Carbon emission calculation method for nonferrous metal industry and medium

Abstract

The invention discloses a carbon emission calculation method for the nonferrous metal industry and a medium. The method comprises the steps that 1, data indexes related to carbon emission of the nonferrous metal industry are collected; 2) calculating a combined correlation value of each data index based on evaluation values of a Kendall coefficient method and a standard deviation method; 3) sorting the data indexes based on the combined correlation value, and selecting the first k data indexes as supplementary variables; 4) acquiring historical carbon emission data and corresponding supplementary variable data and electric quantity data of the nonferrous metal industry in the region, constructing a data set, and constructing a multiple linear regression model; 5) training a multiple linear regression model by using the data set to obtain a non-ferrous metal industry carbon emission calculation model in the region; and 6) acquiring real-time supplementary variable data and electric quantity data in the region, and inputting the data into the non-ferrous metal industry carbon emission calculation model in the region to obtain real-time carbon emission data in the region. According to the method, data support is provided for making a carbon reduction scheme of the nonferrous metal industry in a more targeted manner.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon accounting, and in particular to a carbon emission calculation method and medium for the nonferrous metal industry. Background Art

[0002] With the continued intensification of industrialization and human activities, global warming has become a serious challenge that cannot be ignored. In December 2015, countries reached the Paris Agreement at the Paris Climate Change Conference, which set a dual goal for the global response to climate change: to limit the global average temperature rise to well below 2°C above pre-industrial levels by the end of this century, and to pursue efforts to limit the global temperature rise to 1.5°C above pre-industrial levels. The IPCC's Sixth Assessment Report, Synthesis Report: Climate Change 2023, found that the global average surface temperature rose by 1.1°C from 2011 to 2020 compared to 1850-1900. There is a greater than 50% probability that the global temperature rise will reach or exceed 1.5°C between 2021 and 2040, necessitating urgent action to reduce global greenhouse gas emissions.

[0003] In October 2013, the National Development and Reform Commission (NDRC) issued the "Guidelines for Greenhouse Gas Emissions Accounting and Reporting by Chinese Electrolytic Aluminum Production Enterprises." This guide clarified greenhouse gas (GHG) terminology and stipulated accounting boundaries, methods, quality assurance, and documentation requirements. This approach is of great significance for promoting green and low-carbon development in the electrolytic aluminum industry. However, this method relies on extensive data on energy consumption and product output, and cannot provide real-time measurement of carbon emissions from the electrolytic aluminum industry. Furthermore, to date, there are no comprehensive carbon accounting guidelines for the regional metal industry.

[0004] The energy consumption structure of the nonferrous metals industry in the Xinjiang Uyghur Autonomous Region is likely to be diversified, primarily based on coal and electricity, with natural gas, oil, and clean energy also contributing to a certain proportion. With the improvement of power supply capacity and grid construction in the Xinjiang Uyghur Autonomous Region, the proportion of electricity in the nonferrous metals industry's energy consumption is gradually increasing. Electricity consumption is primarily used in production processes such as electrolysis, smelting, and refining. In some modern smelters, electricity has become the primary energy source.

[0005] Existing methods for calculating industry carbon emissions often use the carbon emission factor method. This involves applying activity data and an emission factor to each emission source based on a carbon emissions inventory, with the product of these two factors serving as the carbon emissions figure. However, this relies on a large amount of statistical data, resulting in limited timeliness and resolution. For carbon emissions calculations in the nonferrous metals industry in the Xinjiang Uyghur Autonomous Region, the carbon emission factor method requires complex activity data, with a high degree of data lag, making it difficult to effectively calculate carbon emissions in real time. Summary of the Invention

[0006] The purpose of the present invention is to provide a carbon emission calculation method and medium for the nonferrous metal industry.

[0007] The technical solution adopted to achieve the technical purpose of the present invention is as follows: a carbon emission calculation method for the non-ferrous metals industry, comprising the following steps:

[0008] 1) Considering the carbon emission structure characteristics of the non-ferrous metals industry in the region, data indicators related to carbon emissions in the non-ferrous metals industry are collected.

[0009] 2) Perform linear correlation analysis on data indicators based on the Kendall coefficient method and calculate the Kendall coefficient of each data indicator.

[0010] 3) Evaluate the importance of data indicators based on the standard deviation method evaluation value and obtain the weight of each data indicator.

[0011] 4) Based on the Kendall coefficient and weight of the data indicators, calculate the combined correlation value of each data indicator.

[0012] 5) Sort the data indicators based on the combined correlation values, and select the top k data indicators as supplementary variables, where k is a positive integer.

[0013] 6) Obtain historical carbon emission data and corresponding supplementary variable data and electricity data of the non-ferrous metal industry in the region, construct a data set, and build a multivariate linear regression model.

[0014] 7) Using the supplementary variable data and electricity data as input and the carbon emission data as output, the data set is used to train a multivariate linear regression model to obtain a carbon emission calculation model for the non-ferrous metal industry in the region.

[0015] 8) Obtaining real-time supplementary variable data and electricity data within the region, and inputting the real-time supplementary variable data and electricity data within the region into the carbon emission calculation model of the nonferrous metal industry in the region to obtain real-time carbon emission data within the region.

[0016] Furthermore, the data indicators include energy consumption and added value by industry, regional main product output, product index, copper output, and aluminum output.

[0017] Furthermore, in step 2), the steps for calculating the Kendall coefficient of each data indicator are as follows:

[0018] 2.1) Obtain n observation values ​​of each data indicator and the corresponding n carbon emission data, and construct the data pair (X ri ,Y ri ), where n is a positive integer, i = 1, 2, ..., n, r is the data index, X riis the i-th observation value of the r-th data indicator, Y ri is the carbon emission data corresponding to the i-th observation value of the r-th data indicator.

[0019] 2.2) For the data pairs of the same data indicator (X ri ,Y ri ) to pair them up and list all pairs of the same data indicator Among them, i1 and i2 are both observation value indexes, and i1≠i2.

[0020] 2.3) All pairs of the same data indicator Divide and obtain harmonious pairing sets and disharmonious pairing sets of the same data indicator.

[0021] 2.4) Based on the harmonious and discordant pairing sets of the same data indicator, the Kendall coefficient of each data indicator is calculated.

[0022] Furthermore, if the pairing middle and Then pair For harmonious pairing.

[0023] If paired middle and Then pair For harmonious pairing.

[0024] If paired middle and Then pair For discordant pairing.

[0025] If paired middle and Then pair For discordant pairing.

[0026] Furthermore, the Kendall coefficient of each data indicator is as follows:

[0027]

[0028] In the formula, r is the data index, τ r is the Kendall coefficient of the rth data indicator. C r is the number of harmonious pairs of the rth data indicator. r is the number of discordant pairs of the rth data indicator. n is the total number of data pairs.

[0029] Furthermore, in step 3), the steps for obtaining the weight of each data indicator are as follows:

[0030] 3.1) Obtain m data indicator values ​​under n0 evaluation objects and construct the matrix X as shown below:

[0031]

[0032] Where x jr It represents the rth data indicator value under the jth evaluation object, where j is the evaluation object index, j = 1, 2, ..., n0, and r is the data indicator index, r = 1, 2, ..., m.

[0033] 3.2) Perform data forward processing on the matrix X to obtain the forward matrix X'.

[0034] 3.3) Normalize the forward matrix X' to obtain the normalized matrix R, as shown below:

[0035]

[0036] Where x j j r represents the rth data indicator value under the jth evaluation object after positive processing, R jr Represents the rth data indicator value under the jth evaluation object after normalization.

[0037] 3.4) Calculate the mean A of each data indicator r and standard deviation S r , as shown below:

[0038]

[0039] 3.5) Calculate the coefficient of variation for each data metric as follows:

[0040]

[0041] Where V r Represents the coefficient of variation of the rth data indicator.

[0042] 3.6) Calculate the weight of each data indicator as follows:

[0043]

[0044] Where, ω r Represents the weight of the rth data indicator. s is the data indicator index, V s Represents the coefficient of variation of the sth data indicator.

[0045] Furthermore, the data in the matrix X includes positive indicators and negative indicators.

[0046] The positive indicator is an indicator whose value is positively correlated with carbon emissions.

[0047] The negative indicator is an indicator whose value is negatively correlated with carbon emissions.

[0048] The forwarding process is as follows:

[0049]

[0050] Where x′ jr Represents the rth data indicator value under the jth evaluation object after positive processing. jr Indicates the value of the rth data indicator under the jth evaluation object. k1 is the specified coefficient. max|x r | represents the maximum absolute value of the rth data indicator.

[0051] Furthermore, the combined correlation values ​​of each data indicator are as follows:

[0052]

[0053] In the formula, r is the data index, is the combined correlation value of the rth data indicator. r Represents the weight of the rth data indicator. τ r is the Kendall coefficient of the rth data indicator.

[0054] Furthermore, the carbon emission calculation model for the nonferrous metals industry in the region is as follows:

[0055] y=b0+b1x1+b2x2+...+b k+1 x k+1 +e (10)

[0056] Where y represents carbon emission data, b0 is a constant term, b1, b2,…, b k+1 are all regression coefficients, x1, x2,…, x k All represent supplementary variable data, x k+1 represents the power data. e is the error term.

[0057] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned carbon emission calculation method for the non-ferrous metal industry.

[0058] The technical effect of the present invention is unquestionable. The present invention constructs a carbon emission calculation model with electricity consumption in the non-ferrous metals industry as the main explanatory variable, accurately understands the carbon emissions of the non-ferrous metals industry, and provides data support for more targeted formulation of carbon reduction plans for the non-ferrous metals industry.

[0059] This paper proposes a carbon emissions calculation method for the nonferrous metals industry in Xinjiang Uyghur Autonomous Region. Based on the primary product composition and industrial structure of the nonferrous metals industry in Xinjiang Uyghur Autonomous Region, copper production is added as a supplementary variable. Based on this, a combined correlation analysis method is used to identify four additional supplementary variables with significant impacts on carbon emissions for use in the model calculation. Finally, a carbon emissions calculation model is constructed using a multivariate linear regression method.

[0060] The present invention has the following advantages: (1) a combination analysis method is used to screen out influencing factors with a high correlation with carbon emissions in the non-ferrous metals industry in Xinjiang Uygur Autonomous Region as supplementary variables to achieve more accurate carbon emissions calculation; (2) based on electricity consumption data as an explanatory variable, combined with copper production and four additional supplementary variables screened out, the carbon emissions of the non-ferrous metals industry in Xinjiang Uygur Autonomous Region are calculated, and the relationship between multiple explanatory variables and supplementary variables and a dependent variable can be processed simultaneously. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 This is a flow chart of the carbon emissions calculation method for the non-ferrous metals industry. DETAILED DESCRIPTION

[0062] The present invention will be further described below with reference to the following examples, but it should not be understood that the scope of the present invention is limited to the following examples. Without departing from the above technical ideas of the present invention, various substitutions and modifications can be made according to common technical knowledge and customary means in the art, and all should be included in the scope of protection of the present invention.

[0063] Example 1:

[0064] A carbon emission calculation method for the nonferrous metals industry includes the following steps:

[0065] 1) Considering the carbon emission structure characteristics of the non-ferrous metals industry in the region, data indicators related to carbon emissions in the non-ferrous metals industry are collected.

[0066] 2) Perform linear correlation analysis on data indicators based on the Kendall coefficient method and calculate the Kendall coefficient of each data indicator.

[0067] 3) Evaluate the importance of data indicators based on the standard deviation method evaluation value and obtain the weight of each data indicator.

[0068] 4) Based on the Kendall coefficient and weight of the data indicators, calculate the combined correlation value of each data indicator.

[0069] 5) Sort the data indicators based on the combined correlation values, and select the top k data indicators as supplementary variables, where k is a positive integer.

[0070] 6) Obtain historical carbon emission data and corresponding supplementary variable data and electricity data of the non-ferrous metal industry in the region, construct a data set, and build a multivariate linear regression model.

[0071] 7) Using the supplementary variable data and electricity data as input and the carbon emission data as output, the data set is used to train a multivariate linear regression model to obtain a carbon emission calculation model for the non-ferrous metal industry in the region.

[0072] 8) Obtaining real-time supplementary variable data and electricity data within the region, and inputting the real-time supplementary variable data and electricity data within the region into the carbon emission calculation model of the nonferrous metal industry in the region to obtain real-time carbon emission data within the region.

[0073] Example 2:

[0074] A method for calculating carbon emissions for the nonferrous metals industry. The main technical content is shown in Example 1. Furthermore, the data indicators include energy consumption and added value by industry, regional main product output, product index, copper output, and aluminum output.

[0075] Example 3:

[0076] A carbon emission calculation method for the nonferrous metals industry, the main technical content of which is shown in any one of Examples 1 to 2. Further, in step 2), the steps of calculating the Kendall coefficient of each data indicator are as follows:

[0077] 2.1) Obtain n observation values ​​of each data indicator and the corresponding n carbon emission data, and construct the data pair (X ri ,Y ri ), where n is a positive integer, i = 1, 2, ..., n, r is the data index, X ri is the i-th observation value of the r-th data indicator, Y ri is the carbon emission data corresponding to the i-th observation value of the r-th data indicator.

[0078] 2.2) For the data pairs of the same data indicator (X ri ,Y ri ) to pair them up and list all pairs of the same data indicator Among them, i1 and i2 are both observation value indexes, and i1≠i2.

[0079] 2.3) All pairs of the same data indicator Divide and obtain harmonious pairing sets and disharmonious pairing sets of the same data indicator.

[0080] 2.4) Based on the harmonious and discordant pairing sets of the same data indicator, the Kendall coefficient of each data indicator is calculated.

[0081] Example 4:

[0082] A method for calculating carbon emissions in the nonferrous metals industry. The main technical content is shown in any one of Examples 1 to 3. Further, if the pair middle and Then pair For harmonious pairing.

[0083] If paired middle and Then pair For harmonious pairing.

[0084] If paired middle and Then pair For discordant pairing.

[0085] If paired middle and Then pair For discordant pairing.

[0086] Example 5:

[0087] A carbon emissions calculation method for the nonferrous metals industry, the main technical content of which is shown in any one of Examples 1 to 4. Furthermore, the Kendall coefficient of each data indicator is as follows:

[0088]

[0089] In the formula, r is the data index, τ r is the Kendall coefficient of the rth data indicator. C r is the number of harmonious pairs of the rth data indicator. r is the number of discordant pairs of the rth data indicator. n is the total number of data pairs.

[0090] Example 6:

[0091] A method for calculating carbon emissions for the nonferrous metals industry, the main technical content of which is shown in any one of Examples 1 to 5. Further, in step 3), the steps for obtaining the weight of each data indicator are as follows:

[0092] 3.1) Obtain m data indicator values ​​under n0 evaluation objects and construct the matrix X as shown below:

[0093]

[0094] Where x jrIt represents the rth data indicator value under the jth evaluation object, where j is the evaluation object index, j = 1, 2, ..., n0, and r is the data indicator index, r = 1, 2, ..., m.

[0095] 3.2) Perform data forward processing on the matrix X to obtain the forward matrix X'.

[0096] 3.3) Normalize the forward matrix X' to obtain the normalized matrix R, as shown below:

[0097]

[0098] Where x′ jr represents the rth data indicator value under the jth evaluation object after positive processing, R jr Represents the rth data indicator value under the jth evaluation object after normalization.

[0099] 3.4) Calculate the mean A of each data indicator r and standard deviation S r , as shown below:

[0100]

[0101]

[0102] 3.5) Calculate the coefficient of variation for each data metric as follows:

[0103]

[0104] Where V r Represents the coefficient of variation of the rth data indicator.

[0105] 3.6) Calculate the weight of each data indicator as follows:

[0106]

[0107] Where, ω r Represents the weight of the rth data indicator. s is the data indicator index, V s Represents the coefficient of variation of the sth data indicator.

[0108] Example 7:

[0109] A method for calculating carbon emissions for the nonferrous metals industry. The main technical content is shown in any one of Examples 1 to 6. Furthermore, the data in the matrix X includes positive indicators and negative indicators.

[0110] The positive indicator is an indicator whose value is positively correlated with carbon emissions.

[0111] The negative indicator is an indicator whose value is negatively correlated with carbon emissions.

[0112] The forwarding process is as follows:

[0113]

[0114] Where x′ jr Represents the rth data indicator value under the jth evaluation object after positive processing. jr Indicates the value of the rth data indicator under the jth evaluation object. k1 is the specified coefficient. max|x r | represents the maximum absolute value of the rth data indicator.

[0115] Example 8:

[0116] A carbon emissions calculation method for the nonferrous metals industry, the main technical content of which is shown in any one of Examples 1 to 7. Furthermore, the combined relevant values ​​of each data indicator are as follows:

[0117]

[0118] In the formula, r is the data index, is the combined correlation value of the rth data indicator. r Represents the weight of the rth data indicator. τ r is the Kendall coefficient of the rth data indicator.

[0119] Example 9:

[0120] A carbon emission calculation method for the non-ferrous metals industry, the main technical content of which is shown in any one of Examples 1 to 8. Furthermore, the carbon emission calculation model for the non-ferrous metals industry in the region is as follows:

[0121] y=b0+b1x1+b2x2+...+b k+1 x k+1 +e (10)

[0122] Where y represents carbon emission data, b0 is a constant term, b1, b2,…, b k+1 are all regression coefficients, x1, x2,…, x k All represent supplementary variable data, x k+1 represents the power data. e is the error term.

[0123] Example 10:

[0124] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a carbon emission calculation method for the non-ferrous metal industry as described in any one of Examples 1-9.

[0125] Example 11:

[0126] A carbon emission calculation method for the nonferrous metals industry includes the following steps:

[0127] 1) Considering the carbon emission structure characteristics of the non-ferrous metals industry in the region, data indicators related to carbon emissions in the non-ferrous metals industry are collected.

[0128] 2) Perform linear correlation analysis on data indicators based on the Kendall coefficient method and calculate the Kendall coefficient of each data indicator.

[0129] 3) Evaluate the importance of data indicators based on the standard deviation method evaluation value and obtain the weight of each data indicator.

[0130] 4) Based on the Kendall coefficient and weight of the data indicators, calculate the combined correlation value of each data indicator.

[0131] 5) Sort the data indicators based on the combined correlation values ​​and select the top 5 data indicators as supplementary variables.

[0132] 6) Obtain historical carbon emission data and corresponding supplementary variable data and electricity data of the non-ferrous metal industry in the region, construct a data set, and build a multivariate linear regression model.

[0133] The change in the dependent variable is often influenced by several important factors. Therefore, two or more influencing factors are needed as independent variables to explain the change in the dependent variable. This is called multiple regression. When there is a linear relationship between multiple independent variables and the dependent variable, the regression analysis performed is multiple linear regression.

[0134] 7) Using the supplementary variable data and electricity data as input and the carbon emission data as output, the data set is used to train a multivariate linear regression model to obtain a carbon emission calculation model for the non-ferrous metal industry in the region.

[0135] 8) Obtaining real-time supplementary variable data and electricity data within the region, and inputting the real-time supplementary variable data and electricity data within the region into the carbon emission calculation model of the nonferrous metal industry in the region to obtain real-time carbon emission data within the region.

[0136] Example 12:

[0137] A method for calculating carbon emissions for the nonferrous metals industry. The main technical content is shown in Example 11. Furthermore, the data indicators include energy consumption and added value by industry, regional main product output, product index, copper output, and aluminum output.

[0138] Example 13:

[0139] A method for calculating carbon emissions for the nonferrous metals industry, the main technical content of which is described in any one of Examples 11 to 12. Furthermore, the Kendall coefficient is a statistical indicator used to measure the consistency of ranking data or rating data given by multiple evaluators. It is obtained by comparing the difference between the ranking or grade given by the evaluator and the true ranking or grade (if any), and calculating the similarity between the two. When multiple (more than two) variable values ​​are arranged in rank order or expressed in rank order, the quantity that describes the degree of consistency between these variables is called the Kendall coefficient.

[0140] The Kendall coefficient ranges from -1 to 1:

[0141] 0: Indicates that the inter-rater ranking or rating agreement is the same as random ranking or random rating.

[0142] 1: Indicates complete consistency.

[0143] -1: Indicates complete inconsistency.

[0144] In step 2), the steps for calculating the Kendall coefficient of each data indicator are as follows:

[0145] 2.1) Obtain n observation values ​​of each data indicator and the corresponding n carbon emission data, and construct the data pair (X ri ,Y ri ), where n is a positive integer, i = 1, 2, ..., n, r is the data index, X ri is the i-th observation value of the r-th data indicator, Y ri is the carbon emission data corresponding to the i-th observation value of the r-th data indicator.

[0146] 2.2) For the data pairs of the same data indicator (X ri ,Y ri ) to pair them up and list all pairs of the same data indicator Among them, i1 and i2 are both observation value indexes, and i1≠i2.

[0147] 2.3) All pairs of the same data indicator Divide and obtain harmonious pairing sets and disharmonious pairing sets of the same data indicator.

[0148] 2.4) Based on the harmonious and discordant pairing sets of the same data indicator, the Kendall coefficient of each data indicator is calculated.

[0149] Example 14:

[0150] A carbon emission calculation method for the nonferrous metals industry, the main technical content of which is shown in any one of Examples 11 to 13. Further, if the pair middle and Then pair For harmonious pairing.

[0151] If paired middle and Then pair For harmonious pairing.

[0152] If paired middle and Then pair For discordant pairing.

[0153] If paired middle and Then pair For discordant pairing.

[0154] Example 15:

[0155] A carbon emissions calculation method for the nonferrous metals industry, the main technical content of which is shown in any one of Examples 11 to 14. Furthermore, the Kendall coefficient of each data indicator is as follows:

[0156]

[0157] In the formula, r is the data index, τ r is the Kendall coefficient of the rth data indicator. C r is the number of harmonious pairs of the rth data indicator. r is the number of discordant pairs of the rth data indicator. n is the total number of data pairs.

[0158] Example 16:

[0159] A method for calculating carbon emissions for the nonferrous metals industry, the main technical content of which is described in any one of Examples 11 to 15. Furthermore, the standard deviation method and the coefficient of variation method weight each evaluation indicator based on the degree of variation between the current value and the target value of each evaluation indicator. If the numerical difference of a certain indicator is large, it can clearly distinguish the evaluated objects, indicating that the indicator has rich discriminative information, and thus the indicator should be given a larger weight. Conversely, if the numerical difference of each evaluated object on a certain indicator is small, then the ability of this indicator to distinguish the evaluated objects is weak, and thus the indicator should be given a smaller weight.

[0160] In step 3), the steps to obtain the weight of each data indicator are as follows:

[0161] 3.1) Obtain m data indicator values ​​under n0 evaluation objects and construct the matrix X as shown below:

[0162]

[0163] Where x jr It represents the rth data indicator value under the jth evaluation object, where j is the evaluation object index, j = 1, 2, ..., n0, and r is the data indicator index, r = 1, 2, ..., m.

[0164] The evaluation objects refer to indicators that affect carbon emissions in the non-ferrous metals industry, including energy consumption and added value by industry, regional major product output, product index, copper production, and aluminum production.

[0165] 3.2) Perform data forward processing on the matrix X to obtain the forward matrix X'.

[0166] 3.3) Normalize the forward matrix X' to obtain the normalized matrix R, as shown below:

[0167]

[0168] Where x′ jr represents the rth data indicator value under the jth evaluation object after positive processing, R jr Represents the rth data indicator value under the jth evaluation object after normalization.

[0169] 3.4) Calculate the mean A of each data indicator r and standard deviation S r , as shown below:

[0170]

[0171] 3.5) Calculate the coefficient of variation for each data metric as follows:

[0172]

[0173] Where V r Represents the coefficient of variation of the rth data indicator.

[0174] 3.6) Calculate the weight of each data indicator as follows:

[0175]

[0176] Where, ω r Represents the weight of the rth data indicator. s is the data indicator index, V s Represents the coefficient of variation of the sth data indicator.

[0177] Example 17:

[0178] A method for calculating carbon emissions for the nonferrous metals industry. The main technical content is shown in any one of Examples 11 to 16. Furthermore, the data in the matrix X includes positive indicators and negative indicators.

[0179] The positive indicator is an indicator whose value is positively correlated with carbon emissions, that is, the larger the value, the greater the carbon emissions.

[0180] The negative indicator is an indicator whose value is negatively correlated with carbon emissions, that is, the larger the value, the smaller the carbon emissions.

[0181] The forwarding process is as follows:

[0182]

[0183] Where x′ jr Represents the rth data indicator value under the jth evaluation object after positive processing. jr Indicates the value of the rth data indicator under the jth evaluation object. k1 is the specified coefficient. max|x r | represents the maximum absolute value of the rth data indicator.

[0184] Example 18:

[0185] A carbon emissions calculation method for the nonferrous metals industry, the main technical content of which is shown in any one of Examples 11 to 17. Furthermore, the combined relevant values ​​of each data indicator are as follows:

[0186]

[0187] In the formula, r is the data index, is the combined correlation value of the rth data indicator. r Represents the weight of the rth data indicator. τ r is the Kendall coefficient of the rth data indicator.

[0188] Example 19:

[0189] A carbon emission calculation method for the non-ferrous metals industry, the main technical content of which is shown in any one of Examples 11 to 18. Furthermore, the carbon emission calculation model for the non-ferrous metals industry in the region is as follows:

[0190] y=b0+b1x1+b2x2+...+b k+1 x k+1 +e (10)

[0191] Where y represents carbon emission data, b0 is a constant term, b1, b2,…, b k+1are all regression coefficients, x1, x2,…, x k All represent supplementary variable data, x k+1 represents the power data. e is the error term.

[0192] Example 20:

[0193] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a carbon emission calculation method for the non-ferrous metal industry as described in any one of Examples 11-19.

[0194] Example 21:

[0195] A carbon emissions calculation method for the nonferrous metals industry, the main technical contents include:

[0196] 1. Selection of Supplementary Variable Indicators for the Nonferrous Metals Industry in the Optimization Model

[0197] Taking into account the structural characteristics of carbon emissions from the nonferrous metals industry in the Xinjiang Uyghur Autonomous Region, data indicators related to carbon emissions from the nonferrous metals industry were collected, including factors such as energy consumption and value-added by industry, regional output of major products, and product indices. These indicators were then grouped into five categories: economic, energy consumption, product output, production index, and other indicators. Furthermore, given that the Xinjiang Uyghur Autonomous Region's copper production ranks among the highest in the country, and copper smelting's contribution to emissions is second only to electrolytic aluminum, copper production was added as a supplementary variable to improve the accuracy of the model's carbon emissions calculations.

[0198] 2. Determining the Input Variables of the Carbon Emission Calculation Model Based on Combination Analysis

[0199] Since there are many factors that affect the carbon emissions of non-ferrous metals and the correlation between each factor and carbon emissions is relatively complex, the present invention designs a combined analysis method using the Kendall coefficient and the standard deviation method to effectively identify the main factors affecting the carbon emissions of the non-ferrous metals industry and quantify the degree of influence of each factor.

[0200] Combined correlation value = |Kendall linear correlation coefficient × standard deviation method evaluation value| (1)

[0201] The influencing factors related to carbon emissions in the non-ferrous metals industry are sorted according to the combined correlation values, and five influencing factors with a strong correlation with carbon emissions are selected as supplementary variables for in-depth application of the model.

[0202] 1) Linear correlation analysis of carbon emission influencing factors based on Kendall coefficient

[0203] The Kendall coefficient is a statistical indicator used to measure the consistency of rankings or ratings given by multiple evaluators. It is calculated by comparing the differences between the rankings or grades given by the evaluators and the true rankings or grades (if any) and calculating the similarity between the two. When multiple (or more) variable values ​​are arranged or expressed in a hierarchical order, the quantity that describes the degree of consistency between these variables is called the Kendall coefficient.

[0204] The Kendall coefficient ranges from -1 to 1:

[0205] 0: indicates that the inter-rater ranking or rating agreement is the same as random ranking or random rating;

[0206] 1: indicates complete agreement;

[0207] -1: Indicates complete inconsistency.

[0208] The calculation of the Kendall coefficient τ is based on the concept of paired comparison. Suppose there are two variables X (influencing factors) and Y (carbon emissions), each with n observations. The Kendall coefficient τ is defined as follows:

[0209]

[0210] Where C is the number of harmonious pairs, that is, the number of pairs that increase or decrease simultaneously in X and Y; D is the number of discordant pairs, that is, the number of pairs in which one increases while the other decreases. is the number of all possible pairings.

[0211] Kendall coefficient calculation steps:

[0212] 1. List all pairings: For n observations, list all possible pairings.

[0213] 2. Compare pairs: For each pair (X i ,Y i ), judge whether it is harmonious or disharmonious.

[0214] 3. Count the number of harmonious and discordant pairs: Count the number of harmonious and discordant pairs.

[0215] 4. Calculate Kendall's coefficient: Use the above formula to calculate Kendall's coefficient τ.

[0216] 2) Importance evaluation of carbon emission influencing factors based on standard deviation method evaluation value The present invention adopts the standard deviation method to study the importance relationship between carbon emissions and its influencing factors.

[0217] The standard deviation method and the coefficient of variation method assign weights to each indicator based on the degree of variation between the current value and the target value of each evaluation indicator. If the numerical difference of an indicator is large, it can clearly distinguish the evaluated objects, indicating that the indicator has rich discrimination information, and thus it should be given a larger weight; conversely, if the numerical difference of each evaluated object on a certain indicator is small, then the ability of this indicator to distinguish the evaluated objects is weak, and thus it should be given a smaller weight. The basic steps are as follows:

[0218] 1. Data collection: Assume that there are m indicators and n objects to be evaluated in a set of data, that is, an n*m matrix, let it be X. ij Represents the data in row i and column j.

[0219]

[0220] 2. Positive indicator data:

[0221] The purpose of indicator positivity is to convert all indicators into positive indicators. Positive indicators are also called "bigger is better" indicators, meaning the higher the value, the better. Negative indicators are also called "smaller is better" indicators, meaning the smaller the value, the better.

[0222] For positive indicators: keep their original data unchanged.

[0223] x' ij =x ij (4)

[0224] For negative indicators: use the following method.

[0225]

[0226] Where k is a specified arbitrary coefficient, and its value can be 0.1, 0.2, etc.; max|x j |Indicates the maximum absolute value of the data (indicator) in the jth column.

[0227] 3. Data standardization:

[0228] Since different indicators have different units, they cannot be directly calculated. The purpose of data standardization is to eliminate the influence of units so that all data can be calculated using the same method. Let the standardized data matrix be R.

[0229]

[0230] 4. Calculate the coefficient of variation:

[0231] Calculate the mean of each indicator:

[0232]

[0233] Calculate the standard deviation (mean square error) of each indicator:

[0234]

[0235] Because the standard deviation can describe the degree of discreteness of the value, that is, the variance of an indicator reflects the resolution ability of the indicator, the standard deviation can be used to define the weight of the indicator.

[0236] Calculate the coefficient of variation for each indicator:

[0237]

[0238] 5. Calculate weights and scores:

[0239]

[0240] 3) Determine the supplementary variables for the model used to calculate carbon emissions from the regional nonferrous metals industry

[0241] The combined correlation values ​​of each supplementary variable were calculated based on the absolute value of the product of the Kendall linear correlation coefficient and the standard deviation method evaluation value between the historical carbon emissions of the nonferrous metals industry in the Xinjiang Uyghur Autonomous Region and each supplementary variable. Based on these results, the five supplementary variables with the highest combined correlation values ​​were selected for use in the carbon emissions model calculations for the nonferrous metals industry in the Xinjiang Uyghur Autonomous Region.

[0242] 3. Constructing a carbon emission calculation model for the nonferrous metals industry in Xinjiang Uyghur Autonomous Region based on a multiple linear regression model

[0243] The change in the dependent variable is often influenced by several important factors. Therefore, two or more influencing factors are needed as independent variables to explain the change in the dependent variable. This is called multiple regression. When there is a linear relationship between multiple independent variables and the dependent variable, the regression analysis performed is multiple linear regression.

[0244] Let y be the dependent variable, x1, x2, ...x k are independent variables, i.e., explanatory variables and supplementary variables. The dependent variables are mainly historical data on energy activities and industrial output, and the explanatory variables are historical and current electricity data. The supplementary variables are other influencing factors of the model, fully considering the impact of other factors other than electricity on the target object. When there is a linear relationship between the independent variables and the dependent variables, the multiple linear regression model is:

[0245] y=b0+b1x1+b2x2+...+b k x k +e(11)

[0246] Among them, b0 is a constant term, b1, b2, ..., bk is the regression coefficient, also known as the influence coefficient, which indicates the degree of influence of each variable. b1 is x1, x2, ...x k When x1 is fixed, the effect of each unit increase in x1 on y is the partial regression coefficient of x1 on y; similarly, b2 is x1, x3, ..., x k When x2 is fixed, the effect of each unit increase in x2 on y is the partial regression coefficient of x2 on y.

Claims

1. A carbon emission calculation method for the nonferrous metals industry, characterized in that: The following steps are involved: 1) Considering the carbon emission structure characteristics of the non-ferrous metals industry in the region, data indicators related to carbon emissions in the non-ferrous metals industry are collected. 2) Perform linear correlation analysis on data indicators based on the Kendall coefficient method and calculate the Kendall coefficient of each data indicator; 3) Evaluate the importance of data indicators based on the standard deviation method evaluation value and obtain the weight of each data indicator; 4) Calculate the combined correlation value of each data indicator based on the Kendall coefficient and weight of the data indicator; 5) Sort the data indicators based on the combined correlation values, and select the top k data indicators as supplementary variables, where k is a positive integer; 6) Obtain historical carbon emission data and corresponding supplementary variable data and electricity data for the nonferrous metals industry in the region, construct a data set, and build a multivariate linear regression model; 7) Using the supplementary variable data and electricity data as input and the carbon emission data as output, the data set is used to train a multivariate linear regression model to obtain a carbon emission calculation model for the nonferrous metals industry in the region; 8) Obtaining real-time supplementary variable data and electricity data within the region, and inputting the real-time supplementary variable data and electricity data within the region into the carbon emission calculation model of the nonferrous metal industry in the region to obtain real-time carbon emission data within the region.

2. A carbon emission calculation method for the nonferrous metals industry according to claim 1, characterized in that: The data indicators include energy consumption and added value by industry, regional major product output, product index, copper production, and aluminum production.

3. The carbon emission calculation method for the nonferrous metals industry according to claim 1, characterized in that: In step 2), the steps for calculating the Kendall coefficient of each data indicator are as follows: 2.1) Obtain n observation values ​​of each data indicator and the corresponding n carbon emission data, and construct the data pair (X ri ,Y ri ), where n is a positive integer, i = 1, 2, ..., n, r is the data index, X ri is the i-th observation value of the r-th data indicator, Y ri is the carbon emission data corresponding to the i-th observation value of the r-th data indicator; 2.2) For the data pairs of the same data indicator (X ri ,Y ri ) to pair them up and list all pairs of the same data indicator Among them, i1 and i2 are both observation value indexes, and i1≠i2; 2.3) All pairs of the same data indicator Divide and obtain harmonious pairing sets and disharmonious pairing sets of the same data indicator; 2.4) Based on the harmonious and discordant pairing sets of the same data indicator, the Kendall coefficient of each data indicator is calculated.

4. A carbon emission calculation method for the nonferrous metals industry according to claim 3, characterized in that: If paired middle and Then pair For harmonious pairing; If paired middle and Then pair For harmonious pairing; If paired middle and Then pair pairing for discord; If paired middle and Then pair For discordant pairing.

5. The carbon emission calculation method for the nonferrous metal industry according to claim 3, characterized in that: The Kendall coefficients for each of the data indicators are as follows: In the formula, r is the data index, τ r is the Kendall coefficient of the rth data indicator; C r is the number of harmonious pairs of the rth data indicator; D r is the number of discordant pairs of the rth data indicator; n is the total number of data pairs.

6. The carbon emission calculation method for the nonferrous metal industry according to claim 1, characterized in that: In step 3), the steps to obtain the weight of each data indicator are as follows: 3.1) Obtain m data indicator values ​​under n0 evaluation objects and construct the matrix X as shown below: Where x jr Represents the rth data indicator value under the jth evaluation object, j is the evaluation object index, j = 1, 2, ..., n0, r is the data indicator index, r = 1, 2, ..., m; 3.2) Perform data forward processing on the matrix X to obtain the forward matrix X'; 3.3) Normalize the forward matrix X' to obtain the normalized matrix R, as shown below: Where x′ jr represents the rth data indicator value under the jth evaluation object after positive processing, R jr Represents the rth data indicator value under the jth evaluation object after standardization; 3.4) Calculate the mean A of each data indicator r and standard deviation S r , as shown below: 3.5) Calculate the coefficient of variation for each data metric as follows: Where V r represents the coefficient of variation of the rth data indicator; 3.6) Calculate the weight of each data indicator as follows: Where, ω r Represents the weight of the rth data indicator; s is the data indicator index, V s Represents the coefficient of variation of the sth data indicator.

7. A carbon emission calculation method for the nonferrous metals industry according to claim 6, characterized in that: The data in the matrix X includes positive indicators and negative indicators; The positive indicator is an indicator whose value is positively correlated with carbon emissions; The negative indicator is an indicator whose value is negatively correlated with carbon emissions; The forwarding process is as follows: Where x′ jr represents the rth data indicator value under the jth evaluation object after positive processing; x jr represents the rth data indicator value under the jth evaluation object; k1 is the specified coefficient; max|x r | represents the maximum absolute value of the rth data indicator.

8. The carbon emission calculation method for the nonferrous metals industry according to claim 1, characterized in that: The combined relevant values ​​of each data indicator are as follows: In the formula, r is the data index, is the combined correlation value of the rth data indicator; ω r Represents the weight of the rth data indicator; τ r is the Kendall coefficient of the rth data indicator.

9. The carbon emission calculation method for the nonferrous metal industry according to claim 1, characterized in that: The carbon emission calculation model for the nonferrous metals industry in the region is as follows: y=b0+b1x1+b2x2+...+b k+1 x k+1 +e (10) Where y represents carbon emission data, b0 is a constant term, b1, b2,…, b k+1 are all regression coefficients, x1, x2,…, x k All represent supplementary variable data, x k+1 represents the electricity data; e is the error term.

10. A computer-readable storage medium, characterized in that A computer program is stored thereon, which, when executed by a processor, implements the steps of a carbon emission calculation method for the non-ferrous metal industry as described in any one of claims 1 to 9.

Images