Power grid enterprise carbon quota collaborative optimization method, system and equipment based on PSO-GWO hybrid algorithm and medium

Abstract

The invention discloses a power grid enterprise carbon quota collaborative optimization method, system and device based on a PSO-GWO hybrid algorithm and a medium, and belongs to the technical field of power grid enterprise carbon quota optimizing.The method comprises the steps that a dynamic carbon emission factor modeling mechanism is established, and a power carbon emission coefficient of a consumption node is obtained; and based on the power carbon emission coefficient, constructing a multi-period carbon quota allocation and scheduling optimization model for a power grid enterprise to restrain the minimization of the total carbon emission amount, based on a PSO-GWO hybrid algorithm, carrying out double-layer carbon quota allocation, and based on an algorithm result, carrying out intelligent optimization scheduling. According to the method, the accuracy and adaptability of carbon quota distribution are effectively improved, accurate distribution of the carbon quota is realized, the problem that a traditional method is prone to falling into a local extreme value is solved, the globality of overall quota optimization is improved, organic fusion of emission reduction cost minimization and a carbon emission peak value regulation and control path is realized, the overall carbon emission reduction cost of an enterprise is reduced, and the economic benefit of the enterprise is improved. And the controllability of a carbon peak target is improved.

Description

Technical Field

[0001] This invention relates to the field of carbon quota optimization technology for power grid enterprises, specifically to a collaborative optimization method, system, equipment, and medium for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm. Background Technology

[0002] As the hub of energy transmission and dispatch, power grid companies' carbon emissions not only affect the allocation of emission reduction responsibilities between power generation and consumption ends, but also directly relate to the achievement of regional and even national carbon emission targets. With the gradual improvement of carbon emission trading markets and quota management policies, power grid companies are facing increasing pressure in carbon quota management, making the need for efficient and intelligent quota allocation and dispatch methods increasingly urgent. Scientific and reasonable carbon quota allocation can not only effectively constrain the carbon emission ceiling of power grid companies, but also incentivize them to optimize dispatch strategies, improve energy efficiency, and enhance carbon emission reduction capabilities. How to develop an adaptive and incentive-based carbon quota allocation mechanism that combines the operating characteristics and dynamic load changes of power grid companies has become one of the core technical challenges in their green transformation.

[0003] Current mainstream carbon quota allocation methods include the historical total method, the historical intensity method, and the baseline method. Among them, the historical intensity method is widely used in the power grid industry because it can take into account both historical emission efficiency and output levels. However, these methods are mainly based on static historical data and are difficult to reflect the dynamic changes in grid load and carbon emissions, resulting in a disconnect between quota allocation and actual operation. In recent years, the academic community has conducted many explorations on the dynamic optimization of carbon quotas. For example, some scholars have used intelligent models such as grey prediction-PSO-BPNN to dynamically predict and allocate carbon emissions in regional power grids, improving the fairness and adaptability of multi-regional quota allocation. Other studies have applied the carbon quota mechanism to scenarios such as microgrids, energy storage, and demand response, proposing time-adaptive quota allocation methods by considering the time-varying factors of carbon emissions, achieving a dual improvement in emission reduction and economic efficiency.

[0004] While related research has made some progress in specific scenarios, there is still a general lack of two-tier system modeling for power grid companies, the central hub of energy and carbon flow. Existing methods often treat quota allocation and operation scheduling separately, lacking unified intelligent algorithms for collaborative optimization, making it difficult to achieve the organic integration of quota design, operating cost control, and carbon peaking targets.

[0005] Existing carbon quota allocation methods for power grid companies rely on static historical data, making it difficult to reflect the dynamic carbon emission characteristics during power grid operation in real time. This leads to a disconnect between quota allocation results and actual dispatching and emission reduction effectiveness. Existing methods often employ single optimization algorithms or traditional linear programming, which are prone to getting trapped in local optima and cannot guarantee the global optimization and fairness of carbon quota allocation. There is a lack of joint modeling of the 24-hour power grid load distribution and time-varying carbon emission factors, failing to fully explore the carbon reduction potential during power grid operation. Quota allocation is disconnected from actual operation and dispatching, the model has insufficient closed-loop performance, and existing solutions generally suffer from low computational efficiency and slow convergence speed when dealing with large-scale data and long-term optimization. Summary of the Invention

[0006] In view of the above-mentioned problems, the present invention is proposed.

[0007] Therefore, this invention aims to propose a two-layer carbon quota optimization allocation method applicable to power grid enterprises, which can dynamically respond to time-varying carbon emission factors and multi-dimensional operation constraints of the power grid, so as to realize the scientific allocation and refined scheduling of carbon emission quotas of power grid enterprises, thereby effectively improving the fairness, economy and carbon peak control capabilities of carbon quota management.

[0008] To address the aforementioned technical problems, this invention provides the following technical solution: a collaborative optimization method for carbon quotas of power grid enterprises based on a PSO-GWO hybrid algorithm, comprising, Establish a dynamic carbon emission factor modeling mechanism to obtain the carbon emission factors of consumption nodes; based on the carbon emission factors, construct a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises to constrain the minimization of total carbon emissions; perform two-layer carbon quota allocation based on the PSO-GWO hybrid algorithm, and perform intelligent optimization scheduling based on the algorithm results.

[0009] As a preferred embodiment of the PSO-GWO hybrid algorithm-based collaborative optimization method for carbon quotas of power grid enterprises described in this invention, the dynamic carbon emission factor modeling mechanism includes constructing a dynamic carbon emission factor modeling mechanism based on the power grid operating status and power flow distribution. For each node at any given time, the carbon emission factor is calculated by combining the emission levels of the local power generation unit, the external active power flow, and the emission factors of the remaining nodes.

[0010] As a preferred embodiment of the PSO-GWO hybrid algorithm-based carbon quota collaborative optimization method for power grid enterprises described in this invention, the construction of a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises includes, based on the carbon emission factor, constructing a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises with the goal of minimizing total carbon emissions. Set the constraints as quota constraints, total quota constraints, operational constraints, and engineering constraints.

[0011] As a preferred embodiment of the power grid enterprise carbon quota collaborative optimization method based on the PSO-GWO hybrid algorithm described in this invention, the PSO-GWO hybrid algorithm includes: introducing a global-local dynamic weight allocation mechanism, constructing a two-layer collaborative feedback architecture, and feeding the quota allocation results back to the intelligent scheduling optimization in real time. A nonlinear convergence weight adjustment is adopted, a hybrid perturbation strategy is set, and a triple mechanism of elite guidance, global perturbation and local fine-tuning is introduced.

[0012] As a preferred embodiment of the power grid enterprise carbon quota collaborative optimization method based on the PSO-GWO hybrid algorithm described in this invention, the two-layer carbon quota allocation includes an exploration phase, an exploration-development conversion mechanism phase, and a development phase. The exploration phase is centered on multi-particle parallel search, combining global particle guidance in PSO and multi-leader cooperation mechanism in GWO. Each solution vector is regarded as the position of a particle / gray wolf. PSO global velocity-location update is adopted. , ,in, For particle velocity, For the individual's historical best position, The optimal position globally. For inertial weights, For acceleration coefficient, , for Uniform random numbers; In the GWO section, the three leaders—Alpha, Beta, and Delta—are used to guide collaborative fixes, and the updates are as follows: in, The parameters are linear convergence parameters. for Uniform random numbers, , , These are the results of collaborative corrections guided by the three leaders: Alpha, Beta, and Delta. , , These are correction factors obtained through guidance from different leaders, which adjust the individual's position update. for The control factor determines the step size and the magnitude of individual updates during the search process. , , The combined influence of the guidance and corrections from the three leaders. For guiding coefficients, In the particle swarm optimization algorithm, the first... The position of each particle. Optimize the first species of gray wolf The position of the wolf. , , These represent the positions of the Alpha, Beta, and Delta wolves, respectively. Representing the optimal individual within a group, it typically guides the entire group towards the optimal direction. Representing the second-best individual in the group, it plays a supporting and guiding role. It represents the third best individual in the group and provides additional guidance and correction.

[0013] Integrating two search schemes ,in, , For adaptive weights, i.e. ; The aforementioned exploration-development conversion mechanism dynamically adjusts the search and convergence weights, with the following inertia weight decreasing formula: Among them, convergence parameters Decrease with iteration; After the development phase converges to the neighborhood of the optimal solution, the focus shifts to the global optimum and high-precision fine-tuning of Alpha Wolf, introducing a target perturbation term. ,in, For perturbation factor, This is the globally optimal solution.

[0014] The beneficial effects of the preferred technical solution in this embodiment of the invention are as follows: by integrating PSO and GWO optimization algorithms, a highly efficient search, accurate optimization, and highly adaptable search mechanism are achieved. It not only avoids local optima but also ensures the stability and accuracy of the solutions, playing a positive role in the refined scheduling and scientific management of carbon quota optimization for power grid enterprises.

[0015] As a preferred embodiment of the PSO-GWO hybrid algorithm-based collaborative optimization method for carbon quotas of power grid enterprises described in this invention, the two-layer carbon quota allocation further includes selecting the top k fitness-optimal solutions and linearly combining them to generate a global guiding vector. : When the fitness change is less than the threshold, a Gaussian perturbation is introduced to escape the local extremum. ,in, For the amplitude of the disturbance, The distribution follows a normal pattern; some individuals are updated based on the global optimum. ,in, An adaptive variation factor; combining inertia weights and nonlinear perturbation step size. : in, Representing the The optimal position for each individual Indicates with individual Relevant weighting factors, Represents an individual The location of the disturbance. Dynamic perturbation terms used to adjust the intensity of perturbations in the search strategy. Represents an individual The location of the variation, Indicates the total number of iterations. This indicates the position of the global optimal solution during the iteration process.

[0016] The beneficial effects of the preferred technical solution in this embodiment of the invention are as follows: By selecting the top k fitness-optimal solutions and linearly combining them to generate a global guiding vector, the global search capability is enhanced, effectively avoiding local optima problems. When the fitness change is less than a threshold, a Gaussian perturbation mechanism is introduced to escape local extrema, and the individual position is further optimized by combining an adaptive mutation factor, thereby improving the diversity of solutions and global optima. Simultaneously, by dynamically adjusting the inertia weight and nonlinear perturbation step size, the algorithm can adaptively switch between global search and local exploitation, improving search efficiency and accuracy. This solution can accurately adjust carbon emission allocation in the carbon quota optimization of power grid enterprises, reduce resource waste, and improve the adaptability and intelligence of carbon emission management.

[0017] As a preferred embodiment of the PSO-GWO hybrid algorithm-based collaborative optimization method for carbon quotas in power grid enterprises described in this invention, the intelligent optimization scheduling includes, under the condition of satisfying constraints, using a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises directly as the fitness function. ,Right now: Where N represents a total of N different entities or devices (corresponding to various nodes, devices or regions in the power grid company, etc.), and T represents the total number of time steps, which is usually used to represent the optimization of carbon emissions or resource consumption over multiple time periods (such as daily, monthly, etc.).

[0018] Initialize the population size, maximum number of iterations, and key parameters for upper and lower layer optimization, and set the initial quota scheme. The upper layer uses a PSO-GWO hybrid mechanism to generate multiple quota allocation solutions. For each quota, the lower layer performs dynamic scheduling optimization based on 24-hour electricity load and time-varying carbon factor. If the lower layer scheme does not meet the quota or operational constraints, information is sent back to adjust the upper layer quota strategy. The current population is optimized using elite collaboration, global perturbation, and dynamic weight adjustment to prevent getting trapped in local optima. It is determined whether the iteration termination condition has been met. If it is met, the optimal quota allocation and scheduling scheme is output. Otherwise, it returns to the step of generating multiple quota allocation solutions and performs quota allocation for different power grid enterprise nodes and time periods.

[0019] The beneficial effects of the preferred technical solution in this embodiment of the invention are as follows: By using a multi-period carbon quota allocation and scheduling optimization model as the fitness function, the optimization problems of carbon emissions and resource consumption of power grid enterprises in different time periods are effectively integrated. Furthermore, through the upper-level optimization mechanism and lower-level dynamic scheduling optimization of the PSO-GWO hybrid algorithm, the quota allocation and operation strategy can be continuously adjusted in multiple iterations to ensure a balance between minimizing carbon emissions and operational efficiency. This method avoids the algorithm getting trapped in local optima through elite collaboration, global perturbation, and dynamic weight adjustment, improving search accuracy and the convergence speed of the global optimum. In addition, this solution can flexibly address the multi-dimensional constraints of power grid enterprises, improving the fairness, economy, and adaptability of carbon quota allocation, and providing power grid enterprises with a more intelligent and efficient carbon emission management strategy.

[0020] Another objective of this invention is to provide a collaborative optimization system for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm.

[0021] To solve the above-mentioned technical problems, the present invention provides the following technical solution: a collaborative optimization system for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm, comprising: a modeling module, a target optimization module, and an optimization module; The modeling module establishes a dynamic carbon emission factor modeling mechanism to obtain the electricity carbon emission coefficient of consumption nodes; The objective optimization module, based on the electricity carbon emission coefficient, constructs a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises to constrain the minimization of total carbon emissions; The optimization module performs two-layer carbon quota allocation based on the PSO-GWO hybrid algorithm and performs intelligent optimization scheduling based on the algorithm results.

[0022] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the steps of the aforementioned method for collaborative optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm.

[0023] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the steps of the aforementioned method for collaborative optimization of carbon quotas for power grid enterprises based on a PSO-GWO hybrid algorithm.

[0024] The beneficial effects of this invention are as follows: By dynamically combining the daily 24-hour load and the carbon factor, this invention effectively improves the accuracy and adaptability of carbon quota allocation, and achieves precise allocation of carbon quotas.

[0025] The PSO-GWO hybrid algorithm proposed in this invention improves search efficiency, overcomes the problem of traditional methods easily getting trapped in local extrema, and enhances the globality of overall quota optimization.

[0026] The dual-level optimization model of this invention achieves the organic integration of minimizing emission reduction costs and regulating carbon emission peak paths, significantly reducing the overall carbon emission reduction costs for enterprises and improving the controllability of carbon peaking targets. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 The following is a flowchart of an overall method for collaborative optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm, provided as an embodiment of the present invention.

[0029] Figure 2 The flowchart of the PSO-GWO algorithm for a collaborative optimization method of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm is provided as an embodiment of the present invention.

[0030] Figure 3 This is a comparison chart of daily carbon emissions, carbon quota targets, actual carbon emissions, and quota savings rate of a power grid enterprise after optimization, based on a PSO-GWO hybrid algorithm, provided as an embodiment of the present invention.

[0031] Figure 4 This is a bar chart showing the pairwise comparison of cumulative carbon emissions and carbon quotas of a power grid enterprise, based on a PSO-GWO hybrid algorithm-based collaborative optimization method for carbon quotas of power grid enterprises, as provided in an embodiment of the present invention.

[0032] Figure 5This invention provides a method for collaborative optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm, showing the daily dynamic quotas and the ratio of optimized emissions to actual emissions for a power grid enterprise. Detailed Implementation

[0033] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0034] Example 1, referring to Figure 1 and Figure 2 This is one embodiment of the present invention, which provides a method for coordinated optimization of carbon quotas for power grid enterprises based on a PSO-GWO hybrid algorithm, comprising: S100. Establish a dynamic carbon emission factor modeling mechanism to obtain carbon emission factors at consumption nodes; S200, Based on carbon emission factors, construct a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises to constrain the minimization of total carbon emissions; S300 uses a two-layer carbon quota allocation based on the PSO-GWO hybrid algorithm and performs intelligent optimization scheduling based on the algorithm results. It should be noted that the existing carbon quota allocation method of power grid companies relies on static historical data, which makes it difficult to reflect the dynamic carbon emission characteristics in real time during power grid operation, resulting in a disconnect between quota allocation results and actual dispatch and emission reduction effectiveness.

[0035] Therefore, to address the aforementioned problems, the S100-S300 steps fully integrate the operating characteristics of power grid companies and the time-varying patterns of load. This allows for intelligent optimization of carbon quota allocation at the upper level and dynamic simulation of actual scheduling at the lower level, with feedback iteration between the two layers achieving coordinated optimization. This technology not only improves the fairness and adaptability of quota allocation but also minimizes overall operating costs and achieves coordinated management of carbon peak paths, providing systematic and intelligent technical support for the green and efficient development of power grid companies.

[0036] Example 2, refer to Figure 1 This is one embodiment of the present invention, which provides a method for coordinated optimization of carbon quotas for power grid enterprises based on a PSO-GWO hybrid algorithm, comprising: This invention provides a two-layer carbon quota allocation and scheduling method based on a PSO-GWO hybrid intelligent optimization algorithm. Compared with existing linear programming and single heuristic optimization schemes, this invention constructs a carbon quota allocation optimization model at the upper layer and embeds daily dynamic scheduling simulation at the lower layer. It employs a collaborative hybrid strategy of PSO (Particle Swarm Optimization) and GWO (Grey Wolf Optimization), combining global search capability with local convergence speed. This approach fully considers the synergy between quota allocation fairness, operating costs, and carbon peaking targets, achieving dynamic closed-loop optimization of quota allocation and actual operation. This method is designed for the entire lifecycle operation data of power grid enterprises, automatically identifying time-varying patterns of load and emissions, and supporting the output of optimal quota allocation and operation schemes under complex constraints, greatly improving the intelligence and scientific level of carbon quota management.

[0037] In this embodiment of the invention, a dynamic carbon emission factor modeling mechanism is established in S100 to obtain the carbon emission factor of the consumption node, specifically as follows: Considering the significant time-varying and spatial differences in the carbon emission factors of power grid enterprises, a dynamic carbon emission factor modeling mechanism based on the power grid operating status and power flow distribution is first established. For each node at any given time, the carbon emission factor is not only related to the emission level of the local power generation unit, but also influenced by the external active power flow and the emission factors of other nodes. The model is as follows: in, Indicates the first The carbon emission coefficient of electricity at each consumption node; Indicates connection to the first Active power output of power plants at each consumption node; It is the carbon emission coefficient of grid-connected power plants; From the first The node to the first Active power input of each node; For the first A set of branches connecting nodes of a given node; It is the first The carbon emission factors of each node.

[0038] In an optional embodiment, the dynamic carbon emission factor modeling in S100 can be achieved by pre-setting fixed carbon emission factors for various power sources (such as coal power, gas power, etc.) and directly allocating static carbon emission factors according to the power generation type connected to the node. This does not change with the power grid operating status. However, this embodiment cannot reflect the real-time operating changes of the power grid, and the accuracy of the carbon emission factors is low.

[0039] In another optional embodiment, the dynamic carbon emission factor modeling in S100 can also be to calculate the historical average carbon emission factor of the entire power grid, allocate the carbon emission factor according to the proportion of electricity consumption of nodes, and ignore the power flow distribution and node differences; however, this embodiment ignores the power flow between nodes and the impact of local power generation, resulting in poor allocation fairness.

[0040] In this embodiment of the invention, S200 constructs a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises based on carbon emission factors to constrain the minimization of total carbon emissions, including the following steps S201-S202: S201. Based on carbon emission factors, construct a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises.

[0041] With the goal of minimizing total carbon emissions, and under the premise of satisfying the quota constraints at each node, the total emission constraints, and operational limitations, the model is expressed as follows: in, For the number of nodes, To optimize the number of time periods in the cycle.

[0042] S202, Constraints include: In an embodiment of the present invention, a) quota constraint: ,in, For the first Node at Carbon emission factors over a period of time For electricity consumption, Assign quotas to nodes.

[0043] In an optional embodiment, the quota constraint in S202 can be to divide the optimization period T into multiple consecutive time periods (such as weekly or monthly), and set a sub-quota for each period. During optimization, first ensure that the total carbon emissions within each period do not exceed the sub-quota for that period, and then ensure that the total emissions for the entire period T do not exceed the total quota. Sub-quota allocation can be based on historical emission ratios or average allocations, but this embodiment has a fixed phase division, which cannot flexibly adapt to real-time load fluctuations and may result in low quota utilization efficiency.

[0044] In another optional embodiment, the quota constraint in S202 can be calculated by using historical data to determine the average carbon emission rate (carbon emission per unit of electricity consumption) for each node, and then setting an overall quota value based on the predicted electricity consumption. During optimization, the total carbon emissions are directly constrained not to exceed this quota value based on the average emission rate, without considering real-time changes in emission factors; however, this embodiment ignores the dynamic nature of emission factors, fails to reflect spatiotemporal differences in actual operation, and has poor fairness.

[0045] In an embodiment of the present invention, b) total quota constraint: ,in, This refers to the total carbon quota for power grid companies.

[0046] In an optional embodiment, the total quota constraint in S202 can be achieved by collecting historical annual carbon emissions or electricity consumption data for each node, calculating the average proportion of each node's historical emissions to the total; and allocating the total carbon quota of the power grid enterprise to... The quota is directly allocated to each node according to this fixed ratio, i.e., the quota for each node. The total quota is multiplied by its historical proportion. After allocation, the quota remains fixed throughout the optimization cycle and is not adjusted according to the operating status; however, this embodiment results in a mismatch between the quota allocation and the current actual emission demand.

[0047] In another alternative embodiment, the total quota constraint in S202 can be the total carbon quota of the power grid enterprise. Directly and evenly allocating quotas to all nodes, meaning each node receives the same quota value, is a simple process that does not require consideration of node differences or operational data, only ensuring that the total is equal. However, this embodiment completely ignores differences in electricity consumption, emission levels, etc. among nodes, making the allocation too coarse. This may result in insufficient quotas for high-emission nodes or wasted quotas for low-emission nodes, leading to poor synergy.

[0048] c) Operational constraints: ,in, , The first Node at Minimum and maximum available load for a given time period.

[0049] d) Engineering constraints such as peak performance, continuity, and safety: , This indicates the maximum load variation per unit time.

[0050] In an embodiment of the present invention, S300 performs a two-layer carbon quota allocation based on the PSO-GWO hybrid algorithm and performs intelligent optimization scheduling based on the algorithm results, including the following steps S301-S302: To address the challenges of time-varying emissions across multiple time periods, the significant conflict between quota allocation fairness and operating costs, and slow convergence of control paths in the optimal allocation of carbon quotas for power grid enterprises, a two-layer carbon quota allocation and scheduling optimization method based on the PSO-GWO hybrid algorithm is proposed. This method integrates the global search advantages of PSO and the local exploitation capabilities of GWO to efficiently and intelligently complete the collaborative optimization of quota allocation and scheduling. The main innovations are reflected in the following four aspects: 1) Introducing a global-local dynamic weight allocation mechanism to enhance the multi-dimensional search capability in the quota allocation stage and the dynamic adaptability at the scheduling level; 2) Constructing a two-layer collaborative feedback architecture to feed the quota allocation results back to scheduling optimization in real time, achieving closed-loop coupling of upper and lower layer decisions; 3) Employing nonlinear convergence weight adjustment to automatically balance the phased priorities of population exploration and development; 4) Setting a hybrid perturbation strategy to dynamically introduce a triple mechanism of elite guidance, global perturbation, and local fine-tuning to avoid getting trapped in local extrema and improve convergence speed.

[0051] S301. Two-layer carbon quota allocation based on PSO-GWO hybrid algorithm includes the exploration phase, the exploration-development conversion mechanism phase, and the development phase. During the exploration phase, the core approach is multi-particle (individual) parallel search, combined with PSO's global particle guidance and GWO's multi-leader cooperative mechanism. Each solution vector is considered as the position of a particle / gray wolf. First, global velocity-position update using PSO is adopted, with the following formula: in, For particle velocity, For the individual's historical best position, The optimal position globally. For inertial weights, For acceleration coefficient, , for Uniform random numbers; In the GWO section, the three leaders—Alpha, Beta, and Delta—are used to guide collaborative fixes, and the updates are as follows: in, The parameters are linear convergence parameters. for Uniform random numbers, , , These are the results of collaborative corrections guided by the three leaders: Alpha, Beta, and Delta. , , These are correction factors obtained through guidance from different leaders, which adjust the individual's position update. for The control factor determines the step size and the magnitude of individual updates during the search process. , , The combined influence of the guidance and corrections from the three leaders. For guiding coefficients, In the particle swarm optimization algorithm, the first... The position of each particle. Optimize the first species of gray wolf The position of the wolf. , , These represent the positions of the Alpha, Beta, and Delta wolves, respectively. Representing the optimal individual within a group, it typically guides the entire group towards the optimal direction. Representing the second-best individual in the group, it plays a supporting and guiding role. It represents the third best individual in the group and provides additional guidance and correction.

[0052] Integrating two search schemes ,in, , For adaptive weights, i.e. ; The aforementioned exploration-development transition mechanism dynamically adjusts the search and convergence weights of the algorithm, improving overall stability. The inertia weight decreasing formula is as follows: Among them, convergence parameters Decrease with each iteration; achieving a transition from global to local: During the development phase, after convergence to the neighborhood of the optimal solution, the focus shifts to the global optimum and high-precision fine-tuning using Alpha Wolf. A target perturbation term is introduced: in, For perturbation factor, This is the globally optimal solution.

[0053] S302. Select the top k fitness-optimal solutions and generate a global guiding vector through a linear combination: When the fitness change is less than the threshold, a Gaussian perturbation is introduced to escape the local extremum: in, For the amplitude of the disturbance, It follows a normal distribution.

[0054] Some individuals are offset and updated based on the global optimum, enhancing diversity: in, It is an adaptive variation factor.

[0055] By combining inertial weights and nonlinear perturbation step sizes, dynamic adaptation is achieved. in, Representing the The optimal position for each individual Indicates with individual Relevant weighting factors, Represents an individual The location of the disturbance. Dynamic perturbation terms used to adjust the intensity of perturbations in the search strategy. Represents an individual The location of the variation, Indicates the total number of iterations. This indicates the position of the global optimal solution during the iteration process.

[0056] S303. Under the condition of satisfying the constraints, the multi-period carbon quota allocation and scheduling optimization model for power grid enterprises is directly used as the fitness function. ,Right now: Where N represents a total of N different entities or devices (corresponding to various nodes, devices or regions in the power grid company, etc.), and T represents the total number of time steps, which is usually used to represent the optimization of carbon emissions or resource consumption over multiple time periods (such as daily, monthly, etc.).

[0057] In an embodiment of the present invention, the population size, maximum number of iterations, and key parameters of the upper and lower optimization layers are initialized, and an initial quota scheme is set. The upper layer uses a PSO-GWO hybrid mechanism to generate multiple quota allocation solutions. For each quota, the lower layer performs dynamic scheduling optimization based on 24-hour electricity load and time-varying carbon factor. If the lower layer scheme does not meet the quota or operational constraints, information is sent back to adjust the upper layer quota strategy. Elite collaboration, global perturbation, and dynamic weight adjustment are used to optimize the current population to prevent it from getting trapped in local optima. It is determined whether the iteration termination condition has been met. If it is met, the optimal quota allocation and scheduling scheme is output; otherwise, the process returns to the step of generating multiple quota allocation solutions. Ultimately, for different power grid enterprise nodes and time periods, the goal is to achieve optimal synergy between quota allocation, dynamic scheduling, and carbon emissions, costs, and safety.

[0058] In an optional embodiment, the intelligent optimization scheduling of S303 can be the parameters for initializing the upper-level carbon quota allocation and the lower-level dynamic scheduling, including particle swarm size, maximum number of iterations, etc.; the upper layer uses the PSO algorithm to generate multiple sets of carbon quota allocation solutions, with each particle representing a quota allocation scheme, and searches are performed through velocity-position updates; the lower layer performs dynamic scheduling simulations for each set of quota schemes based on daily load and carbon emission factor to verify whether the quota constraints and operational limitations are met; if the lower-level scheduling results do not meet the constraints, information is fed back to the upper layer to adjust the optimal positions of individual particles and the global optimal positions, and a new quota solution is generated.

[0059] Iteratively update the particle swarm until the termination condition is met, and output the optimal carbon quota allocation and scheduling scheme.

[0060] In another optional embodiment, the intelligent optimization scheduling of S303 can initialize the parameters for upper-layer quota allocation and lower-layer scheduling, including wolf pack size, number of iterations, etc.; the upper layer uses the GWO algorithm to generate quota allocation solutions and updates the wolf pack positions through the guidance mechanism of Alpha, Beta, and Delta wolves; the lower layer performs dynamic scheduling simulation for each quota scheme to check the compliance of total carbon emissions with operational constraints; if the constraints are not met, the wolf pack positions are adjusted and the search is restarted based on the fitness value; through iterative convergence, the final quota and scheduling scheme are output.

[0061] Example 3, referring to Figures 3-5 This invention provides a method for collaborative optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm. To verify the beneficial effects of this invention, scientific demonstration is carried out through experiments.

[0062] against Figure 3 , Figure 3This study demonstrates the trends in carbon emissions and carbon quota allocation of power grid companies over time during the optimization process. After optimization, the fluctuation range of carbon emissions is significantly reduced, and the allocation of carbon quotas becomes more stable. These results validate that the selected optimization method based on the PSO-GWO hybrid algorithm can effectively balance grid load and carbon emission targets under multi-period scheduling, ensuring the reasonable allocation of carbon quotas across different time periods, thereby improving the stability and controllability of carbon emission management.

[0063] against Figure 4 , Figure 4 This paper presents a comparison of the ratios between actual and optimized quotas, actual and optimized emissions, and dynamic and optimized emissions. Through comparative analysis, the optimized scheme demonstrates better performance under different indicators, especially in carbon emission reduction, where it shows a significant advantage over traditional methods. This figure verifies the adaptability and effectiveness of the optimization method based on the PSO-GWO hybrid algorithm under different parameter settings, ensuring efficient achievement of carbon emission control targets under the ever-changing operating conditions of power grid companies.

[0064] against Figure 5 , Figure 5 This paper presents a comparison between daily dynamic scheduling based on an optimized scheme and traditional carbon emission scheduling. The results show that the optimized scheme, based on a PSO-GWO hybrid algorithm, significantly reduces carbon emissions and decreases the fluctuation range of scheduling results, ensuring the stability of power grid equipment operation. The figure verifies the effectiveness of the optimization method in practical applications, demonstrating that the scheme can reasonably adjust quotas and scheduling strategies under varying grid load conditions, thereby improving the accuracy and economy of carbon emission management.

[0065] Example 4 is an embodiment of the present invention. The above is an illustrative scheme of a collaborative optimization method for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm. It should be noted that the technical solution of a collaborative optimization system for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm and the technical solution of the collaborative optimization method for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm described above belong to the same concept. Details not described in detail in the technical solution of the collaborative optimization system for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm in this embodiment can be found in the description of the technical solution of the collaborative optimization method for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm described above.

[0066] This embodiment provides a collaborative optimization system for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm, including: a modeling module, a target optimization module, and an optimization module; The modeling module establishes a dynamic carbon emission factor modeling mechanism to obtain the electricity carbon emission coefficient of consumption nodes; The objective optimization module, based on the electricity carbon emission coefficient, constructs a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises to constrain the minimization of total carbon emissions; The optimization module performs two-layer carbon quota allocation based on the PSO-GWO hybrid algorithm and performs intelligent optimization scheduling based on the algorithm results.

[0067] This embodiment also provides an electronic device applicable to a collaborative optimization method for carbon quotas of power grid enterprises based on a PSO-GWO hybrid algorithm, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the collaborative optimization method for carbon quotas of power grid enterprises based on a PSO-GWO hybrid algorithm as proposed in the above embodiment.

[0068] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements a collaborative optimization method for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm proposed in the above embodiment.

[0069] The storage medium proposed in this embodiment belongs to the same inventive concept as the method for collaborative optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0070] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0071] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A collaborative optimization method for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm, characterized in that: include, Establish a dynamic carbon emission factor modeling mechanism to obtain carbon emission factors at consumption nodes; Based on carbon emission factors, a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises is constructed to constrain the minimization of total carbon emissions. A two-layer carbon quota allocation is performed based on the PSO-GWO hybrid algorithm, and intelligent optimization scheduling is carried out based on the algorithm results.

2. The method for coordinated optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm as described in claim 1, characterized in that: The dynamic carbon emission factor modeling mechanism includes constructing a dynamic carbon emission factor modeling mechanism based on the power grid operating status and power flow distribution. For each node at any given time, the carbon emission factor is calculated by combining the emission levels of the local power generation unit, the external active power flow, and the emission factors of the remaining nodes.

3. The method for coordinated optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm as described in claim 2, characterized in that: The construction of a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises includes, based on the carbon emission factor, constructing a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises with the goal of minimizing total carbon emissions; Set the constraints as quota constraints, total quota constraints, operational constraints, and engineering constraints.

4. The method for coordinated optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm as described in claim 3, characterized in that: The PSO-GWO hybrid algorithm includes introducing a global-local dynamic weight allocation mechanism, constructing a two-layer collaborative feedback architecture, and feeding back the quota allocation results to the intelligent scheduling optimization in real time. A nonlinear convergence weight adjustment is adopted, a hybrid perturbation strategy is set, and a triple mechanism of elite guidance, global perturbation and local fine-tuning is introduced.

5. The method for coordinated optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm as described in claim 4, characterized in that: The two-layer carbon quota allocation includes, based on the PSO-GWO hybrid algorithm, a two-layer carbon quota allocation including an exploration phase, an exploration-development conversion mechanism phase, and a development phase. The exploration phase is centered on multi-particle parallel search, combining global particle guidance in PSO and multi-leader cooperation mechanism in GWO. Each solution vector is regarded as the position of a particle / gray wolf. PSO global velocity-location update is adopted. , ,in, For particle velocity, For the individual's historical best position, The optimal position globally. For inertial weights, For acceleration coefficient, , for Uniform random numbers; In the GWO section, the three leaders—Alpha, Beta, and Delta—are used to guide collaborative fixes, and the updates are as follows: in, The parameters are linear convergence parameters. for Uniform random numbers, , , These are the results of collaborative corrections guided by the three leaders: Alpha, Beta, and Delta. , , These are correction factors obtained through guidance from different leaders, which adjust the individual's position update. for The control factor determines the step size and the magnitude of individual updates during the search process. , , The combined influence of the guidance and corrections from the three leaders. For guiding coefficients, In the particle swarm optimization algorithm, the first... The position of each particle. Optimize the first species of gray wolf The position of the wolf. , , These represent the positions of the Alpha, Beta, and Delta wolves, respectively. Represents the best individual in the group. The second best individual in the group. This represents the third best individual in the group. Integrating two search schemes ,in, , For adaptive weights, i.e. ; The aforementioned exploration-development conversion mechanism dynamically adjusts the search and convergence weights, with the following inertia weight decreasing formula: Among them, convergence parameters Decrease with iteration; After the development phase converges to the neighborhood of the optimal solution, the focus shifts to the global optimum and high-precision fine-tuning of Alpha Wolf, introducing a target perturbation term. ,in, For perturbation factor, This is the globally optimal solution.

6. The method for coordinated optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm as described in claim 5, characterized in that: The two-layer carbon quota allocation also includes selecting the top k fitness-optimal solutions and generating a global guiding vector through linear combination. : When the fitness change is less than the threshold, a Gaussian perturbation is introduced to escape the local extremum. ,in, For the amplitude of the disturbance, The distribution follows a normal pattern; some individuals are updated based on the global optimum. ,in, An adaptive variation factor; combining inertia weights and nonlinear perturbation step size. : in, Representing the The optimal position for each individual Indicates with individual Relevant weighting factors, Represents an individual The location of the disturbance. Dynamic perturbation terms used to adjust the intensity of perturbations in the search strategy. Represents an individual The location of the variation, Indicates the total number of iterations. This indicates the position of the global optimal solution during the iteration process.

7. The method for coordinated optimization of carbon quotas for power grid enterprises based on the PSO-GWO hybrid algorithm as described in claim 6, characterized in that: The intelligent optimization scheduling includes, under the condition of satisfying constraints, using a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises directly as the fitness function. ,Right now: Where N represents the total number of different entities or devices, and T represents the total number of time steps, which is usually used to represent the optimization of carbon emissions or resource consumption over multiple time periods. Initialize the population size, maximum number of iterations, and key parameters for upper and lower layer optimization, and set the initial quota scheme. The upper layer uses a PSO-GWO hybrid mechanism to generate multiple quota allocation solutions. For each quota, the lower layer performs dynamic scheduling optimization based on 24-hour electricity load and time-varying carbon factor. If the lower layer scheme does not meet the quota or operational constraints, information is sent back to adjust the upper layer quota strategy. The current population is optimized using elite collaboration, global perturbation, and dynamic weight adjustment to prevent getting trapped in local optima. It is determined whether the iteration termination condition has been met. If it is met, the optimal quota allocation and scheduling scheme is output. Otherwise, it returns to the step of generating multiple quota allocation solutions and performs quota allocation for different power grid enterprise nodes and time periods.

8. A power grid enterprise carbon quota collaborative optimization system based on a PSO-GWO hybrid algorithm, employing the power grid enterprise carbon quota collaborative optimization method based on a PSO-GWO hybrid algorithm as described in any one of claims 1 to 7, characterized in that, include: Modeling module, target optimization module, optimization module; The modeling module establishes a dynamic carbon emission factor modeling mechanism to obtain the electricity carbon emission coefficient of consumption nodes; The objective optimization module, based on the electricity carbon emission coefficient, constructs a multi-period carbon quota allocation and scheduling optimization model for power grid enterprises to constrain the minimization of total carbon emissions; The optimization module performs two-layer carbon quota allocation based on the PSO-GWO hybrid algorithm and performs intelligent optimization scheduling based on the algorithm results.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the collaborative optimization method for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the collaborative optimization method for carbon quotas of power grid enterprises based on the PSO-GWO hybrid algorithm as described in any one of claims 1 to 7.

Images