Carbon emission prediction method for ceramic industry based on power data

Through a gray prediction model based on power data, the relationship between carbon emissions and electricity consumption in the ceramic industry is established, and the accuracy and cost of carbon emission prediction in the ceramic industry is solved, and efficient and reliable carbon emission prediction is achieved.

CN114399110BActive Publication Date: 2024-11-01国网福建省电力有限公司营销服务中心 +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210040452.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-01-14
Publication Date
2024-11-01
Estimated Expiration
2042-01-14

AI Technical Summary

Technical Problem

It is difficult to accurately predict carbon emissions in the ceramic industry, resulting in high testing costs and low reliability, and it is impossible to effectively cope with the pressure of carbon emission reduction.

Method used

Using the gray prediction model (GM(0,1) method based on power data, the carbon emission sources in the production process of the ceramic industry are analyzed, and the correlation between carbon emissions and electricity consumption is established, the prediction model is constructed, and the singular values ​​are eliminated. Make predictions later.

Benefits of technology

It realizes accurate prediction of carbon emissions, reduces detection costs, improves detection reliability, and improves fitting accuracy and prediction effect.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114399110B_ABST
    Figure CN114399110B_ABST
Patent Text Reader

Abstract

The present invention relates to a method for predicting carbon emissions in the ceramic industry based on power data, comprising the following steps: step S1: determining the carbon emission source by analyzing the production process of the ceramic industry, and obtaining the carbon emissions of each carbon emission source in the historical years of the ceramic enterprise and the electricity consumption data in each historical year; step S2: pre-processing the data obtained in step S1 to remove singular values; step S3: establishing a correlation relationship between electricity and carbon emissions of each carbon emission source based on the gray prediction model GM (0,1) model, and then obtaining a correlation relationship between electricity and total carbon emissions, and constructing a prediction model; step S4: inputting the data to be tested into the prediction model to obtain the carbon emission prediction value. The present invention can accurately realize the prediction of carbon emissions, effectively reduce the detection cost, and improve the detection reliability.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission calculation, and in particular to a method for predicting carbon emission in the ceramic industry based on power data. Background Art

[0002] In recent years, the problem of global warming has continued to worsen, and carbon emissions from global primary energy consumption have increased from 24.69 billion tons in 2000 to 33.8 billion tons in 2016.

[0003] Ceramics is one of the important raw materials for my country's national economy and urbanization. Since 1993, my country's production of building sanitary ceramics has been ranked first in the world. In 2017, the production of building ceramics reached 10.1 billion m 2 , accounting for about 3 / 5 of the world's total output, and sanitary ceramics accounted for about 1 / 3 of the world's output during the same period. With the continuous increase in the output of building sanitary ceramics, the total carbon emissions have also maintained a rapid growth. In 2010, the building sanitary ceramics industry has reached an annual emission of 126 million tons of CO2. As an energy-intensive industry, the ceramics industry has been listed as a key industry for carbon emission reduction. The ceramics industry faces very severe pressure to reduce emissions and the task is very arduous. Therefore, how to accurately predict carbon emissions will help to better achieve low-carbon development in the ceramics industry. Summary of the invention

[0004] In view of this, the purpose of the present invention is to provide a method for predicting carbon emissions in the ceramic industry based on electricity data, which can accurately predict carbon emissions, effectively reduce detection costs, and improve detection reliability.

[0005] To achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for predicting carbon emissions in the ceramic industry based on power data comprises the following steps:

[0007] Step S1: Determine the carbon emission sources by analyzing the production process of the ceramic industry, and obtain the carbon emissions of each carbon emission source and the electricity consumption data of each historical year of the ceramic enterprise;

[0008] Step S2: preprocess the data obtained in step S1 to remove singular values;

[0009] Step S3: Based on the grey prediction model GM (0,1) model, establish the correlation between electricity and carbon emissions of each carbon emission source, and then obtain the correlation between electricity and total carbon emissions, and construct a prediction model;

[0010] Step S4: Input the data to be measured into the prediction model to obtain the carbon emission prediction value.

[0011] Furthermore, the carbon emission sources include indirect emissions from purchased / transmitted electricity, direct emissions from fossil fuel combustion, and direct emissions from decomposition of raw carbonate minerals.

[0012] Furthermore, the preprocessing includes:

[0013] Clean missing values ​​and format content. After cleaning the data, remove singular values ​​and calculate the comprehensive energy consumption per unit product and the power consumption per unit product according to the energy consumption limit standard for ceramic production enterprises. The definitions are:

[0014]

[0015] Data that is smaller than the limit value is defined as a singular value and removed.

[0016] Furthermore, the step S3 is specifically as follows: based on the processed carbon emission data and electricity consumption data, the carbon emissions of each carbon emission source in each historical year of the ceramic enterprise are used as the system characteristic data sequence, and the electricity consumption data of the ceramic enterprise in each historical year is used as the related factor sequence, and the GM (0,1) model is used to fit the relationship between the indirect emissions P of net purchased / transmitted electricity, the direct emissions F from fossil fuel combustion, and the direct emissions R from the decomposition of raw carbonate minerals and the electricity consumption data E.

[0017] Furthermore, the relationship is obtained as follows:

[0018] (1) Determine the system behavior characteristic data sequence P (0) 、F (0) , R (0) and the related factor sequence E (0) ;

[0019] (2) Calculate the cumulative generated sequence (1-AGO sequence) P (1) 、F (1) , R (1) 、E (1) , that is, the same type of data in the original sequence is accumulated in sequence to form a new sequence. Let the original data of indirect emissions of net purchased electricity be P (0) = {P (0) (1),P (0) (2),......,P (0) (n)},

[0020] The sequence generated by one accumulation is P (1) = {P 1 (1),P 1 (2),......,P 1 (n)}, where:

[0021]

[0022] (3) Generate the model input matrix B and model output vector Y from the 1-AGO sequence;

[0023] (4) According to u=(B T B) -1 B T Solving Y yields the least squares solution u = [a, b] T ;

[0024] (5) Therefore, the relation is as follows: (1) (k) = a + bE (1) (k), k is the data dimension;

[0025] (6) According to P (0) (k) = P (1) (k)-P (1) (k-1), restore the original feature data P (0) (k).

[0026] Furthermore, the electricity consumption data is input into the association relationship of the prediction model, each association relationship obtains a prediction result value, and the three types of predicted carbon emissions are added together to obtain the final carbon emission prediction value.

[0027] Compared with the prior art, the present invention has the following beneficial effects:

[0028] The present invention can accurately predict carbon emissions, effectively reduce detection costs, and improve detection reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 is a flow chart of the method of the present invention;

[0030] Figure 2 is a fitting diagram of the correlation between indirect emissions from net purchased electricity and electricity consumption in one embodiment of the present invention;

[0031] Figure 3 is a fitting diagram of the correlation between direct emissions from fossil fuel combustion and electricity consumption in one embodiment of the present invention;

[0032] Figure 4 is a fitting diagram of the correlation between direct emissions from decomposition of raw carbonate minerals and electricity consumption in one embodiment of the present invention;

[0033] Figure 5 A fitting diagram of the correlation between total carbon emissions and electricity consumption calculated indirectly by decomposing carbon emissions in one embodiment of the present invention;

[0034] Figure 6 This is a fitting diagram of the correlation between the directly calculated total carbon emissions and electricity consumption in one embodiment of the present invention;. DETAILED DESCRIPTION

[0035] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0036] Please refer to Figure 1 The present invention provides a method for predicting carbon emissions in the ceramic industry based on power data, comprising the following steps:

[0037] Step S1: According to the production process of the ceramic industry: powder making-pressing-glazing-firing, the corresponding carbon emission sources of the ceramic industry include: fossil fuel combustion, industrial production process, purchased electricity, and purchased heat. The relevant data in the carbon emission verification report of the tested ceramic enterprise are counted, and the carbon emissions of each carbon emission source in each historical year and the electricity consumption data E and ceramic production data S of the tested ceramic enterprise are calculated. Among them, each carbon emission source includes indirect emissions P from net purchased / transmitted electricity, direct emissions F from fossil fuel combustion, and direct emissions R from decomposition of raw carbonate minerals. The ceramic production data is counted to eliminate singular value data;

[0038] Step S2: Preprocess the statistical data. Since the data in the carbon verification report may have problems such as missing data and format errors, the data needs to be cleaned. The steps include: cleaning missing values ​​and cleaning format content. After cleaning the data, due to statistical errors in the statistical data, some data have a large span, so the data needs to be screened and singular values ​​removed. According to the national industrial energy efficiency guidelines, the energy consumption limit standards for ceramic production enterprises are determined, and the comprehensive energy consumption per unit product and the electricity consumption per unit product are defined as:

[0039]

[0040] Data that is smaller than the limit value is defined as a singular value and removed;

[0041] Step S3: Based on the processed carbon emission data and electricity consumption data, the carbon emission of each carbon emission source in each historical year of the ceramic enterprise is used as the system characteristic data sequence, and the electricity consumption data of each historical year of the ceramic enterprise is used as the related factor sequence. The GM (0,1) model is used to fit the relationship between the indirect emission P of the net purchased / transmitted electricity, the direct emission F of the combustion of fossil fuels, and the direct emission R of the decomposition of raw carbonate minerals and the electricity consumption data E. The specific implementation method is as follows:

[0042] (1) Determine the system behavior characteristic data sequence P (0) 、F (0) , R (0) and the related factor sequence E (0) ;

[0043] (2) Calculate the cumulative generated sequence (1-AGO sequence) P (1)、F (1) , R (1) 、E (1) , that is, the same type of data in the original sequence is accumulated in sequence to form a new sequence. Let the original data of indirect emissions of net purchased electricity be P (0) = {P (0) (1),P (0) (2),......,P (0) (n)},

[0044] The sequence generated by one accumulation is P (1) = {P 1 (1),P 1 (2),......,P 1 (n)}, where:

[0045]

[0046] (3) Generate the model input matrix B and model output vector Y from the 1-AGO sequence;

[0047] (4) According to u=(B T B) -1 B T Solving Y yields the least squares solution u = [a, b] T ;

[0048] (5) Therefore, the relation is as follows: (1) (k) = a + bE (1) (k), k is the data dimension;

[0049] (6) According to P (0) (k) = P (1) (k)-P (1) (k-1), restore the original feature data P (0) (k) Realize the modeling and prediction functions of the algorithm.

[0050] Step S4: Input the electricity consumption data into the established association relationship, obtain a prediction result value for each association relationship, and add the three types of predicted carbon emissions to obtain the final carbon emission prediction value.

[0051] Embodiment 1:

[0052] In this embodiment, taking 40 ceramic enterprises in Hubei Province and Guangdong Province as examples, the specific method is as follows:

[0053] Step S1: Collect the carbon emission data and electricity consumption data of 40 ceramic enterprises in Hubei Province and Guangdong Province for many years, and calculate the comprehensive energy consumption per unit product. According to the national industrial energy efficiency guidelines, the international limit for ceramics is 800 kg of standard coal per ton. According to the conversion relationship between tons of standard coal and tons of carbon dioxide: 1 ton of standard coal = 2.23 tons of carbon dioxide, the comprehensive energy consumption per unit product data is converted, and the enterprise data with less than the limit value are eliminated;

[0054] Step S2: The carbon emissions of each carbon emission source in each historical year of the ceramic enterprise are taken as the system characteristic data sequence, and the electricity consumption data of each historical year of the ceramic enterprise is taken as the related factor sequence. The GM (0,1) model is used to establish the correlation between the carbon emissions of each carbon emission source and the electricity consumption as shown below:

[0055] C=aE+b

[0056] Among them, C is carbon emissions, including: indirect emissions P from net purchased / exported electricity, direct emissions F from fossil fuel combustion, direct emissions R from decomposition of raw carbonate minerals, and total carbon emissions C, and E is electricity consumption.

[0057] According to the carbon emissions of each carbon emission source in each historical year, the total carbon emissions, and the electricity consumption in each historical year, the values ​​of the correlation coefficients a and b are obtained, and the regression analysis determination coefficient R 2 ,The results are shown in Table 1 below. The specific fitting results are shown in Figure 2-6 shown.

[0058] R 2 is the determination coefficient of regression analysis. The closer the value is to l, the higher the fitting degree of the regression equation is. As can be seen from Table 1, the R between carbon emissions from various carbon emission sources and electricity consumption is 2 The R between the total carbon emissions and electricity consumption calculated indirectly by carbon emissions decomposition is higher. 2 Slightly larger than the R between the total carbon emissions and electricity consumption calculated directly 2 , so the method proposed in the present invention can improve the fitting effect.

[0059] Table 1

[0060]

[0061] Step S3: Calculate the deviation between the simulated value of the correlation between electricity and total carbon emissions directly established based on the GM (0,1) model and the true value, as well as the deviation between the simulated value of the correlation between electricity and total carbon emissions established based on the method proposed in the present invention and the true value. The specific results are shown in Table 2 below.

[0062] Table 2

[0063]

[0064]

[0065] It can be seen from Table 2 that the minimum positive deviation between the simulated value and the true value of the correlation between the total carbon emissions and electricity consumption indirectly calculated by carbon emissions decomposition is slightly smaller than the minimum positive deviation between the simulated value and the true value of the correlation between the total carbon emissions and electricity consumption directly calculated, the maximum positive deviation and the maximum negative deviation results are approximately the same, and the minimum negative deviation result is slightly increased. Therefore, the method proposed in the present invention can improve the fitting accuracy and reduce the deviation.

[0066] The above description is only a preferred embodiment of the present invention. All equivalent changes and modifications made according to the scope of the patent application of the present invention should fall within the scope of the present invention.

Claims

1. A method for predicting carbon emissions in the ceramic industry based on power data, characterized in that: The following steps are involved: Step S1: Determine the carbon emission sources by analyzing the production process of the ceramic industry, and obtain the carbon emissions of each carbon emission source and the electricity consumption data of each historical year of the ceramic enterprise; Step S2: preprocess the data obtained in step S1 to remove singular values; Step S3: Based on the grey prediction model GM (0,1) model, establish the correlation between electricity and carbon emissions of each carbon emission source, and then obtain the correlation between electricity and total carbon emissions, and construct a prediction model; Step S4: input the data to be measured into the prediction model to obtain the carbon emission prediction value; The step S3 is specifically as follows: according to the processed carbon emission data and electricity consumption data, the carbon emission of each carbon emission source in each historical year of the ceramic enterprise is used as the system characteristic data sequence, and the electricity consumption data of each historical year of the ceramic enterprise is used as the related factor sequence, and the GM (0,1) model is used to fit the relationship between the indirect emission P of the net purchased / transmitted electricity, the direct emission F of the combustion of fossil fuels, and the direct emission R of the decomposition of raw material carbonate minerals and the electricity consumption data E; The relationship is obtained specifically as follows: (1) Determine the system behavior characteristic data sequence P (0) 、F (0) , R (0) and the related factor sequence E (0) ; (2) Calculate the cumulative generated sequence (1-AGO sequence) P (1) 、F (1) , R (1) 、E (1) , that is, the same type of data in the original sequence is accumulated in sequence to form a new sequence. Let the original data of indirect emissions of net purchased electricity be P (0) = {P (0) (1),P (0) (2),......,P (0) (n)}, The sequence generated by one accumulation is P (1) = {P 1 (1),P 1 (2),......,P 1 (n)}, where: (3) Generate the model input matrix B and model output vector Y from the 1-AGO sequence; (4) According to u=(B T B) -1 B T Solving for Y yields the least squares solution u = [a, b] T ; (5) Therefore, the relation is as follows: (1) (k) = a + bE (1) (k), k is the data dimension; (6) According to P (0) (k) = P (1) (k)-P (1) (k-1), restore the original feature data P (0) (k); The electricity consumption data is input into the association relationship of the prediction model, each association relationship obtains a prediction result value, and the three types of predicted carbon emissions are added together to obtain the final carbon emission prediction value.

2. The method for predicting carbon emissions in the ceramic industry based on power data according to claim 1 is characterized in that: The carbon emission sources include indirect emissions from purchased / transmitted electricity, direct emissions from fossil fuel combustion, and direct emissions from decomposition of raw carbonate minerals.

3. The method for predicting carbon emissions in the ceramic industry based on power data according to claim 1 is characterized in that: The pre-processing specifically includes: Clean missing values ​​and format content. After cleaning the data, remove singular values ​​and calculate the comprehensive energy consumption per unit product and the power consumption per unit product according to the energy consumption limit standard for ceramic production enterprises. The definition is: Data that is smaller than the limit value is defined as a singular value and removed.