Large-scale power grid electricity-carbon coupling analysis optimization method and device and computing equipment

Abstract

The invention relates to the technical field of power grid data processing, in particular to a large-scale power grid electricity-carbon coupling analysis optimization method and device and computing equipment. The method comprises the following steps: acquiring power grid operation data, and extracting a steady-state and fluctuation component matrix of power and carbon emission; according to the steady-state and fluctuation component matrixes of the electric power and the carbon emission, obtaining an electric-carbon coupling steady-state and fluctuation incidence matrix; according to the electric-carbon coupling steady state and fluctuation incidence matrix, an electric-carbon coupling characteristic equation set is constructed and solved, and an electric-carbon coupling coefficient matrix is obtained; based on the electric carbon coupling coefficient matrix, establishing a multi-objective optimization model and determining an optimal solution set and an optimal weight coefficient; and according to the optimal weight coefficient, carrying out comprehensive evaluation on the optimal solution set to obtain a layered electricity-carbon coupling optimization strategy library. According to the method, the hierarchical power-carbon coupling optimization strategy is obtained by constructing the correlation moment of the power-carbon coupling steady state and the fluctuation, and the quantitative influence of the power consumption of the large-scale power grid on the carbon emission is accurately quantified.

Description

Technical Field

[0001] The present application relates to the technical field of power grid data processing, and in particular to a large-scale power grid electric-carbon coupling analysis and optimization method, device and computing equipment. Background Art

[0002] The current power system is facing the dual challenges of energy structure transformation and carbon emission constraints. Traditional fossil fuel power generation is dominant, making the power industry a major source of carbon emissions. In order to achieve green and low-carbon energy development, the power system needs to significantly increase the penetration rate of renewable energy, optimize the power structure, and improve the operating efficiency of the power grid. This requires in-depth analysis and comprehensive understanding of the inherent connection between the power system and carbon emissions.

[0003] Existing research mainly explores the relationship between power systems and carbon emissions from two perspectives. The first is to analyze the corresponding relationship between electricity consumption and carbon emissions from a static perspective, establish a carbon emission intensity model for different power generation methods and transmission and distribution paths, and provide carbon emission constraints for power planning and scheduling. However, this model often uses average parameters, ignoring the dynamic characteristics of the power grid operation, and it is difficult to accurately describe the instantaneous relationship between electricity consumption and carbon emissions. The second is to analyze the impact of fluctuations in power system operating parameters such as frequency and voltage on carbon emissions from a dynamic perspective. This type of research establishes a power grid model, considers factors such as unit response and load characteristics, and simulates changes in carbon emissions in the dynamic process of the power grid. However, this method often focuses on local power grids and is difficult to generalize to large-scale power grids.

[0004] Existing technologies have problems in analyzing the coupling relationship between electricity and carbon emissions, such as difficulty in fully describing the intrinsic relationship between each link of the power system and carbon emissions; inability to accurately quantify the quantitative impact of electricity consumption on carbon emissions; and lack of optimized dispatching strategies for large-scale power grids. These problems further restrict the progress of the power system's transformation to a green and low-carbon one.

[0005] Therefore, it has become an urgent task to develop a large-scale power grid electricity-carbon coupling analysis and optimization scheme, which comprehensively considers the relationship between electricity consumption and carbon emissions and proposes an effective optimization and regulation strategy. Summary of the invention

[0006] In view of this, it is necessary to provide a large-scale power grid electricity-carbon coupling analysis and optimization method, device and computing equipment to solve the problems that the existing technology is difficult to accurately describe the instantaneous relationship between electricity consumption and carbon emissions, and is unable to optimize the scheduling strategy for large-scale power grids.

[0007] A large-scale power grid electricity-carbon coupling analysis and optimization method, comprising:

[0008] Obtain grid operation data and extract the steady-state and fluctuating component matrices of electricity and carbon emissions;

[0009] According to the steady-state and fluctuating component matrices of electricity and carbon emissions, the steady-state and fluctuating correlation matrices of electricity-carbon coupling are obtained;

[0010] According to the steady-state and fluctuation correlation matrix of the electric-carbon coupling, a set of characteristic equations of the electric-carbon coupling is constructed and solved to obtain an electric-carbon coupling coefficient matrix;

[0011] Based on the electric-carbon coupling coefficient matrix, a multi-objective optimization model is established and an optimal solution set and an optimal weight coefficient are determined;

[0012] According to the optimal weight coefficients, the optimal solution set is comprehensively evaluated to obtain a layered electric-carbon coupling optimization strategy library.

[0013] Preferably, it also includes: selecting the optimal scheduling scheme from the layered electric-carbon coupling optimization strategy library through a rolling time domain optimization method, and dynamically optimizing and adjusting the optimal scheduling scheme through an online adaptive algorithm.

[0014] Preferably, the method further comprises: performing simulation verification on the scheduling scheme in the hierarchical electricity-carbon coupling optimization strategy library to evaluate the actual execution effect;

[0015] A dynamic update mechanism for the policy library is established to remove scheduling plans that do not meet the preset performance requirements based on the actual execution results.

[0016] Preferably, based on the electric-carbon coupling coefficient matrix, a multi-objective optimization model is established and an optimal solution set and an optimal weight coefficient are determined, including:

[0017] Establish a multi-objective optimization model and use an improved particle swarm optimization algorithm to determine the Pareto optimal solution set;

[0018] The optimal weight coefficient is determined based on fuzzy decision theory.

[0019] Preferably, the obtaining of grid operation data and extracting steady-state and fluctuating component matrices of electricity and carbon emissions includes:

[0020] Decomposing the power grid operation data according to an empirical mode decomposition method to obtain a plurality of intrinsic mode functions; performing spectrum analysis on the intrinsic mode functions to obtain a steady-state power component and a fluctuation power component;

[0021] Combining the power steady-state components to form a power steady-state component matrix, and combining the power fluctuation components to form a power fluctuation component matrix;

[0022] Using a wavelet decomposition method to perform multi-scale decomposition on the carbon emission data to obtain multiple scale coefficients and wavelet coefficients;

[0023] Reconstruct the steady-state component of carbon emissions based on the scale coefficients, and reconstruct the fluctuating component of carbon emissions based on the wavelet coefficients;

[0024] The carbon emission steady-state components are combined to form a carbon emission steady-state component matrix, and the carbon emission fluctuation components are combined to form a carbon emission fluctuation component matrix.

[0025] Preferably, according to the steady-state and fluctuating component matrices of electricity and carbon emissions, an electricity-carbon coupling steady-state and fluctuating correlation matrix is ​​obtained, including:

[0026] Constructing a steady-state power component characteristic vector, wherein the steady-state power component characteristic vector includes a steady-state voltage component, a steady-state current component, and a steady-state power component;

[0027] Constructing a carbon emission steady-state component characteristic vector, wherein the carbon emission steady-state component characteristic vector includes a carbon emission intensity steady-state component and a carbon emission total steady-state component;

[0028] A kernel canonical correlation analysis method is used to establish a mapping relationship between the characteristic vector of the steady-state component of the electric power and the characteristic vector of the steady-state component of the carbon emission, and an electric-carbon coupling steady-state correlation matrix is ​​obtained;

[0029] Constructing the characteristic vector of power fluctuation component and carbon emission fluctuation component;

[0030] A dynamic time warping algorithm is used to establish a dynamic mapping relationship between the characteristic vector of the power fluctuation component and the characteristic vector of the carbon emission fluctuation component, and an electricity-carbon coupling fluctuation correlation matrix is ​​obtained.

[0031] Preferably, according to the steady-state and fluctuation correlation matrix of the electric-carbon coupling, a set of electric-carbon coupling characteristic equations is constructed and solved to obtain an electric-carbon coupling coefficient matrix, including:

[0032] The Newton iteration method is used to solve the electric-carbon coupling characteristic equation group to obtain the quantitative relationship between electricity consumption and carbon emissions;

[0033] The quantitative relationship is expressed as an electron-carbon coupling coefficient matrix.

[0034] Preferably, it also includes: presetting an emergency response strategy, and incorporating the preset emergency response strategy into a layered electric-carbon coupling optimization strategy library.

[0035] A large-scale power grid electricity-carbon coupling analysis and optimization device, comprising:

[0036] The data acquisition module is used to obtain the grid operation data and extract the steady-state and fluctuating component matrices of electricity and carbon emissions;

[0037] A mapping relationship building module is used to obtain the steady-state and fluctuation correlation matrix of electricity-carbon coupling according to the steady-state and fluctuation component matrices of electricity and carbon emissions;

[0038] A characteristic equation group processing module is used to construct and solve the electric-carbon coupling characteristic equation group according to the electric-carbon coupling steady-state and fluctuation correlation matrix to obtain the electric-carbon coupling coefficient matrix;

[0039] A multi-objective optimization module, used to establish a multi-objective optimization model and determine an optimal solution set and an optimal weight coefficient based on the electric-carbon coupling coefficient matrix;

[0040] The strategy library construction module is used to comprehensively evaluate the optimal solution set according to the optimal weight coefficient to obtain a hierarchical electric-carbon coupling optimization strategy library.

[0041] A computing device comprising:

[0042] processor, and

[0043] A memory having program instructions stored thereon, wherein when the program instructions are executed by the processor, the processor executes the large-scale power grid electricity-carbon coupling analysis and optimization method as described above.

[0044] Compared with the existing technology, the beneficial effect of the present application is that the technical solution provided by the present application obtains a layered electric-carbon coupling optimization strategy by constructing the correlation moment of the steady state and fluctuation of electric-carbon coupling, accurately quantifies the quantitative impact of electricity consumption on carbon emissions for large-scale power grids, selects and adjusts the optimal scheduling plan, and greatly improves the scope of application, optimization accuracy and real-time response capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a flow chart of a large-scale power grid electricity-carbon coupling analysis and optimization method provided in an embodiment of the present application.

[0046] Figure 2 It is a detailed flow chart of the large-scale power grid electricity-carbon coupling analysis and optimization method provided in the embodiment of the present application.

[0047] Figure 3 It is a structural schematic diagram of a large-scale power grid electricity-carbon coupling analysis and optimization device 300 provided in an embodiment of the present application.

[0048] Figure 4 It is a structural schematic diagram of a computing device 400 provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0050] See also Figure 1 to Figure 2 , Figure 1 It is a flow chart of a large-scale power grid electricity-carbon coupling analysis and optimization method provided in an embodiment of the present application; Figure 2 Detailed flow chart of the large-scale power grid carbon coupling analysis and optimization method provided in the embodiment of the present application. The large-scale power grid carbon coupling analysis and optimization method comprises the following steps:

[0051] S101: Obtain grid operation data and extract steady-state and fluctuating component matrices of electricity and carbon emissions.

[0052] Obtaining grid operation data includes collecting power data and carbon emission data from multiple observation points. The power data includes grid load rate, voltage data, current data, and power data. The carbon emission data includes carbon emission intensity, total carbon emission, and the like.

[0053] Specifically, real-time data on grid load rate, phase angle, power quality, voltage, current and power data are collected. The grid load rate data includes peak load data, valley load data and average load data; the power data includes active power data and reactive power data. Carbon emission intensity data and total carbon emission data are collected. The carbon emission intensity data includes carbon emission data per unit of power generation and carbon emission data per unit of power transmission; the total carbon emission data includes total carbon emission per power generation and total carbon emission per power transmission.

[0054] The power data is subjected to multivariate mixed decomposition calculation to obtain a power steady-state component matrix and a power fluctuation component matrix, and the carbon emission data is subjected to multivariate mixed decomposition calculation to obtain a carbon emission steady-state component matrix and a carbon emission fluctuation component matrix. The power data can be decomposed by an empirical mode decomposition method (EMD) to obtain a plurality of intrinsic mode functions; the intrinsic mode functions are subjected to spectrum analysis to identify power steady-state components and power fluctuation components; the power steady-state components are combined to form a power steady-state component matrix, and the power fluctuation components are combined to form a power fluctuation component matrix; the carbon emission data is subjected to multi-scale decomposition by a wavelet decomposition method (WD) to obtain a plurality of scale coefficients and wavelet coefficients; the carbon emission steady-state components are reconstructed based on the scale coefficients, and the carbon emission fluctuation components are reconstructed based on the wavelet coefficients; the carbon emission steady-state components are combined to form a carbon emission steady-state component matrix, and the carbon emission fluctuation components are combined to form a carbon emission fluctuation component matrix.

[0055] S102: According to the steady-state and fluctuating component matrices of electricity and carbon emissions, an electricity-carbon coupling steady-state and fluctuating correlation matrix is ​​obtained.

[0056] A state mapping relationship between the steady-state power component matrix and the steady-state carbon emission component matrix is ​​established to obtain an electric-carbon coupling steady-state correlation matrix. A dynamic mapping relationship between the power fluctuation component matrix and the carbon emission fluctuation component matrix is ​​established to obtain an electric-carbon coupling fluctuation correlation matrix.

[0057] Specifically, a steady-state component characteristic vector of electricity is constructed, and the steady-state component characteristic vector of electricity includes a steady-state component of voltage, a steady-state component of current and a steady-state component of power; a steady-state component characteristic vector of carbon emission is constructed, and the steady-state component characteristic vector of carbon emission includes a steady-state component of carbon emission intensity and a steady-state component of total carbon emission; a kernel canonical correlation analysis method is used to establish a mapping relationship between the steady-state component characteristic vector of electricity and the steady-state component characteristic vector of carbon emission, and an electric-carbon coupling steady-state correlation matrix is ​​obtained; a characteristic vector of electricity fluctuation component and a characteristic vector of carbon emission fluctuation component are constructed; a dynamic time warping algorithm is used to establish a dynamic mapping relationship between the characteristic vector of electricity fluctuation component and the characteristic vector of carbon emission fluctuation component, and an electric-carbon coupling fluctuation correlation matrix is ​​obtained.

[0058] S103: According to the steady-state and fluctuation correlation matrix of the electric-carbon coupling, a set of characteristic equations of the electric-carbon coupling is constructed and solved to obtain an electric-carbon coupling coefficient matrix.

[0059] Based on the electricity-carbon coupling steady-state correlation matrix and the electricity-carbon coupling fluctuation correlation matrix, a group of electricity-carbon coupling characteristic equations are constructed, including the power balance equation, the carbon emission transfer equation, the electricity-carbon coupling intensity equation, and the power grid load distribution equation; the electricity-carbon coupling characteristic equation group is solved to obtain the electricity-carbon coupling coefficient matrix, which describes the quantitative relationship between electricity consumption and carbon emissions.

[0060] Specifically, an electric power balance equation is established based on the electric-carbon coupling steady-state correlation matrix, and the electric power balance equation describes the balance relationship between the power generation power, load power and network loss power of the power system; a carbon emission transfer equation is established based on the electric-carbon coupling fluctuation correlation matrix, and the carbon emission transfer equation describes the transfer relationship between carbon emissions in various links of the power system; an electric-carbon coupling intensity equation is established by combining the electric-carbon coupling steady-state correlation matrix and the electric-carbon coupling fluctuation correlation matrix, and the electric-carbon coupling intensity equation describes the coupling intensity between electricity consumption and carbon emissions; a power grid load distribution equation is established based on the power grid topology structure, and the power grid load distribution equation describes the distribution of power loads at each node of the power grid.

[0061] A matrix decomposition method is used to preprocess the electric-carbon coupling characteristic equation group to eliminate singularities in the equation group. Newton iteration method is used to solve the electric-carbon coupling characteristic equation group to obtain the quantitative relationship between electricity consumption and carbon emissions. The quantitative relationship is expressed as an electric-carbon coupling coefficient matrix, and each element of the electric-carbon coupling coefficient matrix represents the degree of influence of the change in electricity consumption of the corresponding node on carbon emissions.

[0062] S104: Based on the electric-carbon coupling coefficient matrix, a multi-objective optimization model is established and an optimal solution set and an optimal weight coefficient are determined.

[0063] Based on the electricity-carbon coupling coefficient matrix, a multi-objective optimization model is established. The objective function of the multi-objective optimization model includes the objectives of minimizing electricity consumption and minimizing carbon emissions. The constraints include electricity balance constraints, carbon emission constraints, and grid security constraints.

[0064] Specifically, the electricity consumption minimization objective function includes power generation cost, transmission loss and load regulation cost; establishes a carbon emission minimization objective function, which includes power generation carbon emissions, transmission carbon emissions and load regulation carbon emissions; establishes power balance constraints, including node power balance constraints and branch power flow constraints; establishes carbon emission constraints, including total carbon emission constraints and carbon emission intensity constraints; establishes power grid security constraints, including voltage constraints, frequency constraints and power constraints.

[0065] Furthermore, the multi-objective optimization model is solved by using an improved particle swarm optimization algorithm (IPSO), the Pareto Optimal Solution Set (POSS) of power consumption and carbon emissions is obtained through non-dominated sorting, and the optimal weight coefficient is determined based on fuzzy decision theory. Specifically, it includes:

[0066] Initialize the particle swarm parameters, including particle position, velocity and inertia weight; use the adaptive mutation operator to mutate the particle swarm to increase population diversity; calculate the fitness value of each particle, and perform non-dominated sorting based on the power consumption objective function and the carbon emission objective function; use the crowding calculation method to sort the non-dominated solutions of the same level, and maintain external archives; update the global optimal position and individual optimal position of the particle swarm; use the fuzzy decision method to calculate the weight coefficient of the objective function, and the weight coefficient reflects the decision maker's preference for different goals.

[0067] S105: Comprehensively evaluate the optimal solution set according to the optimal weight coefficient to obtain a layered electric-carbon coupling optimization strategy library.

[0068] Based on step S104, the Pareto optimal solution set is comprehensively evaluated according to the optimal weight coefficient, and a hierarchical electricity-carbon coupling optimization strategy library is established. The hierarchical electricity-carbon coupling optimization strategy library includes power dispatching schemes under different time scales and different load levels.

[0069] Specifically, a comprehensive evaluation is performed on each solution in the Pareto optimal solution set. The evaluation indicators include power consumption index, carbon emission index and system stability index. The evaluation results are weighted and summed according to the optimal weight coefficient to obtain a comprehensive score for each solution. The Pareto optimal solutions are classified according to time scale and load level, and a hierarchical electricity-carbon coupling optimization strategy library is constructed.

[0070] Furthermore, a rolling time domain optimization method is adopted to update the electric-carbon coupling coefficient matrix in real time, select the optimal scheduling scheme from the hierarchical electric-carbon coupling optimization strategy library, and dynamically optimize and adjust the optimal scheduling scheme through an online adaptive algorithm.

[0071] Specifically, a rolling time domain optimization method is adopted to establish a rolling time domain optimization model, and the rolling time domain optimization model includes a prediction time domain and a control time domain; at the beginning of each control cycle, the latest power data and carbon emission data are collected, and the electricity-carbon coupling coefficient matrix is ​​updated; according to the current system status and load forecast results, the optimal scheduling scheme is selected from the hierarchical electricity-carbon coupling optimization strategy library; an online adaptive algorithm is used to fine-tune the optimal scheduling scheme to adapt to real-time changes in the system operating status; the optimized scheduling scheme is executed, and the execution effect is recorded for dynamic updating of the strategy library.

[0072] Furthermore, the scheduling schemes in the layered electricity-carbon coupling optimization strategy library can be simulated and verified to evaluate their actual implementation effects; a dynamic update mechanism for the strategy library can be established to regularly remove scheduling schemes with poor performance; in the embodiment of the present application, poor performance refers to schemes whose carbon emissions exceed the baseline by more than 20% or whose electricity consumption efficiency is 15% lower than the average level.

[0073] Furthermore, emergency response strategies can be preset and incorporated into the layered electric-carbon coupling optimization strategy library to deal with potential emergency situations.

[0074] The following is a detailed introduction to the large-scale grid electricity-carbon coupling analysis and optimization method of this application:

[0075] By deploying multiple data acquisition terminals in the power grid system, power data, carbon emission data and power grid operation data are collected. First, data acquisition terminals are installed at m nodes at key locations in the power grid system, including power data acquisition modules, carbon emission data acquisition modules and power grid operation data acquisition modules. The power data acquisition module is responsible for recording the load rate data P of the power grid in real time.L = {P L1 ,P L2 ,…,P Lm}, including the peak load P Lmax , valley load P Lmin and load average Voltage data U={U1,U2,…,U m}, current data I={I1,I2,…,I m}, and active power P = {P1, P2, ..., P m} and reactive power Q={Q1,Q2,…,Q m}. The carbon emission data collection module collects carbon emission intensity data C I ={C I1 ,C I2 ,…,C Im}, including carbon emissions per unit of electricity generation C Ig and carbon emissions per unit of power transmission C It ; Total carbon emissions data C T ={C T1 ,C T2 ,…,C Tm}, including the total carbon emissions from power generation C Tg and the total carbon emissions from power transmission C Tt The power grid operation data acquisition module collects power grid frequency data f={f1,f2,…,f m}、Phase angle data θ={θ1,θ2,…,θ m} and power quality data D={D1,D2,…,D m In this way, multi-dimensional data of each key node of the power grid can be obtained.

[0076] Then, the EMD and wavelet decomposition methods are used to perform multivariate mixed decomposition on the power data and carbon emission data. For the power data, EMD decomposition obtains multiple intrinsic mode functions, and the power steady-state component matrix E is identified through spectrum analysis. s and the power fluctuation component matrix E f In this way, the original power data and carbon emission data are decomposed into two components: steady state and fluctuation. The above decomposition is conducive to studying the long-term relationship and short-term dynamic characteristics between power consumption and carbon emissions respectively.

[0077] Assume that the collected power data and carbon emissions data

[0078] Perform multivariate mixed decomposition calculation: Use the empirical mode decomposition (EMD) method to calculate the power data E t Decompose and obtain multiple intrinsic mode functions {IMF1,IMF2,…,IMFk}. Identify the steady-state components in power data through spectrum analysis and

[0079] Fluctuation Component

[0080] Satisfy E t =E s +E f +ε e For carbon emission data C t , then the wavelet decomposition method is used for multi-scale decomposition to obtain coefficients of different scales {A1, A2, …, A j ; D1, D2, …, D j}. Based on the scale factor A j Reconstructing the steady-state component of carbon emissions Based on the wavelet coefficient D j Reconstructing the Fluctuation Component of Carbon Emissions Satisfy C t =C s +C f +ε c In this way, the original electricity data and carbon emission data can be decomposed into two components: steady-state and fluctuating, laying the foundation for subsequent electricity-carbon coupling analysis.

[0081] Next, the state mapping relationship between the steady-state power component and the steady-state carbon emission component, as well as the dynamic mapping relationship between the power fluctuation component and the carbon emission fluctuation component are established. s , and the dynamic mapping relationship between the power fluctuation component and the carbon emission fluctuation component R f The state mapping relationship uses a nonlinear model containing quadratic terms and time derivative terms, reflecting the complex nonlinear relationship and dynamic change characteristics between power consumption and carbon emissions. The dynamic mapping relationship introduces cubic terms and time derivative terms to better capture the nonlinear characteristics of power grid operation under large disturbances. Through this multi-dimensional modeling method, the intrinsic connection between each link of the power system and carbon emissions can be fully characterized.

[0082] Specifically, first construct the steady-state power component E s The eigenvector of Including the voltage steady-state component U s , the steady-state component of current I s and the steady-state power component P s Similarly, the steady-state component C of carbon emissions is constructed s The eigenvector of Including the steady-state component of carbon emission intensity C Ig and the steady-state component C of total carbon emissions Tg Then, the kernel canonical correlation analysis (KCCA) method was used to establish thes and c s The mapping relationship between them:

[0083]

[0084] Among them, α, β, γ are unknown coefficients, δ is the mapping error term, Represents the time derivative of the steady-state component of power. For the power fluctuation component E f and carbon emission fluctuation component C f The dynamic mapping relationship between them is established using the dynamic time warping (DTW) algorithm:

[0085]

[0086] Among them, λ,μ,ν are unknown coefficients, and ω is the mapping error term. These correlation matrices R s and R f The intrinsic relationship between the steady-state and dynamic characteristics of the power system and carbon emissions is described.

[0087] Based on the above-mentioned electric-carbon coupling steady-state correlation matrix R s and the electric-carbon coupling fluctuation correlation matrix R f , construct the electric-carbon coupling characteristic equations. First, use R s Establish the power balance equation:

[0088]

[0089] Among them, P L is the load power, P i is the output power of the i-th power generation unit, P loss is the network loss power. Secondly, using R f Establish the carbon emission transfer equation:

[0090]

[0091] Where C is the total carbon emissions, k i is the carbon emission coefficient of the i-th power generation unit, η is the network loss carbon emission coefficient, and ξ is the transmission error term. Furthermore, combined with R s and R f The electric-carbon coupling strength equation is established:

[0092]

[0093] Among them, I ec is the intensity of the electric-carbon coupling, σ is the fluctuation coefficient, ω is the angular frequency, and ∈ is the coupling error term. Finally, the grid load distribution equation is established based on the topological structure of the grid to describe the distribution of the power load at each node of the grid. These characteristic equations constitute the mathematical model of the electric-carbon coupling system.

[0094] Solve the aforementioned electric-carbon coupling characteristic equations to obtain the quantitative relationship between electricity consumption and carbon emissions. First, the equations are preprocessed using matrix decomposition to eliminate singularities in the equations:

[0095] Ax = b;

[0096] Among them, A is the coefficient matrix, x is the unknown variable vector, and b is the constant term vector. Then, Newton iteration method is used to solve the equations:

[0097] x (k+1) =x (k) -[J(x (k) )] -1 f(x (k) );

[0098] Among them, x (k) is the solution vector of the kth iteration, J(x (k) ) is the Jacobian matrix, f(x (k) ) is the residual vector. Through iterative calculation, the quantitative impact of changes in electricity consumption on carbon emissions can be obtained, which is expressed as the electricity-carbon coupling coefficient matrix where r ecij Indicates the influence of the change in power consumption of the i-th node on the carbon emissions of the j-th node. In order to further extract the main influencing factors, R ec Perform singular value decomposition (SVD):

[0099] R ec =UΣV T ;

[0100] Among them, U, V are orthogonal matrices, and Σ is a diagonal matrix. Based on U, Σ, V, the state space model of the electric-carbon coupling system is established:

[0101]

[0102] In this way, a quantitative coupling relationship between electricity consumption and carbon emissions can be obtained.

[0103] Based on the electric-carbon coupling coefficient matrix R ec Establish a multi-objective optimization model: First, construct the objective function F1 for minimizing power consumption:

[0104]

[0105] Among them, α i , β is the weight coefficient. Secondly, construct the objective function F2 of minimizing carbon emissions:

[0106]

[0107] Among them, γ is the weight coefficient, C i is the carbon emission of the i-th power generation unit, C loss is the carbon emissions caused by network loss. For constraints, including power balance constraints:

[0108]

[0109] Carbon emission constraints:

[0110]

[0111] And grid security constraints:

[0112] |U i |≤U max ;

[0113] |f i -f0|≤f max ;

[0114] |S i |≤S max ;

[0115] Among them, C cap is the total carbon emission cap. is the upper limit of carbon emission intensity, U max ,f max ,S max are the upper limits of voltage, frequency and power respectively. The multi-objective optimization model constructed in this way can seek a balance between minimizing electricity consumption and minimizing carbon emissions.

[0116] Furthermore, the improved particle swarm optimization algorithm (IPSO) is used to solve the aforementioned multi-objective optimization model. First, the parameters of the particle swarm are initialized, including the particle positions speed And inertia weight θ. Then, the adaptive mutation operator is used to perform mutation operation on the particle swarm:

[0117]

[0118] Among them, ρ is the acceleration factor. Next, the fitness value of each particle is calculated, and non-dominated sorting is performed according to F1 and F2 to obtain the Pareto optimal solution set. While maintaining the external archive, the crowding calculation method is used to sort the non-dominated solutions of the same level. Finally, the fuzzy decision method is used to calculate the weight coefficients ω1, ω2 of the objective function to reflect the decision maker's preference for different goals:

[0119]

[0120] in, are the membership functions of F1 and F2 respectively. This can find the optimal trade-off between minimizing electricity consumption and minimizing carbon emissions.

[0121] Based on the aforementioned Pareto optimal solution set, a hierarchical electric-carbon coupling optimization strategy library is constructed: First, for each solution in the Pareto optimal solution set Conduct comprehensive evaluation, including power consumption indicators Carbon emission indicators and system stability indicators Then, the evaluation results are weighted and summed according to the optimal weight coefficients ω1, ω2 to obtain the comprehensive score J(x j )=ω1J1(x j )+ω2J2(x j )+ω3J3(x j ). Classify the Pareto optimal solutions according to the time scale T and load level L, and build a hierarchical electricity-carbon coupling optimization strategy library For the dispatching schemes in the strategy library, simulation verification is also required to evaluate their actual execution effect. Finally, a dynamic update mechanism for the strategy library is established to regularly remove dispatching schemes with poor performance, where poor performance refers to schemes with carbon emissions exceeding the baseline by more than 20% or power consumption efficiency below the average level by 15%. In this way, an electricity-carbon coupling optimization strategy library covering different time scales and load levels can be formed.

[0122] The electric-carbon coupling coefficient matrix R is updated in real time using the rolling time domain optimization method. ec , and from the Hierarchical Electro-Carbon Coupling Optimization Strategy Library Select the optimal scheduling solution x * , and then dynamically optimize and adjust through the online adaptive algorithm. First, establish a rolling time domain optimization model, including the prediction time domain T p and control time domain T c At the beginning of each control cycle k, the latest power data and carbon emission data are collected and the power-carbon coupling coefficient matrix R is updated. ec (k). Then, according to the current system state x(k) and load forecast results From the library of hierarchical electro-carbon coupling optimization strategies Select the optimal scheduling solution x * :

[0123]

[0124] Next, an online adaptive algorithm is used to * Make fine adjustments to adapt to real-time changes in system operating status:

[0125]

[0126] Among them, θ k is the parameter vector of the kth iteration, μ and ν are the step size factor and momentum factor respectively.

[0127] Finally, execute the optimized scheduling plan and record the execution effect for use in the strategy library This ensures that the electricity-carbon coupling optimization strategy can respond to system changes in real time and improve the applicability of the scheduling scheme.

[0128] The following is an introduction to the large-scale power grid electricity-carbon coupling analysis and optimization method of the present application in conjunction with a specific embodiment:

[0129] For example, a large power grid system consists of multiple power plants and load centers with a total installed capacity of 10 million kilowatts. The power grid company decided to carry out coupling analysis and optimization regulation between the power system and carbon emissions in order to achieve efficient allocation of power resources and significant reduction of carbon emissions. Based on this, the power grid company adopted the large-scale power grid electric-carbon coupling analysis and optimization method proposed in the present invention, and deployed 100 data acquisition terminals at the main nodes of the power grid system. The specific data of the power data matrix and the carbon emission data matrix are shown in Tables 1 and 2. Among them, the power data matrix contains information such as load rate, voltage, current, active power and reactive power of 100 nodes at 240 time points; the carbon emission data matrix records the carbon emission intensity and total carbon emission data of each node at 240 time points.

[0130] Table 1, Electricity data matrix E t :

[0131] time 1 2 … 100 1 0.82 0.75 … 0.91 2 0.84 0.77 … 0.89 … … … … … 240 0.88 0.81 … 0.93

[0132] Table 2, Carbon emission data matrix C t :

[0133] time 1 2 … 100 1 0.62 0.58 … 0.69 2 0.64 0.60 … 0.67 … … … … … 240 0.66 0.63 … 0.71

[0134] The EMD and wavelet decomposition methods are used to perform multivariate mixed decomposition of power data and carbon emission data. For power data, EMD decomposition obtains 10 intrinsic mode functions, and the power steady-state component matrix E is identified through spectrum analysis. s and the power fluctuation component matrix E f For carbon emission data, wavelet decomposition obtains 6 scale coefficients and 6 wavelet coefficients, and the carbon emission steady-state component matrix C is reconstructed based on the scale coefficients. s , reconstruct the carbon emission fluctuation component matrix C based on wavelet coefficients f This decomposes the original electricity data and carbon emission data into two components: steady state and fluctuation.

[0135] In order to establish the intrinsic connection between the power system and carbon emissions, the state mapping relationship between the steady-state power component and the steady-state carbon emission component, as well as the dynamic mapping relationship between the power fluctuation component and the carbon emission fluctuation component, are constructed. Specifically, the eigenvector of the steady-state power component is first extracted. Including the voltage steady-state component U s , the steady-state component of current I s and the steady-state power component P s ; Similarly, extract the characteristic vector of the steady-state component of carbon emissions Including the steady-state component of carbon emission intensity C Ig and the steady-state component C of total carbon emissions Tg Then, the kernel canonical correlation analysis (KCCA) method was used to establish the s and c s The mapping relationship between them is used to obtain the electric-carbon coupling steady-state correlation matrix R s For the power fluctuation component and the carbon emission fluctuation component, the dynamic time warping (DTW) algorithm is used to establish the dynamic mapping relationship between them, and the power-carbon coupling fluctuation correlation matrix R is obtained. f This constructs a model of the intrinsic connection between the power system and carbon emissions.

[0136] Based on R s and R f , and further constructed the electric-carbon coupling characteristic equations. First, using R s The power balance equation is established to describe the power balance relationship between generation, load and network loss:

[0137]

[0138] Among them, P L is the total load power, P i is the output of the i-th generator set, P loss is the network loss power.

[0139] Secondly, using R f The carbon emission transfer equation is established, taking into account the different effects of different power generation methods and grid losses on carbon emissions:

[0140]

[0141] Here, k i is the carbon emission coefficient of the ith generator set, and 0.1 is the grid loss carbon emission coefficient.

[0142] Furthermore, combined with R s and R f The electric-carbon coupling strength equation is established:

[0143]

[0144] Here, ω is the periodic angular frequency of the grid load.

[0145] By solving these electricity-carbon coupling characteristic equations, the quantitative relationship between electricity consumption changes and carbon emissions is obtained, namely, the electricity-carbon coupling coefficient matrix R ec Each element of the matrix represents the degree to which a change in power consumption at a certain node affects the carbon emissions of another node.

[0146] With the electric-carbon coupling coefficient matrix R ec , then a multi-objective optimization model for minimizing electricity consumption and carbon emissions was constructed. The objective function for minimizing electricity consumption is as follows:

[0147]

[0148] Among them, α i is the power generation cost coefficient of the i-th generator set. The objective function of carbon emission minimization is:

[0149]

[0150] C i is the carbon emission of the i-th generator set, C loss Carbon emissions caused by network losses.

[0151] Constraints include:

[0152] Power balance constraints:

[0153] Carbon emission constraints: C Ig ≤0.8;

[0154] Grid security constraints: |U i |≤1.05pu,|f_i-50Hz|0.2Hz,|S i |≤1.1pu;

[0155] The improved particle swarm optimization algorithm (IPSO) was used to solve the multi-objective optimization model, and the Pareto optimal solution set of power consumption and carbon emissions was obtained. Based on the non-dominated sorting, the fuzzy decision method was used to calculate the optimal weight coefficients of the objective function as ω1=0.6, ω2=0.4. Finally, the Pareto solution set was comprehensively evaluated and classified according to time scale and load level, and a hierarchical electricity-carbon coupling optimization strategy library was constructed.

[0156] In actual operation, the rolling time domain optimization method is used to dynamically adjust the scheduling strategy. At the beginning of each control cycle, the electric-carbon coupling coefficient matrix R is first updated. ec, to adapt to changes in system status. Then, according to the current load forecast, the optimal scheduling scheme is selected from the optimization strategy library. Next, the online adaptive algorithm is used to fine-tune and optimize the selected scheduling scheme. The specific update equation is:

[0157]

[0158] Among them, θ k is the parameter vector of the kth iteration, and J is the performance indicator function. Finally, the optimized scheduling scheme is executed and the actual effect is recorded for dynamic update of the strategy library.

[0159] It should be noted that the relevant variables in the embodiments are explained in detail as follows:

[0160]

[0161]

[0162] Based on the same inventive concept, the present application also provides a large-scale power grid electric-carbon coupling analysis and optimization device 300, such as Figure 3 A structural schematic diagram of a large-scale power grid electric-carbon coupling analysis and optimization device 300 is shown, and the large-scale power grid electric-carbon coupling analysis and optimization device 300 includes:

[0163] The data acquisition module 301 is used to obtain the grid operation data and extract the steady-state and fluctuating component matrices of power and carbon emissions;

[0164] A mapping relationship building module 302 is used to obtain an electricity-carbon coupling steady-state and fluctuation correlation matrix according to the steady-state and fluctuation component matrices of electricity and carbon emissions;

[0165] The characteristic equation group processing module 303 is used to construct and solve the electric-carbon coupling characteristic equation group according to the electric-carbon coupling steady-state and fluctuation correlation matrix to obtain the electric-carbon coupling coefficient matrix;

[0166] A multi-objective optimization module 304 is used to establish a multi-objective optimization model and determine an optimal solution set and an optimal weight coefficient based on the electric-carbon coupling coefficient matrix;

[0167] The strategy library construction module 305 is used to comprehensively evaluate the optimal solution set according to the optimal weight coefficient to obtain a layered electric-carbon coupling optimization strategy library.

[0168] When executing each module of the above-mentioned large-scale power grid electricity-carbon coupling analysis and optimization device 200, reference may be made to steps S101-S105 and any optional embodiments thereof, which will not be described in detail in this embodiment.

[0169] Based on the same inventive concept, the present application also provides a computing device 400, Figure 4 4 is a schematic structural diagram of a computing device 400 provided in an embodiment of the present application. The computing device can be used as a large-scale power grid carbon coupling analysis and optimization device to execute the optional embodiments of the large-scale power grid carbon coupling analysis and optimization method described above. The computing device can be a terminal or a chip or chip system inside the terminal. Figure 4 As shown, the computing device 400 includes: a processor 910 , a memory 920 , and a communication interface 930 .

[0170] It should be understood that Figure 4 The communication interface 930 in the computing device 400 shown may be used to communicate with other devices, and may specifically include one or more transceiver circuits or interface circuits.

[0171] The processor 910 may be connected to a memory 920. The memory 920 may be used to store the program code and data. Therefore, the memory 920 may be a storage unit inside the processor 910, or an external storage unit independent of the processor 910, or a component including a storage unit inside the processor 910 and an external storage unit independent of the processor 910.

[0172] Optionally, the computing device 900 may further include a bus. The memory 920 and the communication interface 930 may be connected to the processor 910 via the bus. The bus may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 A line without an arrow is used to represent the bus, but this does not mean that there is only one bus or one type of bus.

[0173] It should be understood that in the embodiment of the present application, the processor 910 may adopt a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. Alternatively, the processor 910 may adopt one or more integrated circuits to execute relevant programs to implement the technical solutions provided in the embodiment of the present application.

[0174] The memory 920 may include a read-only memory and a random access memory, and provides instructions and data to the processor 910. A portion of the processor 910 may also include a nonvolatile random access memory. For example, the processor 910 may also store information on the device type.

[0175] When the computing device 900 is running, the processor 910 executes the computer-executable instructions in the memory 920 to perform any operation step of the above method and any optional embodiment thereof.

[0176] It should be understood that the computing device 400 according to the embodiment of the present application can correspond to the corresponding subjects in the methods according to the embodiments of the present application, and the above-mentioned and other operations and / or functions of each module in the computing device 400 are respectively for implementing the corresponding processes of each method of the present embodiment, which will not be repeated here for the sake of brevity.

[0177] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0178] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

Claims

1. A large-scale power grid electricity-carbon coupling analysis and optimization method, characterized in that: include: Obtain grid operation data and extract the steady-state and fluctuating component matrices of electricity and carbon emissions; According to the steady-state and fluctuating component matrices of electricity and carbon emissions, the steady-state and fluctuating correlation matrices of electricity-carbon coupling are obtained; According to the steady-state and fluctuation correlation matrix of the electric-carbon coupling, a set of characteristic equations of the electric-carbon coupling is constructed and solved to obtain an electric-carbon coupling coefficient matrix; Based on the electric-carbon coupling coefficient matrix, a multi-objective optimization model is established and an optimal solution set and an optimal weight coefficient are determined; According to the optimal weight coefficients, the optimal solution set is comprehensively evaluated to obtain a layered electric-carbon coupling optimization strategy library.

2. The method according to claim 1, characterized in that Also includes: Through the rolling time domain optimization method, the optimal scheduling scheme is selected from the layered electric-carbon coupling optimization strategy library, and the optimal scheduling scheme is dynamically optimized and adjusted through an online adaptive algorithm.

3. The method according to claim 2, characterized in that Also includes: Conduct simulation verification on the scheduling scheme in the hierarchical electricity-carbon coupling optimization strategy library to evaluate the actual execution effect; A dynamic update mechanism for the policy library is established to remove scheduling plans that do not meet the preset performance requirements based on the actual execution results.

4. The method according to claim 1, characterized in that: The method of establishing a multi-objective optimization model based on the electric-carbon coupling coefficient matrix and determining an optimal solution set and an optimal weight coefficient comprises: Establish a multi-objective optimization model and use an improved particle swarm optimization algorithm to determine the Pareto optimal solution set; The optimal weight coefficient is determined based on fuzzy decision theory.

5. The method according to claim 1, characterized in that The obtaining of grid operation data and extracting the steady-state and fluctuating component matrices of electricity and carbon emissions include: Decomposing the power grid operation data according to an empirical mode decomposition method to obtain a plurality of intrinsic mode functions; performing spectrum analysis on the intrinsic mode functions to obtain a steady-state power component and a fluctuation power component; Combining the power steady-state components to form a power steady-state component matrix, and combining the power fluctuation components to form a power fluctuation component matrix; Using a wavelet decomposition method to perform multi-scale decomposition on the carbon emission data to obtain multiple scale coefficients and wavelet coefficients; Reconstruct the steady-state component of carbon emissions based on the scale coefficients, and reconstruct the fluctuating component of carbon emissions based on the wavelet coefficients; The carbon emission steady-state components are combined to form a carbon emission steady-state component matrix, and the carbon emission fluctuation components are combined to form a carbon emission fluctuation component matrix.

6. The method according to claim 1, characterized in that According to the steady-state and fluctuating component matrices of electricity and carbon emissions, the steady-state and fluctuating correlation matrices of electricity-carbon coupling are obtained, including: Constructing a steady-state power component characteristic vector, wherein the steady-state power component characteristic vector includes a steady-state voltage component, a steady-state current component, and a steady-state power component; Constructing a carbon emission steady-state component characteristic vector, wherein the carbon emission steady-state component characteristic vector includes a carbon emission intensity steady-state component and a carbon emission total steady-state component; A kernel canonical correlation analysis method is used to establish a mapping relationship between the characteristic vector of the steady-state component of the electric power and the characteristic vector of the steady-state component of the carbon emission, and an electric-carbon coupling steady-state correlation matrix is ​​obtained; Constructing the characteristic vector of power fluctuation component and carbon emission fluctuation component; A dynamic time warping algorithm is used to establish a dynamic mapping relationship between the characteristic vector of the power fluctuation component and the characteristic vector of the carbon emission fluctuation component, and an electricity-carbon coupling fluctuation correlation matrix is ​​obtained.

7. The method according to claim 1, characterized in that According to the steady-state and fluctuation correlation matrix of the electric-carbon coupling, the electric-carbon coupling characteristic equation group is constructed and solved to obtain the electric-carbon coupling coefficient matrix, including: The Newton iteration method is used to solve the electric-carbon coupling characteristic equation group to obtain the quantitative relationship between electricity consumption and carbon emissions; The quantitative relationship is expressed as an electron-carbon coupling coefficient matrix.

8. The method according to any one of claims 1 to 7, characterized in that: Also includes: Preset emergency response strategies and incorporate the preset emergency response strategies into the layered electric-carbon coupling optimization strategy library.

9. A large-scale power grid electricity-carbon coupling analysis and optimization device, characterized in that: include: The data acquisition module is used to obtain the grid operation data and extract the steady-state and fluctuating component matrices of electricity and carbon emissions; A mapping relationship building module is used to obtain an electricity-carbon coupling steady-state and fluctuation correlation matrix based on the steady-state and fluctuation component matrices of electricity and carbon emissions; A characteristic equation group processing module is used to construct and solve the electric-carbon coupling characteristic equation group according to the electric-carbon coupling steady-state and fluctuation correlation matrix to obtain the electric-carbon coupling coefficient matrix; A multi-objective optimization module, used to establish a multi-objective optimization model and determine an optimal solution set and an optimal weight coefficient based on the electric-carbon coupling coefficient matrix; The strategy library construction module is used to comprehensively evaluate the optimal solution set according to the optimal weight coefficient to obtain a hierarchical electric-carbon coupling optimization strategy library.

10. A computing device, characterized in that: include: processor, and A memory having program instructions stored thereon, wherein when the program instructions are executed by the processor, the processor executes the large-scale power grid electricity-carbon coupling analysis and optimization method as claimed in any one of claims 1 to 8.

Images