CN-121998471-A - Ground city carbon intensity index decomposition method and system based on target planning

CN121998471ACN 121998471 ACN121998471 ACN 121998471ACN-121998471-A

Abstract

The embodiment of the invention provides a ground city carbon strength index decomposition method and system based on target planning, and belongs to the field of low-carbon technology and information processing. The method comprises the steps of collecting multi-source data and preprocessing, clustering regional carbon emission modes based on PCA-K-means, including PCA dimension reduction, K-means clustering and optimal cluster number determination, calculating a carbon emission reduction potential index based on an entropy weight method, carrying out regression modeling on carbon emission influence factors based on LMDI, carrying out decomposition modeling on carbon intensity targets based on linear programming, and outputting results and dynamically adjusting. The method and the system realize scientific, fair and efficient decomposition of the provincial carbon strength target to the surface of the ground, and have low emission reduction cost and wide application prospect.

Inventors

ZHANG YU
BAI YANG
ZHANG CHENGPING
LU XUTING
ZOU ZHIQIANG
LI ZEFENG

Assignees

北京国网信通埃森哲信息技术有限公司
国网江西电力有限公司信息通信分公司

Dates

Publication Date: 20260508
Application Date: 20251212

Claims (10)

1. A ground city carbon intensity index decomposition method based on target planning, the method comprising: Collecting multi-source data and preprocessing; Performing regional carbon emission pattern clustering based on PCA-K-means, including PCA dimension reduction, K-means clustering and optimal cluster number determination; Calculating a carbon emission reduction potential index based on an entropy weight method; performing regression modeling on the carbon emission influence factor based on LMDI; performing carbon strength target decomposition modeling based on linear programming; Outputting the result and dynamically adjusting.
2. The method for decomposing the carbon intensity index of the ground city based on the target planning of claim 1, wherein the collected multi-source data comprises economic data, energy data, population data and carbon emission data of provincial and various ground cities, and the collected data is subjected to extremely poor standardization processing according to a formula (1) to eliminate dimension differences, Wherein, the Is the first First city of the individual The original value of the term's index, Respectively the first The minimum and maximum values of the term index, The normalized value is the value range [0,1].
3. The method for decomposing the ground city carbon intensity index based on the target planning according to claim 2, wherein the PCA dimension reduction comprises constructing a covariance matrix for a preprocessed index data set based on standardized data, performing feature decomposition on the covariance matrix, sorting feature values from large to small, and extracting a front part with a cumulative contribution rate of more than or equal to 85 percent Principal components, build new feature space , wherein, Is the number of ground markets; The K-means clustering includes clustering the market samples in the principal component space with "minimize the squared Euclidean distance of the sample points to the cluster center" as an objective function according to equation (2), Wherein, the Is the first The number of clusters is one, Is the mean of the principal component vectors of the samples within the cluster, Is the first Principal component vectors of the individual samples; the optimal cluster number determination comprises the steps of determining the optimal cluster number through a contour coefficient and cluster stability test The required profile coefficient is more than or equal to 0.6, the stability is more than or equal to 90 percent, and finally the market is divided into Classes have similar carbon emission patterns and state-of-the-art regions.
4. The method for decomposing the carbon intensity index of the ground city based on the target planning of claim 3, wherein the calculating the carbon emission reduction potential index based on the entropy weight method comprises respectively constructing a fair-efficiency two-dimensional evaluation system for each type of region obtained by clustering, objectively weighting by adopting the entropy weight method, calculating the comprehensive carbon emission reduction potential index, Fairness dimension Including average person GDP, average person electricity consumption, historical carbon emission liability coefficient and resident population scale; Efficiency dimension Including a unit GDP carbon emission, a unit energy consumption carbon emission, a second industry ratio, and a renewable energy ratio; The entropy weight calculation includes: calculate the first according to equation (3) First city of the individual The duty cycle of the term index is determined, Wherein, the Is a normalized value; calculate the first according to equation (4) The information entropy of the item index, Wherein if it Definition of Information entropy The value range is 0,1, Smaller indicates higher index discrimination; The index weight is calculated according to equation (5), Wherein, the As the total number of the indexes, ; The comprehensive potential index is calculated according to the formula (6), Wherein, the Is the first Carbon emission reduction potential index of the individual market, , For the fairness weight score to be weighted, The scores are weighted for efficiency.
5. The method for target planning based ground city carbon strength index decomposition of claim 4, wherein said LMDI-based carbon emission impact factor regression modeling comprises LMDI additive decomposition and linear regression modeling, wherein, The LMDI addition, decomposition and selection energy consumption structure Efficiency of energy use Industrial structure Level of economic development As a core driving factor, the carbon emission variation is calculated according to the formula (7) Lossless decomposition into factor contributions: Wherein, the Is that Carbon emission amount at the moment, and, according to the formula (8), each factor contribution amount is calculated, Wherein, the As a result of the certain driving factor, Is a logarithmic mean function; The linear regression modeling was performed with carbon emission variation As a dependent variable Each driving factor contribution is an independent variable vector A linear regression model is established according to the formula (9), Wherein, the As the coefficient of regression of the coefficient of the data, As an error term, By minimizing the sum of squares of the residuals Estimating regression coefficients using determined coefficients Verifying model fitness and 。
6. The target programming-based ground city carbon intensity index decomposition method of claim 5, wherein said linear programming-based carbon intensity target decomposition modeling comprises: Set the first The carbon intensity change rate of the individual markets is a decision variable , For the ratio of the target annual carbon intensity to the reference annual carbon intensity, Indicating that the strength of the carbon was reduced, The smaller the drop amplitude is, the larger the drop amplitude is; Constructing an objective function, taking 'high potential area priority emission reduction' as a principle, constructing a weighted minimization objective function according to a formula (10) to realize 'fairness-efficiency' balance, Wherein, the For the rate of change of the resident population, Is the rate of change of the GDP per unit population, Is the rate of change of energy consumption per unit GDP, The change rate of carbon emission is the unit energy consumption; Is the first Carbon emission reduction potential index of the individual market, For the corresponding variable weights, according to equation (11), by Min-Max normalization of the regression coefficients, A constraint condition is set, wherein, Carrying out total consistency constraint according to a formula (12) to ensure that the sum of the carbon intensity targets of each city meets the provincial level target, Wherein, the Is the first The reference annual carbon emission of the individual city, For the total carbon emission of the provincial reference year, The target change rate of the strength of the provincial carbon is; economic growth constraint is performed according to formula (13) such that the rate of GDP growth per unit population is not lower than the expected annual rate of GDP growth at provincial level Ensures that the economic development is not excessively restricted, Performing resource constraint according to formula (14) to ensure that the unit GDP energy consumption and the unit energy consumption carbon emission are not lower than 90% of the minimum value of the whole province, avoiding the goal exceeding the technical feasibility, Population constraint is carried out according to the formula (15), the population change trend of the city is combined, the resident population change rate interval is set, Non-negative constraint is carried out according to the formula (16) so that the change rate of the decision variable and the influence factor accords with the practical meaning, And adopting Python open source library PuLP to implement solution, setting iteration termination condition as residual error less than 1e-6, retaining four-bit decimal for objective function calculation accuracy, if the number of local markets exceeds 50, adopting interior point method to raise calculation efficiency, setting initial iteration step length to 0.1, and finally outputting optimum carbon intensity change rate of all local markets 。
7. The method of target planning-based ground carbon strength index decomposition according to claim 6, wherein said outputting results includes calculating ground carbon strength decreasing targets Presented in a bar graph and thermodynamic diagram form, and generates an analysis report containing potential index-descending target-constraint satisfaction; The dynamic adjustment comprises the steps of establishing a timing update mechanism, periodically repeating the steps based on the latest economic, energy and carbon emission data, updating the decomposition target, and triggering instant update without waiting for a period if major changes occur.
8. The target planning-based ground city carbon strength index decomposition method of claim 7, wherein said significant change comprises an industrial structure change rate exceeding 20% or a renewable energy installation doubling.
9. A ground city carbon strength index decomposition system based on target planning for implementing the method of any one of claims 1-8, the system comprising: the data acquisition and preprocessing module is used for acquiring and preprocessing multi-source data to form a standardized data set; the regional heterogeneity recognition module is used for carrying out principal component analysis and dimension reduction and K-means clustering on the standardized data set and recognizing the carbon emission pattern category of the ground city; the carbon emission reduction potential quantification module is used for constructing an evaluation system of fairness dimension and efficiency dimension and calculating carbon emission reduction potential indexes of various places and cities through an entropy weight method; the driving factor modeling module is used for executing logarithmic average Diels index addition decomposition and linear regression and quantifying the influence of the carbon emission driving factor; The multi-constraint target planning module is used for constructing and solving a linear planning model taking the carbon intensity change rate as a decision variable and outputting carbon intensity decomposition targets of various places and cities; and the result output and dynamic management module is used for visually displaying the decomposition result and managing the dynamic updating and recalculation of the model.
10. A machine-readable storage medium having stored thereon instructions for causing a machine to perform the method of any of claims 1-8.

Description

Ground city carbon intensity index decomposition method and system based on target planning Technical Field The invention relates to the technical field of low-carbon technology and information processing, in particular to a ground and city carbon strength index decomposition method and system based on target planning. Background The control of the emission of greenhouse gases and the climate change become global consensus, and the control of the intensity of carbon emission is used as a core to promote the establishment of a national, provincial and municipal three-level carbon emission budget management system. Therefore, how to scientifically, reasonably and fairly decompose the control target of the provincial carbon emission intensity into various local markets is a key technical foundation for realizing top-down fine management and control and ensuring provincial and even national target achievement. At present, the carbon emission index decomposition field has a certain technical exploration, but mainly focuses on macroscopic distribution from country to provincial level, and in the aspect of refined decomposition facing provincial and internal municipal level, the prior art scheme has obvious defects, is difficult to adapt to actual operation requirements, and particularly shows that: 1. The decomposition level is not well-suited to the object: The existing research method and practice focus on index decomposition from international or national to province, and lack a differential decomposition scheme specific to the level of each local city in the province. The market in each province has obvious spatial heterogeneity in the aspects of economic development stage, industrial structure, energy consumption structure, resource endowment, technical capability and the like. The use of "one-shot" or simple average decomposition methods often results in target and actual disjointing. 2. The key parameter determination is strong in subjectivity: when the decomposition model is constructed, setting key weight parameters (such as emission reduction responsibility, capacity, potential and other weights of each market) influencing the distribution result, the prior art relies on expert experience judgment or simple historical data averaging, and a quantitative calibration mechanism based on objective data driving is lacking. The subjectivity or rough setting mode makes fairness and scientificity of the decomposition result questioned, and disputes among areas are easily caused, so that the cooperative implementation of targets is not facilitated. 3. Model constraint dimension is single, and multiple targets are difficult to balance: The traditional decomposition method only considers single or few factors such as historical emission or total economic quantity, and cannot systematically incorporate multi-dimensional realistic constraints such as economic development rigidity requirements, resource environment bearing upper limit, technical emission reduction feasibility, industry policy guidance and the like. This results in decomposition schemes that are prone to "averaging" or disjoint from the local actual development capability, failing to achieve an effective balance between the two core principles of fairness and efficiency, thereby affecting the ultimate accessibility of the target. In the prior art and technical literature, although researchers try to improve, for example, a carbon strength decomposition method considering fairness principle is proposed, a model of the carbon strength decomposition method does not perform systematic heterogeneous cluster analysis on various places in provinces, and a differential distribution basis of different characteristic cities cannot be accurately identified. Some attempts are made to optimize the regional emission reduction distribution efficiency by using a data envelope analysis DEA method, but the research of the regional emission reduction distribution efficiency does not introduce a driving factor decomposition method such as a logarithmic average Di index LMDI, so that the actual contribution degree of various factors of various cities in the historical stage to emission change is objectively quantized, and the potential evaluation is not accurate enough. In summary, in the prior art, short plates with insufficient heterogeneity recognition, inexact potential evaluation and weak multi-constraint optimization generally exist, and a complete technical process from ground city difference recognition to emission reduction potential quantification and then to dynamic optimization decomposition under multi-objective multi-constraint is not formed yet. Disclosure of Invention The embodiment of the invention aims to provide a method and a system for decomposing a ground and city carbon intensity index based on target planning, which realize scientific, fair and efficient decomposition of a provincial carbon intensity target to a ground and city level, and have low e