Method for testing bulk density of coal yard of coal-fired power plant based on big data

Abstract

The invention discloses a coal-fired power plant coal yard stacking density testing method based on big data, which comprises the following steps: classifying all coal types according to moisture, ash content, volatile components and compaction times, and then respectively measuring the stacking density D0, ar of the coal type with known weight in a loose free state and the stacking density Dmax, ar under the maximum pressure according to the types; calculating the average stacking density of the coal pile compacted for n times under the actual stacking state condition and the loose free state, and calculating a series of regression coefficients k; fitting a one-time regression model according to the moisture, the ash content, the volatile component and the number of compaction times; calculating a k value according to the moisture, the ash content, the volatile components, the compaction times and the one-time regression model of the coal pile to be detected, and then predicting the stacking density through a formula # imgabs0 #. According to the method, the k value linear regression equation is obtained through big data fitting, so that the accuracy of the predicted bulk density is very high, and the method is a feasible method for testing the stocktaking bulk density of the coal-fired power plant coal yard.

Description

Technical Field

[0001] The present invention belongs to the technical field of coal yard inventory, and in particular relates to a method for testing the stacking density of a coal yard of a coal-fired power plant based on big data. Background Art

[0002] Each modern coal-fired power plant consumes more than one million tons of coal every year. The fuel cost of a thermal power plant accounts for 80% of the total production cost. Regular inventory of fuel is the only way to ensure a reasonable supply. The inventory of coal in the coal yard is related to the calculation of coal consumption and economic indicators of the power plant, and is also the focus and difficulty of fuel management. Therefore, establishing a scientific coal inventory system plays a very important role in reasonable coal storage, ensuring the quality of coal entering the furnace, and reducing consumption. It is also an effective way to reduce power generation costs and improve the economic benefits of power plants.

[0003] At present, the test method for coal yard bulk density is mainly implemented in accordance with DL / T 1878-2018 Coal Storage Yard Inventory Guidelines for Coal-fired Power Plants. The bulk density is mainly obtained by the following three methods. 1. Actual measurement method: This method has a relatively complex operation process, requires a large amount of data, and has a large error. 2. Empirical value method: Estimate the bulk density of the coal yard based on historical data and empirical values. This method is simple and easy to implement, but its accuracy is affected by the accuracy of the empirical value. 3. Simulation method: This method is relatively simple to operate, but it is necessary to consider the difference between simulation conditions and actual conditions. The current regression coefficient k value of the power plant is generally 0.5. The bulk density error is large. Therefore, by using a reasonable k value, the true value of the coal yard bulk density can be obtained more accurately, thereby realizing accurate inventory of the coal yard. Summary of the invention

[0004] The purpose of the present invention is to provide a method for testing the bulk density of coal yards in coal-fired power plants based on big data. A linear regression equation with a regression coefficient k value is obtained by fitting big data. The bulk density measured by this method has high accuracy and is a feasible method for testing the bulk density of coal yards in coal-fired power plants.

[0005] The technical solution of the present invention is: a method for testing the bulk density of coal yards in coal-fired power plants based on big data, the steps are as follows: Step S1: Classify all coal types according to moisture, ash, volatile matter, and compaction times, and then measure the bulk density D of coal types with known mass in loose free state according to the categories. 0,ar and bulk density at maximum pressure D max,ar ; Step S2: Use a coal pan meter to measure the volume of a coal pile of known mass, obtain the volume V of the coal pile, and calculate the stacking density of the coal pile under the actual stacking state conditions. ; Step S3: According to Calculate the regression coefficient k; Step S4: Use a coal pan meter to measure the volume of the coal pile in a loose free state and after compaction for n times, where n=1, 2, 3, ..., N, and N is the total number of compaction times; calculate a series of regression coefficients according to step S3; Step S5: fitting a regression model according to moisture, ash, volatile matter, and compaction times; Step S6: Calculate the k value using a linear regression model based on the moisture, ash, volatile matter, and compaction times of the coal pile to be tested, and then use the formula Predicted bulk density .

[0006] Further preferably, the bulk density D in the loose free state is 0,ar Calculate as follows: ; Where: It represents the mass of the container when it is filled with coal sample in a loose and free state; m0 is the mass of the empty container; V0 is the volume of the container.

[0007] Further preferably, the bulk density D at maximum pressure max,ar Calculate as follows: ; Where: It represents the mass of the container when it is filled with coal samples under maximum pressure; m0 is the mass of the empty container; V0 is the volume of the container.

[0008] Further preferably, the stacking density of the coal pile under the actual stacking state is calculated as follows: ; Where: G re is the mass of the coal pile; V is the volume of the coal pile.

[0009] Further preferably, the linear regression model is: ; In the formula: C is the fitting constant; a, b, c, d are fitting coefficients; M is moisture; A is ash content; is the volatile matter; n is the number of compactions.

[0010] Further preferably, the process conditions and parameters of each compaction operation are the same.

[0011] The present invention measures the volume and bulk density D in a loose free state according to the water content, ash content, volatile matter and number of compaction of the coal of known mass. 0,ar and bulk density at maximum pressure D max,arDetermination. According to the relationship between mass, volume and bulk density, a series of regression coefficients k values ​​can be calculated, and then a linear regression model of k values ​​can be obtained. It is not necessary to measure the bulk density, but to predict the bulk density. This method has a high accuracy in predicting the bulk density, and is a feasible method for testing the bulk density of coal yard inventory in coal-fired power plants. DETAILED DESCRIPTION

[0012] The technical solutions in the embodiments of the present invention will be described clearly and completely below in combination with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0013] A method for testing the bulk density of coal in a coal-fired power plant based on big data, the steps are as follows: Step S1: According to the coal feeding structure of the thermal power plant, all coal types are classified according to moisture, ash, volatile matter, and compaction times, and then the bulk density D of the loose free state of the coal types with known mass is measured by category. 0,ar and bulk density at maximum pressure D max,ar ; Bulk density D in loose free state 0,ar Determination: Slowly add a small amount of water into a container of known volume (0.125m 3 ) The coal sample should be placed in the container. The drop height of the coal sample should be as small as possible, and the maximum should not exceed 0.6m. The coal sample should be placed until the entire coal surface is about 100mm above the top of the container. The coal sample that is above the container should be scraped off with a hard straight plate to make the coal sample surface flush with the top of the container. The bulk density in a loose free state is D 0,ar Calculate as follows: ; Where: It represents the mass of the container when it is filled with coal sample in a loose and free state; m0 is the mass of the empty container; V0 is the volume of the container.

[0014] Bulk density at maximum pressure D max,arDetermination: According to the actual situation on site, place the density box steadily into the coal pit dug by the excavator, and then load the coal sample for density determination into the density box. After it is full, add coal to a height of 200mm~300mm, pad it with thickened plastic film, fill the coal pit with the excavated coal, and pave it flat. Use the crawler-type coal pusher or loader used on site during stacking to walk directly above the density box. Make sure that the tracks or wheels of the coal pusher or loader are facing the density box when walking. When all the tracks pass through or the front and rear wheels pass through the density box, one rolling is completed. Carefully dig out or lift the density box with an excavator. If the amount of coal in the density box after compaction is significantly lower than the upper edge of the box, the operation fails. Use a scraper to scrape off the stored coal above the upper edge of the density box. Use an electronic scale to weigh the mass of the density box filled with coal, accurate to 0.1kg. According to the above steps, the rolling test can be carried out 3 times, 5 times, 7 times, 10 times, ..., N times respectively. The final rolling test number is based on the change of the compaction density before and after not exceeding 5%. The bulk density under the maximum pressure D max,ar Calculate as follows: ; Where: It represents the mass of the container when it is filled with coal samples under maximum pressure; m0 is the mass of the empty container; V0 is the volume of the container.

[0015] Step S2: Use a coal pan meter to measure the volume of a coal pile of known mass to obtain the volume V of the coal pile; calculate the stacking density of the coal pile under actual stacking conditions according to the following formula: : ; Where: G re is the mass of the coal pile; V is the volume of the coal pile.

[0016] Step S3: According to Calculate the regression coefficient k; Step S4: Use a coal pan meter to measure the volume of the coal pile in a loose free state and after compaction for n times, where n=1, 2, 3, ..., N, and N is the total number of compaction times; calculate a series of regression coefficients according to step S3; it should be noted that the process conditions and parameters of each compaction operation are the same; Step S5: Fit a regression model based on moisture, ash, volatile matter, and compaction times: ; In the formula: C is the fitting constant; a, b, c, d are fitting coefficients; M is moisture; A is ash content; is the volatile matter; n is the number of compaction times; Step S6: Calculate the k value using a linear regression model based on the moisture, ash, volatile matter, and compaction times of the coal pile to be tested, and then use the formula Predicted bulk density .

[0017] The present invention selects 28 types of coal entering a thermal power plant, and conducts k value tests under the stacking processes of free loose state, 2 times compaction, and 4 times compaction, as shown in Table 1:

[0018] The linear relationship of the present invention and the Durbin-Watson index are R 2 >0.9、1.5 <DW<2.5。

[0019] Table 2: Linear correlation coefficients and Durbin-Watson DW values.

[0020]

[0021] In Table 2, the linear correlation coefficient R and adjusted R 2 The Durbin-Watson DW value is 1.921, indicating that the first-order autocorrelation of the residuals in the regression analysis is very low, that is, there is no obvious correlation between the error terms of the linear regression model, indicating that the linear regression model of the present invention has high accuracy and reliability.

[0022] The measured data in Table 1 above were subjected to univariate regression fitting, and the fitting constants and coefficients are shown in Table 3. The formula for fitting the linear regression model is: .

[0023] Table 3. Fitting constants and coefficients

[0024] In Table 3, B represents the original impact of the independent variable on the dependent variable, and Beta represents the impact intensity after eliminating the dimension.

[0025] In order to verify the accuracy of the model formula, the stacking density of other types of coal in the power plant was tested, and the k value calculated according to the prediction model formula of the present invention was calculated by the formula Predicted bulk density, the results are shown in Table 4.

[0026] Table 4. Comparison of actual detection values ​​and predicted values

[0027] As can be seen from Table 4, the difference between the actual test results and the predicted results of the model formula is less than the repeatability limit of 0.10 t / m specified in the DL / T1878-2018 standard. 3 requirements.

[0028] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for testing the bulk density of coal in a coal-fired power plant based on big data, characterized in that: Here are the steps: Step S1: Classify all coal types according to moisture, ash, volatile matter, and compaction times, and then measure the bulk density D of coal types with known mass in loose free state according to the categories. 0,ar and bulk density at maximum pressure D max,ar ; Step S2: Use a coal pan meter to measure the volume of a coal pile of known mass, obtain the volume V of the coal pile, and calculate the stacking density of the coal pile under the actual stacking state conditions. ; Step S3: According to Calculate the regression coefficient k; Step S4: Use a coal pan meter to measure the volume of the coal pile in a loose free state and after compaction for n times, where n=1, 2, 3, ..., N, and N is the total number of compaction times; calculate a series of regression coefficients according to step S3; Step S5: fitting a regression model according to moisture, ash, volatile matter, and compaction times; Step S6: Calculate the k value using a linear regression model based on the moisture, ash, volatile matter, and compaction times of the coal pile to be tested, and then use the formula Predicted bulk density .

2. The method for testing the bulk density of coal in a coal-fired power plant based on big data according to claim 1, characterized in that: Bulk density D in loose free state 0,ar Calculate as follows: ； Where: It represents the mass of the container when it is filled with coal sample in a loose and free state; m0 is the mass of the empty container; V0 is the volume of the container.

3. The method for testing the bulk density of coal in a coal-fired power plant based on big data according to claim 1, characterized in that: Bulk density at maximum pressure D max,ar Calculate as follows: ； Where: It represents the mass of the container when it is filled with coal samples under maximum pressure; m0 is the mass of the empty container; V0 is the volume of the container.

4. The method for testing the bulk density of coal in a coal-fired power plant based on big data according to claim 1, characterized in that: The stacking density of coal pile under actual stacking conditions is calculated as follows: ； Where: G re is the mass of the coal pile; V is the volume of the coal pile.

5. The method for testing the bulk density of coal in a coal-fired power plant based on big data according to claim 1, characterized in that: The linear regression model is: ; In the formula: C is the fitting constant; a, b, c, d are fitting coefficients; M is moisture; A is ash content; is the volatile matter; n is the number of compactions.

6. The method for testing the bulk density of coal in a coal-fired power plant based on big data according to claim 1, characterized in that: The process conditions and parameters were the same for each compaction operation.

Images