A carbon quota planning method and system

Abstract

This invention discloses a carbon quota planning method and system, relating to the field of carbon emission reduction technology. It uses minimizing daily emissions as the optimization objective and daily energy consumption sequences as decision variables to construct a lower-level optimization model. An initial particle swarm is built using daily quota sequences as particles. The lower-level optimization model is optimized and solved using each daily quota sequence, resulting in multiple optimized daily energy consumption sequences. The initial particle swarm is then iteratively solved using each daily energy consumption sequence and a pre-set upper-level optimization model to obtain the corresponding carbon quota planning results. This invention solves the technical problem that traditional carbon quota planning often relies primarily on past carbon emission data for allocation, failing to effectively capture daily / weekly fluctuations in load curves, time-varying emission factors caused by production line switching and process cycle time, thus reducing the reliability of carbon quota planning.

Description

Technical Field

[0001] This invention relates to the field of carbon emission reduction technology, and in particular to a carbon quota planning method and system. Background Technology

[0002] Key industries such as power electronics, electric vehicles, lithium batteries, and foreign trade manufacturing are critical areas of energy consumption and carbon emissions. On the one hand, they need to meet the realistic development needs of production scale expansion and load fluctuations, and on the other hand, they also face increasingly stringent compliance pressures such as ESG disclosure and carbon tariffs (CBAM). Therefore, the scientific planning and refined implementation of carbon quotas are becoming increasingly critical.

[0003] Currently, traditional carbon quota planning mainly relies on enterprises' past carbon emission data for carbon quota allocation. However, it cannot effectively capture the daily / weekly fluctuations of the load curve, the time-varying emission factors caused by production line switching and process cycle time, which reduces the reliability of carbon quota planning. Summary of the Invention

[0004] This invention provides a carbon quota planning method and system, which solves the technical problem that traditional carbon quota planning mainly relies on the company's past carbon emission data for carbon quota allocation, but cannot effectively capture the daily / weekly fluctuations of the load curve, the time-varying emission factors caused by production line switching and process cycle time, and reduces the reliability of carbon quota planning.

[0005] The first aspect of this invention provides a carbon quota planning method, comprising: With the goal of minimizing daily emissions and the decision variable of intraday energy consumption sequence, a lower-level optimization model is constructed. An initial particle swarm is constructed using the daily quota sequence as particles, and the lower-level optimization model is optimized and solved using each of the daily quota sequences to obtain multiple optimized intraday energy consumption sequences; The initial particle swarm is iteratively solved using the various intraday energy consumption sequences and the preset upper-level optimization model to obtain the corresponding carbon quota planning results.

[0006] Optionally, the step of optimizing the lower-level optimization model using each of the daily quota sequences to obtain multiple optimized intraday energy consumption sequences includes: Construct an initial intraday energy consumption population for each of the said daily quota sequences, wherein each individual in the initial intraday energy consumption population corresponds to an intraday energy consumption sequence; Each of the intraday energy consumption sequences and the preset hourly carbon data are input into the lower-level optimization model to obtain multiple first fitness values; The first fitness value corresponding to each initial intraday energy consumption population is used for iterative optimization to obtain multiple optimized intraday energy consumption sequences.

[0007] By adopting the above technical solution, a corresponding refined intraday energy consumption sequence can be matched for each daily quota sequence. At the same time, hourly carbon data is combined with intraday energy consumption sequences and input into the lower-level optimization model. The first fitness value is used to complete the quantitative evaluation of each intraday energy consumption sequence, so that the iterative optimization has a clear judgment basis and ensures that the optimization direction always revolves around the core goal of minimizing daily emissions. Then, guided by the first fitness value, iterative optimization is carried out on the initial intraday energy consumption population, gradually screening and optimizing the optimal intraday energy consumption sequence that is suitable for the corresponding daily quota sequence, so as to achieve precise optimization of hourly energy consumption scheduling under daily quota constraints. The overall technical solution achieves precise matching and optimization between the daily quota sequence and the intraday energy consumption sequence. It progressively completes the entire process from constructing multiple candidate schemes, quantitative evaluation to iterative optimization. It can accurately capture hourly carbon factor and energy load changes, so that the results of the lower-level optimization are consistent with the time-varying characteristics of actual production. It can also ensure that each daily quota sequence can obtain the intraday energy consumption scheduling scheme with optimal adaptability. It provides accurate and reliable hourly energy consumption and emission data support for the global iterative optimization of the upper-level carbon quota, and effectively improves the scientificity and accuracy of the lower-level solution in the two-level optimization model.

[0008] Optionally, the step of iteratively optimizing each initial intraday energy consumption population using its corresponding first fitness value to obtain multiple optimized intraday energy consumption sequences includes: Each initial intraday energy consumption population is updated using its first fitness value to obtain multiple new initial intraday energy consumption populations. Determine whether the number of iterations for each of the initial daily energy-consuming populations is greater than or equal to a preset first iteration threshold. When the number of iterations of the initial intraday energy consumption population is less than the first iteration threshold, the process jumps to the step of inputting each intraday energy consumption sequence and the preset hourly carbon data into the lower-level optimization model to obtain multiple first fitness values. When the number of iterations of the initial intraday energy consumption population is greater than or equal to the first iteration threshold, the intraday energy consumption sequence corresponding to the minimum value among the first fitness values ​​associated with the initial intraday energy consumption population is selected as the optimized intraday energy consumption sequence.

[0009] By adopting the above technical solution, the initial daily energy consumption population is updated in a targeted manner based on the first fitness value. This ensures that the population iteration always revolves around the optimization goal of minimizing daily emissions, guaranteeing the directionality and effectiveness of the population update and avoiding aimless iterative adjustments. At the same time, by setting a first iteration threshold and judging the number of population iterations one by one, a clear convergence standard is set for the iteration process of the lower-level optimization. This allows the population to fully iterate and seek optimization within the threshold range, discovering better solutions for the daily energy consumption sequence, while preventing the waste of computational resources caused by excessive iteration, thus improving the solution efficiency of the lower-level optimization. The cyclic solution design when the number of iterations has not reached the threshold realizes the dynamic iterative optimization of the initial daily energy consumption population, allowing the population to continuously update and evolve, constantly approaching the optimal solution. After the number of iterations reaches the threshold, the daily energy consumption sequence corresponding to the minimum first fitness value is selected as the optimization result, which can accurately lock the hourly energy consumption scheduling scheme with the optimal carbon emissions under the corresponding daily quota sequence.

[0010] Optionally, the step of updating each initial intraday energy consumption population using the first fitness value corresponding to each initial intraday energy consumption population to obtain multiple new initial intraday energy consumption populations includes: Based on the first fitness value, individuals in each of the initial daily energy consumption populations are sorted in ascending order, and each sorted initial daily energy consumption population is truncated and selected according to a preset number of divisions to obtain multiple elite daily energy consumption populations and multiple non-elite daily energy consumption populations. Growth traction operations are performed on the corresponding non-elite daily energy-consuming populations based on each of the elite daily energy-consuming populations to obtain multiple growth traction populations. Fire disturbance operations were performed on each of the aforementioned growth-driving populations to obtain multiple fire-disturbed populations; Crossover or mutation operations are performed on each of the aforementioned fire disturbance populations to obtain multiple target fire disturbance populations; A local search is performed on each of the target fire disturbance populations, and each target fire disturbance population after the local search is merged with the corresponding elite daily energy consumption population to obtain multiple new initial daily energy consumption populations.

[0011] By employing the above technical solutions, and sorting and truncating each initial intraday energy-consuming population according to its first fitness value, it is possible to quickly identify elite intraday energy-consuming populations with better carbon emission optimization effects within the initial intraday energy-consuming population, achieving precise retention of superior individuals. Simultaneously, it identifies non-elite populations to be optimized, providing clear optimization targets and directions for subsequent population updates and preventing the loss of high-quality solutions. Growth traction operations, centered on elite intraday energy-consuming populations, guide non-elite intraday energy-consuming populations towards superior energy-consuming characteristics, accelerating the optimization process while preserving basic population diversity and effectively preventing the population from falling into local optima. Fire perturbation operations are implemented on the growth-traction populations, introducing new solution spaces through random perturbation, further enhancing the global search capability of the growth-traction populations and making the optimization of intraday energy-consuming sequences more comprehensive. Combining crossover or mutation operations to optimize the fire-perturbation populations allows for the achievement of superior energy consumption for different individuals. The recombination of features and the introduction of new features continuously enrich the solution space of the population, allowing the population to maintain its evolutionary vitality. Local search of the target fire-perturbed population can focus on the neighborhood of the current optimal solution for fine-tuning, and discover better local feasible solutions, thereby improving the accuracy and stability of the daily energy consumption sequence optimization. The population after local search is then merged with the elite population to obtain a new initial daily energy consumption population, which greatly improves the efficiency and effect of the iterative optimization of the lower-level optimization model, making the optimized daily energy consumption sequence more in line with the time-varying characteristics and daily quota constraints of actual production.

[0012] Optionally, the step of iteratively solving the initial particle swarm using each of the intraday energy consumption sequences and a preset upper-level optimization model to obtain the corresponding carbon quota planning results includes: The daily quota sequence and daily energy consumption sequence corresponding to each particle in the initial particle swarm are respectively input into a preset upper-level optimization model to obtain multiple second fitness values; The initial particle swarm is updated using each of the second fitness values ​​to obtain the corresponding updated particle swarm; Based on the preset annual quota, a consistency callback is performed on the updated particle swarm to obtain a new initial particle swarm. When the number of iterations of the initial particle swarm is less than the preset second iteration threshold, the process jumps to the step of using each of the daily quota sequences to optimize and solve the lower-level optimization model to obtain multiple optimized intraday energy consumption sequences. When the number of iterations of the initial particle swarm is greater than or equal to the second iteration threshold, the daily energy consumption sequence and daily quota sequence corresponding to the minimum value among the second fitness values ​​are selected as the corresponding carbon quota planning results.

[0013] By adopting the above technical solution, the initial particle swarm is updated based on the second fitness value. Combined with the global search characteristics of the particle swarm optimization algorithm, the particle swarm evolves towards a better daily quota sequence, effectively mining the optimal carbon quota allocation scheme globally and avoiding the limitations of local optima. A consistency callback operation is performed based on the annual quota total, strictly controlling the annual total carbon quota constraint and ensuring that the cumulative value of the daily quota sequence corresponding to each particle meets the annual emission ceiling requirements, allowing the carbon quota planning results to meet the actual compliance needs of enterprises. By setting a second iteration threshold and determining the number of iterations of the particle swarm, a scientific convergence standard is set for the upper-level global iteration, enabling the particle swarm to evolve towards a better daily quota sequence. Within the threshold range, the group fully completes the two-layer collaborative iteration, realizing the dynamic matching and optimization of the daily quota sequence and the intraday energy consumption sequence, while avoiding the consumption of computing resources caused by excessive iteration and improving the overall solution efficiency of carbon quota planning. The two-layer cyclical iteration design when the number of iterations does not reach the threshold realizes the deep collaboration and dynamic optimization between the upper-level carbon quota allocation and the lower-level hourly energy consumption scheduling, so that the adjustment of the daily quota sequence can match the optimization results of the lower-level energy consumption scheduling in real time, ensuring the refinement and adaptability of carbon quota planning. After the number of iterations reaches the threshold, the sequence corresponding to the minimum value of the second fitness value is selected as the planning result, which can accurately lock the globally optimal carbon quota allocation scheme and the corresponding hourly energy consumption scheduling scheme.

[0014] Optionally, the lower-level optimization model is specifically: ; in, For the first d The first adaptation value of the day, For the first d day h Energy consumption per hour For the first d day h hourly carbon factor For the first d Daily quota, h Indexed by hour. This is the second penalty coefficient for exceeding the quota. The lower bound of the hour. The upper limit for hours, d For daily indexing.

[0015] A carbon quota planning system provided by a second aspect of the present invention includes: The module is used to build a lower-level optimization model with the goal of minimizing daily emissions and the decision variable of intraday energy consumption sequence. The lower-level optimization module is used to construct an initial particle swarm with the daily quota sequence as particles, and to optimize and solve the lower-level optimization model using each of the daily quota sequences to obtain multiple optimized intraday energy consumption sequences. The upper-level optimization module is used to iteratively solve the initial particle swarm using each of the intraday energy consumption sequences and a preset upper-level optimization model to obtain the corresponding carbon quota planning results.

[0016] A third aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the steps of the carbon quota planning method as described above.

[0017] The fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the carbon quota planning method as described above.

[0018] The fifth aspect of the present invention provides a computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein when the program instructions are executed by a computer, the computer performs the carbon quota planning method as described above.

[0019] As can be seen from the above technical solutions, the present invention has the following advantages: This invention constructs a lower-level optimization model with the goal of minimizing daily emissions and using daily energy consumption sequences as decision variables. An initial particle swarm is built using daily quota sequences as particles. The lower-level optimization model is then optimized using each daily quota sequence to obtain multiple optimized daily energy consumption sequences. The initial particle swarm is iteratively solved using each daily energy consumption sequence and a pre-set upper-level optimization model to obtain the corresponding carbon quota planning results. This overcomes the technical problem of traditional carbon quota planning, which often relies primarily on past carbon emission data for allocation but fails to effectively capture daily / weekly fluctuations in load curves, time-varying emission factors caused by production line switching and process cycles, thus reducing the reliability of carbon quota planning. Compared to traditional carbon quota planning methods, this invention uses the goal of minimizing daily emissions and daily energy consumption sequences as decision variables to construct a lower-level optimization model. This model accurately captures hourly load fluctuations and time-varying carbon factors. Furthermore, based on the lower-level optimization model, a particle swarm optimization algorithm is used to perform a global search on the daily quota sequences to obtain the corresponding carbon quota planning results, thereby improving the reliability of carbon quota planning. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0021] Figure 1 This is a flowchart of the steps of a carbon quota planning method provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart of the steps of a carbon quota planning method provided in Embodiment 2 of the present invention; Figure 3 This is a comparison chart of daily optimized carbon emissions, carbon quota targets, actual carbon emissions, and quota savings rate for a key industry enterprise, provided in Embodiment 2 of the present invention. Figure 4 This is a comparison chart of cumulative carbon emissions and carbon quotas of a key industry enterprise provided in Embodiment 2 of the present invention; Figure 5 This is a diagram showing the daily dynamic quota and optimized emission ratio of a key industry enterprise as provided in Embodiment 2 of the present invention. Figure 6 This is a structural block diagram of a carbon quota planning system provided in Embodiment 3 of the present invention; Figure 7 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0022] This invention provides a carbon quota planning method and system to address the technical problem that traditional carbon quota planning often relies on enterprises' past carbon emission data for allocation, but fails to effectively capture intraday / weekly fluctuations in load curves, time-varying emission factors caused by production line switching and process cycle time, thus reducing the reliability of carbon quota planning.

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that in the optional embodiments of the present invention, the object information and other related data involved require the permission or consent of the object when the embodiments of the present invention are applied to specific products or technologies, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. That is to say, if the embodiments of the present invention involve data related to the object, it needs to be obtained with the authorization and consent of the object, the authorization and consent of the relevant departments, and in compliance with the relevant laws, regulations, and standards of the country and region. If personal information is involved in the embodiments, the acquisition of all personal information requires the consent of the individual. If sensitive information is involved, the separate consent of the information subject is required, and the embodiments also need to be implemented with the authorization and consent of the object.

[0024] Please see Figure 1 , Figure 1 This is a flowchart illustrating the steps of a carbon quota planning method provided in Embodiment 1 of the present invention.

[0025] This invention provides a carbon quota planning method, comprising: Step 101: With the minimum daily emissions as the optimization objective and the daily energy consumption sequence as the decision variable, construct the lower-level optimization model.

[0026] Minimizing daily emissions means optimizing the scheduling of energy consumption for each hour of the 24-hour period to minimize the total carbon emissions of an enterprise within a single natural day.

[0027] The lower-level optimization model refers to a mathematical model that focuses on the daily hourly energy scheduling and carbon emission optimization of an enterprise. It takes the minimum daily emissions as the objective, the daily energy consumption sequence as the decision variable, and integrates various actual production constraints.

[0028] Intraday energy consumption sequence refers to the set of decision variables, that is, the combination of energy consumption values ​​for each hour within a single natural day (24 hours).

[0029] In this embodiment of the invention, the optimization objective is to minimize daily emissions, and the decision variable is the daily energy consumption sequence. A lower-level optimization model is constructed, which focuses on the refined energy consumption scheduling of 24 hours a day. By integrating hourly carbon factors, equipment operating boundaries and over-quota constraints, the model achieves synergistic optimization of daily emissions and energy costs.

[0030] It is worth mentioning that minimizing daily emissions is taken as the core objective of the lower-level optimization model. By optimizing and scheduling hourly energy consumption every day, the model promotes the minimization of daily carbon emissions for enterprises from the source.

[0031] It is worth mentioning that the lower-level optimization model uses the daily energy consumption sequence as the decision variable, which can accurately capture the daily / weekly fluctuations of the enterprise's load curve, the hourly time-varying emission factors brought about by production line switching and process cycle time. It replaces the extensive planning method that relies solely on historical static carbon emission data in the traditional approach, making the allocation of carbon quotas compatible with the dynamic characteristics of the enterprise's actual carbon emissions, and improving the accuracy and reliability of carbon quota planning based on data.

[0032] Step 102: Construct an initial particle swarm using the daily quota sequence as particles, and use each daily quota sequence to optimize and solve the lower-level optimization model to obtain multiple optimized intraday energy consumption sequences.

[0033] The daily quota sequence refers to the set of daily carbon quotas within the optimization period.

[0034] The initial particle swarm refers to a set consisting of multiple daily quota sequences.

[0035] The optimized intraday energy consumption sequence refers to the hourly energy consumption scheduling scheme that optimizes carbon emissions during a natural day.

[0036] In this embodiment of the invention, an initial particle swarm is constructed using daily quota sequences as particles, where the daily quota sequence is a set of daily quota allocation schemes covering all natural days within the optimization period. For each daily quota sequence, an initial intraday energy consumption population corresponding to each natural day is constructed one by one, where each individual in the initial intraday energy consumption population corresponds to an intraday energy consumption sequence. Each intraday energy consumption sequence and preset hourly carbon data are input into the lower-level optimization model to obtain multiple first fitness values. Iterative optimization is performed using the first fitness values ​​corresponding to each initial intraday energy consumption population to obtain multiple optimized intraday energy consumption sequences.

[0037] It is worth mentioning that by constructing an initial particle swarm using the daily quota sequence as particles and conducting iterative solutions, and by leveraging the global search characteristics of the particle swarm algorithm, the daily quota allocation scheme within the optimization cycle can be optimized in all aspects. Under the annual total constraint, the globally optimal daily quota sequence can be found, avoiding the subjectivity and limitations of traditional manual allocation or simple equal distribution methods. This makes the daily allocation of carbon quotas more scientific and reasonable, and achieves balanced management of carbon emissions throughout the entire cycle.

[0038] Step 103: Iteratively solve the initial particle swarm using the daily energy consumption sequences and the preset upper-level optimization model to obtain the corresponding carbon quota planning results.

[0039] The upper-level optimization model refers to a mathematical model that focuses on the global allocation of carbon quotas. It takes the daily quota sequence as the core optimization object, combines constraints such as the total annual quota and quota smoothness, and iteratively optimizes the initial particle swarm to output the globally optimal daily quota allocation scheme.

[0040] Carbon quota planning results refer to the final set of optimized solutions, including the optimized daily quota sequence (globally optimal quota allocation) and the corresponding daily energy consumption sequence (hourly executable operation plan), providing enterprises with a three-level linkage basis for carbon quota management of "annual-daily-hourly".

[0041] In this embodiment of the invention, the daily quota sequence and intraday energy consumption sequence corresponding to each particle in the initial particle swarm are input into a preset upper-level optimization model to obtain multiple second fitness values. Based on each second fitness value, the initial particle swarm is updated using the particle swarm velocity-position update rule to generate an updated particle swarm that balances global search and local optimization. A consistency callback operation is performed on the updated particle swarm based on a preset annual quota total to obtain a new initial particle swarm. When the number of iterations of the initial particle swarm is less than a preset second iteration threshold, step 102 is executed. When the number of iterations of the initial particle swarm is greater than or equal to the second iteration threshold, the intraday energy consumption sequence and daily quota sequence corresponding to the minimum value among the second fitness values ​​are selected as the corresponding carbon quota planning result.

[0042] In this embodiment of the invention, a lower-level optimization model is constructed with the goal of minimizing daily emissions and the daily energy consumption sequence as the decision variable. An initial particle swarm is constructed using the daily quota sequence as particles. The lower-level optimization model is then optimized using each daily quota sequence to obtain multiple optimized daily energy consumption sequences. The initial particle swarm is iteratively solved using each daily energy consumption sequence and a preset upper-level optimization model to obtain the corresponding carbon quota planning results. This overcomes the technical problem that traditional carbon quota planning often relies on past carbon emission data of enterprises for carbon quota allocation, but fails to effectively capture the daily / weekly fluctuations of the load curve, the time-varying emission factors caused by production line switching and process cycle time, thus reducing the reliability of carbon quota planning. Compared with traditional carbon quota planning methods, this invention uses the goal of minimizing daily emissions and the daily energy consumption sequence as the decision variable to construct a lower-level optimization model, accurately capturing hourly load fluctuations and time-varying carbon factors. Based on the lower-level optimization model, a particle swarm algorithm is combined to perform a global search on the daily quota sequence to obtain the corresponding carbon quota planning results, thus improving the reliability of carbon quota planning.

[0043] Please see Figure 2 , Figure 2 This is a flowchart illustrating the steps of a carbon quota planning method provided in Embodiment 2 of the present invention.

[0044] This invention provides a carbon quota planning method, comprising: Step 201: With the minimum daily emissions as the optimization objective and the daily energy consumption sequence as the decision variable, construct a lower-level optimization model.

[0045] It should be noted that the lower-level optimization model is as follows:

[0046] in, For the first d The first adaptation value of the day, For the first d day h Energy consumption per hour For the first d day h hourly carbon factor For the first d Daily quota, h Indexed by hour. This is the second penalty coefficient for exceeding the quota. The lower bound of the hour. The upper limit for hours, d For daily indexing.

[0047] In this embodiment of the invention, the optimization objective is to minimize daily emissions, and the decision variable is the daily energy consumption sequence, which is used to construct a lower-level optimization model.

[0048] It should be noted that the first d The daily quota is allocated by the upper level to the first... d The daily carbon emission cap. The over-quota secondary penalty coefficient is a non-linear penalty weight used to constrain over-quota emissions. By imposing a secondary penalty on the over-quota portion, it guides the energy consumption sequence to conform to the daily quota constraint.

[0049] Step 202: Construct an initial particle swarm using the daily quota sequence as particles, and construct an initial daily energy consumption population for each daily quota sequence, wherein each individual in the initial daily energy consumption population corresponds to a daily energy consumption sequence.

[0050] In this embodiment of the invention, an initial particle swarm is constructed using daily quota sequences as particles. An initial intraday energy consumption population is constructed for each daily quota sequence, wherein each individual in the initial intraday energy consumption population corresponds to an intraday energy consumption sequence. For example, if a daily quota sequence includes carbon quotas for multiple days, an initial intraday energy consumption population is constructed for each day (i.e., a daily quota sequence corresponds to multiple initial intraday energy consumption populations).

[0051] Step 203: Input each intraday energy consumption sequence and the preset hourly carbon data into the lower-level optimization model to obtain multiple first fitness values.

[0052] Hourly carbon data refers to a time-series dataset that includes hourly carbon factors, recording carbon emission intensity information for each hour of the day within the optimization period.

[0053] The first fitness value refers to the quantitative evaluation index obtained after inputting the intraday energy consumption sequence into the lower-level optimization model. The value reflects the carbon emission optimization effect of the corresponding intraday energy consumption sequence, and the smaller the value, the better the optimization effect.

[0054] In this embodiment of the invention, each daily energy consumption sequence and preset hourly carbon data are input into the lower-level optimization model. The hourly carbon data includes the dynamic carbon factor corresponding to each 24-hour period. Combined with the preset daily quota, over-quota penalty coefficient and hourly energy consumption upper and lower bound constraints in the model, the first fitness value corresponding to each daily energy consumption sequence is calculated.

[0055] Step 204: Iteratively optimize each initial intraday energy consumption population using its first fitness value to obtain multiple optimized intraday energy consumption sequences.

[0056] It should be noted that steps 202-204 can match a corresponding refined intraday energy consumption sequence for each daily quota sequence; at the same time, the hourly carbon data is combined with the intraday energy consumption sequence and input into the lower-level optimization model. The first fitness value is used to complete the quantitative evaluation of each intraday energy consumption sequence, so that the iterative optimization has a clear judgment basis and ensures that the optimization direction always revolves around the core goal of minimizing daily emissions; then, guided by the first fitness value, iterative optimization is carried out on the initial intraday energy consumption population, gradually screening and optimizing the optimal intraday energy consumption sequence that is suitable for the corresponding daily quota sequence, so as to achieve precise optimization of hourly energy consumption scheduling under the daily quota constraint. The overall technical solution achieves precise matching and optimization between the daily quota sequence and the intraday energy consumption sequence. It progressively completes the entire process from constructing multiple candidate schemes, quantitative evaluation to iterative optimization. It can accurately capture hourly carbon factor and energy load changes, so that the results of the lower-level optimization are consistent with the time-varying characteristics of actual production. It can also ensure that each daily quota sequence can obtain the intraday energy consumption scheduling scheme with optimal adaptability. It provides accurate and reliable hourly energy consumption and emission data support for the global iterative optimization of the upper-level carbon quota, and effectively improves the scientificity and accuracy of the lower-level solution in the two-level optimization model.

[0057] Further, step 204 includes the following sub-steps: S11. Update each initial intraday energy-consuming population using the first fitness value corresponding to each initial intraday energy-consuming population to obtain multiple new initial intraday energy-consuming populations.

[0058] Furthermore, S11 includes the following sub-steps: S111. Sort the individuals of each initial daily energy-consuming population in ascending order according to the first fitness value, and truncate each sorted initial daily energy-consuming population according to the preset division number to obtain multiple elite daily energy-consuming populations and multiple non-elite daily energy-consuming populations.

[0059] Ascending sorting refers to the sorting method that arranges individuals in the initial daily energy consumption population in ascending order of their first fitness value. Individuals with smaller first fitness values ​​are ranked higher, indicating that their corresponding daily energy consumption sequence has a better carbon emission optimization effect.

[0060] The preset division number refers to the threshold number of individuals used to distinguish between elite and non-elite populations, which is usually set at 30%-50%.

[0061] Cut-off selection refers to a population selection method that selects the top half of the sorted population based on a preset number of divisions to form an elite population, while the remaining individuals form a non-elite population, thus preserving superior individuals and selecting inferior ones.

[0062] Elite daily energy consumption populations refer to the high-quality populations obtained after truncation selection. They consist of individuals with higher ranking and better first fitness values, possessing superior daily energy consumption configuration characteristics, and providing guidance for the optimization of non-elite populations.

[0063] Non-elite daily energy-consuming populations refer to the populations remaining after truncation selection, consisting of individuals with lower ranking and poorer first fitness values, which need to be optimized and updated through operations such as growth traction and fire disturbance.

[0064] In this embodiment of the invention, individuals in each initial daily energy-consuming population are sorted in ascending order according to a first fitness value (i.e., arranged from smallest to largest according to the first fitness value corresponding to the individual). Each sorted initial daily energy-consuming population is then truncated according to a preset number of partitions, resulting in multiple elite daily energy-consuming populations and multiple non-elite daily energy-consuming populations. For example, for a sorted initial daily energy-consuming population, the top 50% of individuals are selected as the elite daily energy-consuming population, and the remaining individuals are selected as the non-elite daily energy-consuming population.

[0065] S112. Perform growth traction operations on the corresponding non-elite daily energy-consuming populations according to each elite daily energy-consuming population to obtain multiple growth traction populations.

[0066] Growth-guided operations refer to population optimization operations that adjust the values ​​of individuals in non-elite daily energy-consuming populations based on the superior characteristics of elite daily energy-consuming populations. The core is to make non-elite individuals move closer to the characteristics of elite individuals while preserving population diversity.

[0067] Growth-driven populations refer to intermediate populations formed by growth-driven operations on non-elite daily energy-consuming populations. Their individuals have begun to resemble the superior energy-consuming characteristics of elite populations, while retaining certain differences.

[0068] In this embodiment of the invention, growth traction operations are performed on the corresponding non-elite daily energy consumption populations according to each elite daily energy consumption population. The average hourly energy consumption of all individuals in the elite daily energy consumption population is taken as the traction center. Combined with the traction noise intensity that decreases with the iteration process, the hourly energy consumption values ​​of individuals in the non-elite daily energy consumption populations are adjusted dimension by dimension, so that the non-elite individuals gradually move closer to the superior characteristics of the elite group, while maintaining a certain population diversity to avoid local convergence, thus obtaining multiple growth traction populations.

[0069] It should be noted that the growth traction operation is expressed as:

[0070] in, For the first i A new individual generated after traction. For the number of iterations k Decreasing disturbance intensity For the first i Individual, For the first k The average hourly energy consumption of all individuals in the elite daily energy-consuming population of the next iteration. k The index is the number of iterations. For random disturbance terms, i For individual indexes.

[0071] S113. Perform fire disturbance operations on each growth-driving population to obtain multiple fire-disturbed populations.

[0072] Fire perturbation operation refers to a population optimization operation that randomly perturbs and adjusts the individuals in the growth-driving population. By dynamically adjusting the hourly energy consumption of some individuals, a new solution space is introduced into the population, thereby improving the population's global search capability and diversity.

[0073] Fire-perturbed populations refer to new populations formed after growth-driven populations are subjected to fire perturbation. Individuals in these populations inherit the excellent basic characteristics of growth-driven populations and obtain new numerical combinations through random perturbation, thereby enhancing the overall diversity and global search capability of the population.

[0074] In this embodiment of the invention, fire disturbance operation is performed on each growth-driving population (i.e., the hourly energy consumption values ​​of some individuals in the growth-driving population are randomly disturbed and adjusted based on the decreasing fire disturbance trigger probability and disturbance amplitude as the iteration process), to obtain multiple fire-disturbed populations.

[0075] It should be noted that the specific expression for the fire disturbance operation is:

[0076] in, For the first i Individuals disturbed by fire. For the amplitude of fire disturbance, The maximum number of iterations, For the first k The probability of triggering the next iteration.

[0077] The fire perturbation amplitude is the range within which the hourly energy consumption value of selected individuals in the growth-guided population is randomly adjusted. It is the core parameter controlling the degree of numerical perturbation of individuals in the fire perturbation operation. Its value gradually decreases as the iteration progresses. In the early stage of the iteration, a larger fire perturbation amplitude is set to introduce more new solution spaces to the population and enhance the global search capability. In the later stage of the iteration, the fire perturbation amplitude is reduced to reduce meaningless numerical adjustments, allowing the population to focus on the discovery of local optima. This ensures the directionality and accuracy of the iterative optimization while preserving the diversity of the population.

[0078] S114. Perform crossover or mutation operations on each fire disturbance population to obtain multiple target fire disturbance populations.

[0079] Crossover operation refers to randomly selecting two individuals from a fire disturbance population, determining the hourly energy consumption dimension to be exchanged through a preset crossover coefficient (such as randomly selecting several hour segments), and exchanging the values ​​of the corresponding dimension to generate a new individual, thereby achieving the fusion of the superior energy consumption characteristics of different individuals.

[0080] Mutation operation refers to randomly selecting some individuals in a fire-disturbed population and randomly adjusting their specific hourly energy consumption values ​​according to a preset mutation rate (such as randomly replacing values ​​within the upper and lower bounds of hourly energy consumption) to introduce a new solution space to avoid population homogenization.

[0081] In the operation of the embodiments of the present invention, crossover or mutation operations are performed on each fire disturbance population (i.e., based on the crossover coefficient and mutation rate which decrease with the iteration process, individuals are randomly selected in the population to perform crossover operations, and excellent gene recombination is achieved by exchanging some hourly energy consumption dimensions, while random mutation adjustments are made to some individuals to introduce new solution space), to obtain multiple target fire disturbance populations.

[0082] It should be noted that the expression for the crossover operation is as follows:

[0083] in, The first body after the crossover operation. The second body after the crossover operation. Cross coefficient, For the first j Individuals disturbed by fire. j Select individual indexes for cross-selection.

[0084] The expression for the mutation operation:

[0085] in, For the first q A mutated individual, For the disturbance intensity, The variability rate Let k be the mutation trigger probability. q Index for variant individuals.

[0086] S115. Perform local searches on each target fire disturbance population, and merge each locally searched target fire disturbance population with the corresponding elite daily energy consumption population to obtain multiple new initial daily energy consumption populations.

[0087] Local search refers to focusing on the neighborhood of the current optimal solution and fine-tuning the individual's decision variables (hourly energy consumption values) with small steps to discover better local feasible solutions, thereby improving the accuracy and stability of the solutions.

[0088] In this embodiment of the invention, a local search is performed on each target fire disturbance population, and the target fire disturbance population after the local search is merged with the corresponding elite daily energy consumption population based on a preset fusion update function to obtain multiple new initial daily energy consumption populations.

[0089] It should be noted that the expression for local search is as follows:

[0090] in, For individuals after local search, As the current optimal individual, For local fine-tuning of step size, This is the first fitness value of the individual after the local search.

[0091] The fusion update function is as follows:

[0092] in, For the first i 1 initial daily energy consumption population t +1 iterations of boundary projection target fire perturbation population For the dimension-by-dimensional boundary projection operator, The first fitness value of the first body after the crossover operation. The first fitness value of the second body after the crossover operation. As the first adaptive weight, For the second adaptive weight, For the first i 1 initial daily energy consumption population t Elite intraday energy-consuming population (+1 iteration) For the new initial intraday energy-consuming population, i 1 represents the population index. t This is the index of the number of population iterations.

[0093] It is worth mentioning that after the population enters the neighborhood of the optimal solution, a perturbation correction oriented towards the global optimum is performed on the new initial intraday energy consumption population, which further improves the accuracy and stability of the later solutions:

[0094] in, For the first t The new initial intraday energy consumption population after +1 iteration correction For the first t +1 iteration for a new initial intraday energy consumption population For perturbation factor, This is the globally optimal solution.

[0095] It should be noted that by sorting and truncating each initial intraday energy-consuming population according to its first fitness value, the elite intraday energy-consuming population with better carbon emission optimization can be quickly screened out, ensuring the precise preservation of superior individuals. Simultaneously, non-elite populations to be optimized are identified, providing clear optimization targets and directions for subsequent population updates and preventing the loss of high-quality solutions. Growth traction operations centered on elite intraday energy-consuming populations guide non-elite intraday energy-consuming populations towards superior energy-consuming characteristics, accelerating the optimization process while preserving basic population diversity and effectively preventing the population from getting trapped in local optima. Fire perturbation operations are implemented on the growth-traction populations, introducing new solution spaces through random perturbation, further enhancing the global search capability of the growth-traction populations and making the optimization of intraday energy-consuming sequences more comprehensive. Combining crossover or mutation operations to optimize the fire-perturbation populations allows for the achievement of superior energy consumption for different individuals. The recombination of features and the introduction of new features continuously enrich the solution space of the population, allowing the population to maintain its evolutionary vitality. Local search of the target fire-perturbed population can focus on the neighborhood of the current optimal solution for fine-tuning, and discover better local feasible solutions, thereby improving the accuracy and stability of the daily energy consumption sequence optimization. The population after local search is then merged with the elite population to obtain a new initial daily energy consumption population, which greatly improves the efficiency and effect of the iterative optimization of the lower-level optimization model, making the optimized daily energy consumption sequence more in line with the time-varying characteristics and daily quota constraints of actual production.

[0096] S12. Determine whether the number of iterations for each initial day's energy-consuming population is greater than or equal to the preset first iteration threshold.

[0097] The first iteration threshold refers to the convergence criterion for the lower-level optimization, which is the pre-set maximum number of iterations for the population (usually 50-200 times).

[0098] In this embodiment of the invention, it is determined whether the number of iterations of each initial day's energy-consuming population has reached a preset first iteration threshold.

[0099] S13. When the number of iterations of the initial intraday energy consumption population is less than the first iteration threshold, the process jumps to the step of inputting each intraday energy consumption sequence and the preset hourly carbon data into the lower-level optimization model to obtain multiple first fitness values.

[0100] In this embodiment of the invention, when the number of iterations of the initial daily energy-consuming population is less than the first iteration threshold, the process jumps to step 203.

[0101] S14. When the number of iterations of the initial intraday energy consumption population is greater than or equal to the first iteration threshold, the intraday energy consumption sequence corresponding to the minimum value in the first fitness value associated with the initial intraday energy consumption population is selected as the optimized intraday energy consumption sequence.

[0102] In this embodiment of the invention, when the number of iterations of the initial intraday energy consumption population reaches the first iteration threshold, the intraday energy consumption sequence corresponding to the minimum value among the first fitness values ​​associated with the initial intraday energy consumption population is selected as the optimized intraday energy consumption sequence.

[0103] It should be noted that the initial daily energy consumption population is updated in a targeted manner based on the first fitness value, ensuring that the population iteration always revolves around the optimization goal of minimizing daily emissions. This guarantees the directionality and effectiveness of the population update and avoids aimless iterative adjustments. At the same time, by setting a first iteration threshold and judging the number of population iterations one by one, a clear convergence standard is set for the iteration process of the lower-level optimization. This allows the population to fully iterate and optimize within the threshold range, discovering better solutions for the daily energy consumption sequence, while preventing the waste of computational resources caused by excessive iteration, thus improving the solution efficiency of the lower-level optimization. The cyclic solution design when the number of iterations has not reached the threshold realizes the dynamic iterative optimization of the initial daily energy consumption population, allowing the population to continuously update and evolve, constantly approaching the optimal solution. After the number of iterations reaches the threshold, the daily energy consumption sequence corresponding to the minimum first fitness value is selected as the optimization result, which can accurately lock the hourly energy consumption scheduling scheme with the optimal carbon emissions under the corresponding daily quota sequence.

[0104] Step 205: Iteratively solve the initial particle swarm using the daily energy consumption sequences and the preset upper-level optimization model to obtain the corresponding carbon quota planning results.

[0105] Furthermore, step 205 includes the following sub-steps: S21. Input the daily quota sequence and daily energy consumption sequence corresponding to each particle in the initial particle swarm into the preset upper-level optimization model to obtain multiple second fitness values.

[0106] The second fitness value refers to the quantitative evaluation index obtained after inputting the daily quota sequence and intraday energy consumption sequence corresponding to the particle into the upper-level optimization model. The smaller the value, the better the global optimization effect of the corresponding daily quota sequence.

[0107] In this embodiment of the invention, the daily quota sequence and daily energy consumption sequence corresponding to each particle in the initial particle swarm are respectively input into the preset upper-level optimization model to obtain multiple second fitness values.

[0108] It should be noted that the upper-level optimization model is as follows:

[0109] in, For the first The second fitness value corresponding to the daily quota sequence For the first Daily quota sequence number d The first fitness value corresponding to the day To smooth out the weighting of quotas, For the consistency of annual total penalty weight, For the first Daily quota sequence number d Daily carbon allowance For the first Daily quota sequence number d -1 day's carbon allowance This refers to the total annual quota. D For the number of days, This is the daily quota sequence.

[0110] S22. Update the initial particle swarm using each second fitness value to obtain the corresponding updated particle swarm.

[0111] The updated particle swarm refers to the particle swarm obtained after the initial particle swarm has been adjusted by the velocity-position update rule. It includes the updated daily quota sequence scheme and has not been verified by the annual total quota constraint.

[0112] In this embodiment of the invention, the initial particle swarm is updated using various second fitness values. Combining the velocity-position update rule of Particle Swarm Optimization (PSO), the second fitness value corresponding to each particle is used as the basis to compare the historical best fitness value of the particle with the global best fitness value. The flight speed and position of the particle are dynamically adjusted so that the particle evolves towards a better daily quota sequence. At the same time, an inertia weight that decreases with iteration is introduced to balance the global exploration ability and local development ability of the population, avoids falling into local optima too early, and obtains the corresponding updated particle swarm.

[0113] It should be noted that the inertia weight is a parameter used to control the degree to which the particle's historical velocity affects its current velocity. It decreases with the number of iterations, with a larger weight in the early stages to strengthen global exploration and a smaller weight in the later stages to focus on local development, thus balancing the breadth and depth of the population search.

[0114] The expression for inertia weight is as follows:

[0115] in, For the first k The inertia weight of the next iteration, For maximum inertia weight, As the baseline inertia weight, For minimum inertia weight, This is the threshold for the second iteration.

[0116] S23. Based on the preset annual quota, perform a consistency callback on the updated particle swarm to obtain a new initial particle swarm.

[0117] The annual total allowance refers to the upper limit of the total carbon emissions allowed for an enterprise during the entire carbon allowance optimization cycle, which is determined by industry standards, emission reduction targets, and other practical requirements.

[0118] Consistency callback refers to the operation of adjusting the daily quota sequence of each particle in the particle swarm based on the total quota. The core purpose is to keep the cumulative value of the daily quota sequence of each particle consistent with the total annual quota, so as to meet the total quota constraint.

[0119] The new initial particle swarm refers to the particle swarm obtained after the update particle swarm has undergone a consistency callback operation, where the daily quota sequence of each particle satisfies the annual quota total constraint.

[0120] In this embodiment of the invention, the preset annual quota and the updated particle swarm are input into a preset consistency callback function to obtain a new initial particle swarm.

[0121] It should be noted that the consistency callback function is as follows:

[0122] in, It is a vector consisting entirely of 1s.

[0123] S24. When the number of iterations of the initial particle swarm is less than the preset second iteration threshold, the process jumps to the step of using the daily quota sequence to optimize the lower-level optimization model and obtain multiple optimized daily energy consumption sequences.

[0124] The second iteration threshold refers to the preset maximum number of iterations for the particle swarm (usually 100-500 times).

[0125] In this embodiment of the invention, when the number of iterations of the initial particle swarm is less than a preset second iteration threshold, then step 202.

[0126] S25. When the number of iterations of the initial particle swarm is greater than or equal to the second iteration threshold, the daily energy consumption sequence and daily quota sequence corresponding to the minimum value among the various second fitness values ​​are selected as the corresponding carbon quota planning results.

[0127] In this embodiment of the invention, when the number of iterations of the initial particle swarm reaches the second iteration threshold, the daily energy consumption sequence and daily quota sequence corresponding to the minimum value among the various second fitness values ​​are selected as the corresponding carbon quota planning results.

[0128] It should be noted that updating the initial particle swarm based on the second fitness value, combined with the global search characteristics of the particle swarm algorithm, allows the particle swarm to evolve towards a better daily quota sequence, effectively discovering the optimal carbon quota allocation scheme globally and avoiding the limitations of local optima. Performing a consistency callback operation based on the annual quota total ensures strict control over the annual total carbon quota constraint, guaranteeing that the cumulative value of the daily quota sequence corresponding to each particle meets the annual emission ceiling requirements, thus ensuring that the carbon quota planning results meet the actual compliance needs of enterprises. By setting a second iteration threshold and determining the number of iterations of the particle swarm, a scientific convergence standard is set for the upper-level global iteration, enabling the particle swarm to evolve towards a better daily quota sequence. Within the threshold range, the two-layer collaborative iteration is fully completed, realizing the dynamic matching and optimization of the daily quota sequence and the intraday energy consumption sequence, while avoiding the consumption of computing resources caused by excessive iteration and improving the overall solution efficiency of carbon quota planning. The two-layer cyclical iteration design when the number of iterations does not reach the threshold realizes the deep collaboration and dynamic optimization between the upper-level carbon quota allocation and the lower-level hourly energy consumption scheduling, so that the adjustment of the daily quota sequence can match the optimization results of the lower-level energy consumption scheduling in real time, ensuring the refinement and adaptability of carbon quota planning. After the number of iterations reaches the threshold, the sequence corresponding to the minimum value of the second fitness value is selected as the planning result, which can accurately lock the globally optimal carbon quota allocation scheme and the corresponding hourly energy consumption scheduling scheme.

[0129] It is worth mentioning that, see Figure 3 As shown, the three curves correspond to actual carbon emissions, dynamic carbon quotas, and optimized carbon emissions, respectively. Actual carbon emissions represent the raw daily emissions data of the enterprise before adopting the proposed solution, influenced by factors such as production load fluctuations and process switching. This curve exhibits significant fluctuations and multiple peak nodes. Dynamic carbon quotas represent the daily quota sequence output by the upper-level algorithm of this invention. This curve is smooth with no significant fluctuations, fully demonstrating the effect of quota smoothing regularization constraints. Optimized carbon emissions represent the actual daily emissions result after hourly scheduling optimization by the lower-level algorithm. This curve closely matches the dynamic carbon quota curve, showing no over-quota situations and overall emissions lower than actual carbon emissions. This directly confirms the core advantage of this invention in achieving dynamic matching of "quota-operation," and also demonstrates the effective control of production operations by engineering feasibility constraints by smoothing emission peaks. (See reference...) Figure 4As shown, the optimized total carbon quota is 654,752 tons, which is the total quota allocated by the upper-level algorithm under the annual total quota constraint, meeting the total quota consistency requirement. The optimized cumulative carbon emissions are 701,039 tons, which is the actual total emissions within the cycle after the two-layer collaborative optimization. The comparison results of the three sets of data show that the actual carbon emissions are reduced by 14,307 tons (a decrease of 2.00%) compared to the optimized carbon quota, the actual carbon emissions are reduced by 60,593 tons (a decrease of 8.47%) compared to the optimized carbon quota, and the optimized carbon quota is reduced by 46,286 tons (a decrease of 6.60%) compared to the optimized carbon quota. This clearly quantifies the direct emission reduction effect, quota optimization space, and quota execution efficiency of this invention, proving that it can achieve a balance between total compliance, operating costs, and emission reduction effects. (See also...) Figure 5 As shown, the optimized emission ratio is consistently lower than the dynamic quota ratio. On most days, the dynamic quota ratio is between 80% and 100%, while the optimized emission ratio averages about 85%. This reflects the rationality of the quota allocation (neither too lenient nor too strict) and verifies the stability of the emission reduction effect, proving that the present invention can effectively tap into emission reduction potential in different production scenarios.

[0130] In this embodiment of the invention, a lower-level optimization model is constructed with the goal of minimizing daily emissions and the daily energy consumption sequence as the decision variable. An initial particle swarm is constructed using the daily quota sequence as particles. The lower-level optimization model is then optimized using each daily quota sequence to obtain multiple optimized daily energy consumption sequences. The initial particle swarm is iteratively solved using each daily energy consumption sequence and a preset upper-level optimization model to obtain the corresponding carbon quota planning results. This overcomes the technical problem that traditional carbon quota planning often relies on past carbon emission data of enterprises for carbon quota allocation, but fails to effectively capture the daily / weekly fluctuations of the load curve, the time-varying emission factors caused by production line switching and process cycle time, thus reducing the reliability of carbon quota planning. Compared with traditional carbon quota planning methods, this invention uses the goal of minimizing daily emissions and the daily energy consumption sequence as the decision variable to construct a lower-level optimization model, accurately capturing hourly load fluctuations and time-varying carbon factors. Based on the lower-level optimization model, a particle swarm algorithm is combined to perform a global search on the daily quota sequence to obtain the corresponding carbon quota planning results, thus improving the reliability of carbon quota planning.

[0131] Please see Figure 6 , Figure 6 This is a structural block diagram of a carbon quota planning system provided in Embodiment 3 of the present invention.

[0132] This invention provides a carbon quota planning system, comprising: Module 301 is used to construct a lower-level optimization model with the goal of minimizing daily emissions and the decision variable of intraday energy consumption sequence. The lower-level optimization module 302 is used to construct an initial particle swarm with the daily quota sequence as particles, and to optimize and solve the lower-level optimization model using the daily quota sequence to obtain multiple optimized intraday energy consumption sequences. The upper-level optimization module 303 is used to iteratively solve the initial particle swarm using various intraday energy consumption sequences and a preset upper-level optimization model to obtain the corresponding carbon quota planning results.

[0133] Furthermore, the lower-level optimization module 302 includes: A submodule is constructed to build the initial intraday energy consumption population for each daily quota sequence, wherein each individual in the initial intraday energy consumption population corresponds to an intraday energy consumption sequence; The first adaptation submodule is used to input each intraday energy consumption sequence and preset hourly carbon data into the lower-level optimization model to obtain multiple first adaptation values; The first analysis submodule is used to iteratively optimize each initial intraday energy consumption population using its first fitness value to obtain multiple optimized intraday energy consumption sequences.

[0134] Furthermore, the first analysis submodule includes: The update unit is used to update each initial day's energy consumption population using the first fitness value corresponding to each initial day's energy consumption population, thereby obtaining multiple new initial day's energy consumption populations. The analysis unit is used to determine whether the number of iterations for each initial day's energy-consuming population is greater than or equal to a preset first iteration threshold. When the number of iterations of the initial intraday energy consumption population is less than the first iteration threshold, the process jumps to the step of inputting each intraday energy consumption sequence and the preset hourly carbon data into the lower-level optimization model to obtain multiple first fitness values. When the number of iterations of the initial intraday energy consumption population is greater than or equal to the first iteration threshold, the intraday energy consumption sequence corresponding to the minimum value in the first fitness value associated with the initial intraday energy consumption population is selected as the optimized intraday energy consumption sequence.

[0135] Furthermore, the update unit includes: The screening subunit is used to sort the individuals of each initial daily energy consumption population in ascending order according to the first fitness value, and to truncate and select each sorted initial daily energy consumption population according to the preset division number, so as to obtain multiple elite daily energy consumption populations and multiple non-elite daily energy consumption populations. The growth traction subunit is used to perform growth traction operations on the corresponding non-elite daily energy-consuming populations according to each elite daily energy-consuming population, resulting in multiple growth traction populations. The fire disturbance subunit is used to perform fire disturbance operations on each growth-driving population to obtain multiple fire disturbance populations. The update sub-unit is used to perform crossover or mutation operations on each fire disturbance population to obtain multiple target fire disturbance populations. Local searches are performed on each target fire disturbance population, and the target fire disturbance populations obtained from each local search are merged with the corresponding elite daily energy consumption populations to obtain multiple new initial daily energy consumption populations.

[0136] Furthermore, the upper-layer optimization module 303 includes: The second analysis submodule is used to input the daily quota sequence and intraday energy consumption sequence corresponding to each particle in the initial particle swarm into the preset upper-level optimization model to obtain multiple second fitness values. The update submodule is used to update the initial particle swarm using each second fitness value to obtain the corresponding updated particle swarm. The consistency callback submodule is used to perform a consistency callback on the updated particle swarm based on the preset annual quota total, so as to obtain a new initial particle swarm. The third analysis submodule is used to jump to the step of optimizing the lower-level optimization model by using the daily quota sequence when the number of iterations of the initial particle swarm is less than the preset second iteration threshold, so as to obtain multiple optimized intraday energy consumption sequences. When the number of iterations of the initial particle swarm is greater than or equal to the second iteration threshold, the daily energy consumption sequence and daily quota sequence corresponding to the minimum value among the various second fitness values ​​are selected as the corresponding carbon quota planning results.

[0137] Furthermore, the lower-level optimization model is as follows: ; in, For the first d The first adaptation value of the day, For the first d day h Energy consumption per hour For the first d day h hourly carbon factor For the first d Daily quota, h Indexed by hour. This is the second penalty coefficient for exceeding the quota. The lower bound of the hour. The upper limit for hours, d For daily indexing.

[0138] Please see Figure 7 , Figure 7 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention.

[0139] An electronic device according to an embodiment of the present invention includes: a memory 401 and a processor 402. The memory 401 stores a computer program. When the computer program is executed by the processor 402, the processor 402 executes the carbon quota planning method as described in any of the above embodiments.

[0140] Memory 401 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Memory 401 has storage space 403 for program code 413 for performing any of the method steps described above. For example, storage space 403 for program code may include individual program codes 413 for implementing the various steps in the methods described above. This program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When run by a computing processing device, this code causes the computing processing device to perform the various steps in the methods described above. This program code can be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When this code is run by a computing processing device, it causes the device to perform the various steps in the carbon quota planning method described above.

[0141] Embodiment 5 of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the carbon quota planning method as described in any of the above embodiments.

[0142] Embodiment 6 of the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer performs the carbon quota planning method as described in any of the above embodiments.

[0143] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0144] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0145] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0146] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0147] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0148] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A carbon quota planning method, characterized in that, include: With the goal of minimizing daily emissions and the decision variable of intraday energy consumption sequence, a lower-level optimization model is constructed. An initial particle swarm is constructed using the daily quota sequence as particles, and the lower-level optimization model is optimized and solved using each of the daily quota sequences to obtain multiple optimized intraday energy consumption sequences; The initial particle swarm is iteratively solved using the various intraday energy consumption sequences and the preset upper-level optimization model to obtain the corresponding carbon quota planning results.

2. The carbon quota planning method according to claim 1, characterized in that, The step of optimizing the lower-level optimization model using the daily quota sequences to obtain multiple optimized intraday energy consumption sequences includes: Construct an initial intraday energy consumption population for each of the said daily quota sequences, wherein each individual in the initial intraday energy consumption population corresponds to an intraday energy consumption sequence; Each of the intraday energy consumption sequences and the preset hourly carbon data are input into the lower-level optimization model to obtain multiple first fitness values; The first fitness value corresponding to each initial intraday energy consumption population is used for iterative optimization to obtain multiple optimized intraday energy consumption sequences.

3. The carbon quota planning method according to claim 2, characterized in that, The step of iteratively optimizing each initial intraday energy consumption population using its corresponding first fitness value to obtain multiple optimized intraday energy consumption sequences includes: Each initial intraday energy consumption population is updated using its first fitness value to obtain multiple new initial intraday energy consumption populations. Determine whether the number of iterations for each of the initial daily energy-consuming populations is greater than or equal to a preset first iteration threshold. When the number of iterations of the initial intraday energy consumption population is less than the first iteration threshold, the process jumps to the step of inputting each intraday energy consumption sequence and the preset hourly carbon data into the lower-level optimization model to obtain multiple first fitness values. When the number of iterations of the initial intraday energy consumption population is greater than or equal to the first iteration threshold, the intraday energy consumption sequence corresponding to the minimum value among the first fitness values ​​associated with the initial intraday energy consumption population is selected as the optimized intraday energy consumption sequence.

4. The carbon quota planning method according to claim 3, characterized in that, The step of updating each initial intraday energy consumption population using the first fitness value corresponding to each initial intraday energy consumption population to obtain multiple new initial intraday energy consumption populations includes: Based on the first fitness value, individuals in each of the initial daily energy consumption populations are sorted in ascending order, and each sorted initial daily energy consumption population is truncated and selected according to a preset number of divisions to obtain multiple elite daily energy consumption populations and multiple non-elite daily energy consumption populations. Growth traction operations are performed on the corresponding non-elite daily energy-consuming populations based on each of the elite daily energy-consuming populations to obtain multiple growth traction populations. Fire disturbance operations were performed on each of the aforementioned growth-driving populations to obtain multiple fire-disturbed populations; Crossover or mutation operations are performed on each of the aforementioned fire disturbance populations to obtain multiple target fire disturbance populations; A local search is performed on each of the target fire disturbance populations, and the target fire disturbance populations after each local search are merged with the corresponding elite daily energy consumption populations to obtain multiple new initial daily energy consumption populations.

5. The carbon quota planning method according to claim 1, characterized in that, The step of iteratively solving the initial particle swarm using each of the intraday energy consumption sequences and a preset upper-level optimization model to obtain the corresponding carbon quota planning results includes: The daily quota sequence and daily energy consumption sequence corresponding to each particle in the initial particle swarm are respectively input into a preset upper-level optimization model to obtain multiple second fitness values; The initial particle swarm is updated using each of the second fitness values ​​to obtain the corresponding updated particle swarm; Based on the preset annual quota, a consistency callback is performed on the updated particle swarm to obtain a new initial particle swarm. When the number of iterations of the initial particle swarm is less than the preset second iteration threshold, the process jumps to the step of using each of the daily quota sequences to optimize and solve the lower-level optimization model to obtain multiple optimized intraday energy consumption sequences. When the number of iterations of the initial particle swarm is greater than or equal to the second iteration threshold, the daily energy consumption sequence and daily quota sequence corresponding to the minimum value among the second fitness values ​​are selected as the corresponding carbon quota planning results.

6. The carbon quota planning method according to claim 1, characterized in that, The lower-level optimization model is specifically as follows: ； in, For the first d The first adaptation value of the day, For the first d day h Energy consumption per hour For the first d day h hourly carbon factor For the first d Daily quota, h Indexed by hour. This is the second penalty coefficient for exceeding the quota. The lower bound of the hour. The upper limit for hours, d For daily indexing.

7. A carbon quota planning system, characterized in that, include: The module is used to build a lower-level optimization model with the goal of minimizing daily emissions and the decision variable of intraday energy consumption sequence. The lower-level optimization module is used to construct an initial particle swarm with the daily quota sequence as particles, and to optimize and solve the lower-level optimization model using each of the daily quota sequences to obtain multiple optimized intraday energy consumption sequences. The upper-level optimization module is used to iteratively solve the initial particle swarm using each of the intraday energy consumption sequences and a preset upper-level optimization model to obtain the corresponding carbon quota planning results.

8. An electronic device, characterized in that, It includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the steps of the carbon quota planning method as described in any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the carbon quota planning method as described in any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium, the computer program including program instructions, wherein when the program instructions are executed by a computer, the computer performs the carbon quota planning method as described in any one of claims 1-6.

Images