A method for predicting carbon concentration and flue gas flow rate of coal-fired boilers

Abstract

The present invention provides a method for predicting carbon concentration and flue gas flow in a coal-fired boiler, belonging to the technical field of coal-fired units. The method comprises: collecting operating data and monitoring data of the coal-fired unit, performing data preprocessing to obtain corresponding first processed data and second processed data, dividing the first processed data and the second processed data into the same operating conditions and the same time period to obtain a data combination under each time period; analyzing the correlation between any two parameters under each data combination, and constructing a correlation matrix; assigning colors according to the correlation and according to color setting standards to obtain a thermal map, and then screening effective parameters; extracting final data related to the effective parameters from each array combination, and inputting the data into a neural network model to construct a prediction model; and predicting carbon concentration and flue gas flow. This method provides strong data support for model construction and provides reliable guarantee for subsequent prediction of carbon concentration and flue gas flow.

Description

Technical Field

[0001] The present invention relates to the technical field of coal-fired units, and in particular to a method for predicting carbon concentration and flue gas flow in a coal-fired boiler. Background Art

[0002] The accelerated industrial development has led to Emissions continue to rise, and It is the main factor causing global climate change. Coal-fired power plants are one of the major sources of global carbon emissions. Emissions account for 40% of the world's total emissions, of which 70% come from coal-fired power plants. The large-scale access of renewable energy to the power grid has increased the importance of coal-fired units as a supplementary energy source in the power system, but the volatility and intermittency of renewable energy require coal-fired units to maintain the stability of the power grid output through peak regulation. Under low-load conditions, the amount of pulverized coal, primary air volume and secondary air volume need to be adjusted to maintain combustion stability, but this will lead to a decrease in furnace temperature and incomplete combustion of fuel, causing more carbon to be converted into CO, thereby increasing the CO concentration in the flue gas during low-load operation. The volume ratio decreases, which poses a challenge to the detection and measurement of carbon emissions. The percentage and CO concentration will affect the capture performance of chemical or physical adsorption. Currently, coal-fired power plants mainly use the IPCC recommended method for carbon accounting, but the traditional calculation method based on emission factors has a lag and is difficult to match the actual production conditions of power plants.

[0003] Therefore, the present invention proposes a method for predicting carbon concentration and flue gas flow of a coal-fired boiler. Summary of the Invention

[0004] The present invention provides a method for predicting the carbon concentration and flue gas flow of a coal-fired boiler, which is used to collect and process the operating data and monitoring data of the coal-fired unit to obtain data combinations under different time periods, and then construct a thermal map by confirming the correlation, screen effective parameters, provide strong data support for model construction, ensure the prediction accuracy of the model, and provide reliable guarantee for subsequent prediction of carbon concentration and flue gas flow.

[0005] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler, comprising:

[0006] Step 1: collecting operating data and monitoring data of the coal-fired unit and performing data preprocessing to obtain corresponding first processed data and second processed data, wherein the operating data and monitoring data include data under steady-state conditions and data under unsteady-state conditions;

[0007] Step 2: Divide the first processed data and the second processed data into the same working condition and the same time period according to the first working period of the steady-state working condition and the second working period of the non-steady-state working condition, and obtain a data combination in each time period;

[0008] Step 3: Analyze the correlation between any two parameters under each data combination and construct a correlation matrix. According to the correlation and the color setting standard, the color is assigned to obtain a heat map, and then the effective parameters are screened;

[0009] Step 4: extracting the final data related to the effective parameters from each data combination and inputting it into the neural network model to construct a prediction model;

[0010] Step 5: Obtain the measurement data of the coal-fired boiler at the current moment and input it into the prediction model to predict the carbon concentration and flue gas flow rate.

[0011] Preferably, performing data preprocessing on the operating data to obtain corresponding first processed data includes:

[0012] Determine the operating data under the same working condition and sort them in chronological order to obtain a first data sequence for each operating parameter;

[0013] Plotting a curve for the first data sequence, and closing an upper and lower interval of the corresponding plotted curve according to an operating standard of a corresponding operating parameter under a corresponding working condition, to obtain a first area of ​​the corresponding plotted curve above the closed interval and a second area of ​​the corresponding plotted curve below the closed interval;

[0014] determining a coordination coefficient corresponding to an operating parameter according to the first area and the second area;

[0015]

[0016] in, represents the coordination coefficient of the y1th operating parameter; 1 represents the corresponding first area; represents the corresponding second area; Indicates the area of ​​the corresponding closed area; min indicates the minimum function;

[0017] Adjusting each parameter value in the first data sequence of the corresponding operating parameter according to the coordination coefficient to obtain a second data sequence;

[0018] First processed data is obtained based on all second data sequences.

[0019] Preferably, adjusting each parameter value in the first data sequence of the corresponding operating parameter according to the coordination coefficient includes:

[0020] Locking a first value in the first data sequence that does not match the corresponding operating standard, and adjusting the first value in combination with the coordination coefficient to obtain a second value;

[0021]

[0022] in, represents the corresponding second value; represents the corresponding first value; Respectively represent the lower boundary value and upper boundary value that match the corresponding operating standard; Represents the coordination coefficient of the y1th operating parameter.

[0023] Preferably, analyzing the correlation between any two parameters under each data combination includes:

[0024] According to the preset operation and monitoring comparison relationship, determine the operation parameter that has a comparison relationship with each monitoring parameter, and obtain a monitoring value sequence based on each monitoring parameter and an analysis operation set based on the operation time corresponding to each monitoring value from the data combination;

[0025] Based on the analysis run sets under any two monitoring parameters at the same running time in the same period, the operation similarity function sim(F1t, F2t) is constructed, where F1t and F2t represent the analysis run sets under any two monitoring parameters respectively;

[0026] Combine the monitored values ​​of any two monitoring parameters at the same operating time to construct the value difference;

[0027] The correlation between the two monitoring parameters at the corresponding operating moments is determined based on the numerical difference and the operating similarity function.

[0028] Preferably, determining the correlation between two monitoring parameters at corresponding operating times includes:

[0029] Calculate the initial properties corresponding to the two monitoring parameters;

[0030]

[0031] in, 、 Respectively represent the monitoring values ​​corresponding to any two monitoring parameters; Respectively represent the standard values ​​of any two monitoring parameters at the corresponding operating time; Indicates the initialization of the two monitoring parameters at the corresponding operating time;

[0032] Analyze all initial characteristics and determine the correlation between any two monitoring parameters;

[0033]

[0034] Where T represents the number of running times involved in the corresponding period; Indicates that based on all variance; Indicates that based on all The variance of .

[0035] Preferably, assigning colors based on the correlation and in accordance with the color setting standard to obtain a heat map includes:

[0036] Assign color labels to each correlation according to the color setting criteria;

[0037] Fill in the color of all the cells corresponding to the correlations according to the color labels to obtain a heat map.

[0038] Preferably, the effective parameters for screening include:

[0039] Filtering the number of covered and filled cells involved in each monitoring parameter from the heat map, and setting an important coefficient for each monitoring parameter in combination with a set relevant range of a corresponding color;

[0040] Parameters greater than the preset coefficients are selected from all important coefficients as valid parameters.

[0041] Preferably, an important coefficient is set for each monitoring parameter, including:

[0042]

[0043] in, represents the important coefficient of the i-th monitoring parameter; Indicates the number of filled cells involved in the corresponding heat map for the i-th monitoring parameter; Indicates the total number of filled cells involved in the heat map; ln represents the logarithmic function sign; It represents the sum of the corresponding correlations of the filled grids involved in the i-th monitoring parameter; Indicates the minimum value in the set range of the color with the largest number of filled grids under different colors corresponding to the i-th monitoring parameter; 2 represents the maximum value in the setting range of the color with the largest number of filled grids under different colors corresponding to the i-th monitoring parameter; Indicates the maximum value within the relevant range of all settings.

[0044] Compared with the prior art, the present invention has the following advantages:

[0045] By collecting and processing the operating data and monitoring data of coal-fired units, we obtain data combinations in different time periods, and then construct a heat map by confirming the correlation, and then screen effective parameters to provide strong data support for model construction, ensure the model's prediction accuracy, and provide reliable guarantees for subsequent predictions of carbon concentration and flue gas flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction is given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 This is a flow chart of a method for predicting carbon concentration and flue gas flow in a coal-fired boiler provided by an embodiment of the present invention;

[0048] Figure 2 This is a load and carbon emission trend diagram of an embodiment of the present invention;

[0049] Figure 3 This is a heat map showing the correlation of parameters according to an embodiment of the present invention;

[0050] Figure 4 is a prediction graph of the PPB model in an embodiment of the present invention;

[0051] Figure 5 In the embodiment of the present invention Volume ratio SHAP analysis chart;

[0052] Figure 6 This is a SHAP analysis diagram of CO concentration in an embodiment of the present invention;

[0053] Figure 7 This is a SHAP analysis diagram of flue gas flow in an embodiment of the present invention;

[0054] Figure 8 It is a structural diagram related to curve drawing of the present invention. DETAILED DESCRIPTION

[0055] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the embodiments described are only some of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0056] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler. Figure 1 Shown, including:

[0057] Step 1: collecting operating data and monitoring data of the coal-fired unit and performing data preprocessing to obtain corresponding first processed data and second processed data, wherein the operating data and monitoring data include data under steady-state conditions and data under unsteady-state conditions;

[0058] Step 2: Divide the first processed data and the second processed data into the same working condition and the same time period according to the first working period of the steady-state working condition and the second working period of the non-steady-state working condition, and obtain a data combination in each time period;

[0059] In this embodiment, the coal-fired unit may be unstable during the startup phase, for example, during the T1 period. In this case, the period is regarded as an unsteady-state period, and the data obtained during the period is data under unsteady-state conditions.

[0060] The coal-fired unit enters a stable operating state after the T1 period. At this time, this stage is regarded as a steady-state period, and the data obtained during this period is the data under steady-state conditions.

[0061] Step 3: Analyze the correlation between any two parameters under each data combination and construct a correlation matrix. According to the correlation and the color setting standard, the color is assigned to obtain a heat map, and then the effective parameters are screened;

[0062] Step 4: extracting the final data related to the effective parameters from each data combination and inputting it into the neural network model to construct a prediction model;

[0063] Step 5: Obtain the measurement data of the coal-fired boiler at the current moment and input it into the prediction model to predict the carbon concentration and flue gas flow rate.

[0064] In this embodiment, the operating data refers to data related to the combustion conditions within the coal-fired unit.

[0065] In this embodiment, the monitoring data refers to data such as carbon concentration and flue gas flow rate.

[0066] In this embodiment, data preprocessing refers to a process of correcting data with deviations, and the first processed data and the second processed data are both implemented using the same processing method.

[0067] In this embodiment, Table 1 summarizes the application of ML in the prediction of boiler gaseous pollutant emissions. Existing research mostly focuses on the prediction of NOx, SOx and oxygen content under steady-state conditions, while the prediction of CO and There are fewer studies on the prediction of volume ratio.

[0068]

[0069] Table 1 Application of ML in the prediction of boiler gaseous pollutant emissions

[0070] In this embodiment, the collected data comes from the DCS operation data and CEMS monitoring data of a 200MW unit in a power plant. The collection time is from 15:14 on September 25 to 13:14 on December 3, 2023, a total of 70 days, and the sampling interval is 1 minute.

[0071] In this embodiment, Figure 2 As shown, it shows the load and carbon emission trends, showing the unit load and The volume ratio is positively correlated and the CO concentration is negatively correlated.

[0072] In this embodiment, the monitoring parameters involved can be shown in the table:

[0073] Table 2 Main variables of boiler carbon emissions

[0074]

[0075] In this example, a heatmap is a visual representation that illustrates the correlation between datasets. It illustrates the correlation matrix between samples and features, with the strength of the correlation represented by color and specific values. This enables the identification of samples or features with high or low correlation. The heatmap used in this study is shown in Figure 2. Figure 3 As shown in the figure, the maximum correlation coefficient between N8 and Y1 is 0.35. Meanwhile, the absolute values ​​of the correlation coefficients for the remaining features are less than 0.25. The maximum correlation coefficient between N3 and Y2 is 0.34, and the absolute values ​​of the correlation coefficients for the remaining features are also less than 0.25. The correlation coefficient between N8 and Y3 is 0.5, representing the maximum value. Meanwhile, the absolute values ​​of the correlation coefficients for the remaining features are less than 0.25. Therefore, the linear correlation coefficients between the 15 features and the three target variables (Y1, Y2, and Y3) are relatively low. Therefore, the linear correlation between individual features in the dataset and the target variables is very small and cannot be directly used as model predictor input variables. Principal component analysis (PCA) was used to reduce the 15-dimensional data to 6 principal components (with a cumulative contribution rate of 92.86%), which were used as model input.

[0076] In this example, model construction is divided into four stages: data preprocessing, training, parameter optimization, and performance evaluation. 80% of the data is used for training, and 20% for testing. PSO optimizes BPNN weights and thresholds, and model performance is evaluated using MAE, RMSE, and MAPE.

[0077] In this embodiment, MAE, RMSE, and MAPE are used to evaluate the model performance, and the formulas are as follows: , , ;

[0078] In this embodiment, Figure 4 As shown, the PPB model (PCA-PSO-BPNN) is effective in predicting The deviation is the smallest when the volume ratio is used, and the MAE, RMSE and MAPE are 0.381, 0.576 and 0.031 respectively, which is better than other models.

[0079] In this embodiment, during the model prediction process, it can be seen that the smoke humidity has an impact on Volume ratio, CO concentration and flue gas flow rate have the greatest impact on prediction and are positively correlated; unit load is negatively correlated with flue gas flow rate, such as Figure 5-7 shown.

[0080] The beneficial effects of the above technical solution are: by collecting and processing the operating data and monitoring data of the coal-fired units, data combinations in different time periods are obtained, and then a thermal map is constructed by confirming the correlation, and then effective parameters are screened to provide strong data support for model construction, ensure the prediction accuracy of the model, and provide reliable guarantees for subsequent predictions of carbon concentration and flue gas flow.

[0081] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler, which performs data preprocessing on operating data to obtain corresponding first processed data, including:

[0082] Determine the operating data under the same working condition and sort them in chronological order to obtain a first data sequence for each operating parameter;

[0083] Plotting a curve for the first data sequence, and closing an upper and lower interval of the corresponding plotted curve according to an operating standard of a corresponding operating parameter under a corresponding working condition, to obtain a first area of ​​the corresponding plotted curve above the closed interval and a second area of ​​the corresponding plotted curve below the closed interval;

[0084] determining a coordination coefficient corresponding to an operating parameter according to the first area and the second area;

[0085]

[0086] in, represents the coordination coefficient of the y1th operating parameter; 1 represents the corresponding first area; represents the corresponding second area; Indicates the area of ​​the corresponding closed area; min indicates the minimum function;

[0087] Adjusting each parameter value in the first data sequence of the corresponding operating parameter according to the coordination coefficient to obtain a second data sequence;

[0088] First processed data is obtained based on all second data sequences.

[0089] In this embodiment, because the operating data is composed of data of different operating parameters at different times, the first data sequence can be directly obtained in chronological order. Therefore, the first data sequence = {operating values ​​of the corresponding operating parameters at each time point}.

[0090] In this embodiment, the curve is drawn in chronological order.

[0091] In this embodiment, the operating standards are pre-set, such as Figure 8 As shown in the figure, u01 is the plotted curve corresponding to the parameter, a1 is the upper boundary of the operating standard, and a2 is the lower boundary of the operating standard. Part A1 corresponds to area S1, and part A2 corresponds to area S2. A3 is a closed interval.

[0092] The beneficial effect of the above technical solution is: by sorting in chronological order to obtain the first data sequence, and then combining it with the drawn curve, the upper and lower areas can be effectively obtained, and then the coordination coefficient can be obtained, which facilitates the adjustment of subsequent values.

[0093] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler, which adjusts each parameter value in a first data sequence of corresponding operating parameters according to the coordination coefficient, including:

[0094] Locking a first value in the first data sequence that does not match the corresponding operating standard, and adjusting the first value in combination with the coordination coefficient to obtain a second value;

[0095]

[0096] in, represents the corresponding second value; represents the corresponding first value; Respectively represent the lower boundary value and upper boundary value that match the corresponding operating standard; Represents the coordination coefficient of the y1th operating parameter.

[0097] In this embodiment, the operating standard and the upper and lower boundary values ​​that match the operating standard are all pre-set.

[0098] The beneficial effect of the above technical solution is: by adjusting the first value based on the upper boundary value and the lower boundary value, the second value can be effectively obtained, thereby ensuring the validity of the subsequent first processed data.

[0099] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler, which analyzes the correlation between any two parameters under each data combination, including:

[0100] According to the preset operation and monitoring comparison relationship, determine the operation parameter that has a comparison relationship with each monitoring parameter, and obtain a monitoring value sequence based on each monitoring parameter and an analysis operation set based on the operation time corresponding to each monitoring value from the data combination;

[0101] Based on the analysis run sets under any two monitoring parameters at the same running time in the same period, the operation similarity function sim(F1t, F2t) is constructed, where F1t and F2t represent the analysis run sets under any two monitoring parameters respectively;

[0102] Combine the monitored values ​​of any two monitoring parameters at the same operating time to construct the value difference;

[0103] The correlation between the two monitoring parameters at the corresponding operating moments is determined based on the numerical difference and the operating similarity function.

[0104] In this embodiment, the comparison relationship is set. For example, the operating parameters 1, 2 and 3 of the coal-fired unit are compared with the monitoring parameter 01. At this time, the data of the operating parameters 1, 2 and 3 related to the monitoring parameter 01 can be retrieved from the data combination. At this time, the corresponding analysis operation set can be obtained, that is, the data in the operation set is part of the first processed data.

[0105] In this embodiment, the numerical difference is the result of subtracting two monitoring numerical values.

[0106] The beneficial effect of the above technical solution is: determining the operating parameters where the monitoring parameters exist through the control relationship, and then constructing the relevant analysis operation set, and subsequently determining the correlation between the two corresponding monitoring parameters through similarity analysis of the relevant analysis operation sets under the two monitoring parameters, providing a basis for subsequent screening of effective variables.

[0107] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler, which determines the correlation between two corresponding monitoring parameters at corresponding operating times, including:

[0108] Calculate the initial properties corresponding to the two monitoring parameters;

[0109]

[0110] in, 、 Respectively represent the monitoring values ​​corresponding to any two monitoring parameters; Respectively represent the standard values ​​of any two monitoring parameters at the corresponding operating time; Indicates the initialization of the two monitoring parameters at the corresponding operating time;

[0111] Analyze all initial characteristics and determine the correlation between any two monitoring parameters;

[0112]

[0113] Where T represents the number of running times involved in the corresponding period; Indicates that based on all variance; Indicates that based on all The variance of .

[0114] The beneficial effect of the above technical solution is: determining the initiality based on the standard value and the monitoring value, and then effectively determining the correlation between the monitoring parameters through analysis of the initiality.

[0115] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler, which assigns colors based on correlation and according to color setting standards to obtain a thermal map, including:

[0116] Assign color labels to each correlation according to the color setting criteria;

[0117] Fill in the color of all the cells corresponding to the correlations according to the color labels to obtain a heat map.

[0118] In this embodiment, the color setting standard includes which colors correspond to different correlation ranges, which is pre-set and can be used directly.

[0119] The beneficial effect of the above technical solution is that it is easy to construct a heat map by filling the filled cells with colors.

[0120] The present invention provides a method for predicting carbon concentration and flue gas flow of a coal-fired boiler, which screens effective parameters, including:

[0121] Filtering the number of covered and filled cells involved in each monitoring parameter from the heat map, and setting an important coefficient for each monitoring parameter in combination with a set relevant range of a corresponding color;

[0122] Parameters greater than the preset coefficients are selected from all important coefficients as valid parameters.

[0123] Preferably, an important coefficient is set for each monitoring parameter, including:

[0124]

[0125] in, represents the important coefficient of the i-th monitoring parameter; Indicates the number of filled cells involved in the corresponding heat map for the i-th monitoring parameter; Indicates the total number of filled cells involved in the heat map; ln represents the logarithmic function sign; It represents the sum of the corresponding correlations of the filled grids involved in the i-th monitoring parameter; Indicates the minimum value in the set range of the color with the largest number of filled grids under different colors corresponding to the i-th monitoring parameter; 2 represents the maximum value in the setting range of the color with the largest number of filled grids under different colors corresponding to the i-th monitoring parameter; Indicates the maximum value within the relevant range of all settings.

[0126] In this embodiment, for example, the setting range of dark blue is: 0.94 to 1.

[0127] The beneficial effect of the above technical solution is: based on the number of filled grids involved in the monitoring parameters and the relevant range of color settings, important coefficients are set for the monitoring parameters, thereby effectively screening the required parameters and providing accurate samples for subsequent model construction.

[0128] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0129] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0130] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for predicting carbon concentration and flue gas flow of a coal-fired boiler, characterized in that: include: Step 1: collecting operating data and monitoring data of the coal-fired unit and performing data preprocessing to obtain corresponding first processed data and second processed data, wherein the operating data and monitoring data include data under steady-state conditions and data under unsteady-state conditions; Step 2: Divide the first processed data and the second processed data into the same working condition and the same time period according to the first working period of the steady-state working condition and the second working period of the non-steady-state working condition, and obtain a data combination in each time period; Step 3: Analyze the correlation between any two parameters under each data combination and construct a correlation matrix. According to the correlation and the color setting standard, the color is assigned to obtain a heat map, and then the effective parameters are screened; Step 4: extracting the final data related to the effective parameters from each data combination and inputting it into the neural network model to construct a prediction model; Step 5: Obtain the measurement data of the coal-fired boiler at the current moment and input it into the prediction model to predict the carbon concentration and flue gas flow rate.

2. The method for predicting carbon concentration and flue gas flow of a coal-fired boiler according to claim 1, characterized in that: Performing data preprocessing on the operating data to obtain corresponding first processed data includes: Determine the operating data under the same working condition and sort them in chronological order to obtain a first data sequence for each operating parameter; Plotting a curve for the first data sequence, and closing an upper and lower interval of the corresponding plotted curve according to an operating standard of a corresponding operating parameter under a corresponding working condition, to obtain a first area of ​​the corresponding plotted curve above the closed interval and a second area of ​​the corresponding plotted curve below the closed interval; determining a coordination coefficient corresponding to an operating parameter according to the first area and the second area; in, represents the coordination coefficient of the y1th operating parameter; 1 represents the corresponding first area; represents the corresponding second area; Indicates the area of ​​the corresponding closed area; min indicates the minimum function; Adjusting each parameter value in the first data sequence of the corresponding operating parameter according to the coordination coefficient to obtain a second data sequence; First processed data is obtained based on all second data sequences.

3. The method for predicting carbon concentration and flue gas flow of a coal-fired boiler according to claim 2, characterized in that: Adjusting each parameter value in the first data sequence of the corresponding operating parameter according to the coordination coefficient includes: Locking a first value in the first data sequence that does not match the corresponding operating standard, and adjusting the first value in combination with the coordination coefficient to obtain a second value; in, represents the corresponding second value; represents the corresponding first value; Respectively represent the lower boundary value and upper boundary value that match the corresponding operating standard; Represents the coordination coefficient of the y1th operating parameter.

4. The method for predicting carbon concentration and flue gas flow of a coal-fired boiler according to claim 1, characterized in that: Analyze the correlation between any two parameters under each data combination, including: According to the preset operation and monitoring comparison relationship, determine the operation parameter that has a comparison relationship with each monitoring parameter, and obtain a monitoring value sequence based on each monitoring parameter and an analysis operation set based on the operation time corresponding to each monitoring value from the data combination; Based on the analysis run sets under any two monitoring parameters at the same running time in the same period, the operation similarity function sim(F1t, F2t) is constructed, where F1t and F2t represent the analysis run sets under any two monitoring parameters respectively; Combine the monitored values ​​of any two monitoring parameters at the same operating time to construct the value difference; The correlation between the two monitoring parameters at the corresponding operating moments is determined based on the numerical difference and the operating similarity function.

5. The method for predicting carbon concentration and flue gas flow of a coal-fired boiler according to claim 4, characterized in that: Determine the correlation between two monitoring parameters at corresponding operating times, including: Calculate the initial properties corresponding to the two monitoring parameters; in, 、 Respectively represent the monitoring values ​​corresponding to any two monitoring parameters; Respectively represent the standard values ​​of any two monitoring parameters at the corresponding operating time; Indicates the initialization of the two monitoring parameters at the corresponding operating time; Analyze all initial characteristics and determine the correlation between any two monitoring parameters; Where T represents the number of running times involved in the corresponding period; Indicates that based on all variance; Indicates that based on all The variance of .

6. The method for predicting carbon concentration and flue gas flow of a coal-fired boiler according to claim 1, characterized in that: The heat map is obtained by assigning colors based on correlation and according to color setting standards, including: Assign color labels to each correlation according to the color setting criteria; Fill in the color of all the cells corresponding to the correlations according to the color labels to obtain a heat map.

7. The method for predicting carbon concentration and flue gas flow of a coal-fired boiler according to claim 6, characterized in that: Valid screening parameters include: Filtering the number of covered and filled cells involved in each monitoring parameter from the heat map, and setting an important coefficient for each monitoring parameter in combination with a set relevant range of a corresponding color; Parameters greater than the preset coefficients are selected from all important coefficients as valid parameters.

8. The method for predicting carbon concentration and flue gas flow of a coal-fired boiler according to claim 7, characterized in that: Set importance coefficients for each monitoring parameter, including: in, represents the important coefficient of the i-th monitoring parameter; Indicates the number of filled cells involved in the corresponding heat map for the i-th monitoring parameter; Indicates the total number of filled cells involved in the heat map; ln represents the logarithmic function sign; It represents the sum of the corresponding correlations of the filled grids involved in the i-th monitoring parameter; Indicates the minimum value in the set range of the color with the largest number of filled grids under different colors corresponding to the i-th monitoring parameter; 2 represents the maximum value in the setting range of the color with the largest number of filled grids under different colors corresponding to the i-th monitoring parameter; Indicates the maximum value within the relevant range of all settings.

Images