustion parameter type set to obtain the current final coal combustion parameter dataset;

[0040] S53. The current final coal quality combustion parameter dataset is used to burn the current replaced coal quality, and a gas sensor is used to measure the carbon emissions after combustion in real time.

[0041] By adjusting the combustion parameters of various types of coal, it is possible to ensure that the carbon content of the coal entering the furnace can be fully burned in the burner. This not only makes full use of the coal fuel, but also reduces the carbon monoxide content caused by incomplete combustion, making it more environmentally friendly.

[0042] Preferably, adjusting the current combustion parameter data for various types of coal in step S52 includes the following steps:

[0043] S521. Obtain the value range of various types of coal combustion parameter data in conjunction with the coal combustion parameter type set to obtain the current coal combustion parameter value range set;

[0044] Construct a pelican population for adjusting coal combustion parameters; set the maximum number of iterations for the pelican population for adjusting coal combustion parameters as follows: And the current iteration number is These represent the maximum number of iterations for adjusting the combustion parameters and the current number of iterations for adjusting the combustion parameters, respectively.

[0045] S522. Generate the initial position of each pelican in the pelican population by adjusting the coal combustion parameters according to the current coal combustion parameter value range set, and obtain the second initial position matrix;

[0046] S523. Construct the fitness function for adjusting the coal combustion parameters of the pelican population;

[0047] S524. Begin iteration. Before iteration, set the current iteration count of the combustion parameter adjustment to 1. During the first iteration, use the fitness function of the coal combustion parameter adjustment pelican population to calculate the fitness value of the initial position of each pelican in the second initial position matrix, obtaining the third fitness value set. Take the largest fitness value in the third fitness value set and the corresponding initial position of the pelican as the third global best fitness and the third global best position, respectively. Update the initial position of each pelican in the second initial position matrix according to the third global best fitness and the third global best position. After the update is completed, increment the current iteration count of the combustion parameter adjustment by 1 and proceed to the next iteration.

[0048] In each iteration, the fitness function of the coal combustion parameter adjustment pelican population is used to calculate the fitness value of each pelican position in the coal combustion parameter adjustment pelican population updated in the previous iteration, resulting in a fourth fitness value set. The largest fitness value in the fourth fitness value set and the corresponding pelican position are taken as the fourth global best fitness and the fourth global best position, respectively. The position of each pelican in the coal combustion parameter adjustment pelican population updated in the previous iteration is updated according to the fourth global best fitness and the fourth global best position. After the update is completed, the current iteration number of the combustion parameter adjustment is incremented by 1 and the next iteration is performed.

[0049] S525, when If the condition is met, stop iterating to obtain the second final global optimal position; otherwise, continue iterating until... Up to this point; the second final global optimal position is used as the current final coal combustion parameter dataset;

[0050] The Pelican Optimization Algorithm, by simulating group predation behavior, can efficiently search the multidimensional parameter space, avoiding the pitfalls of traditional gradient descent methods that are prone to getting trapped in local optima. For key parameters such as burner outlet size, recirculation zone intensity of swirl burners, boiler load, and oxygen concentration, the algorithm can find the global optimum within 20-50 iterations, reducing the carbon content of fly ash to below 2%. It also supports real-time response to changes in combustion conditions (such as load fluctuations and coal quality fluctuations), and quickly matches the optimal parameter combination by adaptively adjusting the position update strategy.

[0051] A real-time monitoring system for carbon emissions from coal-fired power plants includes a module for setting the type of parameters affecting the number of carbon content detection probes, a module for acquiring historical carbon content detection process data, a module for constructing a carbon content detection error mapping model, a module for acquiring current coal quality data entering the furnace, a module for adjusting the number of current LIBS detection probes, a module for updating current coal quality entering the furnace, a module for adjusting current coal quality combustion parameters, and a module for measuring final carbon emissions.

[0052] The present invention has the following beneficial effects:

[0053] 1. In this invention, the accuracy of carbon content detection in coal before it enters the furnace is first optimized by adjusting the number of LIBS detection probes on the coal feed belt, ensuring that the accuracy of carbon content detection in coal before entering the furnace meets the requirements. Then, based on the carbon content data of the coal before entering the furnace and combined with the corresponding power consumption scenario, the coal before entering the furnace is replaced, so that the carbon content of the coal before entering the furnace is more consistent with the current power consumption scenario, thereby helping to save fuel and improve power generation efficiency. Finally, the combustion parameters of the corresponding burner are adjusted according to the carbon content data of the replaced coal before entering the furnace, ensuring complete combustion of the coal, which not only makes full use of fuel but also reduces the proportion of harmful gases in the final carbon emissions.

[0054] 2. In this invention, by collecting real-time data on the carbon content of coal on the coal feed belt, the estimation deviation of carbon content caused by fluctuations in coal quality can be avoided. The real-time carbon content data can be linked with the boiler control system to dynamically adjust parameters such as air-coal ratio and combustion temperature. For example, when a sudden increase in carbon content is detected, the secondary air volume is automatically increased to promote complete combustion, reduce unburned carbon loss, and reduce the emission of harmful gases (carbon monoxide). At the same time, it also provides data basis for optimizing the blending ratio of coal in the subsequent firing process and reducing the carbon emission intensity per unit of power generation.

[0055] 3. In this invention, the coal with different carbon contents fed into the furnace is dynamically updated according to different power consumption scenarios, thereby achieving precise control of carbon intensity. This is beneficial to reducing total carbon emissions, making it more environmentally friendly, and also reducing unit fuel consumption, thus reducing power generation costs.

[0056] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0057] To more clearly illustrate the technical solutions of the embodiments of the invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0058] Figure 1 This is a flowchart illustrating a method for real-time monitoring of carbon emissions from coal-fired power plants according to the present invention.

[0059] Figure 2 This is a schematic diagram illustrating the process of adjusting the number of detection probes in the LIBS system according to the present invention.

[0060] Figure 3 This is a schematic diagram illustrating the process of adjusting combustion parameter data for various types of coal according to the present invention.

[0061] Figure 4 This is a schematic diagram of a module of a real-time monitoring system for carbon emission indicators of a coal-fired power plant according to the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the invention, and not all embodiments. Based on the embodiments of the invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the invention.

[0063] Example 1

[0064] Please see Figure 1-3 This embodiment is a method for real-time monitoring of carbon emissions from coal-fired power plants, including the following steps:

[0065] S1. Collect data from multiple historical studies on the use of the LIBS system to detect the carbon content in coal entering the furnace;

[0066] S1 includes the following steps:

[0067] S11. Several parameter types are defined to correlate the number of detection probes used in the coal feed belt of a coal-fired power plant to detect the carbon content in the coal using a LIBS system, resulting in a set of parameter types affecting carbon content detection in the coal feed belt. This set includes parameters for feed belt bandwidth, feed belt speed, feed belt conveying capacity, and feed belt conveying length. A larger bandwidth results in a larger cross-sectional area of ​​the coal flow, requiring an increased number of probes to ensure full coverage of carbon content detection across the entire width of the coal seam. Large bandwidth can lead to differences in coal flow accumulation between the edge and center (e.g., when the bandwidth is ≥1400mm), requiring additional probes on both sides to monitor stratification. High belt speeds (e.g., >3m / s) necessitate an increase in probe sampling frequency. High-frequency data acquisition can be achieved by increasing the number of parallel probes. Excessive belt speed (>4m / s) can easily cause coal flow fluctuations, requiring additional probes in the middle of the conveying section to capture instantaneous carbon content changes. Long-distance conveying (L>100m) can easily lead to coal flow stratification or segregation, requiring probes to be set in sections (e.g., one set every 50m) to monitor carbon content gradient changes. Additional probes are needed at the head (discharge end) and tail (receiving end) of the conveyor to eliminate detection deviations caused by material impact. Higher conveying volumes correspond to thicker coal seams, requiring vertically layered probe arrangements (e.g., dual probes at the top and bottom) to measure cross-sectional carbon distribution. When Q≥2000t / h, probes should be evenly distributed on both sides and the centerline of the belt (e.g., 3-5 probes per cross-section) to avoid local detection failures caused by coal flow accumulation.

[0068] S12. In conjunction with the aforementioned set of parameters affecting carbon content detection in the coal feed zone, collect data on the number of LIBS probes, carbon content detection data, actual carbon content data, and corresponding parameters affecting carbon content detection in the coal feed zone from multiple historical instances of using the LIBS system to detect carbon content in coal feed. This yields a historical set of LIBS probe counts. Historical coal carbon content detection dataset Historical coal carbon content actual dataset And a data matrix of parameters affecting the detection of carbon content in historical coal entering the furnace; a1i, a2i, and a3i represent the number of detection probes, carbon content detection data, and actual carbon content data of the LIBS system in the i-th group of historical data collected when the carbon content in the coal entering the furnace was detected using the LIBS system. This indicates the total number of data collection sessions conducted using the LIBS system to detect the carbon content in the coal entering the furnace.

[0069] S2. Construct an error mapping model for coal carbon content detection in the coal feed zone, based on the data collected in S1.

[0070] S2 includes the following steps:

[0071] S21. Calculate the relative error between the historical coal carbon content detection dataset and the corresponding data in the historical coal carbon content actual dataset to obtain the historical coal carbon content detection error dataset. a4i represents the error between a2i and a3i; the calculation formula is as follows.

[0072] ;

[0073] S22. Construct a coal carbon content detection error mapping model for coal feed zone using the historical LIBS detection probe quantity set, the historical coal carbon content detection error dataset, and the historical coal feed zone carbon content detection influence parameter data matrix.

[0074] S22 includes the following steps:

[0075] S221. Construct a first initial SVM model and set a first training data ratio; according to the first training data ratio, divide the historical coal carbon content detection error dataset, the historical LIBS detection probe number set, and the historical coal feed carbon content detection influence parameter data matrix to obtain the historical coal carbon content detection error training dataset, the historical LIBS detection probe number training set, the historical coal feed carbon content detection influence parameter training data matrix, the historical coal carbon content detection error test dataset, the historical LIBS detection probe number test set, and the historical coal feed carbon content detection influence parameter test data matrix.

[0076] S222. Set a first training error threshold; input the training set of the number of historical LIBS detection probes, the training data matrix of historical coal feed carbon content detection influence parameters as training data, and the training dataset of historical coal carbon content detection errors as training labels into the first initial SVM model for training; during the training process, when the training error data is less than the first training error threshold, stop training and obtain the first trained SVM model; otherwise, continue training until the training error data is less than the first training error threshold.

[0077] S223. Set a first test accuracy threshold; input the historical LIBS detection probe quantity test set, the historical coal feed carbon content detection influence parameter test data matrix as test data, and the historical coal carbon content detection error test dataset as test labels into the first trained SVM model for testing; after the test is completed, obtain the first test accuracy data; when the first test accuracy data is greater than or equal to the first test accuracy threshold, use the first trained SVM model as the coal feed carbon content detection error mapping model; otherwise, return to S222 to continue training the first trained SVM model until the first test accuracy data is greater than or equal to the first test accuracy threshold.

[0078] S3. Collect parameter data of the coal feed zone during the current coal quality burning process of the coal-fired power plant and the corresponding number of LIBS detection probes, and input them into the coal carbon content detection error mapping model of the coal feed zone for mapping. Then, adjust the current initial number of LIBS detection probes in the coal feed zone according to the mapping results to obtain the current final number of LIBS detection probes in the coal feed zone.

[0079] S3 includes the following steps:

[0080] S31. In conjunction with the set of parameters affecting carbon content detection in the coal feed zone, collect data on the parameters affecting carbon content detection in the coal feed zone during the current coal quality burning process in the coal-fired power plant, as well as the number of LIBS detection probes set on the corresponding coal feed zone, to obtain the current dataset of parameters affecting carbon content detection in the coal feed zone and the current initial data on the number of LIBS detection probes in the coal feed zone.

[0081] S32. Set the current coal carbon content detection error threshold; input the current coal feed carbon content detection influence parameter dataset and the current initial coal feed LIBS detection probe quantity data into the coal feed carbon content detection error mapping model for mapping to obtain the current coal carbon content detection error data.

[0082] If the current coal carbon content detection error data is greater than or equal to the current coal carbon content detection error threshold, the current initial coal feed with LIBS detection probe count data is adjusted until the current coal carbon content detection error data is less than the current coal carbon content detection error threshold, thus obtaining the current final coal feed with LIBS detection probe count data; otherwise, there is no need to adjust the current initial coal feed with LIBS detection probe count data, and the current initial coal feed with LIBS detection probe count data is used as the current final coal feed with LIBS detection probe count data.

[0083] The adjustment of the number of LIBS detection probes in the current initial coal feed into the furnace in step S32 includes the following steps:

[0084] S321. Set the value range of the current initial coal feed with LIBS detection probe count data to obtain the current LIBS detection probe count range. ; , These represent the lower and upper limits of the current initial coal feed data with LIBS detection probes, respectively.

[0085] Construct a pelican population with adjusted LIBS detection probe count in the coal feed area; set the maximum number of iterations for this pelican population. And the current iteration number is , denoted as the maximum number of iterations for probe adjustment and the current number of iterations for probe adjustment, respectively; the search space dimension of the pelican population for adjusting the number of LIBS detection probes in the coal feeder is 1-dimensional;

[0086] S322. Based on the current range of LIBS detection probe counts, generate the number of LIBS detection probes in the coal feed zone and adjust the initial position of each pelican in the pelican population to obtain the first initial position set. b1i indicates that the number of LIBS detection probes in the coal feeder belt is adjusted to adjust the initial position of the i-th pelican in the pelican population. The number of LIBS detection probes in the coal feeder is used to adjust the size of the pelican population; the formula for generating b1i is as follows.

[0087] ;

[0088] In the formula: rand1i represents a random number between 0 and 1 generated for b1i; ceil represents the floor function;

[0089] S323. Construct a fitness function for adjusting the number of LIBS detection probes in the coal feed zone to control the pelican population. ;as follows,

[0090] ;

[0091] In the formula, This means that the number of LIBS detection probes in the coal feed zone and the current dataset of parameters affecting the detection of carbon content in the coal feed zone are input into the coal quality carbon content detection error mapping model in each iteration to obtain the mapped data. A positive number indicates a protection parameter;

[0092] S324. Begin iteration. Before each iteration, set the current iteration count of the probe to 1. During the first iteration, adjust the fitness function of the pelican population by using the number of LIBS detection probes in the coal feeder. Calculate the fitness value of the initial position of each pelican in the first initial position set to obtain the first fitness value set; take the largest fitness value in the first fitness value set and the corresponding initial position of the pelican as the first global best fitness and the first global best position, respectively; update the initial position of each pelican in the first initial position set according to the first global best fitness and the first global best position; after the update is completed, increment the current iteration number of the probe by 1 and proceed to the next iteration.

[0093] In each iteration, the fitness function of the pelican population is adjusted by the number of LIBS detection probes in the coal feeder. The fitness value of each pelican in the coal feed zone LIBS detection probe number adjustment pelican population, obtained from the previous iteration, is calculated to obtain a second fitness value set. The maximum fitness value in the second fitness value set and the corresponding pelican position are used as the second global optimal fitness and the second global optimal position, respectively. The position of each pelican in the coal feed zone LIBS detection probe number adjustment pelican population, obtained from the previous iteration, is updated according to the second global optimal fitness and the second global optimal position. After the update is completed, the current iteration number of the probe adjustment is incremented by 1, and the next iteration is performed.

[0094] S325, when If the first final global optimum is reached, stop the iteration and obtain the first final global optimum position and the first final global optimum fitness; otherwise, continue the iteration until... Up to this point; the first final global optimal fitness is used as the current optimized coal carbon content detection error data; when the current optimized coal carbon content detection error data is less than the current coal carbon content detection error threshold, the first final global optimal position is used as the current final coal feed LIBS detection probe quantity data; otherwise, return to S324 to continue iteration until the current optimized coal carbon content detection error data is less than the current coal carbon content detection error threshold;

[0095] The Pelican optimization algorithm supports real-time data-driven updates and dynamically adjusts the number of active probes by combining LIBS spectral stability parameters (such as relative standard deviation between pulses, RSD). For example, when the uniformity of coal powder particle flow decreases (RSD≥8%), the number of effective probes is automatically increased to improve data redundancy and ensure detection reliability. During the coal quality stabilization stage (RSD≤3%), only the core probe group is activated to extend equipment life and reduce energy consumption. By optimizing the number and layout of LIBS probes through the Pelican optimization algorithm, the overall efficiency of the detection-optimization-control link in coal-fired power plants can be improved while ensuring the accuracy of carbon emission accounting.

[0096] S4. Based on the data of the number of LIBS detection probes in the current coal feed belt, the carbon content of the coal in the current coal feed belt is detected, and the corresponding power consumption scenario type is used to replace the coal in the current coal feed belt, so as to obtain the current coal quality after replacement.

[0097] S4 includes the following steps:

[0098] S41. In conjunction with the current final coal feed belt LIBS detection probe quantity data, detect the carbon content of the coal in the current coal feed belt to obtain the current coal carbon content data; set low carbon content coal, medium carbon content coal, and high carbon content coal and the corresponding carbon content value range to obtain the low carbon content value range, medium carbon content value range, and high carbon content value range.

[0099] Several electricity consumption scenarios were set up for low-carbon, medium-carbon, and high-carbon coal, respectively, resulting in sets of electricity consumption scenarios for low-carbon, medium-carbon, and high-carbon coal. For example, during peak electricity consumption periods, high-carbon, high-calorific-value coal (such as Malang coal accounting for 65%) was prioritized for combustion to increase boiler instantaneous output and reduce unburned carbon loss; during off-peak electricity consumption periods, low-priced, low-calorific-value coal (carbon content ≤ 50%) was blended in, while during peak periods, high-carbon, high-calorific-value coal (carbon content ≥ 65%) was used to shorten combustion time. Reducing unburned carbon loss lowers the carbon emission intensity per unit of electricity generation; blending low-carbon coal (carbon content ≤50%) during off-peak periods directly reduces total carbon emissions; through dynamic coal quality matching, sulfur oxide (SOx) emission concentration can be controlled below 50 mg / m³, meeting ultra-low emission standards; high-carbon coal has lower ash content (Aar ≤20%), which can reduce ash and slag generation by 30%-40% and reduce fly ash carbon content to below 2%; low-carbon coal has a higher ash melting point (ST ≥1350℃), which can suppress the risk of coking.

[0100] S42. Obtain the current electricity consumption scenario and set the coal quality replacement judgment condition set. Wherein, c1 indicates: when the current coal carbon content data is within the low carbon content range and the current electricity consumption scenario is not concentrated in the low carbon content electricity consumption scenario cluster; c2 indicates: when the current coal carbon content data is within the medium carbon content range and the current electricity consumption scenario is not concentrated in the medium carbon content electricity consumption scenario cluster; c3 indicates: when the current coal carbon content data is within the high carbon content range and the current electricity consumption scenario is not concentrated in the high carbon content electricity consumption scenario cluster;

[0101] When there are coal quality replacement judgment conditions in the coal quality replacement judgment condition set, the coal quality on the current coal feed belt is replaced until there are no coal quality replacement judgment conditions in the coal quality replacement judgment condition set, and the current replaced coal quality is obtained and enters S5; otherwise, it is not necessary to replace the coal quality on the current coal feed belt.

[0102] S43. Set several types of parameters in the coal combustion process to obtain a coal combustion parameter type set; the coal combustion parameter type set includes the burner outlet size, such as when the outlet diameter is >800mm, it will cause the ignition distance of the pulverized coal airflow to be extended by 2-3 meters, increasing the loss of unburned carbon; the recirculation zone strength of the swirl burner, such as when the recirculation zone strength of the swirl burner is insufficient, the entrainment of high-temperature flue gas decreases, and the ignition stability decreases; boiler load and oxygen concentration, when the load rate is <60%, the average furnace temperature drops by 200-300℃, the carbon content of fly ash increases by 3-5 percentage points, when the flue gas oxygen concentration is <3%, the oxygen supply is insufficient in the later stage of combustion, and an oxygen-deficient protective layer forms on the surface of large pulverized coal particles, etc.; in conjunction with the coal combustion parameter type set, collect historical coal combustion parameter data, coal carbon content data, and corresponding coal combustion completeness data corresponding to the coal combustion process to obtain a historical coal combustion parameter data matrix and a historical coal carbon content dataset. and historical coal combustion completeness dataset , , These represent the carbon content data and combustion completeness data of the coal collected in the i-th historical data set, respectively, when the coal was burned. This indicates the total number of data sets collected during the historical combustion process of coal.

[0103] A coal combustion completeness mapping model is constructed using the historical coal combustion parameter data matrix, the historical coal carbon content dataset, and the historical coal combustion completeness dataset.

[0104] Preferably, the construction of the coal combustion completeness mapping model in S43 using the historical coal combustion parameter data matrix, the historical coal carbon content dataset, and the historical coal combustion completeness dataset includes the following steps:

[0105] S431. Construct a second initial SVM model and set a second training data ratio; divide the historical coal combustion parameter data matrix, historical coal carbon content dataset, and historical coal combustion completeness dataset according to the second training data ratio to obtain the historical coal combustion parameter training data matrix, historical coal carbon content training dataset, historical coal combustion completeness training dataset, historical coal combustion parameter test data matrix, historical coal carbon content test dataset, and historical coal combustion completeness test dataset.

[0106] S432. Set a second training error threshold; input the historical coal combustion parameter training data matrix, the historical coal carbon content training dataset as training data, and the historical coal combustion completeness training dataset as training labels into the second initial SVM model for training; during the training process, when the training error data is less than the second training error threshold, stop training and obtain the second trained SVM model; otherwise, continue training until the training error data is less than the second training error threshold;

[0107] S433. Set a second test accuracy threshold; input the historical coal combustion parameter test data matrix, the historical coal carbon content test dataset as test data, and the historical coal combustion completeness test dataset as test labels into the second trained SVM model for testing; after the test is completed, obtain the second test accuracy data; when the second test accuracy data is greater than or equal to the second test accuracy threshold, use the second trained SVM model as the coal combustion completeness mapping model; otherwise, return to S432 to continue training the second trained SVM model until the second test accuracy data is greater than or equal to the second test accuracy threshold;

[0108] S5. Adjust the current coal combustion parameter data of various types in conjunction with the carbon content data of the current replaced coal to obtain the current final coal combustion parameter dataset;

[0109] The carbon emissions after combustion are measured in conjunction with the current final coal quality combustion parameter dataset and the current replaced coal quality.

[0110] S5 includes the following steps:

[0111] S51. Obtain the carbon content data of the current replaced coal as described in S42, and get the carbon content data of the current replaced coal.

[0112] S52. Adjust the current coal combustion parameter data of various types in conjunction with the current changed coal carbon content data, coal combustion completeness mapping model and coal combustion parameter type set to obtain the current final coal combustion parameter dataset;

[0113] The adjustment of current combustion parameter data for various types of coal in S52 includes the following steps:

[0114] S521. In conjunction with the coal combustion parameter type set, obtain the value ranges of various types of coal combustion parameter data to obtain the current coal combustion parameter value range set. ;as follows,

[0115] ;

[0116] in, , These represent the lower limit and upper limit of the values ​​for the current i-th type of coal combustion parameter data, respectively, and d represents the total number of coal combustion parameter types set.

[0117] Construct a pelican population for adjusting coal combustion parameters; set the maximum number of iterations for the pelican population for adjusting coal combustion parameters as follows: And the current iteration number is , which are the maximum number of iterations for adjusting the combustion parameters and the current number of iterations for adjusting the combustion parameters, respectively; the search space dimension of the pelican population for adjusting the coal combustion parameters is the same as d;

[0118] S522. Based on the current set of coal combustion parameter values, generate the initial position of each pelican in the coal combustion parameter adjustment pelican population to obtain the second initial position matrix; the calculation formula is as follows.

[0119] ;

[0120] in, The initial position of the j-th pelican in the pelican population adjusted by the coal combustion parameters represents the positional component of the coal combustion parameter data dimension of the i-th type. Indicating targeting Generate random numbers between 0 and 1;

[0121] S523. Construct the fitness function for adjusting the coal combustion parameters of the pelican population. ;as follows,

[0122] ;

[0123] In the formula, This means that a set of coal combustion parameters updated in each iteration process and the current coal carbon content data after replacement are input into the coal combustion completeness mapping model to obtain the data;

[0124] S524. Begin iteration. Before each iteration, adjust the current iteration number to 1 using the combustion parameters. During the first iteration, adjust the fitness function of the pelican population using the coal combustion parameters. Calculate the fitness value of the initial position of each pelican in the second initial position matrix to obtain the third fitness value set; take the largest fitness value in the third fitness value set and the corresponding initial position of the pelican as the third global best fitness and the third global best position, respectively; update the initial position of each pelican in the second initial position matrix according to the third global best fitness and the third global best position; after the update is completed, increment the current iteration number of the combustion parameter adjustment by 1 and proceed to the next iteration.

[0125] In each subsequent iteration, the fitness function of the pelican population is adjusted using the aforementioned coal combustion parameters. Calculate the fitness value of each pelican position in the coal combustion parameter adjustment pelican population obtained from the previous iteration, to obtain the fourth fitness value set; take the largest fitness value in the fourth fitness value set and the corresponding pelican position as the fourth global best fitness and the fourth global best position, respectively; update the position of each pelican in the coal combustion parameter adjustment pelican population obtained from the previous iteration according to the fourth global best fitness and the fourth global best position; after the update is completed, increment the current iteration number of the combustion parameter adjustment by 1 and proceed to the next iteration.

[0126] S525, when If the condition is met, stop iterating to obtain the second final global optimal position; otherwise, continue iterating until... Up to this point; the second final global optimal position is used as the current final coal combustion parameter dataset;

[0127] S53. The current final coal quality combustion parameter dataset is used to burn the current replaced coal quality, and a gas sensor is used to measure the carbon emissions after combustion in real time.

[0128] Example 2

[0129] Please see Figure 4 This embodiment discloses a real-time monitoring system for carbon emission indicators of coal-fired power plants. The system can implement the method of the above embodiment, including a module for setting the parameter type of the number of carbon content detection probes, a module for acquiring historical carbon content detection process data, a module for constructing a carbon content detection error mapping model, a module for acquiring current coal quality data entering the furnace, a module for adjusting the number of current LIBS detection probes, a module for updating current coal quality entering the furnace, a module for adjusting current coal quality combustion parameters, and a final carbon emission measurement module.

[0130] The parameter type setting module for the number of carbon content detection probes sets several parameter types related to the number of detection probes used to detect the carbon content in the coal entering the furnace using the LIBS system, thus obtaining a set of parameter types affecting the carbon content detection in the coal entering the furnace.

[0131] The historical carbon content detection process data acquisition module, in conjunction with the coal feeding carbon content detection influence parameter type set, collects data from multiple historical processes of using the LIBS system to detect carbon content in coal entering the furnace, resulting in a historical LIBS detection probe number set, a historical coal carbon content detection dataset, a historical coal carbon content actual dataset, and a historical coal feeding carbon content detection influence parameter data matrix.

[0132] The carbon content detection error mapping model construction module, together with the historical LIBS detection probe quantity set, the historical coal feed carbon content detection influence parameter data matrix, the historical coal quality carbon content detection dataset, and the historical coal quality carbon content actual dataset, constructs a coal feed carbon content detection error mapping model.

[0133] The current coal quality data acquisition module, in conjunction with the coal feed carbon content detection influence parameter type set, collects parameter data of the coal feed zone during the current coal quality burning process of the coal-fired power plant, as well as the corresponding number of LIBS detection probes, to obtain the current coal feed carbon content detection influence parameter dataset and the current initial coal feed LIBS detection probe number data.

[0134] The current LIBS detection probe quantity adjustment module inputs the current coal feed carbon content detection influence parameter dataset and the current initial coal feed carbon content detection probe quantity data into the coal feed carbon content detection error mapping model for mapping, and then adjusts the current initial coal feed carbon content detection probe quantity data according to the mapping result to obtain the current final coal feed carbon content detection probe quantity data.

[0135] The current coal quality update module, in conjunction with the data on the number of LIBS detection probes in the current final coal feed belt, detects the carbon content of the coal in the current coal feed belt and the corresponding power consumption scenario type to replace the coal in the current coal feed belt, thus obtaining the current replaced coal quality.

[0136] The current coal combustion parameter adjustment module, in conjunction with the carbon content data of the currently replaced coal, adjusts the current coal combustion parameter data of various types to obtain the current final coal combustion parameter dataset;

[0137] The final carbon emission measurement module works in conjunction with the current final coal quality combustion parameter dataset and the current replaced coal quality to measure the carbon emissions after combustion.

[0138] In the description of this specification, references to terms such as "an embodiment," "example," "specific example," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0139] The preferred embodiments of the invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention.

Claims

1. A method for real-time monitoring of carbon emissions from coal-fired power plants, characterized in that, Includes the following steps: S1. Collect data from multiple historical studies on the use of the LIBS system to detect the carbon content in coal entering the furnace; Specifically, this includes: setting several parameter types related to the number of detection probes used in the coal feed zone of a coal-fired power plant to detect the carbon content in the coal using a LIBS system, thereby obtaining coal feed zone parameters that affect carbon content detection; and, in conjunction with these coal feed zone parameters, collecting data on the number of LIBS system detection probes, carbon content detection data, actual carbon content data, and corresponding carbon content detection influence parameter data for multiple historical sets of coal feed zone carbon content detection processes, thereby obtaining a historical LIBS detection probe quantity set, a historical coal carbon content detection dataset, a historical coal carbon content actual dataset, and a historical coal feed zone carbon content detection influence parameter data matrix. S2. Construct an error mapping model for coal carbon content detection in the coal feed zone, based on the data collected in S1. Specifically, this includes: calculating the relative error between the historical coal carbon content detection dataset and the corresponding data in the historical coal carbon content actual dataset to obtain a historical coal carbon content detection error dataset; and constructing a coal carbon content detection error mapping model for the coal feed zone using the historical LIBS detection probe quantity set, the historical coal carbon content detection error dataset, and the historical coal feed zone carbon content detection influence parameter data matrix. S3. Collect parameter data of the coal feed zone during the current coal quality burning process of the coal-fired power plant and the corresponding number of LIBS detection probes, and input them into the coal carbon content detection error mapping model of the coal feed zone for mapping. Then, adjust the current initial number of LIBS detection probes in the coal feed zone according to the mapping results to obtain the current final number of LIBS detection probes in the coal feed zone. S4. Based on the data of the number of LIBS detection probes in the current coal feed belt, the carbon content of the coal in the current coal feed belt is detected, and the corresponding power consumption scenario type is used to replace the coal in the current coal feed belt, so as to obtain the current coal quality after replacement. S5. Adjust the current coal combustion parameter data of various types in conjunction with the carbon content data of the current replaced coal to obtain the current final coal combustion parameter dataset; The carbon emissions after combustion are measured in conjunction with the current final coal quality combustion parameter dataset and the current changed coal quality.

2. The method for real-time monitoring of carbon emission indicators of coal-fired power plants according to claim 1, characterized in that, The coal feed carbon content detection error mapping model described in S22 is constructed based on the SVM model.

3. The method for real-time monitoring of carbon emissions from coal-fired power plants according to claim 1, characterized in that, S3 includes the following steps: S31. In conjunction with the parameters of the coal feed belt that affect carbon content detection, collect the data of the parameters affecting carbon content detection of the coal feed belt during the current coal quality burning process of the coal-fired power plant, as well as the number of LIBS detection probes set on the corresponding coal feed belt, to obtain the current dataset of parameters affecting carbon content detection of the coal feed belt and the current initial data of the number of LIBS detection probes in the coal feed belt. S32. Set the current coal carbon content detection error threshold; input the current coal feed carbon content detection influence parameter dataset and the current initial coal feed LIBS detection probe quantity data into the coal feed carbon content detection error mapping model for mapping to obtain the current coal carbon content detection error data. If the current coal carbon content detection error data is greater than or equal to the current coal carbon content detection error threshold, the current initial coal feed with LIBS detection probe count data is adjusted until the current coal carbon content detection error data is less than the current coal carbon content detection error threshold, thus obtaining the current final coal feed with LIBS detection probe count data; otherwise, there is no need to adjust the current initial coal feed with LIBS detection probe count data, and the current initial coal feed with LIBS detection probe count data is used as the current final coal feed with LIBS detection probe count data.

4. The method for real-time monitoring of carbon emission indicators of coal-fired power plants according to claim 3, characterized in that: In S32, the Pelican optimization algorithm is used to adjust the number of LIBS detection probes in the current initial coal feed.

5. The method for real-time monitoring of carbon emission indicators of coal-fired power plants according to claim 3, characterized in that, S4 includes the following steps: S41. Using the current final coal feed belt LIBS detection probe quantity data, detect the carbon content of the coal in the current coal feed belt to obtain the current coal carbon content data; set low carbon content coal, medium carbon content coal, and high carbon content coal and their corresponding carbon content value ranges to obtain low carbon content value ranges, medium carbon content value ranges, and high carbon content value ranges; set several power consumption scenarios corresponding to low carbon content coal, medium carbon content coal, and high carbon content coal to obtain low carbon content power consumption scenario sets, medium carbon content power consumption scenario sets, and high carbon content power consumption scenario sets; S42. Obtain the current electricity consumption scenario and set a set of coal quality replacement judgment conditions; when there is a coal quality replacement judgment condition in the set of coal quality replacement judgment conditions, replace the coal quality on the current coal feed belt until there is no coal quality replacement judgment condition in the set of coal quality replacement judgment conditions, obtain the current replaced coal quality and proceed to S5; otherwise, it is not necessary to replace the coal quality on the current coal feed belt. S43. Set several types of parameters in the coal combustion process to obtain a coal combustion parameter type set; collect historical coal combustion parameter data, coal carbon content data and corresponding coal combustion completeness data corresponding to the coal combustion in the coal combustion parameter type set to obtain a historical coal combustion parameter data matrix, a historical coal carbon content dataset and a historical coal combustion completeness dataset. A coal combustion completeness mapping model is constructed using the historical coal combustion parameter data matrix, the historical coal carbon content dataset, and the historical coal combustion completeness dataset.

6. The method for real-time monitoring of carbon emission indicators of coal-fired power plants according to claim 5, characterized in that, S5 includes the following steps: S51. Obtain the carbon content data of the current replaced coal as described in S42, and get the carbon content data of the current replaced coal. S52. Adjust the current coal combustion parameter data of various types in conjunction with the current changed coal carbon content data, coal combustion completeness mapping model and coal combustion parameter type set to obtain the current final coal combustion parameter dataset; S53. The current final coal quality combustion parameter dataset is used to burn the current replaced coal quality, and a gas sensor is used to measure the carbon emissions after combustion in real time.

7. The method for real-time monitoring of carbon emission indicators of coal-fired power plants according to claim 6, characterized in that, The adjustment of current combustion parameter data for various types of coal in S52 includes the following steps: S521. Using the coal combustion parameter type set, obtain the value ranges of various types of current coal combustion parameter data to obtain the current coal combustion parameter value range set; construct a coal combustion parameter adjustment pelican population; set the maximum number of iterations for the coal combustion parameter adjustment pelican population to be... And the current iteration number is These represent the maximum number of iterations for adjusting the combustion parameters and the current number of iterations for adjusting the combustion parameters, respectively. S522. Generate the initial position of each pelican in the pelican population by adjusting the coal combustion parameters according to the current coal combustion parameter value range set, and obtain the second initial position matrix; S523. Construct the fitness function for adjusting the coal combustion parameters of the pelican population; S524. Start the iteration. Before the iteration, set the current iteration number of the combustion parameter adjustment to 1. During each iteration, use the fitness function of the coal combustion parameter adjustment pelican population to calculate the fitness value of each pelican position in the coal combustion parameter adjustment pelican population updated in the previous iteration, and update the position of each pelican in the coal combustion parameter adjustment pelican population updated in the previous iteration. After the update is completed, increment the current iteration number of the combustion parameter adjustment by 1 and proceed to the next iteration. S525, when If the condition is met, stop iterating to obtain the second final global optimal position; otherwise, continue iterating until... Up to this point, the second final global optimal position is used as the current final coal combustion parameter dataset.

8. A system for implementing a real-time monitoring method for carbon emission indicators of a coal-fired power plant as described in any one of claims 1-7.

Images