Power grid carbon emission evaluation and optimization method based on fuzzy dynamic clustering

Abstract

The invention discloses a power grid carbon emission evaluation and optimization method based on fuzzy dynamic clustering, and solves the problems that in the prior art, power grid optimization evaluation lacks optimization consideration on CO2 emission, and power grid carbon emission calculation is inaccurate, and the method comprises the following steps: S1, establishing a carbon emission consumption calculation model considering carbon emission of each link of a power system, the influence of power grid damage on carbon emission is considered preferentially; s2, constructing a collaborative model based on fuzzy dynamic clustering between enterprises; s3, introducing a power grid loss carbon emission rate total amount index, and establishing a power grid carbon emission optimization model based on a quantum genetic algorithm; s4, simulation verification is carried out, and the feasibility of the power grid carbon emission optimization model is evaluated; and S5, optimizing the power grid carbon emission by using the power grid carbon emission optimization model. Compared with other reinforcement learning methods, the method has higher convergence, and is suitable for real-time rolling optimization of the large-scale power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission assessment of power systems, and in particular to a method for assessing and optimizing carbon emission of power grids based on fuzzy dynamic clustering. Background Art

[0002] As global climate change becomes increasingly serious, emission reduction efforts are being promoted around the world. As a producer and user, the power sector suffers from large power losses in power generation, transmission, distribution, conversion, and user end. It is a very important issue for each link of the power system to explore energy-saving potential, reduce system losses, and improve system economic benefits. Using the idea of ​​energy conservation and emission reduction to evaluate and optimize the power grid is a complex optimization problem with nonlinear characteristics.

[0003] The existing energy consumption evaluation and indicators such as power purchase volume and network loss are based on the total pollutant emissions or pollutant emission costs. In the optimization and evaluation of the power system, the economic benefit indicators such as reducing coal consumption, plant power consumption rate, and network loss are used as the basis, lacking the optimization consideration of CO2 emissions, and energy conservation and emission reduction cannot be carried out from the perspective of environmental protection. The existing carbon emission rights evaluation methods in the power industry are mostly based on the power generation end. If the carbon emissions at the power generation end are used as the total carbon emissions at the power end, it will lead to an excess of "carbon emission rights" for coal-fired power plants, which is not conducive to the construction of energy infrastructure and the smooth implementation of energy conservation and emission reduction. On the power grid side, since power grid companies undertake the task of power transmission, it is necessary to take the carbon emissions corresponding to the losses generated during the power transmission process as the emission reduction target. At present, the commonly used method for power grid carbon emission evaluation is mainly the "carbon emission conversion coefficient method", which assumes that the carbon dioxide emission coefficient required for a certain energy is a constant, ignoring the transmission link, and can only estimate the carbon emissions in the power grid, and cannot accurately evaluate large-scale power grids. Summary of the invention

[0004] The purpose of the present invention is to overcome the problems in the prior art of lack of optimization consideration of CO2 emissions in power grid optimization evaluation and inaccurate calculation of power grid carbon emissions. A power grid carbon emissions evaluation and optimization method based on fuzzy dynamic clustering is provided, which can scientifically evaluate and optimize the carbon emissions of the power grid and improve the accuracy of the calculation of power grid carbon emissions. Compared with other reinforcement learning methods, it has higher convergence and is suitable for real-time rolling optimization of large-scale power grids.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: a method for evaluating and optimizing carbon emissions of power grid based on fuzzy dynamic clustering, comprising the following steps:

[0006] S1: Establish a carbon emission consumption calculation model that takes into account carbon emissions in all links of the power system, giving priority to the impact of power grid damage on carbon emissions;

[0007] S2: Construct a collaborative model among enterprises based on fuzzy dynamic clustering;

[0008] S3: Introduce the total carbon emission rate index of power grid loss and establish a power grid carbon emission optimization model based on quantum genetic algorithm;

[0009] S4: Perform simulation verification to evaluate the feasibility of the grid carbon emission optimization model;

[0010] S5: Optimizing the carbon emissions of the power grid using the power grid carbon emissions optimization model.

[0011] A carbon emission consumption calculation model that takes into account all aspects of the power grid’s carbon emissions is established, and the impact of power grid damage on carbon emissions is also considered to improve the accuracy of carbon emission assessments. The total carbon emission rate indicator of power grid losses is introduced, and combined with a quantum genetic algorithm, a power grid carbon emission optimization model is established. Dynamic clustering and shared fitness are used with fast convergence speed and good global search performance to achieve the optimal configuration of carbon emissions at the power generation end, reduce the total carbon emissions of the power industry, achieve energy conservation and emission reduction, and is suitable for real-time rolling optimization of large-scale power grids.

[0012] As a preference, the grid carbon emission optimization model adopts the linear superposition value g of the grid active power loss, the total grid loss carbon emission rate and the voltage stability component. 1 The minimum value of is taken as the objective function.

[0013] That is, the objective function of the grid carbon emission optimization model is the product of the total component of the network loss carbon emission rate and its weight coefficient, plus the product of the active network loss component of the grid and its weight coefficient, plus the product of the voltage stability component and its weight coefficient, and the sum of the weight coefficients of the three is 1.

[0014] Preferably, when establishing a carbon emission consumption calculation model, the network loss and the network loss that each user should bear are tracked according to the principle of proportional distribution, and the network loss borne by each load node is determined.

[0015] Line losses are borne by load nodes, and the carbon emissions of each load node are the product of the network loss share borne by node i and the unit's carbon emission intensity.

[0016] Preferably, the quantum genetic algorithm comprises:

[0017] A1: Initialize the population and randomly generate n chromosomes encoded with quantum bits;

[0018] A2: Measure each individual in the initial population, obtain the corresponding definite solution, evaluate the fitness of each definite solution, and record the optimal individual and the corresponding fitness;

[0019] A3: Determine whether the end condition is met. If so, output the optimal solution. If not, continue with the subsequent steps.

[0020] A4: Measure each individual in the population, obtain a definite solution, and evaluate the fitness of the definite solution;

[0021] A5: Use quantum revolving gate to adjust individuals, generate new populations, and record the best individuals and corresponding fitness;

[0022] A6: Increase the number of iterations by 1 and return to step A3.

[0023] The global search improved by quantum genetic algorithm has good performance and fast convergence speed. It can realize scientific assessment and rapid optimization of carbon emissions of power grids, and provide support for carbon verification and carbon neutrality of power grid companies.

[0024] Preferably, in step S4, the grid carbon emission optimization model is simulated and verified using an IEEE57 node system.

[0025] The node system consists of 7 units, 80 lines, 42 load points, 15 adjustable transformers and 3 parallel capacitors.

[0026] Preferably, the fitness function for evaluating the fitness of a determined solution is g′=-g, where ming=g 1 +N, N indicates the number of state variables that exceeds the limit, g indicates the update value, g 1 It represents the linear superposition value of the active power loss of the power grid, the total carbon emission rate of the power grid loss and the voltage stability component.

[0027] The fitness function reflects the global optimal solution. The larger the target value of an individual is, the larger its fitness value should be, so the optimized target function is directly used as the fitness function.

[0028] Preferably, the quantum genetic algorithm further comprises:

[0029] Dynamically adjust the rotation angle of the quantum gate according to the evolution process;

[0030] Temporarily exchange the best solution and the best fitness value between individuals to achieve quantum crossover;

[0031] Temporarily exchange the values ​​of quantum bit probability amplitudes between individuals to achieve quantum mutation;

[0032] When the quantum genetic algorithm has been stable for several generations and the optimal individuals remain stable, disturbances are applied to some individuals in the population, and some individuals are randomly regenerated to achieve quantum catastrophe.

[0033] The quantum mutation operation combines single-point mutation and multi-point mutation to enhance the diversity of genes in the population. The quantum catastrophe operation can make the quantum genetic algorithm get rid of the local optimal solution.

[0034] Preferably, the quantum revolving gate adopts a universal quantum revolving gate:

[0035] V λ is the quantum of the revolving door, and α is the angle of rotation. The quantum revolving door realizes individual adjustment.

[0036] Preferably, the constraints of the grid carbon emission optimization model include reactive power constraints of generator sets and active power constraints of balancing nodes, and also include voltage constraints of load nodes.

[0037] Therefore, the present invention has the following beneficial effects:

[0038] 1. Establish a carbon emission consumption calculation model that takes into account all aspects of grid carbon emission, taking into account the impact of grid damage on carbon emissions, and improving the accuracy of carbon emission assessment;

[0039] 2. Improve the traditional carbon emission evaluation model, introduce the total carbon emission rate indicator of power grid loss, establish a carbon emission optimization model, achieve the optimal configuration of carbon emissions at the power generation end, reduce the total carbon emissions of the power industry, and achieve energy conservation and emission reduction;

[0040] 3. Establish a power grid carbon emission optimization model based on quantum genetic algorithm, which uses dynamic clustering and shared fitness, has fast convergence speed and good global search performance, and is suitable for real-time rolling optimization of large-scale power grids. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 The figure is a flow chart of the steps of the method of the present invention.

[0042] Figure 2 This is a schematic diagram of an example of carbon emissions assessment.

[0043] Figure 3 Schematic diagram of the calculation process of the power grid carbon emission optimization model based on quantum genetic algorithm.

[0044] Figure 4 This is the convergence curve diagram of carbon emission prediction based on quantum genetic algorithm.

[0045] Figure 5 Schematic diagram of the improved quantum genetic algorithm flow chart. DETAILED DESCRIPTION

[0046] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0047] Embodiment 1:

[0048] like Figure 1 In the illustrated embodiment, a method for evaluating and optimizing carbon emissions of power grids based on fuzzy dynamic clustering can be seen, and its operating process is as follows: Step 1, establishing a carbon emission consumption calculation model that takes into account carbon emissions in all links of the power system, giving priority to the impact of power grid damage on carbon emissions; Step 2, constructing a collaborative model between enterprises based on fuzzy dynamic clustering; Step 3, introducing a total carbon emission rate indicator for power grid losses, and establishing a power grid carbon emission optimization model based on a quantum genetic algorithm; Step 4, conducting simulation verification to evaluate the feasibility of the power grid carbon emission optimization model; Step 5, optimizing the power grid carbon emissions using the power grid carbon emission optimization model.

[0049] A carbon emission consumption calculation model that takes into account all aspects of the power grid’s carbon emissions is established, and the impact of power grid damage on carbon emissions is also considered to improve the accuracy of carbon emission assessments. The total carbon emission rate indicator of power grid losses is introduced, and combined with a quantum genetic algorithm, a power grid carbon emission optimization model is established. Dynamic clustering and shared fitness are used with fast convergence speed and good global search performance to achieve the optimal configuration of carbon emissions at the power generation end, reduce the total carbon emissions of the power industry, achieve energy conservation and emission reduction, and is suitable for real-time rolling optimization of large-scale power grids.

[0050] The technical solution of the present application is further illustrated below through specific examples.

[0051] Step 1: Establish a carbon emission consumption calculation model that takes into account carbon emissions in all links of the power system, giving priority to the impact of power grid damage on carbon emissions.

[0052] like Figure 2 As shown, carbon emissions from various links of the power system include residential carbon emissions, tertiary industry carbon emissions and industrial carbon emissions. Residential carbon emissions, tertiary industry carbon emissions and industrial carbon emissions are all calculated through energy consumption and carbon emission factors.

[0053] At the same time, the GIS weighted point density of regions, energy intensity, residential areas and industrial focus points is combined with land use data to map CEs density. The obtained map is combined with residential carbon emissions, tertiary industry carbon emissions and industrial carbon emissions to obtain residential CEs distribution, tertiary industry CEs distribution and industrial CEs distribution. The three are then overlaid and analyzed in time and space to accurately assess carbon emissions.

[0054] Specifically, when establishing a carbon emission consumption calculation model that takes into account carbon emissions in all links of the power system, the network losses and the network losses that each user should bear are tracked according to the principle of proportional distribution, and the grid losses borne by each load node are determined.

[0055] (1) Track network losses and the network losses that each user should bear.

[0056] Using the formula:

[0057] M s =FM r

[0058] M dik =F(i,t)×M r (t)

[0059] Among them, M s The line loss is borne by the load node, M s ∈R n ;M r is the output of the generator node, M r ∈R n ; F is the network loss borne by the load in the proportional allocation matrix from the corresponding generator, M dik is the share of network loss borne by node i.

[0060] (2) Calculate the carbon emissions of each load node.

[0061] Using the formula:

[0062]

[0063] Among them, ξ gt Indicates the carbon emission intensity of a unit, D sr Represents the power grid.

[0064] Step 2: Construct a collaborative model among enterprises based on fuzzy dynamic clustering.

[0065] By calculating the fuzzy similarity between enterprises, the internal relationship between them can be reflected without distortion, which is convenient for analyzing corporate carbon emission data. At the same time, the fuzzy similarity between enterprises is mapped to the Euclidean distance between enterprises, that is, the high-dimensional enterprise data is mapped to a two-dimensional plane. The genetic algorithm is used to continuously optimize the mapping between the two, so that the Euclidean distance between samples gradually approaches their fuzzy similarity, and dynamic fuzzy clustering is realized. The dependence of clustering effectiveness on sample distribution is overcome; at the same time, the flexibility and visualization of clustering are increased, with better clustering effect and faster convergence speed.

[0066] Step 3: Introduce the total carbon emission rate indicator of power grid losses and establish a power grid carbon emission optimization model based on quantum genetic algorithm.

[0067] In this embodiment, the grid carbon emission optimization model uses the linear superposition value g of the grid active power loss, the total grid loss carbon emission rate and the voltage stability component. 1 The minimum value of is taken as the objective function. Specifically, it is:

[0068]

[0069] Where M lc (x) is the total component of the network loss carbon emission rate, M lc (x) is the active network loss component of the power grid, V is the voltage stability component, and They are respectively lc (x) and M sr (x) The corresponding weight coefficient, x={V G , V L , M ref ,c,W c , W G} refers to the generator terminal voltage value V G , load node voltage V L , Balance node active power output M ref , load transformer open circuit voltage regulator ratio c, reactive power compensation capacity W c and the reactive power output of the generator W G .

[0070] The constraints take into account the reactive power of the generator set, the active power of the balancing node, and the voltage of the load node.

[0071] In order to improve the traditional genetic algorithm, this embodiment adopts a quantum genetic algorithm. The quantum genetic algorithm is a genetic algorithm based on the principle of quantum computing. The quantum state vector expression is introduced into the genetic coding, and the quantum logic gate is used to realize the evolution of chromosomes, which achieves better results than the conventional genetic algorithm.

[0072] The quantum genetic algorithm is based on the quantum state vector representation. It applies the probability amplitude representation of quantum bits to the encoding of chromosomes, so that a chromosome can express the superposition of multiple states, and uses quantum logic gates to implement chromosome update operations, thereby achieving the optimal solution of the target.

[0073] The calculation process of the power grid carbon emission optimization model based on quantum genetic algorithm is as follows: Figure 3 As shown:

[0074] (1) Initialize the population P(t0) and randomly generate n chromosomes encoded with quantum bits.

[0075] In quantum genetic algorithms, quantum bits are used to store and express a gene. The gene can be in a "0" state or a "1" state, or any superposition of them. That is, the gene no longer expresses a certain information, but contains all possible information, and any operation on the gene will also act on all possible information at the same time. The use of quantum bit encoding allows a chromosome to express the superposition of multiple states at the same time, making the quantum genetic algorithm have better diversity characteristics than the classical genetic algorithm. The use of quantum bit encoding can also achieve better convergence.

[0076] (2) Measure each individual in the initial population P(t0), obtain the corresponding deterministic solution Q(t0), evaluate the fitness of each deterministic solution, and record the optimal individual and its corresponding fitness.

[0077] The fitness function that evaluates the fitness of a given solution is:

[0078] g′=-g

[0079] in:

[0080] ming=g 1 +N

[0081] ming=g 1 +N represents the out-of-limit state variable constraint processing, N represents the number of state variables exceeding the limit, and g represents the updated value.

[0082] The fitness function reflects the global optimal solution. The larger the target value of an individual is, the larger its fitness value should be, so the optimized target function is directly used as the fitness function.

[0083] The solution is determined to be a binary string of length m, where each bit is 0 or 1, which is selected according to the probability of the quantum bit. The measurement process is to randomly generate a number in the interval [0, 1]. If it is greater than the square of the probability amplitude, the measurement result is 1, otherwise it is 0. Then, the fitness of this set of solutions is evaluated, and the best fitness individual is recorded as the target value for the next evolution.

[0084] Subsequently, the algorithm enters the cyclic iteration phase, and as the iteration proceeds, the population's solution gradually converges to the optimal solution. In each iteration, the population is first measured to obtain a set of definite solutions, and then the fitness value of each solution is calculated. Then, according to the current evolutionary goal and the pre-determined adjustment strategy, the individuals in the population are adjusted using the quantum revolving gate to obtain the updated population, record the current optimal solution, and compare it with the current target value. If it is greater than the current target value, the new optimal solution is used as the target value for the next iteration, otherwise the current target value remains unchanged.

[0085] The subsequent steps are as follows:

[0086] (3) Determine whether the end condition is met. If so, output the optimal solution. If not, continue with the subsequent steps. The end condition is that the number of iterations reaches the maximum or the objective function value converges.

[0087] (4) Measure each individual in the population P(t) to obtain the determined solution Q(t), and evaluate the fitness of the determined solution.

[0088] (5) Using quantum revolving door V λ Adjust the individuals, generate a new population P(t+1), and record the best individual and the corresponding fitness.

[0089] In this embodiment, the quantum gate adopts a universal quantum rotation gate:

[0090]

[0091] Where V λ represents the quantum of the revolving door, and α represents the angle of rotation.

[0092] (6) Iteration number t=t+1, return to step (3).

[0093] The obtained grid carbon emission optimization model is used to simulate and optimize the grid carbon emissions.

[0094] Step 4: Conduct simulation verification to evaluate the feasibility of the grid carbon emission optimization model.

[0095] The IEEE57 node system is used to verify the simulation and optimization effects of the model.

[0096] In this embodiment, the IEEE57 node system consists of 7 units (numbered as generator A-generator F), 80 lines (numbered as L1-L80), 42 load points (numbered as load 1-load 42), 15 adjustable transformers (numbered as transformer 1-transformer 15) and 3 parallel capacitors (numbered as C1-C3).

[0097] It is assumed that the network loss of line L13 (representing line L1 and line L3) comes from generator A and generator B, as follows:

[0098] M lc(13,A) =M lc13 ×100 / (100+142),

[0099] M lc(13,B) =M lc13 ×142 / (100+142).

[0100] The network loss ratio of generator A, the loss ratio related to the output of generator A borne by load 3 and load 4 is as follows:

[0101] F(3,1)=M lc(13，A) / M HA

[0102] F(4,1)=0

[0103] The network loss ratio of generator B, the loss ratio related to the output of generator B borne by load 3 and load 4 is as follows:

[0104] F(3,2)=M lc(13，B) / M HB

[0105] F(4,2)=M lc 24 / M HB

[0106] M HA 、M HB are the active outputs of generator A and generator B respectively. Among them, ξ gt Expressing the carbon emission intensity of a unit, we can get:

[0107]

[0108] Among them, D sr Represents the power grid.

[0109] The simulation results of the objective function convergence analysis based on the quantum genetic algorithm are as follows: Figure 4 As shown, the power grid carbon emission optimization model based on the quantum genetic algorithm provided in this embodiment is adopted. The quantum genetic algorithm can converge to the minimum value within 2 seconds, and can converge within 3 seconds as a whole. The algorithm converges quickly.

[0110] Simulation results show that compared with other algorithms, the quantum genetic algorithm adopts the idea of ​​dynamic clustering and uses shared fitness to converge to a better objective function value more quickly, realize global search of large-scale complex power grids, and quickly transform evaluation optimization problems into optimal solutions.

[0111] Step 5: Utilize the grid carbon emission optimization model to optimize the grid carbon emission.

[0112] Embodiment 2:

[0113] Based on the first embodiment, this embodiment expands the quantum genetic algorithm when establishing a power grid carbon emission optimization model based on the quantum genetic algorithm. Figure 5 As shown, crossover operation and mutation operation are added to the original quantum genetic algorithm.

[0114] Specific:

[0115] (1) Dynamically adjust the rotation angle of the quantum gate according to the evolution process.

[0116] A large rotation angle is set at the beginning of the quantum genetic algorithm, and the rotation angle is gradually reduced as the number of evolutionary generations increases. The adjustment strategy is to measure the individual, evaluate its fitness, compare it with the fitness value of the retained optimal individual, and adjust the corresponding quantum bits in the individual according to the comparison result, so that the rotation angle evolves in the direction that is conducive to the optimal solution.

[0117] (2) Add quantum crossover operation.

[0118] The most representative of individual structural information in quantum genetic algorithms is its evolutionary goal, that is, the individual's current optimal solution and the corresponding fitness value. Therefore, it is possible to consider exchanging the evolutionary goals of individuals to achieve the exchange of information between individuals, thereby achieving the purpose of quantum crossover. Its basic operation is to temporarily exchange the optimal solution and the optimal fitness value between individuals. After the individual accepts the crossover operation, its evolutionary direction will be affected by other individuals, thereby obtaining new evolutionary information.

[0119] (3) Add quantum mutation operation. The role of quantum mutation is to slightly disrupt the current evolutionary direction of an individual to prevent the evolution of the individual from falling into a local optimum. Quantum mutation is achieved by operating chromosome encoding. By exchanging the value of the quantum bit probability amplitude, the evolutionary direction of the individual can be completely reversed. Quantum mutation operation uses a combination of single-point mutation and multi-point mutation to enhance the diversity of genes in the population.

[0120] (4) Adding quantum catastrophe. When the algorithm has been stable for several generations and the optimal individual remains stable, the algorithm may fall into a local optimal solution. At this time, taking quantum catastrophe operation can make it escape from the local optimal solution. The specific method is to apply large disturbances to some individuals in the population and re-randomly generate some individuals.

[0121] The above-described embodiment is only a preferred solution of the present invention and does not limit the present invention in any form. There are other variations and modifications without exceeding the technical solution described in the claims.

Claims

1. A method for evaluating and optimizing carbon emissions of power grids based on fuzzy dynamic clustering, characterized in that: The following steps are involved: S1: Establish a carbon emission consumption calculation model that takes into account carbon emissions in all links of the power system, giving priority to the impact of power grid damage on carbon emissions; S2: Construct a collaborative model among enterprises based on fuzzy dynamic clustering; S3: Introduce the total carbon emission rate index of power grid loss and establish a power grid carbon emission optimization model based on quantum genetic algorithm; S4: Perform simulation verification to evaluate the feasibility of the grid carbon emission optimization model; S5: Optimizing the carbon emissions of the power grid using the power grid carbon emissions optimization model.

2. A method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 1, characterized in that: The grid carbon emission optimization model adopts the minimum value of the linear superposition value g1 of the grid active power loss, the total grid loss carbon emission rate and the voltage stability component as the objective function.

3. A method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 1 or 2, characterized in that: In the step S1, when establishing the carbon emission consumption calculation model, the network loss and the network loss that each user should bear are tracked according to the principle of proportional distribution, and the network loss borne by each load node is determined.

4. The method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 1 is characterized in that: The quantum genetic algorithm comprises: A1: Initialize the population and randomly generate n chromosomes encoded with quantum bits; A2: Measure each individual in the initial population, obtain the corresponding definite solution, evaluate the fitness of each definite solution, and record the optimal individual and the corresponding fitness; A3: Determine whether the end condition is met. If so, output the optimal solution. If not, continue with the subsequent steps. A4: Measure each individual in the population, obtain a definite solution, and evaluate the fitness of the definite solution; A5: Use quantum revolving gate to adjust individuals, generate new populations, and record the best individuals and corresponding fitness; A6: Increase the number of iterations by 1 and return to step A3.

5. A method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 1, 2 or 4, characterized in that: In the step S4, the IEEE57 node system is used to simulate and verify the grid carbon emission optimization model.

6. A method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 4, characterized in that: The fitness function for evaluating the fitness of the determined solution is g′=-g, where ming=g1+N, N represents the number of state variables exceeding the limit, g represents the update value, and g1 represents the linear superposition value of the active power loss of the power grid, the total carbon emission rate of the power grid loss, and the voltage stability component.

7. A method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 4 or 6, characterized in that: The quantum genetic algorithm also includes: Dynamically adjust the rotation angle of the quantum gate according to the evolution process; Temporarily exchange the best solution and the best fitness value between individuals to achieve quantum crossover; Temporarily exchange the values ​​of quantum bit probability amplitudes between individuals to achieve quantum mutation; When the quantum genetic algorithm has been stable for several generations and the optimal individuals remain stable, disturbances are applied to some individuals in the population, and some individuals are randomly regenerated to achieve quantum catastrophe.

8. A method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 4 or 6, characterized in that: The quantum revolving gate adopts a universal quantum revolving gate: V λ is the quantum of the revolving door, and α is the angle of rotation.

9. A method for evaluating and optimizing power grid carbon emissions based on fuzzy dynamic clustering according to claim 1, 2 or 4, characterized in that: The constraints of the grid carbon emission optimization model include reactive power constraints of generator sets and active power constraints of balancing nodes.

Images