Source-load-storage integrated low-carbon optimal scheduling method

Abstract

The invention provides a'source-load-storage 'integrated low-carbon optimal scheduling method, which comprises the following steps of: coupling carbon emission of a power generation side with power flow through carbon emission flow modeling, and realizing transmission of carbon emission responsibility from a source side to a load side; carbon responsibility distribution is dynamically corrected in combination with the carbon emission characteristics of the energy storage system in the charging and discharging process; the scheduling strategy is divided into two stages, thermal power, wind power and energy storage optimization is carried out with cost minimization, and then the carbon responsibility is fairly and reasonably allocated and priced by using an extended Shapley value and a bidirectional stepped carbon cost mechanism; and meanwhile, an improved grey wolf optimization algorithm is introduced to improve global and local search capabilities of the model, so that economical efficiency and low-carbon targets are considered.

Description

Technical Field

[0001] This application relates to the field of power system optimization and dispatching technology, and in particular to an integrated low-carbon optimization and dispatching method for "source-load-storage". Background Technology

[0002] With the increasing severity of global climate change, countries have proposed carbon peaking and carbon neutrality goals. As a major source of carbon emissions, the low-carbon transformation of the power system has become an important research direction in the energy sector. Traditional power system operation and dispatch primarily focus on economic efficiency, reducing operating costs by optimizing thermal power unit output and grid power flow distribution. However, this approach often ignores the spatiotemporal characteristics of carbon emissions; carbon emissions not only fluctuate with changes in power generation and load over time, but also migrate spatially within the grid as electricity is transmitted. This leads to uneven allocation of carbon emission responsibility, making it difficult to accurately reflect the contribution of each stage to carbon emissions, thus weakening the incentive effect for emission reduction. In recent years, with the large-scale integration of new energy sources, the proportion of clean energy such as wind power and photovoltaics in the power system has been continuously increasing. However, due to their intermittent and fluctuating nature, they often require regulation by thermal power units and balancing through energy storage systems to maintain the safe and stable operation of the power system. Energy storage devices can play a role in peak shaving and valley filling during charging and discharging. However, existing research often simplifies this process in carbon emission accounting, failing to effectively track the carbon emission responsibility of energy storage in different spatiotemporal scenarios during charging and discharging, leading to distorted carbon emission attribution. On the other hand, most existing carbon cost mechanisms adopt a unified carbon trading or fixed carbon price approach, which is difficult to couple with the dynamic characteristics of power system operation and cannot fully leverage the guiding role of price signals in emission reduction scheduling. Some scholars have attempted to introduce carbon emission constraints or tiered carbon trading mechanisms into scheduling models. While these have improved the effectiveness of low-carbon operation to some extent, they still have shortcomings in terms of fairness in responsibility allocation, dynamic carbon costs, and coupling with energy storage systems. At the optimization solution level, existing scheduling methods often employ intelligent algorithms such as particle swarm optimization, genetic algorithms, and gray wolf optimization. These algorithms still have limitations in global search capabilities and convergence speed, easily getting trapped in local optima, thus affecting the reliability and practicality of scheduling results. Summary of the Invention

[0003] To address the problem of low efficiency in existing low-carbon dispatching methods for power systems, this invention proposes an integrated low-carbon optimization dispatching method that combines power sources, loads, and storage.

[0004] To achieve the above-mentioned technical effects, the technical solution of the present invention is as follows: An integrated low-carbon optimization scheduling method for "source-load-storage" includes the following steps: Step 1: Establish an integrated model of power flow and carbon emission flow, which includes a first-stage scheduling model and a second-stage scheduling model; Step 2: With the goal of minimizing costs, and based on the constraints of generating units, power flow, energy storage, and source-load balance, establish a first-stage scheduling model for thermal power, wind power, and energy storage. Use the improved Grey Wolf optimization algorithm to solve the first-stage scheduling model and obtain preliminary optimized scheduling results. Step 3: In the second-stage scheduling model, the carbon flow rate, carbon flow density and node carbon potential in the power flow network are calculated based on the preliminary optimized scheduling results. A dynamic carbon emission model for energy storage charging and discharging is constructed based on the carbon flow rate, carbon flow density and node carbon potential. The node load power and node energy storage charging and discharging power are input into the dynamic carbon emission model for energy storage charging and discharging to obtain the node carbon emissions. Step 4: Use the extended Shapley value and calculate the carbon emission responsibility of the load and energy storage nodes based on the node carbon emission amount. Based on the carbon emission responsibility amount, obtain the two-way tiered carbon cost on the positive load side and the negative energy storage side. Step 5: Calculate the comprehensive optimization cost based on the two-way tiered carbon cost and carbon emission responsibility. With the goal of minimizing the comprehensive optimization cost, iterate the second-stage scheduling model in combination with the preliminary optimization scheduling results, and finally output the optimal scheduling scheme.

[0005] Compared with the prior art, the beneficial effects of the technical solution of the present invention are: This invention proposes an integrated low-carbon optimization scheduling method for "source-load-storage". By modeling carbon emission flows, it couples carbon emissions from the generation side with power flow, realizing the transfer of carbon emission responsibility from the source side to the load side. Combining the carbon emission characteristics of the energy storage system during charging and discharging, it dynamically corrects the allocation of carbon responsibility. The scheduling strategy is divided into two stages: first, optimizing thermal power, wind power, and energy storage with cost minimization; then, using the extended Shapley value and a two-way tiered carbon cost mechanism to fairly and reasonably allocate and price carbon responsibility. At the same time, an improved gray wolf optimization algorithm is introduced to enhance the model's global and local search capabilities, thereby balancing economic efficiency and low-carbon goals. Attached Figure Description

[0006] Figure 1 This is a flowchart of an integrated low-carbon optimization scheduling method for "source-load-storage" in one embodiment; Figure 2 This is a schematic diagram of power flow and carbon emission flow in one embodiment; Figure 3 This is a diagram of a two-stage optimization model in one embodiment; Figure 4 Here is a diagram of an improved IEEE 30-node system architecture in one embodiment; Figure 5 Here is a load and wind power forecast diagram in one embodiment; Figure 6 This is a diagram showing the carbon cost range partitioning for load node 1 in one embodiment; Figure 7 This is a diagram showing the carbon cost range partitioning for load node 2 in one embodiment; Figure 8 This is a diagram showing the carbon cost range of an energy storage node in one embodiment; Figure 9 Here is an iterative curve diagram of the CEC test function solution in one embodiment; Figure 10 This is a graph showing the solution results of the objective function in the first stage of one embodiment; Figure 11 This is a diagram showing the carbon potential response of load node 1 under different scenarios in one embodiment. Figure 12 This is a graph showing the carbon potential response of load node 2 under different scenarios in one embodiment; Figure 13 This is a diagram showing the carbon potential response of an energy storage node under different scenarios in one embodiment. Figure 14 This is a carbon emission responsibility map of each sub-alliance in one embodiment. Detailed Implementation

[0007] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0008] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” used in this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0009] It should be understood that although the terms first, second, third, etc., may be used in this invention to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, first information may also be referred to as second information without departing from the scope of this invention, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to a determination."

[0010] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments.

[0011] Example 1 This embodiment proposes an integrated low-carbon optimization scheduling method for "source-load-storage", the flowchart of which is as follows: Figure 1 As shown, it includes the following steps: Step 1: Establish an integrated model of power flow and carbon emission flow, which includes a first-stage scheduling model and a second-stage scheduling model; Furthermore, within the integrated "source-load-storage" framework, carbon emission flow theory is introduced to couple power flow with carbon emissions, establishing a carbon emission flow model for energy storage charging and discharging. This model tracks the accumulation and release of carbon emissions from energy storage under different time and spatial scenarios. A schematic diagram of power flow and carbon emission flow in the power system is shown below. Figure 2 As shown.

[0012] Step 2: With the goal of minimizing costs, and based on the constraints of generating units, power flow, energy storage, and source-load balance, establish a first-stage scheduling model for thermal power, wind power, and energy storage. Use the improved Grey Wolf optimization algorithm to solve the first-stage scheduling model and obtain preliminary optimized scheduling results. Step 3: In the second-stage scheduling model, the carbon flow rate, carbon flow density and node carbon potential in the power flow network are calculated based on the preliminary optimized scheduling results. A dynamic carbon emission model for energy storage charging and discharging is constructed based on the carbon flow rate, carbon flow density and node carbon potential. The node load power and node energy storage charging and discharging power are input into the dynamic carbon emission model for energy storage charging and discharging to obtain the node carbon emissions. Step 4: Use the extended Shapley value and calculate the carbon emission responsibility of the load and energy storage nodes based on the node carbon emission amount. Based on the carbon emission responsibility amount, obtain the two-way tiered carbon cost on the positive load side and the negative energy storage side. Step 5: Calculate the comprehensive optimization cost based on the two-way tiered carbon cost and carbon emission responsibility. With the goal of minimizing the comprehensive optimization cost, iterate the second-stage scheduling model in combination with the preliminary optimization scheduling results, and finally output the optimal scheduling scheme.

[0013] The objective of the first-phase scheduling model is to formulate the optimal day-ahead power generation and energy storage charging and discharging schedule within the "source-load-storage" system framework, in order to minimize overall operating costs, including the power generation costs of thermal power units. Cost of wind curtailment Operating costs of energy storage systems .

[0014] In one optional embodiment, the expression for the first-stage scheduling model is:

[0015]

[0016]

[0017]

[0018]

[0019] in, For the cost of generating electricity from thermal power units, For the cost of wind curtailment, For the operating costs of energy storage systems, This represents the number of thermal power units. For the operating costs of thermal power units, The start-up and shutdown costs of thermal power units, , , For thermal power units The cost coefficient, This represents the start-up and shutdown cost coefficient for thermal power units. For thermal power units in The start / stop status at any given time. The number of wind turbine units. For wind turbines The cost coefficient of wind curtailment For wind turbine units Wind curtailment power at any given time This is the energy storage charging and discharging cost coefficient. For energy storage systems in The amount of charge at any given moment. For energy storage systems in Discharge amount at any given time The total number of charging periods for energy storage. This represents the sum of the energy storage discharge periods; The constraints of the first-stage scheduling model include: The constraints on generating units, power flow, energy storage, and source-load balance specifically include: Power constraints of thermal power units:

[0020] in, To provide power to thermal power units, , These are the upper and lower limits of the active power output of thermal power units; Thermal power unit ramping constraints:

[0021] in, To provide power to thermal power units, , These are the upper and lower limits of the active power output ramp rate of thermal power units; Wind turbine power constraints:

[0022] in, To provide power to wind turbines, , These are the upper and lower limits of the active power output of the wind turbine generator; Line power flow upper and lower limit constraints:

[0023] in, The active power of the power grid branch. This represents the maximum power transmitted through the branch. This represents the minimum power transmitted through the branch. Energy storage operation constraints:

[0024]

[0025] in, For the charging and discharging power of the energy storage system, This represents the maximum charging power of the energy storage system. This represents the maximum discharge power of the energy storage system. For energy storage systems in Capacity status at any given time. , These represent the maximum and minimum values ​​of the energy storage system capacity, respectively. Source-load balance constraints:

[0026] in, To provide power to thermal power units, To provide power to wind turbines, , These represent the charging and discharging power of the energy storage system. For the number of branches, This represents the number of thermal power units. The number of wind turbine units. for Power flowing in at each time node for Power flowing out at each time node for Predict load values ​​at all times.

[0027] In this embodiment, the constructed two-stage optimization model achieves front-to-back coupling through parameter transfer: the first stage focuses on the day-ahead economic dispatch of the "source-load-storage" system, optimizing thermal power output, wind power absorption, and energy storage strategies. The output results, such as nodal output, power flow path, and energy storage status, serve as input parameters for the second stage. The second stage then conducts carbon emission flow tracking, responsibility allocation based on Shapley values, and carbon cost assessment under a two-way tiered carbon pricing mechanism. Although the two stages are structurally independent, they remain linked in the solution process. The first stage focuses on operational economics, while the second stage emphasizes carbon fairness and low-carbon incentives. The two objectives are synergistic and complementary, effectively achieving integrated modeling and solution of power dispatch and carbon responsibility allocation.

[0028] Example 2 This embodiment provides a detailed explanation of how to solve the first-stage scheduling model using the improved gray wolf optimization algorithm, based on Embodiment 1.

[0029] In one alternative embodiment, the first-stage scheduling model is solved using an improved gray wolf optimization algorithm, including the following steps: Step 1: Initialize the first-stage scheduling model; initialize the gray wolf population by randomly generating a group of gray wolves within the search space; Step 2: Calculate the fitness value of each gray wolf according to the fitness function, and determine the three optimal gray wolves based on the fitness values; Step 3: Update the position of each gray wolf based on the optimal gray wolf, Cauchy mutation factor, and nonlinear convergence factor; Step 4: Update the fitness value based on the updated gray wolf position. When the preset maximum number of iterations is reached, output the preliminary optimized scheduling result based on the updated fitness value; otherwise, return to step 2.

[0030] Furthermore, the Grey Wolf Optimizer (GWO) is a heuristic optimization algorithm that simulates the hunting behavior of grey wolves. Grey wolf packs possess a well-defined social structure, including... Wolf (leader) Wolf (second leader) Wolf (support role) and Wolves (lowest-ranking members). In the hunting process, gray wolves cooperate to obtain food through encirclement, surrounding prey, and collective attacks. The GWO algorithm mimics this behavior by initializing multiple candidate solutions, evaluating their fitness, and selecting the optimal one. Wolf, wolves and Wolves. Then, based on these optimal individuals, other wolves update their positions through hunting behavior, gradually bringing the group closer to the optimal solution. The algorithm adjusts the position of each wolf step by step to bring it closer to the target solution until a termination condition is met, such as reaching the maximum number of iterations or finding a sufficiently good solution. The position update of each gray wolf is based on the following formula.

[0031] In one alternative embodiment, the position of each gray wolf is updated; the expression is:

[0032] in, For the first Sekiro in the 1990s The position of the generation, It is a linearly decreasing factor. , A random vector between 0 and 1; Furthermore, in traditional GWO, as iterations proceed, the search range of the wolf pack is determined by parameters. Gradually decreasing the size of the search area helps refine the search, but it can also cause the wolf pack to converge prematurely to a local optimum. When the objective function has multiple local optima, the wolf pack may focus its search around a single solution, ignoring other potential global optima. The algorithm lacks an effective global search mechanism, causing the wolf pack's exploration ability to gradually weaken and making it difficult to escape local optima. Furthermore, the randomness of the GWO (Global Optimum Search) factor... and While increasing the diversity of the search, it may also lead to excessive localization of the search, thereby increasing the risk of getting trapped in local optima. To address this issue, this invention proposes an improved Grey Wolf Optimization Algorithm (IGWO) using Cauchy mutation factor and nonlinear update factor.

[0033] Cauchy mutation is a mutation operation based on the Cauchy distribution, characterized by strong jump potential. Introducing a Cauchy mutation factor can enhance the algorithm's versatility and avoid the trap of local optima. In the basic update process of GWO, the Cauchy mutation operation is added, using the Cauchy distribution to generate random numbers to perturb the current wolf's position. This means that the update of each wolf depends not only on... Wolf, wolves and The location of wolves may also undergo leapfrog variations through Cauchy distribution, leading to a wider search area.

[0034] Cauchy variation factors include jump variations through the Cauchy distribution; their expression is:

[0035] in, Let be a random variable with a Cauchy distribution. , For distribution parameters, Step size factor; The nonlinear convergence factor is expressed as:

[0036] in, For the maximum step size, The maximum number of iterations, To control the rate of decay.

[0037] Furthermore, the nonlinear update factor refers to the search strategy that is progressively adjusted based on the number of iterations, in order to adaptively control the fineness of the search at different stages. Introducing the nonlinear update factor into the position update process allows the search strategy to be adjusted according to the progress of the search at different stages of iteration. and This allows for a broader initial search and a more focused later search.

[0038] By introducing a nonlinear convergence factor and a Cauchy mutation factor, IGWO can not only enhance population diversity through random perturbation, helping the algorithm escape local optima and avoid premature convergence, but also dynamically balance global exploration and local exploitation, with slow decay in the early stage to fully search the solution space and accelerated convergence in the later stage to improve accuracy.

[0039] To effectively solve the two-stage optimization model proposed in this invention, the two-stage optimization model diagram is shown below. Figure 3 As shown, this invention introduces IGWO as the core optimizer in solving the first-stage day-ahead economic scheduling model. First, the population is initialized, and the position of each individual is set as a candidate solution for the scheduling scheme. Then, its fitness value is evaluated according to the economic cost function, and it is sorted according to the gray wolf hierarchy. Next, the population is guided towards the optimal solution through a gray wolf hunting mechanism, position update rules, and mutation factors. The search range is dynamically adjusted in each iteration, and the optimal scheduling scheme is output after the maximum number of iterations or convergence accuracy conditions are met. This scheme serves as the input for the second-stage carbon emission assessment, providing data support for subsequent responsibility allocation and carbon cost calculation.

[0040] For example, on the MATLAB platform, the superiority of the proposed improved Grey Wolf Optimization Algorithm (IGWO) is verified by using test functions, and the improved optimization algorithm is used to solve the model to obtain the optimal solution of the decision variables and the minimum system cost.

[0041] For example, this invention uses an improved IEEE 30-node system for example analysis, and the system structure diagram is as follows: Figure 4 As shown. The predicted power of wind power and load is as follows. Figure 5 As shown.

[0042] As shown in the improved IEEE 30-node system architecture diagram, the system has two load nodes and one energy storage node. Load node 3 is designated as member L1; load node 4 as member L2; and energy storage node 14 as member E. Therefore, the alliance N = {L1, L2, E}, and the entire alliance has six non-empty subsets: {L1}, {L2}, {E}, {L1, L2}, {L1, E}, and {L2, E}. To maintain system power balance, a 1000MW virtual load is uniformly scheduled to participate in operation when each sub-alliance independently accesses the system. This virtual load is used to replace the actual load demand, thereby ensuring the comparability of carbon emission calculations under different alliance combinations. The final system carbon emission is composed of the carbon emission contribution of the sub-alliance minus the carbon emission introduced by the virtual load. Taking time t=1 as an example, the system carbon emission is calculated under different alliance conditions, and the results are shown in Table 1. After allocating carbon responsibility to the three nodes, their carbon responsibility allocation intervals can be calculated. Based on the intervals, the carbon cost is divided into three parts, and the results are as follows. Figures 6 to 8 As shown.

[0043] Table 1 Carbon Emission Responsibilities of Each Sub-Alliance

[0044] Depend on Figure 6 and Figure 7The results show that the carbon responsibility distribution of different nodes exhibits significant temporal fluctuations during the scheduling cycle. Specifically, the total carbon emission cost range for load nodes L1 and L2 increases significantly during peak electricity consumption periods. Taking the period from 0:00 to 7:00 as an example, the peak carbon emission costs for L1 and L2 reach 3940.2 yuan and 3226.9 yuan respectively, increasing by 44.7% and 53.4% ​​compared to the lowest values ​​during off-peak periods (18:00–24:00). This is because the system needs to utilize more thermal power units during high-load periods, leading to an increase in carbon emission intensity per unit of electricity, thus pushing up the carbon emission costs allocated to users. During the daytime high-load period (8:00–17:00), high carbon emissions from thermal power units are unavoidable; therefore, the first-stage carbon emission cost range accounts for a larger proportion, while the second and third-stage ranges are relatively smaller. During this stage, the average first-stage proportion for load nodes L1 and L2 is 86.2% and 86.3% respectively, indicating that users bear a higher carbon emission cost for the redundant electricity generated by high load. Conversely, during off-peak hours in the evening (18:00–24:00), the total carbon emission cost range decreases significantly. This is because reduced output from thermal power units leads to a corresponding decrease in the carbon costs shared by users. At this time, the proportion of the first-stage range is relatively small, while the proportions of the second and third-stage ranges increase significantly: the average proportion of the second-stage range for L1 and L2 during this period is 12.1% and 19.7%, respectively, and the average proportion for the third-stage range is 9.2% and 9.0%, respectively. This means that in addition to necessary costs, extra electricity consumption incurs higher carbon costs, thus creating a price signal that guides users to reduce unnecessary electricity consumption, thereby achieving the goal of curbing and reducing carbon emissions. Figure 8 Therefore, for energy storage nodes, when charging, they are regarded as load nodes, and their carbon liability cost range changes similarly to L1 and L2, with a peak value of 326.2 yuan; when discharging, they are regarded as generating electricity externally, producing a carbon emission reduction effect, with a corresponding carbon cost of -75.7 yuan, which reflects the dual value of energy storage in peak shaving and valley filling and carbon emission reduction.

[0045] To verify the superiority of IGWO in solving the one-stage model, GWO, IGWO, Particle Swarm Optimization (PSO), Moth Flame Optimization (MFO), Sand Cat Swarm Optimization (SCSO), and Caterpillar Fungus Optimizer (CFO) were used to solve the one-stage model and the CEC2017 (Competition on Evolutionary Computation) test function. The results are as follows: Figure 9 and Figure 10 As shown. By Figure 9As can be seen, during the solution process of the test function, IGWO exhibits the fastest convergence speed in the first 100 iterations, and stably converges to 4899.8 in the 300th iteration, demonstrating excellent convergence efficiency and stability. At this point, compared to the traditional GWO, its fitness value decreases by 27.5%. In contrast, although the traditional GWO converges to 6761.5 with a similar number of iterations, its overall performance is significantly inferior to IGWO due to prematurely falling into local optima. Further comparison with other algorithms shows that CFO, SCSO, MFO, and PSO all failed to break through the local optimum trap; at the 500th iteration, IGWO's fitness value is 13.7%, 23.2%, 30.0%, and 46.1% lower than the aforementioned algorithms, respectively. In summary, IGWO significantly outperforms the compared algorithms in terms of optimal solution accuracy, convergence speed, and convergence stability, verifying its outstanding advantages in complex optimization problems. Figure 10 As can be seen, in the first-stage model solution, IGWO consistently balances global search and local exploitation throughout the entire iteration process, converging to 2127.3 in round 461. It rapidly approaches the optimal fitness value in the early stages of iteration and maintains a stable and efficient convergence trend in the later stages, significantly outperforming the compared algorithms. In round 500, IGWO's model solution value is 13.3%, 26.3%, 29.0%, and 17.1% lower than CFO, SCSO, MFO, and PSO, respectively. Its performance advantage is mainly attributed to two improvements: first, the introduced Cauchy mutation factor effectively expands the individual search boundary, enabling the algorithm to perform leapfrog search capabilities, thus overcoming the limitation of traditional GWO's tendency to get trapped in local optima; second, the nonlinear update factor dynamically adjusts the convergence speed, achieving a balance between global exploration and local exploitation. It is the synergistic effect of these two mechanisms that makes IGWO exhibit superior solution performance in model solving.

[0046] In this embodiment, Cauchy mutation factor and nonlinear convergence factor are introduced into the traditional gray wolf optimization algorithm to enhance the algorithm's global search capability and convergence speed. Through population diversity and dynamic convergence mechanism, the algorithm avoids getting trapped in local optima and improves the performance of solving complex optimization problems.

[0047] Example 3 This embodiment further explains the present invention in detail based on Embodiments 1 and 2.

[0048] By constructing a carbon emission flow model and combining an extended Shapley allocation mechanism with a two-way stepped carbon cost model, a fair allocation of carbon emission responsibility and dynamic cost incentives can be achieved. In terms of scheduling strategy, a two-stage optimization framework is adopted. The first stage optimizes thermal power output, wind power consumption, and energy storage charging and discharging with the goal of minimizing operating costs. The second stage performs responsibility allocation and carbon cost assessment based on carbon emission flow tracking results. At the solution level, an improved gray wolf optimization algorithm is introduced to enhance global search and convergence performance. Verified by an IEEE 30-bus system example, this invention demonstrates superior performance in reducing operating costs and carbon emissions, and enhancing the peak-shaving and valley-filling value of energy storage, proving its effectiveness and application value in low-carbon power system scheduling.

[0049] Within the integrated scheduling and optimization framework of "source-load-storage," to more accurately consider the spatiotemporal flow characteristics of carbon emissions, the "carbon emission flow" theory is introduced. This theory combines carbon emissions with power flow in the power system and further considers the impact of energy storage, load-side carbon emission allocation, and marginal carbon costs. Carbon emission flow is essentially a virtual network flow that does not exist in the power system itself; what truly exists in the power network is the power flow. Since carbon emissions are generated during power generation at the source side, a link between power flow and carbon emissions can be established. It is assumed that the carbon dioxide generated at the source side is not directly released into the atmosphere from the power plant, but rather accompanies the power flow as a virtual "carbon flow" to the load side. The load side consumes electricity and needs to pay the carbon emission costs incurred during power generation. Therefore, the responsibility for carbon emissions shifts from a single calculation at the source side to active assumption by the load side. To track carbon emissions from the source side to the load side, the power system carbon emission flow method has emerged. The carbon flow rate represents the amount of carbon emissions corresponding to the energy flow passing through a node or branch per unit time, expressed as... It is expressed in tons of CO2 / h.

[0050] In one alternative embodiment, the carbon flow rate in the power flow network is expressed as:

[0051] in, Carbon flow rate, This refers to the carbon emissions flowing into the node or through the branch. For scheduling time; Furthermore, carbon flux density represents the amount of carbon emissions carried per unit of electrical energy transmitted, expressed as... This indicates that the unit is tons of CO2 / MWh: The expression for carbon flux density in a power flow network is:

[0052] in, Carbon flux density, To correspond to the active power flow in the network; Furthermore, the dynamic node carbon potential represents the load node at... The equivalent carbon emissions generated per unit of electricity consumed during a given time period, compared to those generated on the source side. express.

[0053] The expression for the carbon potential of a node in a power flow network is:

[0054] in, For nodes exist Dynamic nodal carbon potential at any given time. branch road exist Time-period injection node The amount of electricity, branch road carbon flux density, For nodes Generator injected power, This refers to the carbon emission intensity of the generator.

[0055] Traditional carbon emission flow models, failing to consider the dynamic characteristics of energy storage systems, lead to inaccurate carbon emission liability attribution and significant errors in calculating total system emissions. However, by introducing an energy storage model, carbon emissions during the charging and discharging phases can be traced back to different spatiotemporal scenarios, dynamically correcting the node carbon potential and thus accurately calculating the carbon emissions of energy storage in both time and space. The carbon potential of energy storage during charging is equal to the nodal carbon potential of its node, and its carbon emission flow model is as follows: .

[0056] In one optional embodiment, a dynamic carbon emission model for energy storage charging and discharging is constructed based on carbon flow rate, carbon flow density, and nodal carbon potential; wherein, the model expression for the carbon emission flow during energy storage system charging is:

[0057] in, For nodes The dynamic node carbon potential. Dynamic node carbon potential for charging energy storage systems; Furthermore, when the energy storage device discharges, it releases electrical energy into the grid, which can be equivalent to a generator set. The carbon flow accumulated inside the device begins to be released along with the electrical charge. Let the energy storage element start from... Start charging immediately until Discharges continuously.

[0058]

[0059] in, , They are respectively , It continuously stores the remaining electricity accumulated internally. , They are respectively , The internal carbon flow is constantly stored. , These represent the electricity and carbon flow accumulated by the energy storage device during the charging process, respectively.

[0060] Based on the carbon potential of nodes in the power flow network, the carbon potential during discharge is equivalent to the ratio of the accumulated carbon flow to the electrical charge during the charging process. The energy conversion rate during charging and discharging is considered. The carbon emission flow model for energy storage discharge is obtained as follows:

[0061] After further discretization of the above formula, the following formula is obtained: The model expression for the carbon emission flow during the discharge of an energy storage system is:

[0062] in, Carbon potential at dynamic discharge nodes in energy storage systems , From respectively arrive The total number of discrete time intervals and the length of each time interval. This refers to the charging and discharging efficiency of the energy storage system.

[0063] Furthermore, considering the linear relationship between node carbon emissions, node carbon potential, and power consumption, and by incorporating the energy storage system into the carbon emission flow model, the carbon emissions of the node during charging and discharging of the energy storage system can be obtained. The specific calculation relationship is as follows: carbon emissions during the charging period are calculated by multiplying the charging carbon potential by the node's total power consumption demand; during the discharging period, the emissions are calculated by subtracting the carbon debt "borne" by the energy storage system from the node's own load based on the current carbon potential, and adding the additional emissions caused by losses.

[0064] In one optional embodiment, the node load power and node energy storage charging and discharging power are input into the energy storage charging and discharging dynamic carbon emission model to obtain the node carbon emissions; the expression is as follows:

[0065] in, , The distribution consists of nodes containing energy storage. Carbon emissions during charging and discharging , , They are nodes current The load power, energy storage charging power, and energy storage discharging power at any given time. This refers to the charging and discharging efficiency of the energy storage system.

[0066] Furthermore, carbon emission flow theory reveals the intrinsic link between electricity consumption and carbon emissions by transferring carbon emission responsibility from the generation side to the load side. Building on this, this invention further expands the scope of carbon emission responsibility attribution by incorporating energy storage systems into the carbon emission flow tracking and responsibility allocation framework. However, differences in geographical location, spatiotemporal distribution of electricity, and grid constraints between load nodes and energy storage nodes lead to a highly unbalanced distribution of system carbon emission responsibility in both spatial and temporal dimensions. To achieve a fairer and more reasonable allocation of responsibility and design of a carbon pricing mechanism, this invention establishes an extended Shapley value allocation model based on cooperative game theory, including energy storage nodes as one of the participating entities. This model systematically quantifies the marginal contribution of each load and energy storage node to system carbon emissions, thereby defining their reasonable carbon emission responsibility range. Finally, this invention further proposes a two-way tiered carbon cost allocation model, dividing the carbon responsibility range of each node into multiple levels. The negative carbon responsibility generated by energy storage nodes discharging during high-carbon periods is transformed into a negative carbon cost range, thus making the value of carbon reduction explicit. Ultimately, this mechanism can accurately measure the carbon emission costs of electricity consumption and regulation behavior, promoting the effective implementation of the low-carbon principle of "whoever emits, bears the responsibility; whoever reduces emissions, benefits" in the electricity market. The Shapley value method, based on cooperative game theory, calculates the average marginal contribution of each member across all possible alliance combinations, allocating node carbon emission responsibility according to its marginal impact on the overall system. It determines the amount of carbon emission responsibility allocated to load nodes without considering the participation of energy storage in the allocation. .

[0067] In an optional embodiment, the calculation of the carbon emission responsibility of the load and energy storage nodes using the extended Shapley value and based on the node carbon emissions further includes the following steps: The expression for the carbon emission responsibility allocated to load nodes without considering energy storage participation is:

[0068] in, This refers to the carbon emission responsibility allocated to load nodes when energy storage is not considered in the sharing process. For the set of load nodes, For the sub-alliance, For nodes Sub-alliance, For the Alliance The marginal carbon emissions generated For the Alliance Total carbon emissions; The expression for considering energy storage's participation in the extended Shapley value for allocating carbon emission responsibilities is as follows:

[0069] in, To account for the carbon emission responsibility allocated to load nodes when energy storage participates in the cost-sharing process, It is a collection of energy storage nodes. It is a collection that includes loads and energy storage nodes. For the sub-alliance, For nodes Sub-alliance, For the Alliance The marginal carbon emissions generated For the Alliance Total carbon emissions.

[0070] Furthermore, the carbon emission responsibilities borne by each member of the sub-alliance should be within a certain range, not exceeding the maximum value of the marginal effect of that member. It is also not less than the minimum value of its marginal effect. ,Right now:

[0071] in, Carbon emissions are allocated to nodes.

[0072] Furthermore, based on the aforementioned analytical model, the regulation of carbon emission responsibility on the positive load side is set through extreme values ​​within a range. benchmark value and mean Define the scope of dynamic regulation; set a benchmark value for carbon emission responsibility regulation on the reverse energy storage side. This is used to define the negative carbon cost range for energy storage discharge. The specific setting method is as follows: The mathematical model for the two-way carbon emission price range is as follows:

[0073] in, For nodes exist Carbon emissions at any given moment; , , , These are the carbon emission cost coefficients for the first, second, third, and fourth phase load nodes, respectively. , These are the negative carbon cost coefficients for the first and second phases of energy storage nodes, respectively. This represents the first-stage range of carbon emissions from energy storage nodes. , , These represent the range values ​​of carbon emissions for the first, second, and third stages of the load node, respectively.

[0074] Calculated based on tiered carbon pricing Time Node Total carbon emission cost .

[0075] In one alternative embodiment, the expression for the two-way carbon emission cost range is:

[0076] in, Two-way carbon emission cost range For nodes exist Carbon emissions at any given moment. , , , These are the carbon emission cost coefficients for the first, second, third, and fourth phase load nodes, respectively. , These are the negative carbon cost coefficients for the first and second phases of energy storage nodes, respectively. This represents the first-stage range of carbon emissions from energy storage nodes. , , These represent the range values ​​of carbon emissions for the first, second, and third stages of the load node, respectively.

[0077] In one optional embodiment, the expression for the second-stage scheduling model is:

[0078] in, This represents the cost range for both carbon emissions and emissions. The constraints of the second-stage scheduling model are:

[0079] in, For node carbon emissions. , The inflow sets of the branches are respectively Outflow set exist Inflow and outflow nodes at all times Carbon emissions.

[0080] For example, this invention sets up three different scenarios to verify the effectiveness of the proposed two-stage model that considers the spatiotemporal characteristics of carbon emissions. Scenario 1: Without considering the spatiotemporal characteristics of carbon emissions, only the cost and carbon emissions are calculated from power generation measurement; Scenario 2: Considering the spatiotemporal characteristics of carbon emissions, but not considering the Shapley value for carbon responsibility allocation; Scenario 3: Considering the spatiotemporal characteristics of carbon emissions, and considering the Shapley value method for allocating nodal carbon emissions.

[0081] To assess the changing trends of carbon emission responsibilities at each node in the "load-storage" system under different scenarios, this invention compares and analyzes the changes in carbon potential of the load node and the energy storage node under three scenarios. The results are as follows: Figures 11 to 13 As shown.

[0082] Depend on Figure 11 Figure 12 and Figure 13 As can be seen, in Scenario 1, since carbon emissions are not allocated from the generation side to the load and energy storage nodes for calculation, the carbon potential of each node is 0. In this case, the system's carbon emissions and operating costs cannot be attributed according to electricity consumption, making it difficult to fully reflect the impact of spatiotemporal differences on carbon emissions. In Scenario 2, a significant correlation is shown between carbon potential changes and load power. Specifically, the Pearson correlation coefficient between the load curve and the carbon potential curve of load node 1 is 0.9708, and the correlation coefficient for load node 2 is 0.9378, indicating that the carbon potential can effectively reflect load changes and has a strong load response characterization capability, thus effectively supporting the attribution of node carbon emissions under the spatiotemporal characteristics of carbon emissions. For energy storage nodes, when charging power increases, the carbon potential rises; when charging power decreases, the carbon potential decreases. During discharge, an increase in discharge power corresponds to an increase in negative carbon potential, and a decrease in discharge power leads to a decrease in negative carbon potential, exhibiting a response pattern consistent with power changes. In Scenario 3, the correlation between node carbon potential and load power is further enhanced. Compared to Scenario 2, the Pearson correlation coefficient between carbon potential and power at load node 1 increased from 0.9708 to 0.9827, an increase of 1.23%; the correlation coefficient at load node 2 increased from 0.9378 to 0.9664, an increase of 3.08%. The carbon potential variation pattern of the energy storage node remained consistent with Scenario 2. However, during the peak electricity consumption period from 8:00 to 17:00, the carbon potential of both load nodes 1 and 2 in Scenario 2 was generally higher than that in Scenario 3. This difference is because Scenario 3 introduced the Shapley value method, which divided the carbon emission cost range on the electricity consumption node side, thus subjecting the generation side to carbon cost constraints during power supply. As power generation increases, carbon costs rise accordingly. Since the system's optimization objective is to minimize total operating costs, the model will more rationally schedule units under constraints, thereby effectively reducing system carbon emissions and operating costs.

[0083] To evaluate the carbon emission reduction effect at the "source" side under different scenarios, this invention compared and analyzed the total daytime output of six thermal power units, and obtained the following results: Figure 14 As shown.

[0084] As shown in the figure, in Scenario 1, the total output of units 2 and 5, which are closer to the load node, is 15790.1 MW and 14201.6 MW, respectively; the total output of unit 13, which is closer to the energy storage node, is 9170.3 MW; and the outputs of the remaining units 1, 8, and 11 are 9818.5 MW, 21533.1 MW, and 23015.8 MW, respectively. Under this scenario, because the spatiotemporal characteristics of carbon emissions are not considered, the scheduling potential of low-carbon units is not fully realized, and network power flow transmission losses are not at the optimal level, resulting in significant room for optimization of the total carbon emissions of the system. In Scenario 2, the total output of Units 2 and 5, located closer to the load node, increases to 16,531.1 MW and 15,223.6 MW, respectively, representing increases of 4.7% and 7.2% compared to Scenario 1. The output of Unit 13, located closer to the energy storage node, increases to 10,145.3 MW, an increase of 10.6%. Meanwhile, the outputs of the remaining Units 1, 8, and 11 decrease to 9,724.0 MW, 19,467.1 MW, and 22,415.8 MW, respectively, representing decreases of 0.9%, 9.6%, and 2.6% compared to Scenario 1. This indicates that under the spatiotemporal constraints of carbon emissions, the system tends to prioritize scheduling units closer to the load-storage node or with lower carbon emission factors, thereby effectively reducing system carbon emissions while meeting load power demands. In Scenario 3, the output of Units 2 and 5, located closer to the load node, further increases to 16,999.1 MW and 15,802.6 MW, respectively, representing increases of 7.7% and 11.3% compared to Scenario 1, and 2.8% and 3.8% compared to Scenario 2. The output of Unit 13, located closer to the energy storage node, increases to 10,800.3 MW, a 17.8% increase compared to Scenario 1 and a 6.5% increase compared to Scenario 2. Meanwhile, the output of Units 1, 8, and 11 decreases to 9,600.0 MW, 19,007.1 MW, and 21,102.8 MW, respectively, representing decreases of 2.2%, 11.7%, and 8.3% compared to Scenario 1, and further decreases of 1.3%, 2.4%, and 5.9% compared to Scenario 2. In this scenario, in addition to considering the spatiotemporal characteristics of carbon emissions, the Shapley value allocation method is introduced, thereby further strengthening carbon cost constraints. Through an optimization strategy aimed at minimizing carbon emissions and operating costs, the system achieves more rational unit scheduling, resulting in better emission reduction and economic benefits.

[0085] The system operating costs and carbon emissions generated by optimizing the solution for each scenario are shown in Table 2.

[0086] Table 2 Carbon Emissions and Operating Costs under Different Scenarios

[0087] Table 2 shows that the operating cost consists of the operation and maintenance cost and unit power generation cost of the power generation unit, the revenue from external discharge of the energy storage unit, and the carbon emission cost. Since the main carbon emissions in the system originate from thermal power units, their emissions are recorded as the overall carbon emissions of the system. In Scenario 1, the system operating cost is 9.063 million yuan, and the carbon emission is 25,436.6 tons. This scenario does not consider the spatiotemporal characteristics of carbon emissions, nor does it adopt a carbon responsibility allocation mechanism based on Shapley values; therefore, both the operating cost and carbon emission levels are relatively high. This is mainly due to two reasons: firstly, the carbon emission responsibility on the power generation side has not been transferred to the load side, and thermal power units lack effective emission reduction constraints; secondly, the carbon emission cost is calculated using a fixed carbon price, failing to generate a dynamic price signal, resulting in insufficient emission reduction incentives. In Scenario 2, the system operating cost is 8.958 million yuan, and the carbon emission is 24,650.1 tons. Compared with Scenario 1, the operating cost decreases by 1.2%, and the carbon emission decreases by 3.1%. In this scenario, carbon emissions and carbon costs are linked to node power changes, which to some extent suppresses the operation of high-emission units, achieving a dual reduction in cost and emissions. However, although this scenario allocates carbon emission responsibility to the load side, it still does not introduce Shapley value allocation, and carbon costs still use a fixed carbon price, which cannot reflect market fluctuations. The linkage mechanism between price and emissions is insufficient, and there is still room for improvement in emission reduction and cost reduction effects. In Scenario 3, the system operating cost is reduced to 8.892 million yuan, and carbon emissions are 24,033.9 tons. Compared with Scenario 1, operating costs and carbon emissions decrease by 1.9% and 5.5%, respectively; compared with Scenario 2, they decrease by 0.7% and 2.5%, respectively. This scenario, based on carbon emission flow tracking, introduces the Shapley value method to accurately allocate node carbon responsibility and refines the carbon emission price range for each node. As a result, power generation and energy storage units can complete the power dispatch curve response with lower overall cost and carbon emission levels based on the range price signal, thereby achieving more reasonable unit dispatch and obtaining the best emission reduction and economic effects.

[0088] In this embodiment, a two-stage optimization model that takes into account the spatiotemporal characteristics of carbon emissions is proposed within the integrated "source-load-storage" framework. The second stage introduces the Shapley allocation mechanism for carbon emission flow tracking and extension, incorporates the carbon emission responsibility of energy storage into the calculation, and proposes a two-way tiered carbon cost model to achieve a fair allocation of responsibility and cost, and measure the carbon emission cost of each node.

[0089] The various embodiments in this invention are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the device embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions in the method embodiments. The device embodiments described above are merely exemplary. The modules described as separate components may or may not be physically separate. When implementing the present invention, the functions of each module can be implemented in one or more software and / or hardware. Alternatively, some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0090] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the implementation of the present invention. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively describe all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A "source-load-storage" integrated low-carbon optimal scheduling method, characterized in that, The method comprises the following steps: Step 1: establishing an integrated power flow and carbon emission flow model, wherein the integrated power flow and carbon emission flow model comprises a first-stage scheduling model and a second-stage scheduling model; Step 2: establishing a first-stage scheduling model of thermal power, wind power and energy storage based on the minimization of cost and the constraints of units, power flow, energy storage and source-load balance, and solving the first-stage scheduling model by using an improved grey wolf optimization algorithm to obtain a preliminary optimal scheduling result; Step 3: in the second-stage scheduling model, calculating the carbon flow rate, carbon flow density and node carbon potential in the power flow network according to the preliminary optimal scheduling result, constructing an energy storage charging and discharging dynamic carbon emission model according to the carbon flow rate, carbon flow density and node carbon potential, and inputting the node load power and node energy storage charging and discharging power into the energy storage charging and discharging dynamic carbon emission model to obtain the node carbon emission; Step 4: calculating the carbon emission responsibility of the load and energy storage nodes by using an extended Shapley value and according to the node carbon emission, and obtaining the two-way ladder carbon cost of the positive load side and the reverse energy storage side according to the carbon emission responsibility; Step 5: calculating a comprehensive optimization cost based on the two-way ladder carbon cost and the carbon emission responsibility, and iteratively processing the second-stage scheduling model based on the minimization of the comprehensive optimization cost and the preliminary optimal scheduling result to finally output an optimal scheduling scheme.

2. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 1, characterized in that, The first-stage scheduling model expression is: in, For the cost of generating electricity from thermal power units, For the cost of wind curtailment, For the operating costs of energy storage systems, This represents the number of thermal power units. For the operating costs of thermal power units, The start-up and shutdown costs of thermal power units, , , For thermal power units The cost coefficient, This represents the start-up and shutdown cost coefficient for thermal power units. For thermal power units in The start / stop status at any given time. The number of wind turbine units. For wind turbines The cost coefficient of wind curtailment For wind turbine units Wind curtailment power at any given time This is the energy storage charging and discharging cost coefficient. For energy storage systems in The amount of charge at any given moment. For energy storage systems in Discharge amount at any given time The total number of charging periods for energy storage. This represents the sum of the energy storage discharge periods; The unit, power flow, energy storage and source-load balance constraints specifically comprise: Thermal power unit power constraint: wherein, is the power output of the thermal power unit, , are the upper and lower limits of the active power output of the thermal power unit, respectively. Thermal power unit ramping constraint: wherein, is the thermal power output of the thermal power unit, , are the upper and lower active power ramping limits of the thermal power unit, respectively. Wind power unit power constraint: wherein, Pout is the wind turbine output power, , Pmin and Pmax are the wind turbine active power upper and lower limit values, respectively. Line power flow upper and lower limit constraint: wherein is the active power of the grid branch, is the maximum value of the branch transmission power, is the minimum value of the branch transmission power; Energy storage operation constraint: wherein, is the charging power of the energy storage system, is the maximum charging power of the energy storage system, is the maximum discharging power of the energy storage system, is the state of capacity of the energy storage system at is the state of capacity of the energy storage system at , are the maximum and minimum values of the energy storage system capacity, respectively; Source-load balance constraint: wherein, is the output of the thermal power unit, is the output of the wind power unit, , are the charging and discharging power of the energy storage system, respectively, is the number of branches, is the number of thermal power units, is the number of wind power units, is the power flowing into the node at time t, is the power flowing out of the node at time t, is the predicted load value at time t.

3. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 2, characterized in that, The improved grey wolf optimization algorithm for solving the first-stage scheduling model comprises the following steps: Step 1: initializing the first-stage scheduling model; initializing the grey wolf population and randomly generating a set of grey wolves in the search space; Step 2: calculating the fitness value of each grey wolf according to the fitness function, and determining three optimal grey wolves according to the fitness value; Step 3: updating the position of each grey wolf according to the optimal grey wolf, Cauchy mutation factor and nonlinear convergence factor; Step 4: updating the fitness value according to the updated grey wolf position, and when the preset maximum iteration number is reached, outputting the preliminary optimal scheduling result according to the updated fitness value, or returning to step 2 for execution.

4. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 3, characterized in that, The expression for updating the position of each grey wolf is: wherein, is the first Wolf in the position of the is a linear decay factor, , is a random vector within 0 to 1; The Cauchy mutation factor comprises a jump mutation through Cauchy distribution; the expression is: wherein, is a random variable of Cauchy distribution, , is a distribution parameter, is a step factor; The nonlinear convergence factor; the expression is: wherein, is the maximum step size, is the maximum number of iterations, is the speed of control decay.

5. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 1, characterized in that, The expression for the carbon flow rate in the power flow network is: wherein, is the carbon flow rate, is the carbon emissions into a node or through a branch, is the dispatch time; The expression for the carbon flow density in the power flow network is: wherein, is the carbon flow density, is the active power flow in the corresponding network; The expression for the node carbon potential in the power flow network is: wherein, is a node at the dynamic node carbon potential at a time instant, is a branch injecting electric power into the node at a time interval, is a branch carbon flow density, is a node generator injecting power, is a generator carbon emission intensity.

6. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 5, characterized in that, The energy storage charging and discharging dynamic carbon emission model is constructed according to the carbon flow rate, carbon flow density and node carbon potential; wherein the model expression for the carbon emission flow when the energy storage system is charging is: wherein, is the dynamic node carbon potential of the node is the dynamic node carbon potential of the energy storage system charging node;​ The model expression for the carbon emission flow when the energy storage system is discharging is: wherein, a dynamic discharge node carbon potential of the energy storage system, , are respectively a total number of discrete time periods and a unit time period length from to time, is a charge-discharge efficiency of the energy storage system.

7. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 6, characterized in that, The node carbon emission is obtained by inputting the node load power and node energy storage charging and discharging power into the energy storage charging and discharging dynamic carbon emission model; The expression is: in, , The distribution consists of nodes containing energy storage. Carbon emissions during charging and discharging , , They are nodes current The load power, energy storage charging power, and energy storage discharging power at any given time. This refers to the charging and discharging efficiency of the energy storage system.

8. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 1, characterized in that, The calculation of the carbon emission responsibility of the load and energy storage nodes by using the extended Shapley value and according to the node carbon emission further comprises the following steps: The expression of the carbon emission responsibility of the load node without considering the participation of energy storage in allocation is: wherein, is the amount of carbon emission responsibility allocated to the load node when the energy storage participation is not considered, is the set of load nodes, is the sub-alliance, is the sub-alliance without the node , is the sub-alliance produces the marginal amount of carbon emission, is the total amount of carbon emission of the sub-alliance . The expression of the carbon emission responsibility of the load node considering the participation of energy storage in allocation is: wherein, is the amount of carbon emission responsibility allocated to the load node considering the participation of energy storage, is the set of energy storage nodes, is the set of nodes containing both load and energy storage nodes, is the sub-alliance, is the sub-alliance without node , is the sub-alliance generated marginal carbon emission amount, is the sub-alliance total carbon emission amount.

9. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 8, characterized in that, The expression of the two-way carbon emission cost interval is: wherein, Two-way carbon emission cost interval, For node In Carbon emissions at the moment, , , , The first, second, third and fourth stage load node carbon emission cost coefficient respectively, , The first and second stage energy storage node negative carbon cost coefficient respectively, The first stage energy storage node carbon emission interval value, , , The first, second and third stage carbon emission interval value of the load node respectively.

10. The "source-load-storage" integrated low-carbon optimal scheduling method according to claim 1, characterized in that, The expression of the second-stage scheduling model is: wherein, is a two-way carbon emission cost interval; The constraint condition of the second-stage scheduling model is: wherein, is the carbon emission of the node, , are the carbon emissions of the incoming and outgoing sets of branches, , are the carbon emissions of the incoming and outgoing sets of branches, at the time instant, flowing into and out of the node.

Images