A green electricity regulating method and system for electricity calculation collaboration

Abstract

The application discloses a green electricity regulation method and system for algorithm power coordination, relates to the technical field of data analysis, and solves the technical problem that the existing technology lacks information interaction and collaborative optimization mechanism, so that the inherent space-time regulation potential of algorithm power load cannot be identified and called by a power system, resulting in low accuracy and efficiency of the green electricity regulation method; an algorithm power carbon emission regulation capacity list is determined based on target area data; scheduling optimization is performed based on green electricity data and the algorithm power carbon emission regulation capacity list to generate algorithm power scheduling instructions, with the core target of minimizing green electricity mismatch and carbon emission, and optimization calculation is performed in combination with the regulation potential in the list and real-time green electricity data to generate algorithm power scheduling instructions that can be directly issued, so that the algorithm power system is changed from a passive load to an active regulation resource of the power grid, thereby improving the accuracy of green electricity consumption and the regulation efficiency of the overall system.

Description

Technical Field

[0001] This application belongs to the field of data analysis technology, specifically a green energy regulation method and system for computer-electricity coordination. Background Technology

[0002] With the explosive growth of artificial intelligence and the digital economy, computing power has become a new key productive force after heat and electricity. However, the ever-increasing energy consumption challenges behind it have also given rise to a new type of infrastructure project: computing-power synergy. Essentially, computing-power synergy is the deep integration of computing infrastructure and new power systems. It aims to break the traditional passive energy supply model of "electricity follows computing" through digital technology and intelligent algorithms, and to build a resource optimization allocation system that allows for two-way interaction and dynamic matching between computing load and power supply. When renewable energy generation is at its peak, computing load actively absorbs green electricity; when power supply is tight, computing tasks are flexibly reduced or migrated to other locations, truly achieving computing following electricity and peak shaving and valley filling. The green electricity regulation of computing-power synergy has profound strategic significance for promoting the green and low-carbon transformation of the energy structure and achieving high-quality synergistic development of the digital economy and the energy industry.

[0003] Existing technologies lack information interaction and collaborative optimization mechanisms when performing green energy regulation, which prevents the power system from recognizing and utilizing the inherent spatiotemporal regulation potential of computing power load, resulting in low accuracy and efficiency of green energy regulation methods. Therefore, further improvements are needed for green energy regulation methods. Summary of the Invention

[0004] This application aims to solve at least one of the technical problems existing in the prior art; to this end, this application proposes a green electricity regulation method and system for computing-electricity collaboration, which solves the technical problem that the lack of information interaction and collaborative optimization mechanism in the prior art makes it impossible for the power system to identify and call upon the inherent spatiotemporal regulation potential of computing load, resulting in low accuracy and efficiency of green electricity regulation methods.

[0005] To achieve the above objectives, the first aspect of this application provides a green energy regulation method for computer-electricity coordination, comprising: Acquire target area data; the target area data includes green electricity data and computing power data corresponding to the target area; A list of computing power carbon emission regulation capabilities will be determined based on data from the target region. Based on green electricity data and a list of computing power carbon emission adjustment capabilities, scheduling optimization is performed to generate computing power scheduling instructions.

[0006] This application, through the aforementioned steps, establishes a foundation for information exchange by simultaneously acquiring green electricity data and computing power data, breaking down information barriers between computing power and the power system. Based on the task attribute analysis in the computing power data, it generates a list of computing power carbon emission regulation capabilities, transforming the spatiotemporal flexibility of computing power load into quantifiable dynamic carbon emission reduction factors, enabling the power system to accurately identify and assess the regulation capabilities of the computing power side. With minimizing green electricity mismatch and carbon emissions as the core objective, it combines the regulation potential in the list with real-time green electricity data for optimized calculations, generating directly identifiable computing power dispatch instructions. This transforms the computing power system from a passive load into an active regulation resource for the power grid, thereby improving the accuracy of green electricity consumption and the overall system regulation efficiency.

[0007] Furthermore, the process of determining the computing power carbon emission adjustment capacity list based on target area data includes: Extract target area data; the target area data includes green electricity data and computing power data corresponding to the target area; the green electricity data includes the green electricity power corresponding to several time points; the computing power data includes computing power task information that is running and planned to run in several computing centers, the computing power task information includes task ID and its corresponding computing power resources required by the task, task priority, adjustment type and time constraints; the time constraints include the expected execution duration, earliest start time and latest completion time; An adjustable potential analysis is performed on the computing power task information to obtain an elastic task list; the elastic task list includes task ID, computing center ID, adjustment type, adjustment time window, and adjustable computing power capacity; Obtain the carbon intensity corresponding to different computing center IDs in the target region at different time periods; The dynamic carbon emission reduction factor is calculated based on carbon intensity and the elastic task list, and the dynamic carbon emission reduction factor is added to the elastic task list to obtain the computing power carbon emission adjustment capability list; the computing power carbon emission adjustment capability list includes task ID, computing center ID, adjustment type, adjustment time window, dynamic carbon emission reduction factor and adjustment computing power capacity.

[0008] Furthermore, the process of performing adjustable potential analysis on the computing power task information to obtain a list of elastic tasks includes: For each task ID corresponding to a different adjustment type, calculate its corresponding adjustment time window and adjustment computing capacity; the adjustment types include time shift, interruptible, and transferable. When the adjustment type is time shift, the adjustment time window satisfies [EST, LFT-D]; where EST represents the earliest start time, LFT represents the latest finish time, and D represents the expected execution time; the adjustment computing power capacity represents the computing power resources required by the task corresponding to the task ID; When the adjustment type is interruptible, the adjustment time window satisfies [EST, LFT]; the adjustment computing power capacity is represented by the computing power resources required by the task corresponding to the task ID; When the adjustment type is portable, the adjustment time window satisfies [EST_c1, LFT_c1]∪[EST_c2, LFT_c2]; where c1 and c2 represent the numbers corresponding to different computing center IDs; the adjustment computing power capacity represents the computing power resources required by the task corresponding to the task ID; The list of elastic tasks is determined based on task ID, computing center ID, adjustment type, adjustment time window, and adjustment computing capacity.

[0009] Furthermore, the calculation of the dynamic carbon reduction factor based on carbon intensity and a resilient task list includes: Extract the compute center ID and adjustment type corresponding to the task ID in the elastic task list; When the adjustment type is time shift, the carbon intensity of the task ID in the original time period and the target time period of the current computing center ID is extracted; the dynamic carbon reduction factor is the difference between the carbon intensity of the original time period and the carbon intensity of the target time period. When the adjustment type is interruptible, extract the carbon intensity of the task ID during the interruption period of the current computing center ID; the dynamic carbon reduction factor is the carbon intensity during the interruption period. When the adjustment type is portable, the carbon intensity of the task ID in the original computing center ID and the target computing center ID during the migration period is extracted; the dynamic carbon reduction factor is the difference between the carbon intensity of the original computing center ID and the carbon intensity of the target computing center ID during the migration period.

[0010] This application first refines the time-series power data of green electricity and multi-dimensional information of computing power tasks, and then analyzes the adjustability potential of computing power tasks. For three types—time-shiftable, interruptible, and transferable—specific adjustable time windows and computing power capacity are quantified based on their time constraints, generating a structured list of flexible tasks. Subsequently, by obtaining the real-time carbon intensity of different computing centers at different times, and combining it with the adjustment type and spatiotemporal information in the list of flexible tasks, the dynamic carbon emission reduction factor corresponding to each task under each adjustment operation is accurately calculated. This is then integrated into a list that includes both adjustment capabilities and quantitative indicators of carbon emission benefits. This transforms the originally abstract computing power flexibility into a list of computing power carbon emission adjustment capabilities that can be accurately measured, evaluated, and compared, providing a high-precision decision-making basis for subsequent scheduling optimization, thereby significantly improving the accuracy and collaborative efficiency of green electricity adjustment.

[0011] Furthermore, the process of generating computing power scheduling instructions based on green electricity data and a computing power carbon emission adjustment capability list includes: Extract green electricity data and a list of computing power carbon emission regulation capabilities; An instruction optimization model is constructed based on green electricity data and a list of computing power carbon emission regulation capabilities; the instruction optimization model includes defining decision variables, constructing an objective function, and defining constraints. The computing power scheduling instructions are obtained by solving the instruction optimization model; the solution operation adopts global optimization based on genetic algorithm to solve the model.

[0012] Furthermore, the decision variables include execution decision variables, interruption decision variables, and migration decision variables; The execution decision variables satisfy , indicating whether the i-th task ID is executed by the c-th computing center ID at time t; where i represents the number corresponding to the task ID, t represents the time identifier, and c represents the number corresponding to the computing center ID; The interruption decision variables satisfy , indicating whether the i-th task ID was interrupted at time t; The migration decision variables satisfy , indicating whether the i-th task ID has migrated from computing center c1 to computing center c2 at time t.

[0013] Furthermore, the objective function includes a main objective function and an auxiliary objective function; The main objective function satisfies: ;in, Let be the total power consumption of the computing center c at time t, satisfying Where p represents the power consumption of the i-th task ID at time t in the c-th computing center ID; i represents the number corresponding to the task ID; t represents the time identifier; c represents the number corresponding to the computing center ID; C represents the total number of computing center IDs; N represents the total number of task IDs; and T represents the total length of time t. This is expressed as the green power of the c-th computing center ID at time t; This is expressed as the carbon intensity of the c-th computational center ID at time t; Represented as a positive part function; and It is represented as a weighting coefficient, used to balance green electricity utilization and carbon emission targets, and the specific value is set based on experience.

[0014] Furthermore, the auxiliary objective function satisfies: ;in, This represents the task priority corresponding to the i-th task ID; This represents the actual completion time of the i-th task ID; This represents the latest completion time of the i-th task ID.

[0015] Furthermore, the constraints include task integrity constraints, resource capacity constraints, time shift constraints, interruptibility constraints, and portability constraints. The task integrity constraint means that each task ID needs to complete the task execution of the expected execution time within its corresponding adjustment time window. The resource capacity constraint means that the adjusted computing power capacity of the computing center ID at time t does not exceed the available capacity of its corresponding resource type. The time shift constraint means that when the adjustment type is time shift, the execution time of the task ID is continuous. The interruptible constraint means that when the adjustment type is interruptible, the number of interruptions for a task ID is less than or equal to the interruption threshold. The interruption threshold is set based on experience, and the interruption threshold varies for different task IDs. The transferable constraint requires that, when the adjustment type is transferable, the migration time point of the task ID must be within its corresponding adjustment time window.

[0016] This application achieves precise scheduling by constructing a complete mathematical optimization model: First, it clearly defines the decision variables for task execution, interruption, and migration, transforming the scheduling problem into a computable mathematical form; second, it establishes an optimization objective centered on minimizing the supply-demand deviation of green electricity and carbon emissions, supplemented by minimizing the delay of high-priority tasks, ensuring that scheduling guarantees service quality while pursuing low carbon emissions; furthermore, it incorporates operational constraints specific to task integrity, resource capacity, and various adjustment types to ensure the feasibility of the solution; finally, it employs a genetic algorithm for efficient solution, thereby generating globally optimal computing power scheduling instructions, achieving dynamic and precise matching between computing power load and the spatiotemporal distribution of green electricity, fundamentally improving the efficiency of green electricity consumption and the overall carbon efficiency of the system.

[0017] Another aspect of this application provides a green energy regulation system for computer-electricity coordination, comprising: a data acquisition module and a data analysis module; the data acquisition module and the data analysis module are connected together; The data acquisition module acquires data from the target area through a data acquisition device; the target area data includes green electricity data and computing power data corresponding to the target area. The data analysis module includes a list integration unit and a result generation unit; The list integration unit: determines the list of computing power carbon emission adjustment capabilities based on target area data; The result generation unit generates computing power scheduling instructions based on green electricity data and a list of computing power carbon emission adjustment capabilities through scheduling optimization.

[0018] Compared with the prior art, the beneficial effects of this application are: 1. This application determines a list of computing power carbon emission regulation capabilities based on target area data; it generates computing power scheduling instructions by optimizing scheduling based on green electricity data and the computing power carbon emission regulation capability list. By simultaneously acquiring green electricity data and computing power data, it establishes a foundation for information interaction, breaking down information barriers between computing power and the power system; it analyzes the task attributes in the computing power data and generates a computing power carbon emission regulation capability list, transforming the spatiotemporal flexibility of computing power load into a quantifiable dynamic carbon reduction factor, enabling the power system to accurately identify and assess the regulation capabilities of the computing power side; with minimizing green electricity mismatch and carbon emissions as the core objective, it combines the regulation potential in the list with real-time green electricity data for optimization calculation, generating computing power scheduling instructions that can be directly issued, transforming the computing power system from a passive load into an active regulation resource of the power grid, thereby improving the accuracy of green electricity consumption and the overall system regulation efficiency.

[0019] 2. This application first refines the time-series power data of green electricity and multi-dimensional information of computing power tasks, and then conducts an adjustment potential analysis of the computing power tasks. For three types—time-shiftable, interruptible, and transferable—specific adjustable time windows and computing power capacity are quantified based on their time constraints, generating a structured list of flexible tasks. Subsequently, by obtaining the real-time carbon intensity of different computing centers at different times, and combining it with the adjustment type and spatiotemporal information in the list of flexible tasks, the dynamic carbon emission reduction factor corresponding to each task under each adjustment operation is accurately calculated. This is then integrated into a list that includes both adjustment capabilities and quantitative indicators of carbon emission benefits. This transforms the originally abstract computing power flexibility into a list of computing power carbon emission adjustment capabilities that can be accurately measured, evaluated, and compared, providing a high-precision decision-making basis for subsequent scheduling optimization, thereby significantly improving the accuracy and collaborative efficiency of green electricity adjustment.

[0020] 3. This application achieves precise scheduling by constructing a complete mathematical optimization model: First, it clearly defines the decision variables for task execution, interruption, and migration, transforming the scheduling problem into a computable mathematical form; second, it establishes an optimization objective centered on minimizing the supply-demand deviation of green electricity and carbon emissions, supplemented by minimizing the delay of high-priority tasks, ensuring that scheduling guarantees service quality while pursuing low carbon emissions; furthermore, it incorporates operational constraints specific to task integrity, resource capacity, and various adjustment types to ensure the feasibility of the solution; finally, it employs a genetic algorithm for efficient solution, thereby generating globally optimal computing power scheduling instructions, achieving dynamic and precise matching between computing power load and the spatiotemporal distribution of green electricity, fundamentally improving the efficiency of green electricity consumption and the overall carbon efficiency of the system. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of this application 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 this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart of a green energy regulation method for computing-powered coordination according to this application; Figure 2 This is a schematic diagram of the principle of a green electricity regulation system for computing-electricity coordination according to this application. Detailed Implementation

[0023] The technical solutions of this application will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0024] Please see Figure 1 The first aspect of this application provides a green energy regulation method for computing-powered coordination, comprising: Acquire data for the target area; the target area data includes green electricity data and computing power data corresponding to the target area. A list of computing power carbon emission regulation capabilities will be determined based on data from the target region. Based on green electricity data and a list of computing power carbon emission adjustment capabilities, scheduling optimization is performed to generate computing power scheduling instructions.

[0025] The method for determining the computing power carbon emission adjustment capability list based on target area data in this embodiment includes: Extract target area data; target area data includes green electricity data and computing power data corresponding to the target area; green electricity data includes green electricity power at several time points; computing power data includes computing power task information that is running and planned to run in several computing centers, computing power task information includes task ID and its corresponding computing power resources required by the task, task priority, adjustment type and time constraints; time constraints include estimated execution time, earliest start time and latest finish time; An adjustable potential analysis is performed on the computing power task information to obtain an elastic task list; the elastic task list includes task ID, computing center ID, adjustment type, adjustment time window, and adjustable computing power capacity; Obtain the carbon intensity corresponding to different computing center IDs in the target area at different time periods; in this embodiment, carbon intensity refers to the carbon emissions per 1 kWh of electricity consumed; The dynamic carbon emission reduction factor is calculated based on carbon intensity and the elastic task list, and the dynamic carbon emission reduction factor is added to the elastic task list to obtain the computing power carbon emission adjustment capability list. The computing power carbon emission adjustment capability list includes task ID, computing center ID, adjustment type, adjustment time window, dynamic carbon emission reduction factor, and adjustment computing power capacity.

[0026] In this embodiment, the adjustable potential analysis of computing power task information yields a list of elastic tasks, including: For each task ID corresponding to a different adjustment type, calculate its corresponding adjustment time window and adjustment computing capacity; adjustment types include time shift, interruptible, and portable. When the adjustment type is time shift, the adjustment time window satisfies [EST, LFT-D]; where EST represents the earliest start time, LFT represents the latest finish time, and D represents the estimated execution time; the adjustment computing power capacity represents the computing power resources required by the task corresponding to the task ID; When the adjustment type is interruptible, the adjustment time window satisfies [EST, LFT]; the adjusted computing power capacity is represented by the computing power resources required by the task corresponding to the task ID. When the adjustment type is portable, the adjustment time window satisfies [EST_c1, LFT_c1]∪[EST_c2, LFT_c2]; where c1 and c2 represent the numbers corresponding to different computing center IDs; the adjustment computing capacity represents the computing resources required by the task corresponding to the task ID; The list of elastic tasks is determined based on task ID, computing center ID, adjustment type, adjustment time window, and adjustment computing capacity.

[0027] The calculation of dynamic carbon reduction factors based on carbon intensity and a resilient task list in this embodiment includes: Extract the compute center ID and adjustment type corresponding to the task ID in the elastic task list; When the adjustment type is time shift, extract the carbon intensity of the task ID in the original time period and the target time period in the current computing center ID; the dynamic carbon reduction factor is the difference between the carbon intensity of the original time period and the carbon intensity of the target time period. When the adjustment type is interruptible, extract the carbon intensity of the task ID during the interruption period of the current computing center ID; the dynamic carbon reduction factor is the carbon intensity during the interruption period. When the adjustment type is portable, the carbon intensity of the task ID in the original computing center ID and the target computing center ID during the migration period is extracted; the dynamic carbon reduction factor is the difference between the carbon intensity of the original computing center ID and the carbon intensity of the target computing center ID during the migration period.

[0028] In this embodiment, the scheduling optimization based on green electricity data and the computing power carbon emission adjustment capability list generates computing power scheduling instructions, including: Extract green electricity data and a list of computing power carbon emission regulation capabilities; An instruction optimization model is constructed based on green electricity data and a list of computing power carbon emission regulation capabilities. The instruction optimization model includes defining decision variables, constructing an objective function, and defining constraints. The computing power scheduling instructions are obtained by solving the instruction optimization model; the solution operation adopts global optimization based on genetic algorithm to solve the model.

[0029] In this embodiment, the decision variables include execution decision variables, interruption decision variables, and migration decision variables; The decision variables for execution satisfy , indicating whether the i-th task ID is executed by the c-th computing center ID at time t; where i represents the number corresponding to the task ID, t represents the time identifier, and c represents the number corresponding to the computing center ID; Disruption decision variables satisfy , indicating whether the i-th task ID was interrupted at time t; Migration decision variables satisfy , indicating whether the i-th task ID migrates from computing center c1 to computing center c2 at time t; in this embodiment, c1 and c2 belong to c.

[0030] The objective function in this embodiment includes a main objective function and an auxiliary objective function; The main objective function satisfies: ;in, Let be the total power consumption of the computing center c at time t, satisfying Where p represents the power consumption of the i-th task ID at time t in the c-th computing center ID; i represents the number corresponding to the task ID; t represents the time identifier; c represents the number corresponding to the computing center ID; C represents the total number of computing center IDs; N represents the total number of task IDs; and T represents the total length of time t. This is expressed as the green power of the c-th computing center ID at time t; This is expressed as the carbon intensity of the c-th computational center ID at time t; Represented as a positive part function; and The values ​​are represented as weighting coefficients, used to balance green electricity utilization and carbon emission targets. The specific values ​​are set based on experience, and in this embodiment, they are set to 0.6 and 0.4. In this embodiment, the main objective function is used to maximize green electricity utilization while minimizing carbon emissions.

[0031] The auxiliary objective function in this embodiment satisfies: ;in, This represents the task priority corresponding to the i-th task ID; This represents the actual completion time of the i-th task ID; Let represent the latest completion time of the i-th task ID; in this embodiment, the auxiliary objective function is used to minimize the delay in task completion time.

[0032] The constraints in this embodiment include task integrity constraints, resource capacity constraints, time shift constraints, interruptibility constraints, and portability constraints. Task integrity constraints mean that each task ID must complete the task execution of its expected duration within its corresponding adjustment time window; Resource capacity constraint refers to the fact that the adjusted computing power capacity of a computing center ID at time t does not exceed the available capacity of its corresponding resource type; The time shift constraint means that when the adjustment type is time shift, the execution time of the task ID is continuous. The interruptible constraint means that when the adjustment type is interruptible, the number of interruptions for a task ID is less than or equal to the interruption threshold. The interruption threshold is set based on experience, and the interruption threshold varies for different task IDs. The migration constraint applies until the adjustment type is migrationable, at which point the task ID migration time point needs to be within its corresponding adjustment time window.

[0033] Please see Figure 2 Another aspect of this application provides a green energy regulation system for computer-electricity coordination, comprising: a data acquisition module and a data analysis module; the data acquisition module and the data analysis module are connected. Data acquisition module: Acquires data from the target area through data acquisition equipment; the target area data includes green electricity data and computing power data corresponding to the target area; the data acquisition equipment in this embodiment includes several sensors, etc. The data analysis module includes a list integration unit and a result generation unit; List Integration Unit: Determines a list of computing power carbon emission adjustment capabilities based on target region data; Result generation unit: Based on green electricity data and the list of computing power carbon emission adjustment capabilities, it generates computing power scheduling instructions through scheduling optimization.

[0034] Some of the data in the above formula are calculated by removing dimensions and taking their numerical values. The formula is the closest to the real situation obtained by software simulation of a large amount of collected data. The preset parameters and preset thresholds in the formula are set by those skilled in the art according to the actual situation or obtained through simulation of a large amount of data.

[0035] The working principle of this application is as follows: First, target area data is acquired. Second, a list of computing power carbon emission regulation capabilities is determined based on the target area data. Third, scheduling optimization is performed based on green electricity data and the computing power carbon emission regulation capability list to generate computing power scheduling instructions. By simultaneously acquiring green electricity data and computing power data, an information interaction foundation is established, breaking down the information barrier between computing power and the power system. Fourth, based on the task attribute analysis in the computing power data, a computing power carbon emission regulation capability list is generated, transforming the spatiotemporal flexibility of computing power load into a quantifiable dynamic carbon reduction factor, enabling the power system to accurately identify and assess the regulation capability of the computing power side. Fifth, with minimizing green electricity mismatch and carbon emissions as the core objective, optimization calculations are performed by combining the regulation potential in the list with real-time green electricity data to generate directly issued computing power scheduling instructions. This transforms the computing power system from a passive load into an active regulation resource of the power grid, thereby improving the accuracy of green electricity consumption and the overall system regulation efficiency. This avoids the problem in existing technologies where the lack of information interaction and collaborative optimization mechanisms prevents the power system from recognizing and utilizing the inherent spatiotemporal regulation potential of computing power load, resulting in low accuracy and efficiency of green electricity regulation methods.

[0036] The above embodiments are only used to illustrate the technical methods of this application and are not intended to limit it. Although this application 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 methods of this application without departing from the spirit and scope of the technical methods of this application.

Claims

1. A green energy regulation method for computer-based power coordination, characterized in that, include: Obtain data for the target area; The target area data includes green electricity data and computing power data corresponding to the target area; A list of computing power carbon emission regulation capabilities will be determined based on data from the target region. Based on green electricity data and a list of computing power carbon emission adjustment capabilities, scheduling optimization is performed to generate computing power scheduling instructions.

2. The green energy regulation method for computer-electricity coordination according to claim 1, characterized in that, The process of determining the computing power carbon emission adjustment capacity list based on target area data includes: Extract target area data; the target area data includes green electricity data and computing power data corresponding to the target area; the green electricity data includes the green electricity power corresponding to several time points; the computing power data includes computing power task information that is running and planned to run in several computing centers, the computing power task information includes task ID and its corresponding computing power resources required by the task, task priority, adjustment type and time constraints; the time constraints include the expected execution duration, earliest start time and latest completion time; An adjustable potential analysis is performed on the computing power task information to obtain an elastic task list; the elastic task list includes task ID, computing center ID, adjustment type, adjustment time window, and adjustable computing power capacity; Obtain the carbon intensity corresponding to different computing center IDs in the target region at different time periods; The dynamic carbon emission reduction factor is calculated based on carbon intensity and the elastic task list, and the dynamic carbon emission reduction factor is added to the elastic task list to obtain the computing power carbon emission adjustment capability list; the computing power carbon emission adjustment capability list includes task ID, computing center ID, adjustment type, adjustment time window, dynamic carbon emission reduction factor and adjustment computing power capacity.

3. A green energy regulation method for computer-electricity coordination according to claim 2, characterized in that, The process of performing adjustable potential analysis on computing power task information to obtain an elastic task list includes: For each task ID corresponding to a different adjustment type, calculate its corresponding adjustment time window and adjustment computing capacity; the adjustment types include time shift, interruptible, and transferable. When the adjustment type is time shift, the adjustment time window satisfies [EST, LFT-D]; where EST represents the earliest start time, LFT represents the latest finish time, and D represents the expected execution time; the adjustment computing power capacity represents the computing power resources required by the task corresponding to the task ID; When the adjustment type is interruptible, the adjustment time window satisfies [EST, LFT]; the adjustment computing power capacity is represented by the computing power resources required by the task corresponding to the task ID; When the adjustment type is portable, the adjustment time window satisfies [EST_c1, LFT_c1]∪[EST_c2, LFT_c2]; where c1 and c2 represent the numbers corresponding to different computing center IDs; the adjustment computing power capacity represents the computing power resources required by the task corresponding to the task ID; The list of elastic tasks is determined based on task ID, computing center ID, adjustment type, adjustment time window, and adjustment computing capacity.

4. A green energy regulation method for computer-electricity coordination according to claim 2, characterized in that, The calculation of the dynamic carbon reduction factor based on the carbon intensity and resilience task list includes: Extract the compute center ID and adjustment type corresponding to the task ID in the elastic task list; When the adjustment type is time shift, the carbon intensity of the task ID in the original time period and the target time period of the current computing center ID is extracted; the dynamic carbon reduction factor is the difference between the carbon intensity of the original time period and the carbon intensity of the target time period. When the adjustment type is interruptible, extract the carbon intensity of the task ID during the interruption period of the current computing center ID; the dynamic carbon reduction factor is the carbon intensity during the interruption period. When the adjustment type is portable, the carbon intensity of the task ID in the original computing center ID and the target computing center ID during the migration period is extracted; the dynamic carbon reduction factor is the difference between the carbon intensity of the original computing center ID and the carbon intensity of the target computing center ID during the migration period.

5. A green energy regulation method for computer-electricity coordination according to claim 1, characterized in that, The process of generating computing power scheduling instructions based on green electricity data and a list of computing power carbon emission adjustment capabilities includes: Extract green electricity data and a list of computing power carbon emission regulation capabilities; An instruction optimization model is constructed based on green electricity data and a list of computing power carbon emission regulation capabilities; the instruction optimization model includes defining decision variables, constructing an objective function, and defining constraints. The computing power scheduling instructions are obtained by solving the instruction optimization model; the solution operation adopts global optimization based on genetic algorithm to solve the model.

6. A green energy regulation method for computer-electricity coordination according to claim 5, characterized in that, The decision variables include execution decision variables, interruption decision variables, and migration decision variables; The execution decision variables satisfy , indicating whether the i-th task ID is executed by the c-th computing center ID at time t; where i represents the number corresponding to the task ID, t represents the time identifier, and c represents the number corresponding to the computing center ID; The interruption decision variables satisfy , indicating whether the i-th task ID was interrupted at time t; The migration decision variables satisfy , indicating whether the i-th task ID has migrated from computing center c1 to computing center c2 at time t.

7. A green energy regulation method for computer-electricity coordination according to claim 5, characterized in that, The objective function includes a main objective function and an auxiliary objective function; The main objective function satisfies: ;in, Let be the total power consumption of the computing center c at time t, satisfying Where p represents the power consumption of the i-th task ID at time t in the c-th computing center ID; i represents the number corresponding to the task ID; t represents the time identifier; c represents the number corresponding to the computing center ID; C represents the total number of computing center IDs; N represents the total number of task IDs; and T represents the total length of time t. This is expressed as the green power of the c-th computing center ID at time t; This is expressed as the carbon intensity of the c-th computational center ID at time t; Represented as a positive part function; and It is represented as a weighting coefficient.

8. A green energy regulation method for computer-electricity coordination according to claim 7, characterized in that, The auxiliary objective function satisfies: ;in, This represents the task priority corresponding to the i-th task ID; This represents the actual completion time of the i-th task ID; This represents the latest completion time of the i-th task ID.

9. A green energy regulation method for computer-electricity coordination according to claim 5, characterized in that, The constraints include task integrity constraints, resource capacity constraints, time shift constraints, interruptibility constraints, and portability constraints. The task integrity constraint means that each task ID needs to complete the task execution of the expected execution time within its corresponding adjustment time window. The resource capacity constraint means that the adjusted computing power capacity of the computing center ID at time t does not exceed the available capacity of its corresponding resource type. The time shift constraint means that when the adjustment type is time shift, the execution time of the task ID is continuous. The interruptible constraint means that when the adjustment type is interruptible, the number of interruptions for the task ID is less than or equal to the interruption threshold. The transferable constraint requires that, when the adjustment type is transferable, the migration time point of the task ID must be within its corresponding adjustment time window.

10. A green energy regulation system for computer-powered coordination, applied to the green energy regulation method for computer-powered coordination as described in any one of claims 1-9, characterized in that, include: A data acquisition module and a data analysis module; the data acquisition module and the data analysis module are connected to each other; The data acquisition module acquires data from the target area through a data acquisition device. The target area data includes green electricity data and computing power data corresponding to the target area; The data analysis module includes a list integration unit and a result generation unit; The list integration unit: determines the list of computing power carbon emission adjustment capabilities based on target area data; The result generation unit generates computing power scheduling instructions based on green electricity data and a list of computing power carbon emission adjustment capabilities through scheduling optimization.

Images