Thermal power plant unit carbon emission characteristic analysis method based on correlation technology

Abstract

The invention provides a thermal power plant unit carbon emission characteristic analysis method based on a correlation technology. The analysis method comprises the following operation steps: step 1, obtaining a reference value; step 2, material accounting; step 3, field detection; step 4, correlation analysis: combining a material accounting method and an on-site detection method to carry out data analysis work; and step 5, correlation modeling: according to a result obtained in the step 4, aiming at three-dimensional characteristics of data, dividing a behavior matrix of the sample into a time dimension and an index dimension. According to the method, a material accounting method and an on-site detection method are combined to realize abnormal data discrimination, and clustering analysis is carried out by using a correlation model, so that a more accurate evaluation carbon emission index can be obtained.

Description

Technical Field

[0001] The present invention relates to the field of carbon emission analysis, and in particular to a method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology. Background Art

[0002] Carbon emissions is a general term or abbreviation for greenhouse gas emissions. Carbon dioxide is the most important gas in greenhouse gases, so the term carbon is used as a representative. Although it is not accurate, the fastest way for the public to understand is to simply understand "carbon emissions" as "carbon dioxide emissions."

[0003] The fuel itself is an organic hydrocarbon, so if it cannot undergo a combustion chemical reaction with oxygen in the air to become water and carbon dioxide (but carbon dioxide is being considered a greenhouse gas that is harmful to the global atmospheric environment), it may become a harmful substance and be emitted.

[0004] Carbon emission measurement standards are an important tool for assessing environmental impacts. They quantify the impact of greenhouse gas emissions on the global climate system. These standards are designed to provide a way to assess and compare carbon footprints from different sources or processes, thereby facilitating the development and implementation of emission reduction strategies. Carbon emissions vary depending on factors such as power plant size, unit capacity, equipment type and operating conditions. Large or high-load units generally emit more carbon dioxide. Carbon emissions from the same unit at different operating loads will also fluctuate.

[0005] However, my country is a major coal-fired power industry country, and its energy structure mainly relies on coal. The power generation sector contributes greatly to my country's greenhouse gas emissions. In recent years, my country has implemented a series of energy-saving and emission-reduction measures, such as encouraging the development of new energy and controlling coal consumption for power supply, which have played an important role in reducing carbon dioxide emissions. Real-time, accurate and comprehensive measurement of carbon emissions related to power generation constitutes the basic cornerstone for identifying the trajectory and current status of carbon emissions in the power industry. This in turn helps to identify potential carbon emission reduction pathways, promote the transition to a low-carbon power economy, and provide important support for the development of the carbon market. Therefore, we must continue to pay attention to the power generation industry, and the carbon emission monitoring work of power plants that use fuel combustion as a source of power generation still needs to continue.

[0006] There are currently two main methods for carbon emission analysis. Method 1: accounting method, starting from the fuel level, directly calculates according to parameters such as fuel consumption, elemental carbon content, and carbon oxidation rate. Standard carbon accounting methods mainly include emission factor method and material balance method. Method 2: on-site detection method, compares and analyzes the carbon emission data obtained by real-time monitoring method and emission factor method.

[0007] However, the experiment shows that the calculation results of the accounting method and the on-site detection method have certain deviations from the actual carbon emissions. According to the real-time monitoring data, the data of the on-site detection method is more stable and accurate, and the material accounting method is easily interfered by human factors in the verification process;

[0008] At the same time, the above two methods only rely on the total amount of carbon emissions when assessing carbon emission levels, and combine them with the total amount of electricity production and supply for comprehensive consideration, which leads to deviations in the analysis results. In addition, the method of using output as the benchmark for emission intensity has limitations, especially when it is impossible to measure each unit individually. This method is only suitable for assessing the emission level of the entire power plant area, and cannot be accurate to a single unit. Summary of the invention

[0009] The technical problem to be solved by the present invention is to overcome the defect of low accuracy of carbon emission characteristic evaluation and analysis results in the prior art, and to provide a carbon emission characteristic analysis method for thermal power plant units based on correlation technology.

[0010] The present invention solves the above technical problems through the following technical solutions:

[0011] The present invention provides a method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology, and the analysis method comprises the following steps:

[0012] Step 1: Obtaining a reference value: Taking the power supply as a reference standard, obtaining a reference value of the emission intensity of carbon dioxide;

[0013] Step 2: Material accounting: collect data on fuel consumption and emission factors of thermal power plant units, and calculate emission based on the above data;

[0014] Step 3: On-site testing: testing key auxiliary parameters at the end of the thermal power plant unit emission to further determine the concentration of particulate matter under standard conditions and the overall emission of flue gas;

[0015] Step 4: Correlation analysis, combining the material accounting method with the on-site detection method to carry out data analysis;

[0016] Step 5: Correlation modeling. Based on the results of step 4, focusing on the three-dimensional characteristics of the data, the sample behavior matrix is ​​divided into time dimension and indicator dimension. Based on the modeling concept of slope correlation, the increments of each point in the time dimension and indicator dimension are initially standardized, and the composition ratio and composition difference of the increments are comprehensively used to define the correlation coefficient.

[0017] In this technical solution, the carbon emissions of different units under different load and stable load conditions are analyzed through correlation modeling. By comparing the predicted values ​​with the actual values ​​in terms of correlation, the results show that the method of the present invention can better realize the analysis of the carbon emission characteristics of the generator set compared with the prior art.

[0018] Preferably, in step 1, the reference value calculation formula is:

[0019]

[0020] Where Sgd is the carbon emission intensity of power supply, Wgd is the power supply in MWh, and a is the heat supply ratio in %.

[0021] Preferably, in step 2, the monitoring of fossil fuel parameters required in the fuel combustion emission accounting process is mainly calculated according to the following formula:

[0022]

[0023] Where i is the fuel type; AD i is the activity data of the i-th fuel, in GJ; EF i is the carbon dioxide emission factor of the ith fuel, in tCO2 / GJ.

[0024] Preferably, the AD i The calculation formula is:

[0025] AD i =NCV i ×FC i

[0026] Where NCV i is the average low calorific value of the i-th fuel; FC i is the net consumption of the ith fuel.

[0027] Preferably, the EF i The calculation formula is:

[0028]

[0029] In the formula, CC i is the carbon content per unit calorific value of the i-th fuel, in tC / GJ; OF i is the carbon oxidation rate of the ith fuel, expressed in %.

[0030] Preferably, in step three, the key auxiliary parameters are the temperature, pressure, flow rate, humidity and oxygen content of the flue gas.

[0031] Preferably, in step 3, the emission of carbon dioxide in the flue gas is expressed as follows in terms of volume concentration under standard conditions:

[0032]

[0033] Where, θ is the mean volume concentration of carbon dioxide emissions under standard conditions per unit time; Q i is the carbon dioxide concentration of the ith record; n is the number of data records per unit time.

[0034] Preferably, in step 3, in order to calculate the volume of carbon dioxide passing through the flue cross section per unit time, it is necessary to calculate the standard dry flue gas emission per unit time Q at the measuring point. g , the calculation formula is:

[0035]

[0036] Among them, t is the running time of the measuring point, s is the cross-sectional area of ​​the flue at the end of the measuring point, and ν gas The flue gas flow velocity at the end of the flue, Pa, is the sum of the atmospheric pressure at the measuring point and the flue pressure at the end of the flue at the measuring point, T g is the temperature of flue gas at the end of the measuring point; H g It is used to measure the humidity of flue gas at the end of the flue.

[0037] Preferably, the carbon dioxide emission per unit time of the computer group is calculated based on the average value of the carbon dioxide emission volume fraction per unit time of the measuring point and the standard dry flue gas emission. m The formula is:

[0038]

[0039] Preferably, the specific steps of step 4 are:

[0040] S1, multivariate linear fitting, taking the dry oxygen content and carbon emission of the on-site detection method in the training data set as independent variables, and the carbon emission value of the material accounting method as the dependent variable, and carrying out multivariate linear fitting;

[0041] S2, assessment input, substitute the dry basis oxygen content and the carbon emission by on-site detection method into the assessment;

[0042] S3. Data comparison and analysis: obtain the measurement value of the material accounting algorithm model based on multivariate linear regression, and compare and analyze it with the reference value to verify the feasibility of the method and compare it with the reference value.

[0043] On the basis of being in accordance with the common sense in the art, the above-mentioned preferred conditions can be arbitrarily combined to obtain the preferred embodiments of the present invention.

[0044] The positive and progressive effects of the present invention are:

[0045] The present invention uses correlation modeling to analyze the carbon emissions of different units under different load and stable load conditions. By comparing the predicted value with the actual value in terms of correlation, the carbon emission characteristics of the generator set can be better analyzed.

[0046] The present invention combines the material accounting method with the on-site detection method to realize abnormal data discrimination, and performs cluster analysis with a correlation model, thereby obtaining a more accurate assessment of carbon emission indicators. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 The figure is a schematic diagram of the overall process of a method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to an embodiment of the present invention.

[0048] Figure 2 for Figure 1 The flowchart of specific operation steps of the correlation analysis of the carbon emission characteristics analysis method of thermal power plant units based on correlation technology is shown.

[0049] Figure 3 for Figure 1 The schematic diagram of carbon emission intensity in March under different loads of thermal power plant units using the carbon emission characteristics analysis method based on correlation technology is shown.

[0050] Figure 4 for Figure 1 The schematic diagram of carbon emission intensity in July under different loads of thermal power plant units using the carbon emission characteristics analysis method based on correlation technology is shown.

[0051] Figure 5 for Figure 1 The schematic diagram of carbon emission intensity of unit B under different unit loads of the thermal power plant unit carbon emission characteristics analysis method based on correlation technology is shown.

[0052] Figure 6 for Figure 1 The schematic diagram of carbon emission intensity of unit A under stable operating conditions of the carbon emission characteristics analysis method of thermal power plant units based on correlation technology is shown.

[0053] Figure 7 for Figure 1 The schematic diagram of carbon emission intensity of unit C under stable operating conditions of the carbon emission characteristics analysis method of thermal power plant units based on correlation technology is shown.

[0054] Figure 8 for Figure 1 The diagram shown is a comparative analysis diagram of the correlation of three methods for analyzing carbon emission characteristics of thermal power plant units based on correlation technology. DETAILED DESCRIPTION

[0055] The present invention is further described below by way of examples, but the present invention is not limited to the scope of the examples.

[0056] Figures 1 to 8 The structure diagram of an embodiment of the method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology of the present invention is shown. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology includes the following steps:

[0057] Step 1: Obtaining a reference value: Taking the power supply as a reference standard, obtaining a reference value of the emission intensity of carbon dioxide;

[0058] Step 2: Material accounting: collect data on fuel consumption and emission factors of thermal power plant units, and calculate emission based on the above data;

[0059] Step 3: On-site testing: testing key auxiliary parameters at the end of the thermal power plant unit emission to further determine the concentration of particulate matter under standard conditions and the overall emission of flue gas;

[0060] Step 4: Correlation analysis, combining the material accounting method with the on-site detection method to carry out data analysis;

[0061] Step 5: Correlation modeling. Based on the results of step 4, focusing on the three-dimensional characteristics of the data, the sample behavior matrix is ​​divided into time dimension and indicator dimension. Based on the modeling concept of slope correlation, the increments of each point in the time dimension and indicator dimension are initially standardized, and the composition ratio and composition difference of the increments are comprehensively used to define the correlation coefficient.

[0062] In this technical solution, the carbon emissions of different units under different load and stable load conditions are analyzed through correlation modeling. By comparing the predicted values ​​with the actual values ​​in terms of correlation, the results show that the method of the present invention can better realize the analysis of the carbon emission characteristics of the generator set compared with the prior art.

[0063] In step 1, the reference value calculation formula is:

[0064]

[0065] Where Sgd is the carbon emission intensity of power supply, Wgd is the power supply in MWh, and a is the heat supply ratio in %.

[0066] In step 2, the monitoring of fossil fuel parameters required in the fuel combustion emission accounting process is mainly calculated according to the following formula:

[0067]

[0068] Where i is the fuel type; ADi is the activity data of the i-th fuel, in GJ; EF i is the carbon dioxide emission factor of the ith fuel, in tCO2 / GJ.

[0069] The AD i The calculation formula is:

[0070] AD i =NCV i ×FC i

[0071] Where NCV i is the average low calorific value of the i-th fuel; FC i is the net consumption of the ith fuel.

[0072] The EF i The calculation formula is:

[0073]

[0074] In the formula, CC i is the carbon content per unit calorific value of the i-th fuel, in tC / GJ; OF i is the carbon oxidation rate of the ith fuel, expressed in %.

[0075] In step three, the key auxiliary parameters are the temperature, pressure, flow rate, humidity and oxygen content of the flue gas.

[0076] In step 3, the emission of carbon dioxide in the flue gas is expressed in terms of volume concentration under standard conditions:

[0077]

[0078] Where, θ is the mean volume concentration of carbon dioxide emissions under standard conditions per unit time; Q i is the carbon dioxide concentration of the ith record; n is the number of data records per unit time.

[0079] In step 3, in order to calculate the volume of carbon dioxide passing through the flue cross section per unit time, it is necessary to calculate the standard dry flue gas emission per unit time at the measuring point Q g , the calculation formula is:

[0080]

[0081] Among them, t is the running time of the measuring point, s is the cross-sectional area of ​​the flue at the end of the measuring point, and ν gas The flue gas flow velocity at the end of the flue, Pa, is the sum of the atmospheric pressure at the measuring point and the flue pressure at the end of the flue at the measuring point, T gis the temperature of flue gas at the end of the measuring point; H g It is used to measure the humidity of flue gas at the end of the flue.

[0082] According to the average value of the carbon dioxide emission volume fraction per unit time at the measurement point and the standard dry flue gas emission, the carbon dioxide emission per unit time of the computer group is calculated. m The formula is:

[0083]

[0084] The specific steps of step 4 are:

[0085] S1, multivariate linear fitting, taking the dry oxygen content and carbon emission of the on-site detection method in the training data set as independent variables, and the carbon emission value of the material accounting method as the dependent variable, and carrying out multivariate linear fitting;

[0086] S2, assessment input, substitute the dry basis oxygen content and the carbon emission by on-site detection method into the assessment;

[0087] S3. Data comparison and analysis: obtain the measurement value of the material accounting algorithm model based on multivariate linear regression, and compare and analyze it with the reference value to verify the feasibility of the method and compare it with the reference value.

[0088] In step five, based on the modeling concept of slope correlation, the increments of each point in the time dimension and the indicator dimension are initially standardized, and the composition ratio and composition difference of the increments are comprehensively used to define the correlation coefficient, so as to make fuller use of the information covered by the data series and enhance the effectiveness of clustering.

[0089] Definition 1: The value of the sth index of the i-th sample in data X at time t is recorded as x i (s, t) where i = 1, 2, ..., N; s = 1, 2, ..., m; t = 1, 2, ..., n, then

[0090]

[0091] is the behavior matrix of sample i (i=1,2,…,N), then the data can be expressed as:

[0092] X={X1(s,t),…,X N (s,t)}

[0093] (s=1,2,…,m; t=1,2,…,n).

[0094] Definition 2: If the dimensions of the multi-dimensional indicators of the sample data are inconsistent, the behavior matrix of the sample must be initialized. Suppose the observed value of the sth indicator of the i-th sample in the data at time t is x i(s, t), the sth indicator observation value at the initial time t0 is x i (s,t0). If x i (s,t) satisfies

[0095] x i (s,t)D=x i (s,t) / x i (s,t0)

[0096] Then D is called the initialization operator.

[0097] Definition 3: Let the behavior matrix of sample i be

[0098]

[0099] Where i = 1, 2, ..., N, so that

[0100] H i (s,t)=x i (s,t)Dx i (s-1,t)D(s=2,3,...,m; t=1,2,...,n)

[0101] W i (s,t)=x i (s,t)Dx i (s,t-1)D(s=1,2,...,m; t=2,3,...,n)

[0102] Then it is called H i (s, t) is the increment of sample i from index s-1 to index s at time t, W i (s,t) is the increment of the s index of sample i from time point t-1 to time point t.

[0103]

[0104] (i=1,2,…,N; t=1,2,…,n)

[0105]

[0106] (i=1,2,…,N; s=1,2,…,m)

[0107] C i (t) represents the average absolute value of the increment of the indicator dimension of object i at time t; D i (s) is the average of the absolute values ​​of the increments of the s indicator of object i in each time period.

[0108]

[0109] (i=1,2,…,N; s=2,3,…,m; t=1,2,…,n)

[0110]

[0111] (i=1,2,…,N; s=1,2,…,m; t=2,3,…,n)

[0112] U i (s, t) represents the average increment of object i from index s-1 to s at time t, V i (s,t) represents the average increment of the s index of object i from time point t-1 to time point t.

[0113] Definition 4: Let data X i and X j , if U i (s,t),U j If (s, t) are not 0 at the same time, then

[0114]

[0115] If U i (s,t),U j (s, t) are both 0,

[0116]

[0117] Then it is called For X i , X j Correlation coefficient in the indicator dimension, (s=2,3,…,m; t=1,2,…,n).

[0118] Similarly, V i (s,t),V j If (s, t) are not 0 at the same time, then

[0119]

[0120] If V i (s,t),V j (s, t) are both 0,

[0121]

[0122] Then it is called For X i , X j Correlation coefficient in the time dimension, (s=1,2,…,m; t=2,3,…,n).

[0123] in

[0124]

[0125]

[0126] Definition 5: Suppose matrix data X i , X j , then it is called For X i , X j In the correlation of indicator dimension, For X i , X j The degree of correlation in the time dimension. For X i With X j The comprehensive correlation between them.

[0127] The accuracy of the above method is verified by examples below.

[0128] Under the premise of stable operation of the unit, four power generation equipment with different capacities, A, B, C, and D, are selected for analysis to explore the relationship characteristics between them and the change of carbon emission waves. In order to consider the different electricity consumption in different seasons, this paper selects the carbon emissions of these four units in March and July for measurement and analysis. The capacities of the four units A, B, C, and D are 500MW, 600MW, 1000MW, and 1000MW respectively.

[0129] In order to explore the impact of different loads on the carbon emission intensity of the unit, the carbon emissions of four units with different capacities were monitored in March and July using the method proposed in this paper. In the experiment, the average of the carbon emissions per unit of power generation under each load value represented the carbon emission intensity of each unit under the load. The results are shown in Figure 2. Figure 3 , Figure 4 and Figure 5 shown.

[0130] from Figure 3 and Figure 4 From the overall trend, we can see that the overall trend of the four units in different months is that with the increase of load, the carbon emission intensity decreases. In March and July, the carbon emission intensity values ​​of units C and D were between 0.78 and 0.85. The capacity of these two units is 1000MW, but the emission intensity of unit C is lower because unit C uses carbon capture agents to reduce the emission of carbon-containing substances in the exhaust gas.

[0131] exist Figure 5 July is the peak month for electricity consumption, which is significantly higher than that in March. From the carbon emission intensity of unit B, we can see that at the same load rate, as the power generation load increases, the carbon emission intensity gradually decreases.

[0132] The above has considered the carbon emission under different loads, but did not distinguish whether the unit is in a stable operating condition. This section selects generator sets A and C that have been running steadily for more than 3 hours as the research objects. These two units do not have frequent startup and shutdown. Under stable load conditions, the carbon emission intensity of generator sets A and C is as follows: Figure 6 and Figure 7 shown.

[0133] from Figure 6 and Figure 7 It can be seen that, except for some abnormal values, the carbon emissions of the units are concentrated in a certain range. The points of unit C are closer together, which shows that the smaller the change in carbon emission intensity at the same load point, the more stable the operating state of the unit, and the more stable its carbon emission characteristics will be. In addition, it can be concluded that there are differences in the carbon emission characteristics of different units. Compared with unit C, the change trend of the carbon emission intensity of unit A is linear and relatively smooth. That is, when the load of the generator set is low, unit A is more suitable for grid peak adjustment, which is more helpful to control carbon emissions.

[0134] The carbon emission indicators of four groups of power generation units were analyzed and compared by the method proposed in this paper, the time series prediction method and the PCA-BP method. The time series prediction method is actually a regression prediction method, which belongs to quantitative prediction. The basic principle is to recognize the continuity of the development of things, and to infer the development trend of things by conducting statistical analysis on past time series data. The training algorithm for time series prediction uses the LM algorithm. The number of hidden neurons is set to 10, the delay is 5, 70% of the samples are used as training data, and 15% of the remaining data are taken as verification and test data sets. The PCA-BP method is to reveal the intrinsic structure of many variables through the statistical method of principal component analysis, select several principal components to achieve the purpose of data dimensionality reduction, and then use the BP neural network for fitting. The correlation between the predicted values ​​and the actual values ​​of the aforementioned methods is compared. The results of the three methods are shown in the figure. Figure 8 shown.

[0135] Depend on Figure 8 It can be seen that compared with the time series prediction method and the PCA-BP method, the correlation degree of the method proposed in this paper shows significant advantages in different units under the above four different units. The time series prediction method is prone to deviations due to the influence of outliers, etc., because it predicts based on the statistical laws of past time series. The PCA-BP method is particularly important in the principal component analysis stage by selecting the form of dimensionality reduction of the principal component. If the principal component is biased, it will also affect the subsequent fitting process of the model.

[0136] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that this is only for illustration and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.

Claims

1. A method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology, characterized by: The analytical method comprises the following steps: Step 1: Obtaining a reference value: Taking the power supply as a reference standard, obtaining a reference value of the emission intensity of carbon dioxide; Step 2: Material accounting: collect data on fuel consumption and emission factors of thermal power plant units, and calculate emission based on the above data; Step 3: On-site testing: testing key auxiliary parameters at the end of the thermal power plant unit emission to further determine the concentration of particulate matter under standard conditions and the overall emission of flue gas; Step 4: Correlation analysis, combining the material accounting method with the on-site detection method to carry out data analysis; Step 5: Correlation modeling. Based on the results of step 4, focusing on the three-dimensional characteristics of the data, the sample behavior matrix is ​​divided into time dimension and indicator dimension. Based on the modeling concept of slope correlation, the increments of each point in the time dimension and indicator dimension are initially standardized, and the composition ratio and composition difference of the increments are comprehensively used to define the correlation coefficient.

2. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 1, characterized in that: In step 1, the reference value calculation formula is: Where Sgd is the carbon emission intensity of power supply, Wgd is the power supply in MWh, and a is the heat supply ratio in %.

3. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 1, characterized in that: In step 2, the monitoring of fossil fuel parameters required in the fuel combustion emission accounting process is mainly calculated according to the following formula: Where i is the fuel type; AD i is the activity data of the i-th fuel, in GJ; EF i is the carbon dioxide emission factor of the ith fuel, in tCO2 / GJ.

4. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 3, characterized in that: The AD i The calculation formula is: AD i =NCV i ×FC i Where NCV i is the average low calorific value of the i-th fuel; FC i is the net consumption of the ith fuel.

5. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 3, characterized in that: The EF i The calculation formula is: In the formula, CC i is the carbon content per unit calorific value of the i-th fuel, in tC / GJ; OF i is the carbon oxidation rate of the ith fuel, expressed in %.

6. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 1, characterized in that: In step three, the key auxiliary parameters are the temperature, pressure, flow rate, humidity and oxygen content of the flue gas.

7. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 1, characterized in that: In step 3, the emission of carbon dioxide in the flue gas is expressed in terms of volume concentration under standard conditions: Where, θ is the mean volume concentration of carbon dioxide emissions under standard conditions per unit time; Q i is the carbon dioxide concentration of the ith record; n is the number of data records per unit time.

8. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 1, characterized in that: In step 3, in order to calculate the volume of carbon dioxide passing through the flue cross section per unit time, it is necessary to calculate the standard dry flue gas emission per unit time at the measuring point Q g , the calculation formula is: Among them, t is the running time of the measuring point, s is the cross-sectional area of ​​the flue at the end of the measuring point, and ν gas The flue gas flow velocity at the end of the flue, Pa, is the sum of the atmospheric pressure at the measuring point and the flue pressure at the end of the flue at the measuring point, T g is the temperature of flue gas at the end of the measuring point; H g It is used to measure the humidity of flue gas at the end of the flue.

9. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 8, characterized in that: According to the average value of the carbon dioxide emission volume fraction per unit time at the measurement point and the standard dry flue gas emission, the carbon dioxide emission per unit time of the computer group is calculated. m The formula is:

10. The method for analyzing carbon emission characteristics of thermal power plant units based on correlation technology according to claim 1, characterized in that: The specific steps of step 4 are: S1, multivariate linear fitting, taking the dry oxygen content and carbon emission of the on-site detection method in the training data set as independent variables, and the carbon emission value of the material accounting method as the dependent variable, and carrying out multivariate linear fitting; S2, assessment input, substitute the dry basis oxygen content and the carbon emission by on-site detection method into the assessment; S3. Data comparison and analysis: obtain the measurement value of the material accounting algorithm model based on multivariate linear regression, and compare and analyze it with the reference value to verify the feasibility of the method and compare it with the reference value.

Images