Regional energy consumption and carbon emission early warning method and system based on electric energy carbon coupling

Abstract

The invention relates to a regional energy consumption and carbon emission early warning method and system based on electric energy carbon coupling, and the method comprises the steps: firstly, carrying out the screening to obtain an electric quantity characteristic parameter and a non-electric quantity characteristic parameter which have the strongest correlation with regional energy consumption and regional carbon emission, and forming an electricity-energy-carbon characteristic parameter library; then, constructing an electricity-energy-carbon preliminary coupling model by adopting an autoregressive distributed hysteresis model, and optimizing a hysteresis period of the preliminary coupling model by adopting a genetic algorithm by taking the minimum relative error of the predicted regional energy consumption and carbon emission as an optimization target to obtain an electricity-energy-carbon coupling model; and finally, inputting the data of the electric quantity characteristic parameters obtained in real time into the coupling model, measuring and calculating to obtain regional energy consumption and carbon emission, and carrying out regional energy consumption and carbon emission early warning. According to the method, the energy consumption and the carbon emission of the region can be measured and analyzed more conveniently and more timely, and green and low-carbon development of the region is assisted through real-time energy consumption and carbon emission early warning.

Description

Technical Field

[0001] The present invention belongs to the technical field of electric power, and specifically relates to a method and system for warning regional energy consumption and carbon emissions based on the coupling of electric energy and carbon. Background Art

[0002] In the global carbon emission process, the monitoring of regional energy consumption and carbon emissions is an important basis for formulating emission reduction policies, evaluating emission reduction effects, and promoting green and low-carbon development. By monitoring the regional energy consumption and carbon emissions in real time, abnormal changes in energy consumption and carbon emissions can be found in a timely manner, providing effective assistance for regions to improve energy utilization efficiency and reduce carbon emission intensity.

[0003] At present, there are still problems in the monitoring technology of regional energy consumption and carbon emissions, such as high monitoring costs, difficult data collection, and uneven data quality, which are difficult to meet the monitoring requirements under different scenarios and different needs. At the same time, due to the very slow update of energy consumption data, which lags behind by 1-2 years, the monitoring results of traditional methods are lagging, and it is difficult to carry out monitoring and evaluation of energy consumption and carbon emissions in a timely manner, which is not conducive to the precise control of regional energy by local governments. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for warning regional energy consumption and carbon emissions based on the coupling of electric energy and carbon in view of the above problems existing in the prior art.

[0005] To achieve the above objectives, the technical solution of the present invention is as follows:

[0006] In the first aspect, the present invention proposes a method for warning regional energy consumption and carbon emissions based on the coupling of electric energy and carbon, including:

[0007] S1. Screen out the electrical quantity characteristic parameters and non-electrical quantity characteristic parameters with the strongest correlation with the regional energy consumption, and the electrical quantity characteristic parameters and non-electrical quantity characteristic parameters with the strongest correlation with the regional carbon emissions, and form an electric-energy-carbon characteristic parameter library;

[0008] S2. Based on the data of each characteristic parameter in the electric-energy-carbon characteristic parameter library, use the autoregressive distributed lag model to construct an electric-energy-carbon preliminary coupling model;

[0009] S3. Taking the minimum relative error of the predicted regional energy consumption and carbon emissions as the optimization objective, use the genetic algorithm to optimize the lag period of the electric-energy-carbon preliminary coupling model to obtain an electric-energy-carbon coupling model;

[0010] S4. Input the data of the electrical quantity characteristic parameters obtained in real time into the electric-energy-carbon coupling model, calculate the regional energy consumption and carbon emissions, and carry out early warnings on the regional energy consumption and carbon emissions according to the calculation results.

[0011] In S2, the preliminary electricity-energy-carbon coupling model is as follows:

[0012]

[0013] In the above formula, E n+1 , C n+1 are the regional energy consumption and carbon emissions in period n + 1 respectively, is the value of the p-th electricity characteristic parameter related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library in period q, is the value of the p-th non-electricity characteristic parameter related to the energy consumption in the electricity-energy-carbon characteristic parameter library in period q, is the value of the s-th electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library in period t, is the value of the s-th non-electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library in period t, are respectively coefficients of, e e , e c are the random deviations of energy consumption and carbon emissions respectively. a1 and a2 are the numbers of electricity characteristic parameters and non-electricity characteristic parameters related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library respectively. b1 and b2 are the numbers of electricity characteristic parameters and non-electricity characteristic parameters related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library respectively. q and t are lag periods.

[0014] S3 includes:

[0015] S31. Initialize the population, and the individuals in the population represent the combinations of characteristic parameters at each lag period;

[0016] S32. Calculate the fitness value of each individual in the population, and judge whether the fitness of the individual meets the optimization goal. If it meets, output the best individual and its optimal solution; if not, enter S33; where the fitness is the negative absolute value of the measurement error of the electricity-energy-carbon coupling model;

[0017] S33. Select parents according to the fitness value. The individuals with high fitness have a high probability of being selected, and the individuals with low fitness are eliminated. Then, perform crossover and mutation in turn. After generating a new generation of population, return to S32 for the next iteration until the optimal solution is generated.

[0018] S4 includes:

[0019] S41. Real-time obtain the electricity characteristic parameter data in period n + 1, and input it into the electricity-energy-carbon coupling model to calculate the regional energy consumption and carbon emissions;

[0020] S42. Calculate the energy consumption warning coefficient and carbon emission warning coefficient for period n + 1 according to the following formula:

[0021]

[0022] In the above formula, are respectively the energy consumption warning coefficient and carbon emission warning coefficient in period T of period n + 1, T t is the total number of moments in period n + 1, E tar is the energy consumption amount and carbon emission amount at moment T in period n + 1, E n+1,T , C n+1,T are respectively the energy consumption target and carbon emission target in period n + 1; tar , C tar

[0023] S43. Conduct warnings on the regional energy consumption amount and carbon emission amount according to the warning coefficient, specifically:

[0024] If and Conduct warnings on both the regional energy consumption amount and the regional carbon emission amount simultaneously;

[0025] If and Only conduct a warning on the regional energy consumption amount;

[0026] If and Only conduct a warning on the regional carbon emission amount.

[0027] The above-mentioned S1 includes:

[0028] S11. Conduct accounting on the regional historical energy consumption amount and carbon emission amount according to the following formula:

[0029]

[0030] In the above formula, E n , C n are respectively the regional energy consumption amount and carbon emission amount in period n, γ k , γ en are respectively the standard coal conversion coefficient of the kth kind of energy and the standard coal conversion coefficient of electricity, G n,k , are respectively the physical quantity of the kth kind of energy consumed in region during period n and the energy consumption for raw materials, are respectively the power generation of renewable energy and the fth imported power in region during period n, λ k , are respectively the carbon emission factors of the kth kind of energy and the fth imported power, K and F are respectively the number of types of energy consumed in the region and the number of types of imported power;​

[0031] S12. Screen out the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional energy consumption, and the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional carbon emissions based on the Spearman correlation coefficient.

[0032] In a second aspect, the present invention proposes a regional energy consumption and carbon emission early warning system based on the electricity-carbon coupling, including an electricity-energy-carbon characteristic parameter library construction module, an electricity-energy-carbon preliminary coupling model construction module, a lag period optimization module, and an energy consumption and carbon emission early warning module.

[0033] The electricity-energy-carbon characteristic parameter library construction module is used to screen out the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional energy consumption, and the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional carbon emissions, and form an electricity-energy-carbon characteristic parameter library.

[0034] The electricity-energy-carbon preliminary coupling model construction module is used to construct an electricity-energy-carbon preliminary coupling model based on the data of each characteristic parameter in the electricity-energy-carbon characteristic parameter library by using an autoregressive distributed lag model.

[0035] The lag period optimization module is used to optimize the lag period of the electricity-energy-carbon preliminary coupling model by using a genetic algorithm with the minimum relative error of the predicted regional energy consumption and carbon emissions as the optimization goal, and obtain an electricity-energy-carbon coupling model.

[0036] The energy consumption and carbon emission early warning module is used to input the data of the electricity characteristic parameters obtained in real time into the electricity-energy-carbon coupling model, calculate the regional energy consumption and carbon emissions, and conduct early warnings on the regional energy consumption and carbon emissions according to the calculation results.

[0037] The electricity-energy-carbon preliminary coupling model is as follows:

[0038]

[0039] In the above formula, E n+1 , C n+1 are the regional energy consumption and carbon emissions in period n + 1 respectively. is the value of the p-th electricity characteristic parameter related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library in period q. is the value of the p-th non-electricity characteristic parameter related to the energy consumption in the electricity-energy-carbon characteristic parameter library in period q. is the value of the s-th electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library in period t. is the value of the s-th non-electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library in period t. respectively are the coefficients of, e e and e c are the random deviations of energy consumption and carbon emissions respectively. a1 and a2 are the numbers of electricity characteristic parameters and non - electricity characteristic parameters related to regional energy consumption in the electricity - energy - carbon characteristic parameter library respectively. b1 and b2 are the numbers of electricity characteristic parameters and non - electricity characteristic parameters related to regional carbon emissions in the electricity - energy - carbon characteristic parameter library respectively. q and t are the lag periods.

[0040] The lag period optimization module optimizes the lag period of the preliminary electricity - energy - carbon coupling model by using the following genetic algorithm:

[0041] A1. Initialize the population, and the individuals in the population represent the combinations of characteristic parameters at each lag period;

[0042] A2. Calculate the fitness value of each individual in the population, and judge whether the fitness of the individual meets the optimization goal. If it meets, output the best individual and its optimal solution; if not, go to A3; where the fitness is the negative absolute value of the measurement error of the electricity - energy - carbon coupling model;

[0043] A3. Select parents according to the fitness value. The individuals with high fitness have a high probability of being selected, and the individuals with low fitness are eliminated. Then, perform crossover and mutation in turn. After generating a new generation of population, return to A2 for the next iteration until the optimal solution is generated.

[0044] The energy consumption and carbon emission warning module includes an energy consumption and carbon emission real - time measurement unit, a warning coefficient calculation unit, and a warning unit;

[0045] The energy consumption and carbon emission real - time measurement unit is used to obtain the electricity characteristic parameter data at time period n + 1 in real time and input it into the electricity - energy - carbon coupling model to measure the regional energy consumption and carbon emissions;

[0046] The warning coefficient calculation unit is used to calculate the energy consumption warning coefficient and carbon emission warning coefficient at time period n + 1 according to the following formula:

[0047]

[0048] In the above formula, are the energy consumption warning coefficient and carbon emission warning coefficient at time T in time period n + 1 respectively, T t is the total number of moments in time period n + 1, E tar and C n+1,T are the energy consumption and carbon emissions at time T in time period n + 1 respectively, E n+1,T and C tar 、Ctar They are respectively the energy consumption target and carbon emission target for period n+1;

[0049] The early warning unit is used to conduct early warning on the regional energy consumption and carbon emissions according to the early warning coefficient, specifically:

[0050] If and Conduct early warning on both the regional energy consumption and regional carbon emissions simultaneously;

[0051] If and Only conduct early warning on the regional energy consumption;

[0052] If and Only conduct early warning on the regional carbon emissions.

[0053] The electricity-energy-carbon characteristic parameter library construction module includes a historical energy consumption and carbon emission accounting unit and a parameter screening unit;

[0054] The historical energy consumption and carbon emission accounting unit is used to calculate the regional historical energy consumption and carbon emissions according to the following formula:

[0055]

[0056] In the above formula, E n and C n are respectively the regional energy consumption and carbon emissions in period n, γ k and γ en are respectively the standard coal conversion coefficient of the kth kind of energy and the standard coal conversion coefficient of electricity, G n,k and are respectively the physical quantity of the kth kind of energy consumed in the region in period n and the energy consumption for raw materials, are respectively the power generation of renewable energy in the region in period n and the fth imported electricity, λ k and are respectively the carbon emission factors of the kth kind of energy and the fth imported electricity, K and F are respectively the number of types of energy consumed in the region and the number of types of imported electricity;

[0057] The parameter screening unit is used to screen out the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional energy consumption and the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional carbon emissions based on the Spearman correlation coefficient.

[0058] Compared with the prior art, the beneficial effects of the present invention are:

[0059] A method for early warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon first screens out the electric quantity characteristic parameters and non-electric quantity characteristic parameters with the strongest correlation with the regional energy consumption, and the electric quantity characteristic parameters and non-electric quantity characteristic parameters with the strongest correlation with the regional carbon emissions to form an electric-energy-carbon characteristic parameter library. Then, based on the data of each characteristic parameter in the electric-energy-carbon characteristic parameter library, an autoregressive distributed lag model is used to construct an initial electric-energy-carbon coupling model. Next, with the goal of minimizing the relative errors of the predicted energy consumption and carbon emissions, a genetic algorithm is used to optimize the lag period of the initial electric-energy-carbon coupling model to obtain an electric-energy-carbon coupling model. Finally, the data of the electric quantity characteristic parameters obtained in real time are input into the electric-energy-carbon coupling model to calculate the regional energy consumption and carbon emissions, and early warnings of the regional energy consumption and carbon emissions are carried out according to the calculation results. This method gives full play to the advantages of strong real-time and high resolution of electric power data, effectively solves the data lag problem in the monitoring of regional energy consumption and carbon emissions through the electric-energy-carbon coupling model, realizes more convenient and real-time measurement and analysis of regional energy consumption and carbon emissions, and helps the regional green and low-carbon development through real-time energy consumption and carbon emission early warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a flowchart of the method described in Embodiment 1.

[0061] Figure 2 It is a structural diagram of the system described in Embodiment 2.

[0062] Figure 3 It is a structural diagram of the system described in Embodiment 3. DETAILED DESCRIPTION OF THE INVENTION

[0063] The present invention will be further described in detail below in conjunction with the drawings and specific embodiments.

[0064] Embodiment 1:

[0065] A method for early warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon, as Figure 1 shown, the specific steps are as follows:

[0066] 1. Collect the annual consumption data of various energy sources and conduct accounting of the historical energy consumption and carbon emissions in the region.

[0067] The annual energy consumption data includes the physical quantities of energy consumption such as coal, oil, and natural gas. Considering the energy consumption for raw materials and the consumption of renewable energy, there is:

[0068]

[0069] In the above formula, E n 、C nThey are the regional energy consumption and carbon emissions in period n, respectively, γ k and γ en They are the standard coal conversion coefficients of the k-th type of energy and the standard coal conversion coefficient of electricity, respectively, G n,k and They are the physical quantity of the k-th type of energy consumed in the region in period n and the energy consumption for raw materials, respectively They are the power generation of renewable energy in the region in period n and the f-th type of imported electricity, respectively, λ k and They are the carbon emission factors of the k-th type of energy and the f-th type of imported electricity, respectively. K and F are the number of types of energy consumed in the region and the number of types of imported electricity, respectively

[0070] The standard coal conversion coefficients and carbon emission factors of typical fossil fuel consumption are shown in Table 1

[0071] Table 1 Standard coal conversion coefficients and carbon emission factors of typical fossil fuel consumption

[0072]

[0073] 2. Construct a characteristic parameter library for energy consumption and carbon emissions. The parameters in this characteristic parameter library include the electricity consumption of each industry in the region, the energy consumption of each industry in the region, the carbon emissions of each industry in the region, and the regional GDP, which are obtained by dividing each industry according to the classification method in the statistical yearbook

[0074] 3. Calculate the Spearman correlation coefficients between each characteristic parameter in the characteristic parameter library of energy consumption and carbon emissions and energy consumption and carbon emissions, respectively

[0075]

[0076] In the above formula, ρ 1,j and ρ 2,j They are the Spearman correlation coefficients between the j-th characteristic parameter and the regional energy consumption and carbon emissions, respectively. x j,n is the value of the j-th characteristic parameter in the n-th year, and y 1,n and y 2,n They are the energy consumption and carbon emissions of the region in the n-th year, respectively They are the averages of the j-th characteristic parameter, energy consumption, and carbon emissions in N years, respectively

[0077] 4. Arrange the parameters in the characteristic parameter library according to the magnitude of the Spearman correlation coefficient, and select a1 electricity characteristic parameters (i.e., the electricity consumption of a1 industries) and a2 non-electricity characteristic parameters with the strongest correlation with the regional energy consumption, and b1 electricity characteristic parameters (i.e., the electricity consumption of b1 industries) and b2 non-electricity characteristic parameters with the strongest correlation with the regional carbon emissions to form an electricity-energy-carbon characteristic parameter library

[0078] 5. Determine whether the data volume of each characteristic parameter in the electricity-energy-carbon characteristic parameter library is sufficient to support the construction of the electricity-energy-carbon preliminary coupling model. If not, use the cubic spline interpolation method to perform time series interpolation processing on each parameter in the electricity-energy-carbon characteristic parameter library to expand the data in the electricity-energy-carbon characteristic parameter library, specifically including:

[0079] 5.1. If a certain parameter T in the electricity-energy-carbon database has N data points (z n , T n ) in the time series, that is, there are (N - 1) data intervals, construct the cubic spline interpolation function S n , z n+1 for each small interval [z n (z), and there is:

[0080] S n (z) = λ n,1 (z - z n ) 3 + λ n,2 (z - z n ) 2 + λ n,3 (z - z n ) + λ n,4

[0081] In the above formula, λ n,1 , λ n,2 , λ n,3 , λ n,4 are undetermined coefficients;

[0082] 5.2. According to the requirements of the interpolation function boundary conditions, solve the function expressions of each small interval and form the cubic spline interpolation function S(z) for the entire interval [z n , z n+1 , and there is:

[0083] S(z) = {S n (z), z ∈ [z n , z n+1 , n = 1, 2,..., N - 1}.

[0084] 6. Based on the data of each characteristic parameter in the electricity-energy-carbon characteristic parameter library, construct an electricity-energy-carbon preliminary coupling model using the autoregressive distributed lag model.

[0085] The energy consumption E at time period n + 1 n+1Calculated from a1 electricity characteristic parameters with the strongest correlation with the regional energy consumption from time period n + 1 - i to time period n + 1 and a2 non - electricity characteristic parameter historical data with the strongest correlation with the regional energy consumption from time period n + 1 - i to time period n, the carbon emission C at time period n + 1 n+1 Calculated from b1 electricity characteristic parameters with the strongest correlation with the regional carbon emission from time period n + 1 - i to time period n + 1 and b2 non - electricity characteristic parameter historical data with the strongest correlation with the regional carbon emission from time period n + 1 - i to time period n, the preliminary electricity - energy - carbon coupling model is as follows:

[0086]

[0087] In the above formula, E n+1 and C n+1 are the regional energy consumption and carbon emission at time period n + 1 respectively, is the value of the p - th electricity characteristic parameter related to the regional energy consumption in the electricity - energy - carbon characteristic parameter library at time period q, is the value of the p - th non - electricity characteristic parameter related to the energy consumption in the electricity - energy - carbon characteristic parameter library at time period q, is the value of the s - th electricity characteristic parameter related to the regional carbon emission in the electricity - energy - carbon characteristic parameter library at time period t, is the value of the s - th non - electricity characteristic parameter related to the regional carbon emission in the electricity - energy - carbon characteristic parameter library at time period t, are respectively 's coefficients, obtained by solving using the least - squares method, e e and e c are the random deviations of energy consumption and carbon emission respectively. a1 and a2 are the numbers of electricity characteristic parameters and non - electricity characteristic parameters related to the regional energy consumption in the electricity - energy - carbon characteristic parameter library respectively. b1 and b2 are the numbers of electricity characteristic parameters and non - electricity characteristic parameters related to the regional carbon emission in the electricity - energy - carbon characteristic parameter library respectively. q and t are lag periods.

[0088] 7. Taking the minimum relative error of the regional energy consumption and carbon emission predicted by the preliminary electricity - energy - carbon coupling model as the optimization goal, the genetic algorithm is used to optimize the lag period of the preliminary electricity - energy - carbon coupling model, thereby constructing the electricity - energy - carbon coupling model, including:

[0089] 7.1. Initialize the population, and the individuals in the population represent the combinations of characteristic parameters under different lag periods;

[0090] 7.2. Calculate the fitness values of each individual in the population, and determine whether the fitness of the individual meets the optimization goal. If it meets, output the best individual and its optimal solution; if not, proceed to 7.3. Among them, the fitness is the negative absolute value of the measurement error of the electricity-energy-carbon coupling model.

[0091] 7.3. Select parents based on the fitness values. Individuals with higher fitness have a higher probability of being selected, and individuals with lower fitness are eliminated. Then, perform crossover and mutation in sequence. After generating a new generation of population, return to 7.2 for the next iteration until the optimal solution is obtained.

[0092] 8. Obtain the electricity characteristic parameter data of time period n + 1 in real time, and input it into the electricity-energy-carbon coupling model to calculate the regional energy consumption and carbon emissions.

[0093] 9. Calculate the energy consumption warning coefficient and carbon emission warning coefficient of time period n + 1 according to the following formula:

[0094]

[0095] In the above formula, are the energy consumption warning coefficient and carbon emission warning coefficient in the T t time period of time period n + 1 respectively. T tar is the total number of moments in time period n + 1. E n+1,T , C n+1,T are the energy consumption and carbon emissions at the T moment in time period n + 1 respectively. E tar , C tar are the energy consumption target and carbon emission target of time period n + 1 respectively.

[0096] 10. Conduct early warnings on the regional energy consumption and carbon emissions according to the warning coefficients. Specifically:

[0097] If and simultaneously conduct early warnings on the regional energy consumption and regional carbon emissions;

[0098] If and only conduct an early warning on the regional energy consumption;

[0099] If and only conduct an early warning on the regional carbon emissions.

[0100] Example 2:

[0101] A regional energy consumption and carbon emission early warning system based on electricity-carbon coupling, as Figure 2As shown, it includes an electricity-energy-carbon characteristic parameter library construction module, an electricity-energy-carbon preliminary coupling model construction module, a lag period optimization module, and an energy consumption and carbon emission warning module.

[0102] The electricity-energy-carbon characteristic parameter library construction module is used to screen out the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional energy consumption, and the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional carbon emission, so as to form an electricity-energy-carbon characteristic parameter library, including a historical energy consumption and carbon emission accounting unit and a parameter screening unit.

[0103] The historical energy consumption and carbon emission accounting unit is used to calculate the regional historical energy consumption and carbon emission according to the following formula:

[0104]

[0105] In the above formula, E n , C n are the regional energy consumption and carbon emission in period n respectively, γ k , γ en are the standard coal conversion coefficients of the kth energy and electricity respectively, G n,k , are the physical quantity of the kth energy consumed in the region in period n and the energy consumption for raw materials respectively, are the power generation of renewable energy in the region in period n and the imported electricity of the fth type respectively, λ k , are the carbon emission factors of the kth energy and the imported electricity of the fth type respectively, and K and F are the number of energy types consumed in the region and the number of imported electricity types respectively.

[0106] The parameter screening unit is used to screen out the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional energy consumption, and the electricity characteristic parameters and non-electricity characteristic parameters with the strongest correlation with the regional carbon emission based on the Spearman correlation coefficient.

[0107] The electricity-energy-carbon preliminary coupling model construction module is used to construct the following electricity-energy-carbon preliminary coupling model by using the autoregressive distributed lag model based on the data of each characteristic parameter in the electricity-energy-carbon characteristic parameter library:

[0108]

[0109] In the above formula, E n+1 , C n+1 are the regional energy consumption and carbon emission in period n + 1 respectively, is the value of the pth electricity characteristic parameter related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library in period q, is the value of the p-th non-electricity characteristic parameter related to energy consumption in the electricity-energy-carbon characteristic parameter library for period q, is the value of the s-th electricity characteristic parameter related to regional carbon emissions in the electricity-energy-carbon characteristic parameter library for period t, is the value of the s-th non-electricity characteristic parameter related to regional carbon emissions in the electricity-energy-carbon characteristic parameter library for period t, are respectively coefficients of, e e and e c are respectively the random deviations of energy consumption and carbon emissions, a1 and a2 are respectively the numbers of electricity characteristic parameters and non-electricity characteristic parameters related to regional energy consumption in the electricity-energy-carbon characteristic parameter library, b1 and b2 are respectively the numbers of electricity characteristic parameters and non-electricity characteristic parameters related to regional carbon emissions in the electricity-energy-carbon characteristic parameter library, and q and t are lag periods.

[0110] The lag period optimization module is used to optimize the lag period of the preliminary electricity-energy-carbon coupling model with the minimum relative error of the predicted regional energy consumption and carbon emissions as the optimization goal, and obtain the electricity-energy-carbon coupling model by using the following genetic algorithm:

[0111] A1. Initialize the population, and the individuals in the population represent the combinations of characteristic parameters under each lag period;

[0112] A2. Calculate the fitness values of each individual in the population, and judge whether the fitness of the individual meets the optimization goal. If it meets, output the best individual and its optimal solution; if not, go to A3; where the fitness is the negative absolute value of the measurement error of the electricity-energy-carbon coupling model;

[0113] A3. Select parents according to the fitness values. Individuals with high fitness have a high probability of being selected, and individuals with low fitness are eliminated. Cross and mutate in turn to generate a new generation of population, and then return to A2 for the next iteration until the optimal solution is generated.

[0114] The energy consumption and carbon emission warning module is used to input the data of the electricity characteristic parameters obtained in real time into the electricity-energy-carbon coupling model, calculate the regional energy consumption and carbon emissions, and conduct early warnings on the regional energy consumption and carbon emissions according to the calculation results, including an energy consumption and carbon emission real-time calculation unit, a warning coefficient calculation unit, and a warning unit.

[0115] The energy consumption and carbon emission real-time calculation unit is used to obtain the data of the electricity characteristic parameters in period n + 1 in real time and input them into the electricity-energy-carbon coupling model to calculate the regional energy consumption and carbon emissions.

[0116] The early warning coefficient calculation unit is used to calculate the energy consumption early warning coefficient and carbon emission early warning coefficient for period n + 1 according to the following formula:

[0117]

[0118] In the above formula, are respectively the energy consumption early warning coefficient and carbon emission early warning coefficient for the T t period in period n + 1, T tar is the total number of moments in period n + 1, E n+1,T , C n+1,T are respectively the energy consumption and carbon emission at moment T in period n + 1, E tar , C tar are respectively the energy consumption target and carbon emission target for period n + 1.

[0119] The early warning unit is used to conduct early warning on the regional energy consumption and carbon emission according to the early warning coefficient, specifically:

[0120] If and conduct early warning on both the regional energy consumption and the regional carbon emission simultaneously;

[0121] If and only conduct early warning on the regional energy consumption;

[0122] If and only conduct early warning on the regional carbon emission.

[0123] Embodiment 3:

[0124] As Figure 3 shown, the difference from Embodiment 2 is that:

[0125] The system described in this embodiment further includes a data expansion module. The data expansion module is used to perform time series interpolation processing on each parameter in the electricity-energy-carbon characteristic parameter library by using the cubic spline interpolation method when the data volume of each characteristic parameter in the electricity-energy-carbon characteristic parameter library cannot sufficiently support the construction of the electricity-energy-carbon preliminary coupling model, so as to expand the data in the electricity-energy-carbon characteristic parameter library.

[0126] The cubic spline interpolation method specifically includes:

[0127] B1. If a certain parameter T in the electricity-energy-carbon database has N data points (z n , T n ) in the time series, that is, there are (N - 1) data intervals, construct the cubic spline interpolation function S for each small interval [z n , z n+1 ​n (z), there is:

[0128] S n (z) = λ n,1 (z - z n ) 3 + λ n,2 (z - z n ) 2 + λ n,3 (z - z n ) + λ n,4

[0129] In the above formula, λ n,1 , λ n,2 , λ n,3 , λ n,4 are coefficients to be determined;

[0130] B2. According to the requirements of the interpolation function boundary conditions, solve the function expressions for each small interval and form the cubic spline interpolation function S(z) for the entire interval [z n , z n+1 , there is:

[0131] S(z) = {S n (z), z ∈ [z n , z n+1 , n = 1, 2,..., N - 1}.

Claims

1. A method for warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon, characterized in that the method includes: S1. Screen the electric quantity characteristic parameters and non-electric quantity characteristic parameters with the strongest correlation with the regional energy consumption, and the electric quantity characteristic parameters and non-electric quantity characteristic parameters with the strongest correlation with the regional carbon emissions to form an electric-energy-carbon characteristic parameter library; S2. Based on the data of each characteristic parameter in the electric-energy-carbon characteristic parameter library, use the autoregressive distributed lag model to construct an initial electric-energy-carbon coupling model; S3. Taking the minimum relative error of the predicted regional energy consumption and carbon emissions as the optimization goal, use the genetic algorithm to optimize the lag period of the initial electric-energy-carbon coupling model to obtain the electric-energy-carbon coupling model; S4. Input the data of the electric quantity characteristic parameters obtained in real time into the electric-energy-carbon coupling model, calculate the regional energy consumption and carbon emissions, and conduct warnings on the regional energy consumption and carbon emissions according to the calculation results.

2. A method for warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon according to claim 1, characterized in that in S2, the initial electric-energy-carbon coupling model is: In the above formula, E n+1 and C n+1 are the regional energy consumption and carbon emissions in period n + 1 respectively, is the value of the p-th electricity characteristic parameter related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library for period q, is the value of the p-th non-electricity characteristic parameter related to the energy consumption in the electricity-energy-carbon characteristic parameter library for period q, is the value of the s-th electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library for period t, is the value of the s-th non-electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library for period t, are respectively coefficients of, e e and e c are the random deviations of energy consumption and carbon emissions respectively. a1 and a2 are the numbers of the electricity characteristic parameter and non-electricity characteristic parameter related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library respectively. b1 and b2 are the numbers of the electricity characteristic parameter and non-electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library respectively. q and t are lag periods.

3. A method for warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon according to claim 2, characterized in that S3 includes: S31. Initialize the population, and the individuals in the population represent the combination of characteristic parameters at each lag period; S32. Calculate the fitness value of each individual in the population, and judge whether the fitness of the individual meets the optimization goal. If it meets, output the best individual and its optimal solution; if not, enter S33; wherein, the fitness is the negative absolute value of the measurement error of the electric-energy-carbon coupling model; S33. Select parents according to the fitness value. The individuals with high fitness have a high probability of being selected, and the individuals with low fitness are eliminated. Cross and mutate in turn. After generating a new generation of population, return to S32 for the next iteration until the optimal solution is generated.

4. A method for warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon according to any one of claims 1-3, characterized in that S4 includes: S41. Obtain the data of the electric quantity characteristic parameters in period n + 1 in real time, and input them into the electric-energy-carbon coupling model to calculate the regional energy consumption and carbon emissions; S42. Calculate the energy consumption warning coefficient and carbon emission warning coefficient in period n + 1 according to the following formula: In the above formula, are respectively the energy consumption warning coefficient and carbon emission warning coefficient in period n + 1 for T t period, T tar is the total number of moments in period n + 1, E n+1,T , C n+1,T are respectively the energy consumption and carbon emission at moment T in period n + 1, E tar , C tar are respectively the energy consumption target and carbon emission target in period n + 1; S43. Conduct warnings on the regional energy consumption and carbon emissions according to the warning coefficient, specifically: If and Warn about the regional energy consumption and regional carbon emissions simultaneously; If and only issue a warning about the regional energy consumption volume; If and only give early warnings about regional carbon emissions.

5. A method for warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon according to any one of claims 1-3, characterized in that S1 includes: S11. Calculate the regional historical energy consumption and carbon emissions according to the following formula: In the above formula, E n and C n are the regional energy consumption and carbon emissions in period n, respectively. γ k and γ en are the standard coal conversion coefficients of the k-th energy source and electricity, respectively. G n,k and are the physical quantity of the k-th energy source consumed in region in period n and the energy consumption for raw materials, respectively. are the power generation of renewable energy and the imported electricity of the f-th type in region in period n, respectively. λ k and are the carbon emission factors of the k-th energy source and the imported electricity of the f-th type, respectively. K and F are the number of energy types consumed in the region and the number of types of imported electricity, respectively; S12. Screen the electric quantity characteristic parameters and non-electric quantity characteristic parameters with the strongest correlation with the regional energy consumption, and the electric quantity characteristic parameters and non-electric quantity characteristic parameters with the strongest correlation with the regional carbon emissions based on the Spearman correlation coefficient.

6. A system for warning of regional energy consumption and carbon emissions based on the coupling of electric energy and carbon, characterized in that The system includes an electricity - energy - carbon characteristic parameter library construction module, an electricity - energy - carbon preliminary coupling model construction module, a lag period optimization module, and an energy consumption and carbon emission warning module; The electricity - energy - carbon characteristic parameter library construction module is used to screen out the electricity characteristic parameters and non - electricity characteristic parameters with the strongest correlation with the regional energy consumption, and the electricity characteristic parameters and non - electricity characteristic parameters with the strongest correlation with the regional carbon emission, so as to form an electricity - energy - carbon characteristic parameter library; The electricity - energy - carbon preliminary coupling model construction module is used to construct an electricity - energy - carbon preliminary coupling model based on the data of each characteristic parameter in the electricity - energy - carbon characteristic parameter library by using an autoregressive distributed lag model; The lag period optimization module is used to optimize the lag period of the electricity - energy - carbon preliminary coupling model by using a genetic algorithm with the goal of minimizing the relative errors of the predicted regional energy consumption and carbon emission, so as to obtain an electricity - energy - carbon coupling model; The energy consumption and carbon emission warning module is used to input the data of the electricity characteristic parameters obtained in real - time into the electricity - energy - carbon coupling model, calculate the regional energy consumption and carbon emission, and conduct warnings on the regional energy consumption and carbon emission according to the calculation results.

7. The regional energy consumption and carbon emission warning system based on electricity - carbon coupling according to claim 6, characterized in that The electricity - energy - carbon preliminary coupling model is: In the above formula, E n+1 and C n+1 are the regional energy consumption and carbon emissions in period n + 1 respectively, is the value of the p-th electricity characteristic parameter related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library for period q, is the value of the p-th non-electricity characteristic parameter related to the energy consumption in the electricity-energy-carbon characteristic parameter library for period q, is the value of the s-th electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library for period t, is the value of the s-th non-electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library for period t, are respectively coefficients of, e e and e c are the random deviations of energy consumption and carbon emissions respectively. a1 and a2 are the numbers of the electricity characteristic parameter and non-electricity characteristic parameter related to the regional energy consumption in the electricity-energy-carbon characteristic parameter library respectively. b1 and b2 are the numbers of the electricity characteristic parameter and non-electricity characteristic parameter related to the regional carbon emissions in the electricity-energy-carbon characteristic parameter library respectively. q and t are lag periods.

8. The regional energy consumption and carbon emission warning system based on electricity - carbon coupling according to claim 7, characterized in that The lag period optimization module uses the following genetic algorithm to optimize the lag period of the electricity - energy - carbon preliminary coupling model: A1. Initialize the population, and the individuals in the population represent the combination of characteristic parameters under each lag period; A2. Calculate the fitness value of each individual in the population, and judge whether the fitness of the individual meets the optimization goal. If it meets, output the best individual and its optimal solution; if not, enter A3; where the fitness is the negative absolute value of the measurement error of the electricity - energy - carbon coupling model; A3. Select parents according to the fitness value. The individuals with high fitness have a high probability of being selected, and the individuals with low fitness are eliminated. Then, cross - over and mutation are carried out in turn to generate a new generation of population, and then return to A2 for the next iteration until the optimal solution is generated.

9. The regional energy consumption and carbon emission warning system based on electricity - carbon coupling according to any one of claims 6 - 8, characterized in that The energy consumption and carbon emission warning module includes an energy consumption and carbon emission real - time measurement unit, a warning coefficient calculation unit, and a warning unit; The energy consumption and carbon emission real - time measurement unit is used to obtain the data of the electricity characteristic parameters in period n + 1 in real - time and input them into the electricity - energy - carbon coupling model to calculate the regional energy consumption and carbon emission; The warning coefficient calculation unit is used to calculate the energy consumption warning coefficient and carbon emission warning coefficient in period n + 1 according to the following formula: In the above formula, are respectively the energy consumption warning coefficient and the carbon emission warning coefficient in period n + 1 for T t period, T tar is the total number of moments in period n + 1, E n+1,T , C n+1,T are respectively the energy consumption and carbon emission at moment T in period n + 1, E tar , C tar are respectively the energy consumption target and the carbon emission target in period n + 1; The warning unit is used to conduct warnings on the regional energy consumption and carbon emission according to the warning coefficient, specifically: If and At the same time, give early warnings for both regional energy consumption and regional carbon emissions; If and only give early warnings for regional energy consumption; If and only issue early warnings for regional carbon emissions.

10. A regional energy consumption and carbon emission early warning system based on the coupling of electric energy and carbon, according to any one of claims 6 - 8, characterized in that the electric - energy - carbon characteristic parameter library construction module includes a historical energy consumption and carbon emission accounting unit and a parameter screening unit; the historical energy consumption and carbon emission accounting unit is used to calculate the regional historical energy consumption and carbon emission according to the following formula: In the above formula, E n , C n are the regional energy consumption and carbon emissions in period n respectively, γ k , γ en are the standard coal conversion coefficients of the k-th energy and the standard coal conversion coefficient of electricity respectively, G n,k , are the physical quantity of the k-th energy consumed in the region in period n and the energy consumption for raw materials respectively, are the power generation of renewable energy in the region in period n and the f-th imported electricity respectively, λ k , are the carbon emission factors of the k-th energy and the f-th imported electricity respectively; K and F are the number of energy types consumed in the region and the number of types of imported electricity respectively; the parameter screening unit is used to screen out the electric quantity characteristic parameters and non - electric quantity characteristic parameters with the strongest correlation with the regional energy consumption and the electric quantity characteristic parameters and non - electric quantity characteristic parameters with the strongest correlation with the regional carbon emission based on the Spearman correlation coefficient.

Images