Enterprise carbon emission monitoring method and system considering relationship between equipment power and material consumption

Abstract

The invention provides an enterprise carbon emission monitoring method considering the relationship between equipment power and material consumption, and the method comprises the following steps: S1, firstly, installing power utilization data collection devices at the power incoming line of an enterprise and each production equipment, and monitoring the power utilization power of the production equipment of the enterprise; s3, according to the power of each piece of production equipment of the enterprise acquired in the step S1, calculating electric equipment which has a strong correlation with the material; s4, calculating the carbon emission of each material by combining the carbon emission factor of each material with a material carbon emission accounting formula in an energy-carbon method; s5, the indirect carbon emission of the enterprise is obtained according to the fact that the power utilization data of the enterprise is multiplied by the local regional power carbon factor; and S6, according to the direct and indirect carbon emissions of the enterprise obtained in the step S4 and the step S5, obtaining the total carbon emission of the enterprise, and realizing enterprise carbon emission monitoring. According to the technical scheme, accurate enterprise carbon emission monitoring can be carried out without additionally arranging expensive carbon monitoring equipment, and data support is provided for environmental protection departments.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission monitoring, and in particular to a method and system for monitoring enterprise carbon emissions by considering the relationship between equipment power and material consumption. Background Art

[0002] Global warming is a major issue currently facing humanity, with carbon dioxide emissions being the main cause. Therefore, establishing a carbon emissions monitoring method based on corporate electricity consumption can effectively improve the carbon emissions monitoring efficiency of companies with electrified production equipment, helping such companies to explore feasible carbon transformation plans in the future and achieve a low-carbon and environmentally friendly corporate strategy.

[0003] Currently, traditional carbon emission monitoring methods include the carbon emission factor method and the online monitoring method. The carbon emission factor method primarily estimates a company's carbon emissions by calculating the consumption of various energy sources during its production process and multiplying it by the corresponding emission factor. This estimate is not accurate enough. The online monitoring method, on the other hand, primarily uses a continuous emission monitoring system (CEMS) installed at the company's gas exhaust outlet to monitor the carbon dioxide content in the exhaust gas in real time, obtaining the company's real-time and continuous carbon dioxide emissions. The carbon emission factor method has low real-time performance, low monitoring efficiency, and high labor costs. The online monitoring method also requires high precision and accuracy of the monitoring equipment, resulting in expensive installation costs that are unaffordable for most companies.

[0004] In existing solutions, most methods for monitoring carbon emissions through electricity data only consider the relationship between the total electricity power of the enterprise and the enterprise's carbon emission data, or only consider the relationship between the electricity power of the enterprise's equipment and the enterprise's carbon emission data. This method ignores the relationship between the equipment and the main sources of carbon emissions in the enterprise's production process, including carbon-containing raw materials and chemical fuels. The results of carbon emission monitoring are less accurate and lack persuasiveness. Currently, there are methods that achieve enterprise carbon emission monitoring by constructing the relationship between the enterprise's production material consumption and the enterprise's total electricity consumption. It does not take into account the relationship between the material and the production equipment that consumes the material, and the method lacks interpretability. Therefore, the present invention takes into account the correlation between equipment electricity power and production material consumption, constructs an enterprise "electricity-energy-carbon" model, and achieves more accurate enterprise carbon emission monitoring. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a method and system for enterprise carbon emission monitoring that takes into account the relationship between equipment power and material consumption, thereby achieving accurate enterprise carbon emission monitoring without installing expensive carbon monitoring equipment and providing data support to environmental protection departments.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for monitoring enterprise carbon emissions taking into account the relationship between equipment power and material consumption, comprising the following steps:

[0007] Step S1: First, install electricity data collection devices at the company's power line and at each production equipment to monitor the power consumption of the company's production equipment. Obtain the company's fuel and material consumption data, carbon emission factor, fuel low heating value, and company production process topology from the company's annual production report, and input the local regional electricity carbon factor. Power consumption data is collected every 3 minutes, resulting in a total of 480 data points over a 24-hour day. The company's production process topology includes the connection relationships between each device and the relationship between the materials consumed by each device.

[0008] Step S2: Using the dynamic time warping algorithm (DTW) to calculate the correlation between the consumption data of each material and the power of the main equipment that consumes them, quantify the statistical correlation coefficient, and screen out the equipment with strong correlation with each material;

[0009] Step S3: Based on the power of each production equipment collected in step S1, combined with step S2, the power-consuming equipment with a strong correlation with the material is calculated. Furthermore, the material relationship consumed by each strongly correlated equipment is further combined, and a regression relationship model between the strongly correlated production equipment and each material after superimposing the power is constructed using the Levenberg-Marquardt algorithm.

[0010] Step S4: Based on the constructed regression model, the power consumption data of each device of the enterprise is input to obtain the consumption data of each production material. The carbon emission factor of each material is combined with the material carbon emission accounting formula in the energy-carbon method to calculate the carbon emission of each material, and the direct carbon emissions of the enterprise are obtained by superimposing them.

[0011] Step S5: Multiply the enterprise's electricity consumption data by the local area electricity carbon factor to obtain the enterprise's indirect carbon emissions;

[0012] Step S6: Add the direct and indirect carbon emissions of the enterprise obtained in step S4 and step S5 to obtain the total carbon emissions of the enterprise, thereby realizing carbon emission monitoring of the enterprise.

[0013] In a preferred embodiment, the DTW algorithm steps in step S2 are as follows:

[0014] Step S201: Assume that the power of each production equipment collected in step S1 is P:

[0015]

[0016] Where: p i,j represents the power value of the jth point in the power sequence of the i-th production equipment;

[0017] Step S202: Assume that the input enterprise material consumption data is Y:

[0018]

[0019] In the formula: y i,j represents the value of the j-th point in the i-th material consumption data sequence;

[0020] Step S203: For a certain material Y l , its consumption data sequence is Y l ={y l,1 , y l,2 ,..., y l,n}, where there are a total of k devices consuming this material. For a certain device h among these k devices, its power sequence is P h ={p h,1 , p h,2 ,..., p<00​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​​h 'The total number of elements;

[0027] Step S206: Next, define a cumulative distance matrix L of dimension M×N, and record the overall minimum cumulative distance between the two sets of time series data through iterative calculation. The specific calculation formula is as follows:

[0028]

[0029] Where: S[i,j] represents the overall cumulative distance, and the initial value is set to S[1,1]=D[1,1]=d(y′ l,1 ,p' h,1 );

[0030] Step S207: Continuously iterate the calculation based on the above formula until i=M, j=N. At this time, S[M,N] is the overall cumulative distance between the material consumption data and the equipment power data, recorded as the distance coefficient d(Y l ',P h '); This parameter is used to characterize the similarity of the changing trends between two time series. The smaller its value is, the smaller the cumulative distance between the two curves in the time series is, that is, the more similar the changing trends of the two curves in the period are.

[0031] Calculate material consumption data Y separately l 'The distance coefficient between each production equipment that consumes the material, and get the distance coefficient matrix W i :

[0032] W i =[d(Y l ',P1')...d(Y l ',P h ')...d(Y l ',P k ')] (6)

[0033] Where, d(Y l ',P h ') represents material consumption data Y l The power consumption P of the equipment h h The distance coefficient between

[0034] Step S208: Finally, the distance coefficient W is obtained i Perform normalization processing of the complementary value:

[0035]

[0036] After performing the above processing on all elements, the new correlation coefficient matrix W is obtained i ′, where d'(Y l ',P h') is larger, it means that the material consumption data Y l The power consumption P of the equipment h h The stronger the correlation between them, the better for material Y l The same correlation analysis is performed on the k devices, and the devices with a correlation coefficient greater than 0.7 are selected as the input devices for the subsequent material regression model.

[0037] In a preferred embodiment, the Levenberg-Marquardt algorithm in step S3 includes the following steps:

[0038] Step S301: For the material Y calculated in step 2 l For r devices with strong correlation, the power of these r devices is summed up as the power input P of the subsequent regression model;

[0039] P=P1+P2+...+P r (8)

[0040] Step S302: Assume that the integrated equipment power data P and the material data Y consumed by it are [(p1,y1),(p2,y2),...,(p n ,y n )], construct the function f(x)=yx(p), and let the parameter matrix be X=[x1,x2,x3,...,x n ] T , construct the least squares problem:

[0041]

[0042] Step S303: Perform a first-order Taylor expansion on f(X) and remove the higher-order terms, and substitute it into F(X) in formula (9) to obtain formula (10):

[0043]

[0044] Where: J is the Jacobian matrix, is the partial derivative of F(X) with respect to x, and Δx is the parameter iteration increment;

[0045] Step S304: Define equation (10) as a function of Δx, and introduce the damping coefficient μΔx into equation (10) T Δx, and taking the partial derivative of formula (10) with respect to Δx, we get the following formula:

[0046]

[0047] Step S305: Define J(x) T J(x) is H, -J(x) T f(x) is g, as shown in Equations (12) and (13):

[0048] H=J(x) T J(x) (12)

[0049] g=-J(x) T f(x) (13)

[0050] Step S306: Set equation (11) to 0, and substitute equations (6) and (7) into equation (5) to obtain the following iterative equation:

[0051] HΔx+μΔx=g (14)

[0052] Step S307: Solve equation (14) to obtain Δx. If Δx≤ε, the obtained x at the end of the iteration is the regression model parameter. Otherwise, let x k+1 =x k +Δx Repeat steps S303 to S307.

[0053] In a preferred embodiment, the complete carbon emission monitoring steps in S4 are as follows:

[0054] Step S401: Based on the regression model constructed in step S3, the material consumption data of each enterprise can be obtained by inputting the power data of each device Y=[Y1,Y2,Y3,...,Y n ], the consumption data of each material can be superimposed to obtain the consumption FC of each material in a period of time;

[0055] Step S402: The material includes fossil fuels and carbon-containing raw materials. The carbon emissions of the material are calculated using the carbon emission calculation method for fossil fuels and carbon-containing raw materials. The formula is shown in the following formula (15) (16):

[0056] E 燃料 =NCV×FC×EF (15)

[0057] Where: E 燃料 represents the carbon emissions of fossil fuels, NCV is the average lower calorific value of fossil fuels, and EF represents the carbon emission factor of fossil fuels, with the unit being tons of carbon dioxide per unit of fossil fuel consumed;

[0058] E 原料 =FC×EF (16)

[0059] Where: E 原料 It represents the carbon emissions of carbon-containing raw materials, and EF represents the carbon emission factor of carbon-containing raw materials, with the unit being tons of carbon dioxide / unit of corresponding consumption of carbon-containing raw materials.

[0060] Step S403: Obtain the carbon emission E of each material based on the carbon emission of the material calculated in S402

[0061] E={E1,E2,...,E o ,...,E n} (18)

[0062] Where: E i Indicates the carbon emissions of the oth material;

[0063] Step S404: The total carbon emissions of the enterprise include the enterprise's direct carbon emissions and the enterprise's indirect carbon emissions. The enterprise's direct carbon emissions refer to formula (19):

[0064] E 直接 =E1+E2+...+E n (19)

[0065] In a preferred embodiment, the total carbon emissions of the enterprise in step S5 are calculated as follows:

[0066] Step S501: Calculate the indirect carbon emissions of the enterprise. Input the total electricity power S of the enterprise and the regional electricity carbon factor CFP of the region where the enterprise is located. Refer to formula (21) to calculate the indirect carbon emissions caused by the enterprise's electricity consumption.

[0067] S={s1,s2,...,s i ,...,s n} (20)

[0068] E 间接 =S×CFP (21).

[0069] In a preferred embodiment, the total carbon emissions of the enterprise in step S6 are calculated as follows:

[0070] Step S601: Total carbon emissions of the enterprise E 总 The calculation formula is as follows:

[0071] E 总 =E 直接 +E 间接 (twenty two).

[0072] The present invention also provides an enterprise carbon emission monitoring system that considers the relationship between equipment power and material consumption, and runs the above-mentioned enterprise carbon emission monitoring method that considers the relationship between equipment power and material consumption.

[0073] Compared with the existing technology, the present invention has the following beneficial effects: before establishing the regression model, the present invention screens out equipment with a strong correlation with the material through the correlation analysis method, making the regression model more accurate and reliable; in the process of training the model, the production material consumption data of the equipment is taken into account, which can better and more accurately monitor the carbon emissions of the enterprise; only simple power monitoring equipment needs to be installed on the electrical equipment within the enterprise, and no additional carbon monitoring equipment needs to be installed to realize the carbon emission monitoring of the enterprise, which is low cost and simple to install. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a schematic diagram of the process of the present invention;

[0075] Figure 2 Specific flow chart of the present invention;

[0076] Figure 3 Schematic diagram of the regression model construction in the present invention. DETAILED DESCRIPTION

[0077] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0078] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present application belongs.

[0079] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application; as used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form, and it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components and / or their combinations.

[0080] A method for monitoring enterprise carbon emissions considering the relationship between equipment power and material consumption, referring to Figure 1-3 ,include:

[0081] (1) Before establishing the regression model of material and equipment power consumption, the association analysis method was used to screen out equipment with a strong correlation with the material for subsequent regression model construction, making the model construction more accurate.

[0082] (2) An enterprise "electricity-energy-carbon" model was established to analyze the relationship between the materials consumed in the enterprise's production process and the power consumption of the equipment, which can achieve accurate carbon emission monitoring through power data.

[0083] (3) The above algorithms and models can be used to accurately monitor corporate carbon emissions without installing expensive carbon monitoring equipment. The installation is simple and data support can be provided to environmental protection departments.

[0084] The specific steps include:

[0085] Step S1: First, install electricity data collection devices at the company's power lines and at each production facility to monitor the power consumption of the company's production equipment. Data on fuel and material consumption, carbon emission factors, fuel low calorific value, and the company's production process topology are collected from the company's annual production report. The local regional electricity carbon factor is also input. Power consumption data is collected every 3 minutes, yielding a total of 480 data points over a 24-hour day. The company's production process topology includes the connections between each device and the materials consumed by each device.

[0086] Step S2: Dynamic Time Warping (DTW) is used to calculate the correlation between the consumption data of each material and the power of the main equipment that consumes them, quantify the statistical correlation coefficient, and screen out the equipment with strong correlation with each material.

[0087] Step S3: Based on the power of each production equipment of the enterprise collected in step S1, combined with step S2, the power-consuming equipment with a strong correlation with the material is calculated. Then, combined with the material relationship consumed by each strongly correlated equipment, a regression relationship model between the strongly correlated production equipment and each material after superimposing the power is constructed using the Levenberg-Marquardt algorithm.

[0088] Step S4: According to the constructed regression model, the power consumption data of each equipment of the enterprise can be input to obtain the consumption data of each production material. The carbon emission factor of each material is combined with the material carbon emission accounting formula in the energy-carbon method to calculate the carbon emission of each material, and the direct carbon emissions of the enterprise are obtained by superimposing them.

[0089] Step S5: The enterprise's indirect carbon emissions are obtained by multiplying the enterprise's electricity consumption data by the local area electricity carbon factor.

[0090] Step S6: Add the direct and indirect carbon emissions of the enterprise obtained in step S4 and step S5 to obtain the total carbon emissions of the enterprise, thereby realizing carbon emission monitoring of the enterprise.

[0091] The DTW algorithm steps described in step S2 are as follows:

[0092] Step S201: Assume that the power of each production equipment collected in step S1 is P:

[0093]

[0094] Where: p i,j represents the power value of the j-th point in the power sequence of the i-th production device;

[0095] Step S202: Let the input enterprise material consumption data be Y:

[0096]

[0097] Where: y i,j represents the value of the j-th point in the i-th material consumption data sequence;

[0098] Step S203: For a certain material Y l , its consumption data sequence is Y l = {y l,1 , y l,2 ,..., y l,n}, where there are a total of k devices consuming this material. For a certain device h among these k devices, its power sequence is P h = { p, p h,2 ,..., p h,n};

[0099] Step S204: Perform normalization on the material consumption data and the device power data. The unified normalization formula is shown in Equation (3). Let the data after normalization be Y l ' and P h '

[0100] <{}>0000375<{}>

[0101] Where: X represents the original time-series data, X' represents the time-series data after normalization, X min represents the minimum value in this set of time-series data, X max represents the maximum value in this set of time-series data

[0102] Step S205: Define a matrix D with a dimension of M×N for the material consumption data Y l ' and the device power P h '. The calculation formula for the elements in the matrix is shown in Equation (4):

[0103]

[0104] Where: d(y′ l,i , p' h,j ) is the Euclidean distance between two elements; y′ l,i is the i-th element of the material consumption data Y l ', 0 < i < M, M is the total number of elements included in the material consumption data Y l '; p'​h,j For the electric power P of device h h The j-th element of ', 0 < j < N, where N is the total number of elements of the electric power P of device h h ';

[0105] Step S206: Secondly, define an accumulated distance matrix L with dimensions M×N, and record the overall minimum accumulated distance between two sets of time series data through iterative calculation. The specific calculation formula is as follows:

[0106]

[0107] In the formula: S[i,j] represents the overall accumulated distance, and the initial value is set as S[1,1] = D[1,1] = d(y′ l,1 , p' h,1 ).

[0108] Step S207: Continuously perform iterative calculation based on the above formula until i = M and j = N. At this time, S[M,N] is the overall accumulated distance between the material consumption data and the device power data, denoted as the distance coefficient d(Y l ', P h '). This parameter is used to characterize the similarity degree of the change trends between two time series. The smaller its value, the smaller the accumulated distance between the two curves in time series, that is, the more similar the change trends of the two during this period.

[0109] Calculate the distance coefficients between the material consumption data Y l ' and each production device consuming this material respectively, and obtain the distance coefficient matrix W i :

[0110] W i = [d(Y l ', P1')... d(Y l ', P h ')... d(Y l ', P k ')] (6)

[0111] In the formula, d(Y l ', P h ') represents the distance coefficient between the material consumption data Y l and the electric power consumption P of device h h .

[0112] Step S208: Finally, perform normalization processing on the obtained distance coefficient W i :

[0113]

[0114] After performing the above processing on all elements, the new correlation coefficient matrix W is obtained i ′, where d'(Y l ',P h ') is larger, it means that the material consumption data Y l The power consumption P of the equipment h h The stronger the correlation between them, the better for material Y l The same correlation analysis is performed on the k devices, and the devices with a correlation coefficient greater than 0.7 are selected as the input devices for the subsequent material regression model.

[0115] The Levenberg-Marquardt algorithm described in further step S3 comprises the following steps:

[0116] Step S301: For the material Y calculated in step 2 l For r devices with strong correlation, the power of these r devices is summed up as the power input P of the subsequent regression model;

[0117] P=P1+P2+...+P r (8)

[0118] Step S302: Assume that the integrated equipment power data P and the material data Y consumed by it are [(p1,y1),(p2,y2),...,(p n ,y n )], construct the function f(x)=yx(p), and let the parameter matrix be X=[x1,x2,x3,...,x n ] T , the least squares problem can be constructed:

[0119]

[0120] Step S303: Perform a first-order Taylor expansion on f(X) and remove the higher-order terms, and substitute it into F(X) in formula (9) to obtain formula (10):

[0121]

[0122] Where: J is the Jacobian matrix, is the partial derivative of F(X) with respect to x, and Δx is the parameter iteration increment;

[0123] Step S304: Define equation (10) as a function of Δx, and introduce the damping coefficient μΔx into equation (10) T Δx, and taking the partial derivative of formula (10) with respect to Δx, we get the following formula:

[0124]

[0125] Step S305: Define J(x)T J(x) is H, -J(x) T f(x) is g, as shown in Equations (12) and (13):

[0126] H=J(x) T J(x) (12)

[0127] g=-J(x) T f(x) (13)

[0128] Step S306: Set equation (11) to 0, and substitute equations (6) and (7) into equation (5) to obtain the following iterative equation:

[0129] HΔx+μΔx=g (14)

[0130] Step S307: Solve equation (14) to obtain Δx. If Δx≤ε, the obtained x at the end of the iteration is the regression model parameter. Otherwise, let x k+1 =x k +Δx Repeat steps S303-S307.

[0131] The complete carbon emission monitoring steps in S4 are as follows:

[0132] Step S401: Based on the regression model constructed in step S3, the material consumption data of the enterprise can be obtained by inputting the power data of each device Y=[Y1,Y2,Y3,…,Y n ], the consumption data of each material can be superimposed to obtain the consumption FC of each material in a period of time;

[0133] Step S402: The material includes fossil fuels and carbon-containing raw materials. The carbon emissions of the material are calculated using the carbon emission calculation method for fossil fuels and carbon-containing raw materials. The formula is shown in the following formula (15) (16):

[0134] E 燃料 =NCV×FC×EF (15)

[0135] Where: E 燃料 It represents the carbon emissions of fossil fuels, NCV is the average lower calorific value of fossil fuels, and EF is the carbon emission factor of fossil fuels, with the unit being tons of carbon dioxide per corresponding unit of fossil fuel consumption.

[0136] E 原料 =FC×EF (16)

[0137] Where: E 原料 It represents the carbon emissions of carbon-containing raw materials, and EF represents the carbon emission factor of carbon-containing raw materials, with the unit being tons of carbon dioxide / corresponding unit of carbon-containing raw materials consumed.

[0138] Step S403: Obtain the carbon emission E of each material based on the carbon emission of the material calculated in S402

[0139] E={E1,E2,...,E o ,...,E n} (18)

[0140] Where: E i Indicates the carbon emissions of the oth material;

[0141] Step S404: The total carbon emissions of the enterprise include the enterprise's direct carbon emissions and the enterprise's indirect carbon emissions. The enterprise's direct carbon emissions refer to formula (19):

[0142] E 直接 =E1+E2+...+E n (19)

[0143] Furthermore, the total carbon emissions of the enterprise described in step S5 are calculated as follows:

[0144] Step S501: Calculate the indirect carbon emissions of the enterprise. Input the total electricity power S of the enterprise and the regional electricity carbon factor CFP of the region where the enterprise is located. Refer to formula (21) to calculate the indirect carbon emissions caused by the enterprise's electricity consumption.

[0145] S={s1,s2,...,s i ,...,s n} (20)

[0146] E 间接 =S×CFP (21)

[0147] Furthermore, the total carbon emissions of the enterprise described in step S6 are calculated as follows:

[0148] Step S601: The formula for calculating the total carbon emissions of an enterprise is as follows:

[0149] E 总 =E 直接 +E 间接 (twenty two)

[0150] The present invention is different from the traditional carbon emission factor method and online monitoring method. The present invention obtains the power consumption of different production equipment within the enterprise by adding power consumption monitoring equipment to the electrical equipment within the enterprise; through the correlation analysis algorithm, the equipment with a strong correlation with each material is screened out for subsequent modeling and analysis; the power consumption of different production equipment with a strong correlation with each material is superimposed and used with the material data as training data to establish a regression model; according to the established regression model, the consumption data of each material is obtained through the equipment power consumption data, and then the total carbon emissions of the enterprise are obtained by combining the carbon emission factors of each material and superimposing them.

Claims

1. A method for monitoring enterprise carbon emissions that considers the relationship between equipment power and material consumption, characterized in that: The following steps are involved: Step S1: First, install electricity data collection devices at the company's power line and at each production equipment to monitor the power consumption of the company's production equipment. Obtain the company's fuel and material consumption data, carbon emission factor, fuel low heating value, and company production process topology from the company's annual production report, and input the local regional electricity carbon factor. Power consumption data is collected every 3 minutes, resulting in a total of 480 data points over a 24-hour day. The company's production process topology includes the connection relationships between each device and the relationship between the materials consumed by each device. Step S2: Using the dynamic time warping algorithm (DTW) to calculate the correlation between the consumption data of each material and the power of the main equipment that consumes them, quantify the statistical correlation coefficient, and screen out the equipment with strong correlation with each material; Step S3: Based on the power of each production equipment collected in step S1, combined with step S2, the power-consuming equipment with a strong correlation with the material is calculated. Furthermore, the material relationship consumed by each strongly correlated equipment is further combined, and a regression relationship model between the strongly correlated production equipment and each material after superimposing the power is constructed using the Levenberg-Marquardt algorithm. Step S4: Based on the constructed regression model, the power consumption data of each device of the enterprise is input to obtain the consumption data of each production material. The carbon emission factor of each material is combined with the material carbon emission accounting formula in the energy-carbon method to calculate the carbon emission of each material, and the direct carbon emissions of the enterprise are obtained by superimposing them. Step S5: Multiply the enterprise's electricity consumption data by the local area electricity carbon factor to obtain the enterprise's indirect carbon emissions; Step S6: Add the direct and indirect carbon emissions of the enterprise obtained in step S4 and step S5 to obtain the total carbon emissions of the enterprise, thereby realizing carbon emission monitoring of the enterprise.

2. The enterprise carbon emission monitoring method considering the relationship between equipment power and material consumption according to claim 1 is characterized in that: The DTW algorithm steps described in step S2 are as follows: Step S201: Assume that the power of each production equipment collected in step S1 is P: Where: p i,j represents the power value of the jth point in the power sequence of the i-th production equipment; Step S202: Assume that the input enterprise material consumption data is Y: Where: y i,j Represents the value of the jth point in the i-th material consumption data sequence; Step S203: For a certain material Y l , whose consumption data sequence is Y l ={y l,1 ,y l,2 ,...,y l,n }, there are k devices that consume the material. For a device h among these k devices, its power sequence is P h ={p h,1 ,p h,2 ,...,p h,n }; Step S204: normalize the material consumption data and the equipment power data. The unified normalization formula is shown in formula (3). Let the normalized data be Y l 'with P h ' Where: X represents the original time series data, X' represents the normalized time series data, X min Indicates the smallest value in this set of time series data, X max Indicates the maximum value in this set of time series data; Step S205: Material consumption data Y l 'With device power P' h Define a matrix D with a dimension of M×N, where the calculation formula of the elements in the matrix is ​​shown in formula (4): where: d(y' l,i , p' h,j ) is the Euclidean distance between two elements; y' l,i is the i-th element of the material consumption data Y' l , 0 < i < M, where M is the total number of elements of the material consumption data Y' l ; p' h,j is the j-th element of the electric power P' h of equipment h, 0 < j < N, where N is the total number of elements of the electric power P' h of equipment h; Step S206: Next, define a cumulative distance matrix L of dimension M×N, and record the overall minimum cumulative distance between the two sets of time series data through iterative calculation. The specific calculation formula is as follows: Where: S[i,j] represents the overall cumulative distance, and the initial value is set to S[1,1]=D[1,1]=d(y' l,1 ,p' h,1 ); Step S207: Continuously iterate the calculation based on the above formula until i=M, j=N. At this time, S[M,N] is the overall cumulative distance between the material consumption data and the equipment power data, recorded as the distance coefficient d(Y l ',P' h ); This parameter is used to characterize the similarity of the changing trends between two time series. The smaller its value is, the smaller the cumulative distance between the two curves in the time series is, that is, the more similar the changing trends of the two curves in the period are. Calculate material consumption data Y separately l 'The distance coefficient between each production equipment that consumes the material, and get the distance coefficient matrix W i : W i =[d(Y' l ,P'1)...d(Y' l ,P' h )...d(Y' l ,P' k )] (6) Where, d(Y' l ,P' h ) represents material consumption data Y l The power consumption P of the equipment h h The distance coefficient between Step S208: Finally, the distance coefficient W is obtained i Perform normalization processing of the complementary value: After performing the above processing on all elements, the new correlation coefficient matrix W is obtained i ′, where d'(Y' l ,P' h ) is larger, the material consumption data Y l The power consumption P of the equipment h h The stronger the correlation between them, the better for material Y l The same correlation analysis is performed on the k devices, and the devices with a correlation coefficient greater than 0.7 are selected as the input devices for the subsequent material regression model.

3. The enterprise carbon emission monitoring method considering the relationship between equipment power and material consumption according to claim 1 is characterized in that: The Levenberg-Marquardt algorithm described in step S3 includes the following steps: Step S301: For the material Y calculated in step 2 l For r devices with strong correlation, the power of these r devices is summed up as the power input P of the subsequent regression model; P=P1+P2+...+P r (8) Step S302: Assume that the integrated equipment power data P and the material data Y consumed by it are [(p1,y1),(p2,y2),…,(p n ,y n )], construct the function f(x)=yx(p), and let the parameter matrix be X=[x1,x2,x3,…,x n ] T , construct the least squares problem: Step S303: Perform a first-order Taylor expansion on f(X) and remove the higher-order terms, and substitute it into F(X) in formula (9) to obtain formula (10): Where: J is the Jacobian matrix, is the partial derivative of F(X) with respect to x, and Δx is the parameter iteration increment; Step S304: Define equation (10) as a function of Δx, and introduce the damping coefficient μΔx into equation (10) T Δx, and taking the partial derivative of formula (10) with respect to Δx, we get the following formula: Step S305: Define J(x) T J(x) is H, -J(x) T f(x) is g, as shown in Equations (12) and (13): H=J(x) T J(x) (12) g=-J(x) T f(x) (13) Step S306: Set equation (11) to 0, and substitute equations (6) and (7) into equation (5) to obtain the following iterative equation: HΔx+μΔx=g (14) Step S307: Solve equation (14) to obtain Δx. If Δx≤ε, the obtained x at the end of the iteration is the regression model parameter. Otherwise, let x k+1 =x k +Δx Repeat steps S303 to S307.

4. The enterprise carbon emission monitoring method considering the relationship between equipment power and material consumption according to claim 1 is characterized in that: The complete carbon emission monitoring steps in S4 are as follows: Step S401: Based on the regression model constructed in step S3, the material consumption data of the enterprise can be obtained by inputting the power data of each device Y=[Y1,Y2,Y3,…,Y n ], the consumption data of each material can be superimposed to obtain the consumption FC of each material in a period of time; Step S402: The material includes fossil fuels and carbon-containing raw materials. The carbon emissions of the material are calculated using the carbon emission calculation method for fossil fuels and carbon-containing raw materials. The formula is shown in the following formula (15) (16): AND 燃料 =NCV×FC×EF (15) Where: E 燃料 represents the carbon emissions of fossil fuels, NCV is the average lower calorific value of fossil fuels, and EF represents the carbon emission factor of fossil fuels, with the unit being tons of carbon dioxide per unit of fossil fuel consumed; AND 原料 =FC×EF (16) Where: E 原料 It represents the carbon emissions of carbon-containing raw materials, and EF represents the carbon emission factor of carbon-containing raw materials, with the unit being tons of carbon dioxide / unit of corresponding consumption of carbon-containing raw materials. Step S403: Obtain the carbon emission E of each material based on the carbon emission of the material calculated in S402 E={E1,E2,...,E o ,...,THE n } (18) Where: E i Indicates the carbon emissions of the oth material; Step S404: The total carbon emissions of the enterprise include the enterprise's direct carbon emissions and the enterprise's indirect carbon emissions. The enterprise's direct carbon emissions refer to formula (19): THE 直接 =E1+E2+…+E n (19) 5. The enterprise carbon emission monitoring method considering the relationship between equipment power and material consumption according to claim 1 is characterized in that: The total carbon emissions of the enterprise described in step S5 are calculated as follows: Step S501: Calculate the indirect carbon emissions of the enterprise. Input the total electricity power S of the enterprise and the regional electricity carbon factor CFP of the region where the enterprise is located. Refer to formula (21) to calculate the indirect carbon emissions caused by the enterprise's electricity consumption. S={s1,s2,...,s i ,...,s n } (20) AND 间接 =S×CFP (21)。 6. The enterprise carbon emission monitoring method considering the relationship between equipment power and material consumption according to claim 1 is characterized in that: The total carbon emissions of the enterprise described in step S6 are calculated as follows: Step S601: Total carbon emissions of the enterprise E 总 The calculation formula is as follows: AND 总 =And 直接 +E 间接 (22)。 7. An enterprise carbon emission monitoring system that considers the relationship between equipment power and material consumption, characterized by: Run the enterprise carbon emission monitoring method according to any one of claims 1 to 6 above, which considers the relationship between equipment power and material consumption.

Images