A multi-computing center cooperative scheduling method and device, equipment and storage medium

Abstract

The application relates to the technical field of power systems, and provides a multi-computing power center cooperative scheduling method and device, equipment and a storage medium. The method comprises the following steps: receiving photovoltaic output data and computing power load data reported by each alliance computing power center in a preset green electricity sharing alliance; calculating the green electricity surplus and deficiency state of each alliance computing power center based on the photovoltaic output data and the computing power load data, and integrating the green electricity surplus and deficiency state into an alliance-level green electricity supply and demand panoramic view; identifying the green electricity supply and demand matching relationship of a current scheduling period based on the alliance-level green electricity supply and demand panoramic view; generating a mutual aid scheduling scheme based on the green electricity supply and demand matching relationship; performing the computing power task migration mutual aid scheme or the green electricity right and interest settlement mutual aid scheme, and obtaining mutual aid execution results of each scheduling period; calculating the green electricity utilization rate of the whole alliance and the mutual aid contribution degree of each alliance computing power center based on the mutual aid execution results, and generating feedback information. The application can realize cross-center green electricity cooperative consumption.

Description

Technical Field

[0001] This application relates to the field of power equipment technology, and in particular to a multi-computing center collaborative scheduling method, a multi-computing center collaborative scheduling device, a corresponding electronic device, and a corresponding computer-readable storage medium. Background Technology

[0002] With the rapid development of industries such as artificial intelligence, big data, and cloud computing, the energy consumption of computing power centers continues to rise, and data centers have become a significant component of the overall electricity consumption of society. Under the "dual carbon" target (carbon reduction and emission reduction), promoting the use of green electricity in computing power centers has become an industry consensus. More and more computing power centers are building distributed photovoltaic and other renewable energy power generation facilities on their own sites or in the surrounding areas to achieve direct green electricity supply and local consumption, thereby reducing carbon emission intensity.

[0003] The current mainstream green energy consumption model is based on a single computing center, meaning each computing center independently manages the balance between its own photovoltaic output and computing load. When photovoltaic output is insufficient, the computing center purchases conventional electricity from the grid to supplement it; when photovoltaic output exceeds its own load demand, surplus green energy can only be fed into the grid at a low price or directly curtailed. There is a lack of horizontal sharing and allocation mechanisms for green energy resources among the computing centers, which prevents the maximization of green energy utilization. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, and storage medium for collaborative scheduling of multiple computing centers, which can realize the collaborative consumption of green electricity across centers, extend the boundary of green electricity consumption from a single data center to the entire alliance, significantly improve the utilization rate of green electricity within the alliance, and reduce the investment in self-built energy storage in individual centers.

[0005] In one aspect, this application provides a multi-computing center collaborative scheduling method, applied to a multi-computing center collaborative scheduling system, wherein the multi-computing center collaborative scheduling system is equipped with a centralized scheduling platform, and the method includes: The centralized scheduling platform receives photovoltaic output data and computing load data reported by each computing center in the pre-set green electricity sharing alliance. The pre-set green electricity sharing alliance is organized and generated by multiple computing centers with local renewable energy power generation capabilities. In each scheduling cycle, the green electricity surplus and shortage status of each computing center in the alliance is calculated based on the photovoltaic output data and the computing load data, and the green electricity surplus and shortage status is integrated into an alliance-level green electricity supply and demand panoramic view. Based on the alliance-level green electricity supply and demand panoramic view, the green electricity supply and demand matching relationship of the current scheduling cycle is identified; A mutual assistance scheduling scheme is generated based on the green electricity supply and demand matching relationship. The mutual assistance scheduling scheme includes a computing power task migration mutual assistance scheme and a green electricity rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme is used to instruct the computing power load to be transferred from the green electricity shortage party to the green electricity surplus party. The green electricity rights settlement mutual assistance scheme is used to instruct the surplus green electricity and its corresponding carbon emission reduction rights of the green electricity surplus party to be converted into green electricity shortage party according to the settlement rules agreed in the alliance. The computing power task migration mutual assistance scheme or the green electricity rights settlement mutual assistance scheme is executed accordingly to obtain the mutual assistance execution results for each scheduling cycle; Based on the mutual assistance execution results, the overall green electricity utilization rate of the alliance and the mutual assistance contribution of each alliance computing center are calculated, and feedback information is generated. The feedback information is used to continuously optimize the scheduling strategy and serves as the basis for the allocation of rights and interests and settlement reconciliation within the alliance.

[0006] On the other hand, this application provides a multi-computing center collaborative scheduling device, applied to a multi-computing center collaborative scheduling system, wherein the multi-computing center collaborative scheduling system is equipped with a centralized scheduling platform, and the device includes: The data receiving module is used to receive photovoltaic output data and computing load data reported by each computing power center in the preset green electricity sharing alliance through the centralized scheduling platform; the preset green electricity sharing alliance is organized and generated by multiple computing power centers with local renewable energy power generation capabilities; in each scheduling cycle, the green electricity surplus and shortage status of each computing power center is calculated based on the photovoltaic output data and the computing load data, and the green electricity surplus and shortage status is integrated into an alliance-level green electricity supply and demand panoramic view; The matching relationship generation module is used to identify the green electricity supply and demand matching relationship for the current scheduling period based on the alliance-level green electricity supply and demand panoramic view; The mutual assistance scheduling scheme generation module is used to generate a mutual assistance scheduling scheme based on the green electricity supply and demand matching relationship. The mutual assistance scheduling scheme includes a computing power task migration mutual assistance scheme and a green electricity rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme is used to instruct the computing power load to be transferred from the green electricity shortage party to the green electricity surplus party. The green electricity rights settlement mutual assistance scheme is used to instruct the surplus green electricity and its corresponding carbon emission reduction rights of the green electricity surplus party to be converted into green electricity shortage party according to the settlement rules agreed upon within the alliance. The mutual assistance execution module is used to execute the computing power task migration mutual assistance scheme or the green electricity rights settlement mutual assistance scheme accordingly, and obtain the mutual assistance execution results for each scheduling cycle; The feedback information generation module is used to calculate the overall green electricity utilization rate of the alliance and the mutual contribution of each alliance computing center based on the mutual assistance execution results, and generate feedback information; the feedback information is used to continuously optimize the scheduling strategy and serve as the basis for the allocation of rights and interests and settlement reconciliation within the alliance.

[0007] In another aspect, this application also provides an electronic device, including: a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements any of the multi-computing center collaborative scheduling methods described in the present application.

[0008] In another aspect, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the multi-computing center collaborative scheduling method described in any one of the claims.

[0009] In another aspect, this application also provides a computer program product containing instructions that, when run on a computer, cause the computer to execute the multi-computing center collaborative scheduling method described in the above aspects.

[0010] The multi-computing center collaborative scheduling method, apparatus, equipment, and storage medium provided in this application receive photovoltaic output data and computing load data reported by each computing center in a pre-set green electricity sharing alliance through a centralized scheduling platform. In each scheduling cycle, the green electricity surplus / shortage status of each alliance computing center can be calculated based on the photovoltaic output data and computing load data. This surplus / shortage status is then integrated into an alliance-level green electricity supply and demand panorama view. Based on this panorama view, the green electricity supply and demand matching relationship for the current scheduling cycle can be identified. A mutual assistance scheduling scheme is then generated based on this matching relationship, and mutual assistance schemes for computing task migration or green electricity rights settlement are implemented. The corresponding execution yields the mutual assistance execution results for each scheduling cycle. Finally, based on the mutual assistance execution results, the overall green energy utilization rate of the alliance and the mutual assistance contribution of each alliance computing center are calculated. Feedback information is generated for continuous optimization of the scheduling strategy and serves as the basis for the allocation of rights and interests and settlement reconciliation within the alliance. The mutual assistance scheduling scheme can include a computing power task migration mutual assistance scheme and a green energy rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme instructs the transfer of computing power load from green energy shortage parties to green energy surplus parties. The green energy rights settlement mutual assistance scheme instructs the surplus green energy and its corresponding carbon emission reduction rights of the green energy surplus parties to be converted into green energy shortage parties according to the settlement rules agreed upon within the alliance. By constructing a green energy sharing alliance composed of multiple computing centers, a centralized scheduling platform is set up to collect green energy surplus and shortage information of each member in real time. Cross-center collaborative green energy consumption is achieved through two mutual assistance methods: computing power task migration and green energy rights settlement. This expands the green energy consumption boundary from a single data center to the entire alliance, significantly improving the green energy utilization rate within the alliance while reducing the investment in self-built energy storage for individual entities. Attached Figure Description

[0011] Figure 1 This is an overall structural block diagram of a multi-computing center collaborative scheduling system provided in an embodiment of this application; Figure 2 This is a flowchart illustrating the steps of a multi-computing center collaborative scheduling method provided in an embodiment of this application; Figure 3 This is a schematic diagram illustrating the implementation process of multi-computing center collaborative scheduling based on the green electricity surplus mutual assistance alliance provided in this application embodiment; Figure 4 This is a schematic diagram illustrating the working principle of local data collection and reporting provided in the embodiments of this application; Figure 5 This is a schematic diagram illustrating the working principle of the mutual assistance scheduling decision module provided in the embodiments of this application; Figure 6 This is a schematic diagram illustrating the working principle of mutual assistance execution and settlement provided in the embodiments of this application; Figure 7 This is a flowchart illustrating the specific implementation steps provided in the embodiments of this application; Figure 8 This is a diagram showing the relationship between key technical features provided in the embodiments of this application; Figure 9 This is a structural block diagram of a multi-computing center collaborative scheduling device provided in an embodiment of this application; Figure 10 This is a structural block diagram of an electronic device provided in an embodiment of this application; Figure 11 This is a structural block diagram of a computer-readable storage medium provided in an embodiment of this application. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. 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 skilled in the art without creative effort are within the scope of protection of this application.

[0013] To facilitate understanding of the multi-computing center collaborative scheduling method provided in the embodiments of this application by those skilled in the art, the technical terms that may be involved are explained as follows: Computing power center: refers to data center facilities that deploy large-scale computing resources (including server clusters, storage devices, network devices, etc.) and provide computing services to the outside world. They are usually equipped with a certain scale of local renewable energy power generation system (such as distributed photovoltaics) and can independently undertake the computing and delivery of computing power tasks. It is the basic participating unit of the Green Electricity Mutual Aid Alliance in the embodiments of this application.

[0014] Green Power Surplus refers to the portion of renewable energy remaining after a computing center's local renewable energy (such as photovoltaic power generation systems) generates more electricity than its current computing load requires during a given period. In traditional models, this surplus is typically either fed into the grid at low prices or directly curtailed, resulting in a waste of green electricity resources.

[0015] The Green Power Mutual-Aid Alliance refers to a green power resource sharing organization formed by multiple computing centers based on a protocol. Members of the alliance share their respective green power surplus and shortage information through a centralized scheduling platform, and achieve cross-center collaborative consumption of green power through computing power task migration or internal settlement of green power rights, thereby extending the boundary of green power utilization from a single center to the entire alliance.

[0016] Computing Task Migration refers to the process within a green energy mutual aid alliance of transferring computing tasks awaiting execution or currently queued at one computing center to another computing center with surplus green energy through a scheduling platform. This method matches the distribution of green energy supply by moving computing loads, achieving indirect cross-center consumption of green energy.

[0017] Green Power Credit Settlement refers to a settlement mechanism within a green power mutual aid alliance where, when one computing center has a surplus of green power while another has a shortage, the surplus party's green power rights (including green power volume and carbon emission reduction rights) are transferred to the shortage party at an agreed price or conversion rule through a scheduling platform. This allows the shortage party to obtain a share of green power consumption on its books, thereby increasing the overall green power consumption rate of the alliance.

[0018] Among the technologies related to green electricity consumption, a single computing center green electricity self-consumption scheduling scheme is adopted. Its working principle involves deploying a local energy management system in a single computing center to monitor photovoltaic (PV) output and computing load in real time. When PV output is sufficient, green electricity is prioritized; when PV output is insufficient, electricity is purchased from the grid; and when PV output is surplus, excess electricity is either fed into the grid at a low price or used to charge local energy storage. The application scenario of this scheme is limited to the scope of a single data center, and key parameters include the ratio of PV installed capacity to peak load and the matching of energy storage capacity to average daily surplus electricity.

[0019] In the aforementioned single-computing-center green energy self-consumption scheduling scheme, the self-consumption boundary of a single center is limited. Peak photovoltaic output and low computing load often occur simultaneously, leading to a large amount of green energy being fed into the grid at low prices or being curtailed, resulting in low green energy utilization. This indicates a drawback: the green energy consumption boundary is limited to a single center, leading to low overall utilization. Specifically, under the existing single-center self-consumption model, each computing center independently faces the time-series mismatch between photovoltaic output and computing load. When the photovoltaic output of a certain computing center exceeds its own load demand, even if other computing centers are experiencing green energy shortages at the same time, the surplus green energy cannot be shared between centers and can only be fed into the grid at low prices or curtailed. This fragmented consumption model results in a green energy utilization rate at the regional or cluster level that is significantly lower than the theoretical optimal level.

[0020] Furthermore, to improve self-consumption rates, a single data center needs to invest heavily in energy storage facilities, resulting in high construction and operation costs and long investment recovery periods. This leads to the disadvantages of high investment pressure and poor economic efficiency for single-center energy storage. To alleviate the timing mismatch between photovoltaic output and load, a single computing center must be equipped with a large-capacity energy storage system to store surplus green electricity. However, the construction and operation costs of energy storage systems are high, and the storage capacity must be designed based on peak surplus, resulting in limited actual utilization. This leads to persistently high economic costs per unit of green electricity consumption, limiting further increases in the scale of green electricity consumption.

[0021] Furthermore, the green energy surplus and shortage information of each computing center is isolated, making it impossible to achieve optimized allocation of green energy resources on a larger scale. This indicates a lack of cross-center green energy surplus and shortage information sharing and coordinated scheduling mechanisms. Specifically, the energy management systems of each existing computing center operate independently, lacking convergence and interaction channels for green energy surplus and shortage information, and further lacking cross-center coordinated scheduling decision-making capabilities. This means that even if multiple computing centers are geographically adjacent or belong to the same operating entity, they cannot leverage their complementary green energy characteristics to improve overall consumption efficiency.

[0022] This application's embodiments can solve problems such as low green electricity utilization, high energy storage investment pressure, and lack of cross-center coordination and scheduling mechanisms under the existing single-center self-consumption model. By organizing multiple computing centers with local renewable energy generation capabilities into a green electricity sharing alliance, and using a centralized scheduling platform to uniformly collect, aggregate, and analyze the green electricity supply and demand status information of each alliance member, alliance-level green electricity consumption optimization is achieved based on two mutual assistance methods: computing task migration and green electricity rights settlement.

[0023] Reference Figure 1 This diagram illustrates the overall structural block diagram of a multi-computing center collaborative scheduling system provided in an embodiment of this application. The overall system structure can include three levels and eight functional modules, which work collaboratively with each other through data interaction and control commands.

[0024] Optionally, the first level is the alliance member level, which contains two core modules: a local energy data acquisition module and a local computing load monitoring module.

[0025] Each alliance member's computing center deploys a local energy data acquisition module, primarily responsible for collecting real-time output data from the local photovoltaic power generation system, including current instantaneous output, cumulative power generation, and power generation curves. It should be noted that the acquisition cycle is consistent with the scheduling platform's scheduling cycle to ensure data timeliness; and the local energy data acquisition module reports the collected data to the scheduling platform through a standardized data interface.

[0026] Each alliance member's computing center deploys a local computing load monitoring module, primarily responsible for real-time monitoring of local computing load power consumption, including current total power consumption, power consumption percentage of each task queue, and load change trends. The local computing load monitoring module also maintains basic attribute information of local task queues, including task type, portability flag, estimated execution time, and data dependencies. This information provides the scheduling platform with necessary decision-making basis for determining whether a task is suitable for migration. This application's embodiments do not impose any limitations on this.

[0027] Optionally, the second layer is the centralized scheduling platform layer, which is the core of the system and includes four functional modules: green electricity surplus and shortage information aggregation module, mutual assistance scheduling decision module, computing power task migration and execution module, and green electricity rights settlement module.

[0028] The green energy surplus / shortage information aggregation module primarily receives photovoltaic output data and computing load data reported by all alliance members, and calculates the green energy surplus / shortage status of each member in real time. Specifically, the calculation method is as follows: subtract the current computing load power consumption from the current photovoltaic output of each member; a positive difference indicates a green energy surplus, while a negative difference indicates a green energy shortage. This module can integrate the surplus / shortage status information of all members to form an alliance-level panoramic view of green energy supply and demand for subsequent scheduling decisions.

[0029] The mutual assistance scheduling decision module is the core decision engine of the scheduling platform. Based on a comprehensive view of green electricity supply and demand, it identifies members with surplus green electricity and members with shortages, establishing supply-demand matching relationships. During the matching process, this module comprehensively evaluates multiple factors, such as the available surplus amount of the surplus party, the shortage amount of the shortage party, the portability of the tasks to be migrated, the network transmission conditions for task migration, and the costs and benefits of the two mutual assistance methods. Ultimately, it generates a mutual assistance scheduling scheme, clearly specifying the supply-demand matching relationship for each pair. Each pair corresponds to a mutual assistance path, which clarifies the direction of green electricity flow (from the surplus party to the shortage party), the mutual assistance method (computing power task migration or green electricity rights settlement), and the amount of mutual assistance. The surplus party in the supply-demand matching relationship is the green electricity supplier in the mutual assistance path, and the shortage party is the green electricity receiver in the mutual assistance path. The pairing relationship between the two directly determines the flow path of green electricity resources or rights, whether to use computing power task migration or green electricity rights settlement, and the specific migration task list or settlement amount of green electricity. This application does not limit the scope of the embodiments.

[0030] The computing power task migration execution module is primarily responsible for executing the portion of the scheduling scheme involving computing power task migration. Specifically, it can issue task migration instructions to computing power centers with shortages, issue task reception instructions to computing power centers with surpluses, coordinate the transmission and handover of task data, and monitor the execution status of the migration tasks at the target centers. After the migration is completed, this module can provide feedback on the migration execution results to the scheduling decision module.

[0031] The green electricity rights settlement module is primarily responsible for executing the green electricity rights settlement portion of the dispatch plan. Specifically, based on the settlement rules agreed upon by the alliance, it converts the surplus green electricity and corresponding rights of the party with surplus green electricity into those with shortage green electricity, generates settlement records, and notifies both parties for confirmation. These settlement records include information such as the settlement period, the amount of green electricity settled, the settlement price or conversion ratio, and the corresponding carbon emission reduction rights. This module maintains the alliance's internal green electricity rights ledger, ensuring the traceability of settlement data.

[0032] Optionally, the third level is the feedback and evaluation layer, which includes two modules: an execution result collection module and an effect evaluation and strategy optimization module.

[0033] The execution result acquisition module primarily collects actual green energy consumption data from each member after the end of each scheduling cycle, including the actual amount of green energy used, the additional green energy consumed due to mutual assistance, and the actual execution status of migration tasks. The collected results can be used to verify the actual effectiveness of the scheduling scheme.

[0034] The performance evaluation and strategy optimization module primarily calculates evaluation indicators such as the overall green energy utilization rate of the alliance, the mutual contribution of each member, and the cost-benefit ratio of mutual assistance based on the execution result data. The evaluation results generate statistical reports for alliance management and are also fed back to the mutual assistance dispatch decision-making module to optimize decision parameters and matching strategies for subsequent dispatch cycles.

[0035] The decision parameters can specifically include the following categories: The first category is supply and demand matching threshold parameters, including the balance state judgment threshold for the green electricity surplus / deficit difference (i.e., when the absolute value of the difference between photovoltaic power output and computing power load power consumption is less than this threshold, it is marked as a balanced state and does not participate in mutual assistance pairing), and the minimum mutual assistance amount threshold (i.e., when the mutual assistance amount of a single pairing is lower than this threshold, it will not be executed to avoid the management overhead of small mutual assistance exceeding the benefits); The second category is task migration condition parameters, including the upper limit threshold for the amount of data that can be migrated, the upper limit constraint of network bandwidth utilization, and the maximum allowable delay increment for a single migration; The third category is mutual assistance method selection weight parameters, including the priority weight coefficient of computing power task migration method relative to green electricity equity settlement method, and the ratio threshold of migration cost to green electricity consumption benefit (when the migration cost-benefit ratio exceeds this threshold, it switches to equity settlement method); The fourth category is historical feedback adjustment parameters, including the sliding average window length of the success rate of various task migrations, and the adjustment range of automatically tightening migration conditions when the success rate is lower than the set threshold. Matching strategy refers to the sorting rules and matching methods for supply and demand, such as whether to introduce geographical proximity as an additional consideration for matching, or whether to allow a surplus party to simultaneously allocate mutual aid to multiple shortage parties. This application does not impose any limitations on these aspects.

[0036] like Figure 1 The data flow relationship of the eight functional modules shown can be represented as follows: the local energy data acquisition module and the local computing power load monitoring module at the alliance member level collect photovoltaic output and computing power load data of each member, and report them to the green electricity surplus and shortage information aggregation module at the dispatch platform level; the aggregation module transmits the integrated supply and demand panorama to the mutual assistance dispatch decision module, which generates a dispatch plan; the dispatch plan is then distributed to the computing power task migration execution module and the green electricity rights settlement module for implementation; the execution results are collected by the execution result acquisition module and transmitted to the effect evaluation and strategy optimization module, forming a closed-loop feedback. All modules work collaboratively to support the complete process of alliance-level green electricity mutual assistance dispatch.

[0037] Reference Figure 2 The diagram illustrates a flowchart of a multi-computing center collaborative scheduling method provided in an embodiment of this application, which may specifically include the following steps: Step S201: Receive photovoltaic power output data and computing load data reported by each computing center in the preset green electricity sharing alliance through the centralized scheduling platform.

[0038] like Figure 3 As shown, a green electricity sharing alliance can be constructed first. Specifically, the green electricity sharing alliance can be generated by multiple computing centers with local renewable energy generation capabilities.

[0039] In some embodiments of this application, multiple computing centers can join a green electricity sharing alliance based on a protocol, with each member agreeing on green electricity information sharing standards, mutual assistance priority rules, and rights settlement methods.

[0040] Specifically, the mutual assistance priority rules can include the following three levels: The first level is the surplus / shortage priority, which sorts each member's green energy surplus or shortage from largest to smallest, prioritizing pairing the member with the largest surplus with the member with the largest shortage to maximize the mutual assistance in a single pairing and reduce the number of matching rounds; the second level is the mutual assistance method priority, which prioritizes the computing power task migration method (achieving physical absorption) in the same supply and demand pairing, and only chooses the green energy rights settlement method when the migration conditions are not met; the third level is the task migration priority, which prioritizes the migration of batch offline tasks with small data transmission volume, short execution time, and low latency sensitivity to reduce migration costs and improve migration success rate. The above three levels of priority rules together constitute the priority system of the alliance's mutual assistance scheduling and other basic terms for alliance operation.

[0041] Each computing center within the alliance needs to deploy local data acquisition nodes to report its real-time photovoltaic output data, real-time computing load data, and green electricity surplus / deficit status to the centralized dispatch platform. The formation of the alliance does not change the independent operating status of each computing center, but rather establishes horizontal collaborative relationships at the level of green electricity consumption.

[0042] In practical applications, such as Figure 7 As shown, the establishment and initial configuration of the Green Electricity Mutual Aid Alliance involves multiple computing centers with local photovoltaic power generation capabilities signing a Green Electricity Mutual Aid Alliance Agreement, clarifying the rights and obligations of each member. The agreement may include: each member must install and maintain a local energy data acquisition module and a computing load monitoring module, and report data according to the agreed scheduling cycle; basic rules for green electricity rights settlement within the alliance, including settlement pricing methods and carbon emission reduction rights conversion ratios; and basic specifications for computing task migration, including data security protection requirements and task execution result delivery standards. After the alliance agreement is signed, the centralized scheduling platform can complete the configuration initialization, register the basic information of all alliance members, establish communication connections with each member's data acquisition nodes, set scheduling cycle parameters, create a green electricity rights ledger, and establish an initial account for each member.

[0043] Furthermore, such as Figure 3As shown, information can be aggregated based on a centralized scheduling platform. Specifically, the scheduling platform is the core hub of this application embodiment, responsible for receiving green energy surplus and shortage information reported by all alliance members, facilitating the subsequent formation of an alliance-level panoramic view of green energy supply and demand.

[0044] Optionally, within each scheduling cycle, the scheduling platform can calculate the surplus or shortage of green electricity for each member of the alliance in real time, identify the current green electricity supply and demand matching relationship, and then generate a mutual assistance scheduling plan. The scheduling platform has unified scheduling decision-making capabilities, and can comprehensively consider factors such as the surplus or shortage of green electricity for each member, computing power task characteristics, network transmission conditions, and mutual assistance costs, to select the optimal mutual assistance method and path.

[0045] In some embodiments of this application, the photovoltaic output data for the current scheduling period is read from the photovoltaic power generation system by each alliance computing power center; the computing load data for the current scheduling period is read from the local computing power management system by each alliance computing power center. Furthermore, the centralized scheduling platform can receive task queue information reported by each alliance computing power center within a preset green electricity sharing alliance, and each task to be executed in the task queue information is determined and marked with a portability identifier.

[0046] Combination such as Figure 1 The implementation process of the local energy data acquisition module shown can be represented as follows: the local energy data acquisition module is deployed on the photovoltaic power generation system side of each alliance member's computing center. This module connects to the photovoltaic inverter or photovoltaic monitoring system via a data interface and periodically reads the photovoltaic system's operating parameters according to a scheduling cycle consistent with the scheduling platform. The core data collected includes the average power generation and cumulative power generation within the current scheduling cycle. After collection, the module standardizes and encapsulates the data, adds a timestamp and member identification information, and reports it to the green energy surplus / deficit information aggregation module of the scheduling platform via a network channel. During the data reporting process, the module has a data verification function, marking abnormal data points to avoid the accuracy of scheduling decisions being affected by collection errors.

[0047] Combination such as Figure 1The implementation process of the local computing load monitoring module shown can be represented as follows: the local computing load monitoring module is deployed on the computing management system side of each alliance member's computing center. This module obtains the current total computing load power consumption and task queue information in real time through an interface with the local computing resource management system. For total power consumption data, the module reads the power metering data from the data center's power distribution system; for task queue information, the module extracts the attributes of each queued task from the local task scheduling system, focusing on extracting the task's portability identifier. The criteria for determining portability include: whether the task is a batch offline computing type, whether the task dataset size is within the portability range, and whether the task depends on local proprietary hardware resources. This module synchronously reports the power consumption data and task queue information to the scheduling platform.

[0048] like Figure 4 The schematic diagram illustrates the internal workflow of the two acquisition modules at the alliance member layer. Photovoltaic system output data is collected, standardized, and verified. Simultaneously, the portability identifiers of each task in the task queue are extracted from the collected computing power load power consumption data. Both types of data are then processed and uniformly reported to the green electricity surplus / deficit information aggregation module of the scheduling platform, serving as the input data source for subsequent scheduling decisions.

[0049] In practical applications, photovoltaic power output data, computing load data, and task queue information can be encapsulated into a unified data message and reported to the centralized scheduling platform through network channels.

[0050] In specific implementations, such as Figure 7 As shown, after the alliance starts operating, each member's local energy data acquisition module and computing load monitoring module operate synchronously according to the scheduling cycle. At the acquisition point of each scheduling cycle, the local energy data acquisition module can read the power output data of the current cycle from the photovoltaic power generation system, perform standardization processing, and data verification. At the same time, the local computing load monitoring module reads the current total power consumption data and task queue information from the local computing management system, and determines and marks the portability of each task to be executed. After the two types of data are processed locally, they can be encapsulated into a unified data message, with a timestamp and member identifier attached, and reported to the scheduling platform through the network channel. The data receiving port of the scheduling platform can receive and confirm the data reported by each member and check its integrity to ensure that the data is complete and usable.

[0051] In step S202, during each scheduling cycle, the green electricity surplus / deficit status of each alliance computing center is calculated based on photovoltaic power output data and computing load data, and the green electricity surplus / deficit status is integrated into an alliance-level green electricity supply and demand panoramic view.

[0052] The state of green electricity surplus and shortage can be divided into a state of green electricity surplus, a state of green electricity shortage, and a state of balance.

[0053] In some embodiments of this application, the photovoltaic output data and computing load data of each alliance computing center can be read one by one to calculate the green electricity surplus / deficit difference of each alliance computing center. Then, the green electricity surplus / deficit status of the corresponding alliance computing centers can be marked based on the green electricity surplus / deficit difference, and then integrated into an alliance-level green electricity supply and demand panoramic view based on the green electricity surplus / deficit status.

[0054] The green electricity surplus / deficit difference can be the difference between photovoltaic (PV) output and computing power load power consumption. In one scenario, a consortium computing center whose PV output exceeds its computing power load power consumption can be marked as having a green electricity surplus and the surplus amount recorded. In another scenario, a consortium computing center whose PV output is less than its computing power load power consumption can be marked as having a green electricity shortage and the shortage amount recorded. In yet another scenario, a consortium computing center whose PV output equals its computing power load power consumption can be marked as being in a balanced state. Finally, based on the aforementioned information, a comprehensive view of consortium-level green electricity supply and demand for the current scheduling period can be obtained.

[0055] Combination such as Figure 1 The implementation process of the green electricity surplus / shortage information aggregation module, as shown, can be represented as follows: After receiving the photovoltaic output and computing load data reported by all alliance members, the module calculates the green electricity surplus / shortage value for each member. The calculation logic is as follows: subtract the computing load power consumption of the member in the current scheduling cycle from its photovoltaic output power to obtain the green electricity surplus / shortage difference. A positive difference indicates that the member is in a green electricity surplus state, and the absolute value of the difference is the surplus green electricity available for mutual assistance; a negative difference indicates that the member is in a green electricity shortage state, and the absolute value of the difference is the green electricity that the member hopes to obtain through mutual assistance. The aggregation module integrates the surplus / shortage status of all members into an alliance-level green electricity supply and demand panoramic view. This view displays key information such as the member's identifier, current green electricity surplus / shortage status, surplus / shortage amount, and number of transferable tasks in a list format. The panoramic view is updated once in each scheduling cycle to ensure that scheduling decisions are based on the latest supply and demand status.

[0056] In practical applications, such as Figure 7As shown, after receiving the reported data from all alliance members for the current scheduling period, the green electricity surplus / shortage information aggregation module of the scheduling platform initiates the panoramic view generation process. This module reads the photovoltaic output and computing load data of each member one by one, calculating the green electricity surplus / shortage difference. For each member, if the photovoltaic output is greater than the computing load power consumption, it is marked as having a green electricity surplus state, and the surplus amount is recorded; if the photovoltaic output is less than the computing load power consumption, it is marked as having a green electricity shortage state, and the shortage amount is recorded; if the two are roughly equal, it is marked as being in a balanced state. The aggregation module summarizes the calculation results of all members to generate an alliance-level panoramic view of green electricity supply and demand for the current scheduling period. This view includes information such as each member's identifier, surplus / shortage status, surplus / shortage amount, and number of transferable tasks, stored in structured data form and transmitted to the mutual aid scheduling decision module.

[0057] Step S203: Identify the green electricity supply and demand matching relationship for the current scheduling cycle based on the alliance-level green electricity supply and demand panoramic view.

[0058] After receiving the alliance-level green energy supply and demand overview for the current scheduling period, a decision-making process can be initiated to generate, for example... Figure 3 The mutual assistance dispatch decision is shown. In some embodiments of this application, the green electricity supply and demand matching relationship of the current dispatch cycle can first be identified, so as to generate a mutual assistance dispatch decision based on the identified green electricity supply and demand matching relationship.

[0059] Specifically, such as Figure 7 As shown, firstly, the list of green electricity surplus members and the list of green electricity shortage members can be extracted from the alliance-level green electricity supply and demand panorama view. In one case, if there are no green electricity shortage members or no green electricity surplus members, there is no need for mutual assistance scheduling in the current cycle, and the decision module outputs an empty plan. In another case, if there are both surplus and shortage members, the decision module enters the supply and demand matching stage.

[0060] During the matching phase, the decision module can sort the green electricity surplus members in descending order based on their surplus amount, and sort the green electricity shortage members in descending order based on their shortage amount. Then, it can sequentially pair the green electricity surplus members with the green electricity shortage members in descending order to obtain the green electricity supply and demand matching relationship for the current scheduling cycle. Optionally, the top-ranked surplus members can be paired with the top-ranked shortage members sequentially; during pairing, the smaller of the surplus and shortage amounts can be used as the mutual assistance amount for this pairing. Specifically, if the surplus amount of the surplus member is greater than the shortage amount of the shortage member, the remaining surplus amount of the surplus member can continue to be paired with the next shortage member; conversely, the opposite is also true. This application embodiment does not impose any limitations on this.

[0061] Step S204: Generate a mutual assistance dispatch scheme based on the green electricity supply and demand matching relationship.

[0062] The mutual assistance scheduling scheme includes a computing power task migration mutual assistance scheme and a green electricity rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme instructs the transfer of computing power load from green electricity shortage parties to green electricity surplus parties, achieving physical absorption of green electricity. The green electricity rights settlement mutual assistance scheme instructs the surplus green electricity and its corresponding carbon emission reduction rights from green electricity surplus parties to green electricity shortage parties according to the settlement rules agreed upon within the alliance, achieving rights-based benefits.

[0063] In some embodiments of this application, after supply and demand matching is completed, the decision module can select a mutual assistance method for each pairing. Specifically, such as... Figure 7 As shown, task portability can be determined by checking whether the total power consumption of portable tasks with portability indicators in the task queue information reported by the green electricity shortage party meets the load transfer requirements corresponding to the mutual assistance amount of this pairing. In one case, if the portable tasks meet the load transfer requirements and network transmission conditions permit, the computing power task migration method can be selected, and the target task to be migrated can be selected from the portable tasks according to the preset priority to generate a computing power task migration mutual assistance scheme. In another case, if the portable tasks do not meet the load transfer requirements or network transmission conditions do not permit, the green electricity rights settlement method can be selected to generate a green electricity rights settlement mutual assistance scheme.

[0064] Combination such as Figure 1 The implementation process of the mutual assistance scheduling decision module shown can be represented in three stages.

[0065] The first stage is supply and demand matching. Specifically, this module filters all green energy surplus members and green energy shortage members from the panoramic view, sorts them according to the surplus amount from largest to smallest and the shortage amount from largest to smallest, and then performs supply and demand matching. The matching principle is to prioritize matching members with large surpluses with members with large shortages in order to reduce the number of matching rounds and improve the overall mutual assistance efficiency.

[0066] The second phase involves selecting the mutual assistance method. Specifically, for each supply-demand pairing, the scheduling decision module determines whether to adopt the computing power task migration method or the green electricity rights settlement method. The core basis for this judgment is the number and characteristics of migrateable tasks in the current task queue of the shortage party. If the shortage party has a sufficient number of migrateable tasks, and the data transmission volume of these tasks is acceptable under network bandwidth constraints, then the computing power task migration method is preferred, because this method can achieve the physical absorption of green electricity. If the shortage party lacks migrateable tasks, or the migration conditions are not met, then the green electricity rights settlement method is chosen as an alternative.

[0067] The third stage is scheme generation. Specifically, after the scheduling decision module determines all supply and demand matching methods, it generates a complete mutual assistance scheduling scheme. For computing power task migration methods, the scheme clearly specifies the task identifier, source center, and target center to be migrated; for green electricity rights settlement methods, it clearly specifies the amount of green electricity to be settled, the transferor, and the transferee. After the scheme is generated, it is distributed to the computing power task migration execution module and the green electricity rights settlement module, respectively.

[0068] like Figure 5 The schematic diagram illustrates the internal workflow of the mutual assistance scheduling decision-making module. This module first receives a panoramic view of green electricity supply and demand generated by the aggregation module, then performs supply and demand matching. Next, based on the portability of tasks and network conditions, it determines the appropriate mutual assistance method for each pairing, generating a complete mutual assistance scheduling plan. Finally, it sends the execution instructions from the plan to the corresponding execution modules. The entire decision-making process is completed once per scheduling cycle.

[0069] Step S205: Execute the computing power task migration mutual assistance scheme or the green electricity rights settlement mutual assistance scheme accordingly to obtain the mutual assistance execution results for each scheduling cycle.

[0070] like Figure 3 The computing power task migration shown is the first mutual assistance method in this application embodiment. When one computing power center (denoted as Center A) in the alliance has a surplus of green electricity, while another computing power center (denoted as Center B) has a shortage of green electricity, the scheduling platform can migrate some of the computing power tasks to be executed at Center B to Center A for execution. By transferring the computing power load from the green electricity shortage side to the green electricity surplus side, the surplus green electricity of Center A can be absorbed by the migrated computing power tasks, while Center B's load decreases due to task relocation, reducing its demand for conventional electricity. Computing power task migration is suitable for batch computing tasks that are not sensitive to latency and have relatively controllable data transmission volume. When selecting tasks to be migrated, the scheduling platform needs to comprehensively evaluate the portability of the task, migration cost, and green electricity absorption benefits.

[0071] In some embodiments of this application, such as Figure 7 As shown, the execution of computing power task migration can be represented by the computing power task migration execution module receiving the task migration instruction and processing each migration task one by one. For each task to be migrated, the following process can be executed: a task pause instruction is sent to the green energy shortage party (as the source center), the source center retrieves the task from its local execution queue, and at this time, the input data and execution environment configuration information required by the target task to be migrated can be packaged into a data packet, and the data packet is transmitted to the green energy surplus party (as the target center) through the network; after receiving the data packet, the target center can allocate computing power resources and load the task execution environment locally to execute the target task to be migrated and obtain the mutual assistance execution result.

[0072] In practical applications, the migration execution module can continuously monitor the execution status of tasks at the target center, including whether the task started normally, the execution progress, and whether any anomalies occurred. After the task is completed, the target center can send the execution result data back to the source center; and after the migration execution module confirms that the result has been sent back, it can record all the status information of this migration, including the amount of data migrated, the migration time, the task execution time, the amount of green electricity consumed, etc., and submit this information to the execution result acquisition module.

[0073] like Figure 3 The green electricity rights settlement shown is the second mutual assistance method in this application embodiment. When migrating computing power tasks is not feasible or the migration cost is too high (e.g., tasks are latency-sensitive or the data volume is too large to be migrated), the scheduling platform can use the green electricity rights settlement method to achieve mutual assistance. Under this method, the surplus green electricity of the green electricity surplus party and its corresponding carbon emission reduction rights are converted to the green electricity shortage party according to the settlement rules agreed upon within the alliance. The shortage party obtains the corresponding green electricity consumption share on its books, thereby increasing the proportion of green electricity in its energy consumption. Green electricity rights settlement does not involve the actual relocation of computing power tasks, but rather achieves green electricity sharing within the alliance through rights transfer.

[0074] In some embodiments of this application, such as Figure 7 As shown, the execution of green electricity rights settlement can be described as follows: after receiving the rights settlement instruction, the green electricity rights settlement module performs operations according to the settlement rules agreed upon by the alliance. For each settlement, the current account information of the green electricity surplus party (i.e., the transferor) and the green electricity shortage party (i.e., the transferee) can be read from the preset green electricity rights ledger, and the agreed settlement amount of green electricity can be transferred from the account of the green electricity surplus party to the account of the green electricity shortage party. At the same time, according to the settlement rules agreed upon within the alliance (i.e., the agreed carbon emission reduction rights conversion rules), the amount of carbon emission reduction rights involved in this settlement of green electricity can be calculated, and the corresponding carbon emission reduction rights transfer record can be made in the preset green electricity rights ledger, thereby obtaining the mutual assistance execution result.

[0075] In practical applications, after settlement is completed, this module can generate settlement vouchers and send them to the transferor and transferee for confirmation. The settlement vouchers can record complete information such as settlement time, settlement period, settled green electricity volume, settlement price or conversion basis, and carbon emission reduction rights transfer amount; after confirmation by both parties, the settlement vouchers can be archived and stored in the historical database of the dispatch platform.

[0076] Combination such as Figure 1The implementation process of the computing power task migration execution module shown can be represented as follows: after receiving the task migration instruction from the scheduling scheme, the module initiates the task migration process. First, the module sends a task pause and data packaging instruction to the source center (the party with a shortage of green electricity). The source center packages the execution data and input data of the specified task for transmission. Then, the module coordinates the data transmission between the source center and the target center (the party with surplus green electricity), transmitting the task data packet to the target center. After receiving the data, the target center loads and starts the execution of the task on its local computing resources. The migration execution module continuously monitors the execution progress of the migration task at the target center. When the task is completed, the execution result is sent back to the source center. Key status information throughout the migration process is recorded for subsequent execution result collection.

[0077] Step S206: Calculate the overall green electricity utilization rate of the alliance and the mutual contribution of each alliance computing center based on the mutual assistance execution results, and generate feedback information.

[0078] The results of the mutual assistance implementation can provide information on the actual green energy consumption of each alliance's computing power centers.

[0079] like Figure 3 The closed-loop feedback and effect evaluation shown herein involve the dispatch platform continuously running the above process according to a set dispatch cycle. At the end of each dispatch cycle, the dispatch platform can collect the actual green energy consumption of each member, calculate the overall green energy utilization rate of the alliance and the mutual contribution of each member, and generate feedback information, which can be specifically formed into a statistical report. This feedback information can be used to continuously optimize the dispatch strategy and serve as the basis for the allocation of rights and interests and settlement reconciliation within the alliance.

[0080] In some embodiments of this application, such as Figure 7 As shown, after the mutual assistance execution of each scheduling cycle is completed, the execution result collection module can summarize all execution data for the current cycle, including the actual green electricity consumed by each member (including local self-consumption and mutual assistance consumption), the execution status of task migration, and the completion status of equity settlement. Based on this data, the effect evaluation and strategy optimization module calculates evaluation indicators such as the overall green electricity utilization rate of the alliance, the improvement rate of green electricity consumption rate of each member, and the cost-effectiveness ratio of mutual assistance operations. The evaluation results generate statistical reports for the alliance management to review and use as a reference for decision-making. At the same time, the effect evaluation module feeds back the evaluation indicators to the mutual assistance scheduling decision-making module to optimize the matching strategy and method selection logic for subsequent scheduling cycles. For example, if the execution success rate of a certain type of task migration remains high, the migration conditions for that type of task can be appropriately relaxed in subsequent scheduling; if the equity settlement frequency among some members is too high, it can be suggested that these members adjust their local photovoltaic installations or load configurations. Through continuous closed-loop feedback, the efficiency of the alliance's green electricity mutual assistance scheduling gradually improves.

[0081] Combination such as Figure 1 The implementation process of the green electricity rights settlement module shown can be described as follows: after receiving the rights settlement instruction from the dispatch plan, the green electricity rights settlement module executes the transfer operation of green electricity rights according to the settlement rules agreed upon by the alliance. This module maintains a green electricity rights ledger within the alliance, recording data such as each member's green electricity production, self-consumption, mutual transfer volume, and mutual transfer receipt volume. When a settlement is executed, the module transfers the surplus green electricity of the transferor to the transferee in the ledger, while recording the settlement price and the corresponding carbon emission reduction rights. After the settlement is completed, the module sends a settlement notification to both members and stores the settlement record in the historical database of the dispatch platform.

[0082] like Figure 6 The schematic diagram illustrates the parallel working logic of the computing power task migration execution module and the green electricity rights settlement module. Each module executes its respective mutual assistance instructions according to the scheduling scheme: the task migration path involves task data packaging and transmission, task execution at the target center, and status feedback; the rights settlement path involves green electricity rights ledger transfer and settlement record generation. Both paths ultimately feed back their execution status to the execution result acquisition module, providing a data foundation for subsequent evaluation.

[0083] like Figure 3 The implementation process illustrated demonstrates the five core components of the overall technical solution and their logical relationships. First, multiple computing centers form a green energy mutual assistance alliance. Then, a centralized scheduling platform aggregates green energy surplus and shortage information from each member. Based on this, the scheduling platform makes mutual assistance scheduling decisions, determining the optimal mutual assistance method and path. Subsequently, mutual assistance is implemented through either computing power task migration or green energy rights settlement. Finally, the scheduling platform collects actual execution results for closed-loop feedback and effect evaluation, continuously optimizing the scheduling strategy. Through the coordinated operation of these five components, the technical goal of expanding the green energy consumption boundary from a single computing center to the alliance level can be achieved. The green energy surplus and shortage of each member within the alliance can be matched and complemented on a larger scale, significantly improving the overall green energy utilization rate. Simultaneously, the mutual assistance effect at the alliance level reduces the dependence of individual computing centers on large-capacity energy storage systems, optimizing economic costs. In other words, a complete scheduling closed loop can be formed based on these five components, ensuring the continuous and efficient operation of alliance-level green energy consumption.

[0084] like Figure 7The flowchart illustrating the specific implementation steps shows seven steps and their sequential logical relationships. After the alliance is established and initialized, it enters a periodic operation phase: each member collects and reports local data, the scheduling platform generates a comprehensive supply and demand view, the mutual assistance scheduling decision module generates a scheduling plan, and mutual assistance is implemented through two paths: task migration execution and equity settlement execution. After feedback and strategy optimization, the execution results enter the next scheduling cycle. The entire process forms a continuously operating closed loop, ensuring the continuous and effective operation of the alliance-level green electricity mutual assistance scheduling.

[0085] like Figure 8 The key technical feature relationships shown illustrate the logical relationships between the three key technical features of this invention. The alliance-level supply and demand panorama provides the data foundation for green electricity supply and demand matching. The matching results enter the task portability determination stage; based on the determination results, either the computing power task migration mode (achieving physical absorption of green electricity) or the green electricity rights settlement mode (achieving rights-level absorption) is entered; the execution results of both modes ultimately converge in the mutual benefit effect summary evaluation stage, forming a closed loop. These three technical features support each other and together constitute the core technical system of this invention.

[0086] Firstly, by constructing a comprehensive alliance-level green energy supply and demand view, the surplus and shortage information of green energy, originally scattered within various computing centers, is aggregated onto a unified scheduling platform. This forms a real-time overview of the green energy supply and demand status of all members within the alliance, enabling the scheduling platform to discover complementary green energy relationships among members at the alliance level—relationships invisible in a single-center self-consumption mode. It should be noted that the update frequency of the panoramic view is consistent with the scheduling cycle, ensuring that scheduling decisions are always based on the latest supply and demand status. The panoramic view includes information on each member's photovoltaic output, computing load, green energy surplus and shortage, number of transferable tasks, and other dimensions, providing complete input data for scheduling decisions.

[0087] Secondly, the alliance employs a collaborative mechanism combining two mutual aid modes: computing power task migration and green electricity rights settlement. Computing power task migration, by transferring computing load from green electricity shortage areas to green electricity surplus areas, achieves physical absorption of green electricity, making it the most direct form of mutual aid. However, this method is limited by constraints such as task portability and data transmission conditions, and is not applicable to all scenarios. Green electricity rights settlement, on the other hand, overcomes physical constraints, achieving green electricity sharing through rights-level transfer, making it more widely applicable, although its green electricity absorption effect is reflected only on paper. The collaborative use of these two modes ensures effective green electricity mutual aid within the alliance in various practical scenarios. The scheduling decision module flexibly selects the most suitable mutual aid mode based on the specific circumstances of each supply-demand pairing, maximizing the mutual aid effect.

[0088] Third, a systematic design for judging the portability of computing tasks is implemented. When reporting task queue information, the local computing load monitoring module marks each task to be executed with a portability identifier. The portability determination comprehensively considers factors such as task type (batch offline tasks are prioritized for portability), data dependencies (tasks without strong dependencies on local proprietary data are prioritized for portability), and latency sensitivity (tasks not sensitive to latency are prioritized for portability). When selecting specific tasks to migrate, the scheduling decision module further evaluates the relationship between the network transmission cost of migration and the benefits of green electricity consumption to ensure the economic rationality of the migration operation.

[0089] In this embodiment, a centralized scheduling platform receives photovoltaic output data and computing load data reported by each computing power center within a pre-defined green electricity sharing alliance. In each scheduling cycle, the green electricity surplus / shortage status of each alliance's computing power center can be calculated based on the photovoltaic output data and computing load data. This surplus / shortage status is then integrated into an alliance-level panoramic view of green electricity supply and demand. Based on this panoramic view, the green electricity supply and demand matching relationship for the current scheduling cycle can be identified. A mutual assistance scheduling scheme is then generated based on this matching relationship, and corresponding mutual assistance schemes for computing power task migration or green electricity rights settlement are executed to obtain the results for each scheduling cycle. The mutual assistance execution results over a period of time are used to calculate the overall green energy utilization rate of the alliance and the mutual assistance contribution of each alliance computing center. This generates feedback information for continuous optimization of scheduling strategies and serves as the basis for internal rights allocation and settlement reconciliation within the alliance. The mutual assistance scheduling scheme can include a computing power task migration mutual assistance scheme and a green energy rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme instructs the transfer of computing power load from green energy shortage parties to green energy surplus parties. The green energy rights settlement mutual assistance scheme instructs the surplus green energy and its corresponding carbon emission reduction rights of the green energy surplus parties to be converted into green energy shortage parties according to the settlement rules agreed upon within the alliance. By constructing a green energy sharing alliance composed of multiple computing centers, a centralized scheduling platform is set up to collect green energy surplus and shortage information of each member in real time. Cross-center collaborative green energy consumption is achieved through two mutual assistance methods: computing power task migration and green energy rights settlement. This expands the green energy consumption boundary from a single data center to the entire alliance, significantly improving the green energy utilization rate within the alliance while reducing the investment in self-built energy storage for individual centers.

[0090] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of this application.

[0091] Reference Figure 9 This diagram illustrates a structural block diagram of a multi-computing center collaborative scheduling device provided in an embodiment of this application. The device is applied to a multi-computing center collaborative scheduling system, which includes a centralized scheduling platform and may specifically include the following modules: The data receiving module 901 is used to receive photovoltaic output data and computing load data reported by each computing center in the preset green electricity sharing alliance through the centralized scheduling platform; the preset green electricity sharing alliance is organized and generated by multiple computing centers with local renewable energy power generation capabilities; The supply and demand panorama view generation module 902 is used to calculate the green electricity surplus and shortage status of each alliance computing center based on photovoltaic power output data and computing load data in each scheduling cycle, and integrate the green electricity surplus and shortage status into an alliance-level green electricity supply and demand panorama view. The matching relationship generation module 903 is used to identify the green electricity supply and demand matching relationship of the current scheduling cycle based on the alliance-level green electricity supply and demand panoramic view; The mutual assistance dispatch scheme generation module 904 is used to generate mutual assistance dispatch schemes based on the green electricity supply and demand matching relationship. The mutual assistance dispatch schemes include a computing power task migration mutual assistance scheme and a green electricity rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme is used to instruct the computing power load to be transferred from the green electricity shortage party to the green electricity surplus party. The green electricity rights settlement mutual assistance scheme is used to instruct the surplus green electricity and its corresponding carbon emission reduction rights of the green electricity surplus party to be converted into green electricity shortage parties according to the settlement rules agreed upon within the alliance. The mutual assistance execution module 905 is used to execute the mutual assistance scheme for computing power task migration or the mutual assistance scheme for green electricity rights settlement, and obtain the mutual assistance execution results for each scheduling cycle. The feedback information generation module 906 is used to calculate the overall green electricity utilization rate of the alliance and the mutual contribution of each alliance computing center based on the mutual assistance execution results, and generate feedback information. The feedback information is used to continuously optimize the scheduling strategy and serves as the basis for the allocation of rights and interests and settlement reconciliation within the alliance.

[0092] In some embodiments of this application, the data receiving module 901 is further configured to receive task queue information reported by each computing power center of the preset green electricity sharing alliance through a centralized scheduling platform; and to determine and mark each task to be executed in the task queue information as having a portability identifier.

[0093] In some embodiments of this application, the green electricity surplus / shortage status includes a green electricity surplus status, a green electricity shortage status, and a balance status; the supply and demand panoramic view generation module 902 may include the following sub-modules: The green electricity surplus / deficit difference calculation submodule is used to read the photovoltaic output data and computing load data of each alliance computing center one by one, and calculate the green electricity surplus / deficit difference of each alliance computing center. The green electricity surplus / deficit status marking submodule is used to mark alliance computing centers whose green electricity surplus / deficit difference indicates that their photovoltaic output is greater than their computing power load power consumption as green electricity surplus state and record the green electricity surplus amount; to mark alliance computing centers whose green electricity surplus / deficit difference indicates that their photovoltaic output is less than their computing power load power consumption as green electricity shortage state and record the green electricity shortage amount; and to mark alliance computing centers whose green electricity surplus / deficit difference indicates that their photovoltaic output is equal to their computing power load power consumption as balanced state. The supply and demand panorama view generation submodule is used to summarize and obtain the alliance-level green electricity supply and demand panorama view for the current scheduling cycle.

[0094] In some embodiments of this application, the matching relationship generation module 903 may include the following sub-modules: The matching relationship generation submodule is used to extract the green electricity surplus member list and the green electricity shortage member list from the alliance-level green electricity supply and demand panorama view respectively; sort the green electricity surplus members in descending order according to the surplus amount of each green electricity surplus member in the green electricity surplus member list, and sort the green electricity shortage members in descending order according to the shortage amount of each green electricity shortage member in the green electricity shortage member list; and pair the green electricity surplus members and the green electricity shortage members in descending order in turn to obtain the green electricity supply and demand matching relationship for the current scheduling period.

[0095] In some embodiments of this application, the mutual assistance scheduling scheme generation module 904 may include the following sub-modules: The mutual assistance scheduling scheme generation submodule is used to check whether the total power consumption of the transferable tasks with transferability identifiers in the task queue information reported by the green electricity shortage party meets the load transfer requirements corresponding to the mutual assistance amount of this pairing; if the transferable tasks meet the load transfer requirements and the network transmission conditions allow, the computing power task migration method is selected, and the target task to be migrated is selected from the transferable tasks according to the preset priority to generate a computing power task migration mutual assistance scheme; if the transferable tasks do not meet the load transfer requirements or the network transmission conditions do not allow, the green electricity rights settlement method is selected to generate a green electricity rights settlement mutual assistance scheme.

[0096] In some embodiments of this application, the mutual assistance execution module 905 may include the following sub-modules: The computing power task migration submodule is used to send a task pause command to the green energy shortage party, package the input data and execution environment configuration information required for the target task to be migrated into a data packet, and transmit the data packet to the green energy surplus party through the network; the green energy surplus party allocates computing power resources locally and loads the task execution environment to execute the target task to be migrated and obtain the mutual assistance execution result.

[0097] In some embodiments of this application, the mutual assistance execution module 905 may include the following sub-modules: The green electricity rights settlement submodule is used to read the current account information of the green electricity surplus party and the green electricity shortage party from the preset green electricity rights ledger, transfer the agreed settlement green electricity from the account of the green electricity surplus party to the account of the green electricity shortage party; and, according to the settlement rules agreed within the alliance, calculate the carbon emission reduction rights involved in this settlement of green electricity, and make corresponding carbon emission reduction rights transfer records in the preset green electricity rights ledger, thereby obtaining the mutual assistance execution result.

[0098] In this embodiment, a centralized scheduling platform receives photovoltaic output data and computing load data reported by each computing power center within a pre-defined green electricity sharing alliance. In each scheduling cycle, the green electricity surplus / shortage status of each alliance's computing power center can be calculated based on the photovoltaic output data and computing load data. This surplus / shortage status is then integrated into an alliance-level panoramic view of green electricity supply and demand. Based on this panoramic view, the green electricity supply and demand matching relationship for the current scheduling cycle can be identified. A mutual assistance scheduling scheme is then generated based on this matching relationship, and corresponding mutual assistance schemes for computing power task migration or green electricity rights settlement are executed to obtain the results for each scheduling cycle. The mutual assistance execution results over a period of time are used to calculate the overall green energy utilization rate of the alliance and the mutual assistance contribution of each alliance computing center. This generates feedback information for continuous optimization of scheduling strategies and serves as the basis for internal rights allocation and settlement reconciliation within the alliance. The mutual assistance scheduling scheme can include a computing power task migration mutual assistance scheme and a green energy rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme instructs the transfer of computing power load from green energy shortage parties to green energy surplus parties. The green energy rights settlement mutual assistance scheme instructs the surplus green energy and its corresponding carbon emission reduction rights of the green energy surplus parties to be converted into green energy shortage parties according to the settlement rules agreed upon within the alliance. By constructing a green energy sharing alliance composed of multiple computing centers, a centralized scheduling platform is set up to collect green energy surplus and shortage information of each member in real time. Cross-center collaborative green energy consumption is achieved through two mutual assistance methods: computing power task migration and green energy rights settlement. This expands the green energy consumption boundary from a single data center to the entire alliance, significantly improving the green energy utilization rate within the alliance while reducing the investment in self-built energy storage for individual centers.

[0099] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0100] This application also provides an electronic device, which is described in reference to... Figure 10 The provided electronic device 1000 includes a memory 1010, a processor 1020, and a computer program 1011 stored in the memory 1010 and capable of running on the processor 1020. When the computer program 1011 is executed by the processor, it implements the various processes of the above-described multi-computing center collaborative scheduling method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0101] This application also provides a computer-readable storage medium, see embodiments thereof. Figure 11 The computer-readable storage medium 1100 provided stores a computer program 1011. When the computer program 1011 is executed by the processor, it implements the various processes of the above-described multi-computing center collaborative scheduling method embodiment and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0102] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0103] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of the embodiments of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, apparatus, product, or device that includes a series of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products, or devices. The division of modules in the embodiments of this application is merely a logical division; in actual applications, there may be other division methods. For example, multiple modules may be combined or integrated into another device, or some features may be ignored or not performed. Additionally, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interface, and the indirect coupling or communication connection between modules may be electrical or other similar forms, none of which are limited in the embodiments of this application. Furthermore, the modules or sub-modules described as separate components may or may not be physically separated, may or may not be physical modules, or may be distributed among multiple circuit modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiments of this application.

[0104] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0105] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus, devices, and modules can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0106] In the several embodiments provided in this application, it should be understood that the disclosed apparatus, devices, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another device, 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, indirect coupling or communication connection between devices or modules, and may be electrical, mechanical, or other forms.

[0107] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0108] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium.

[0109] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product.

[0110] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable system. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can store or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)).

[0111] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A system that specifies functions in one or more boxes.

[0112] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including an instruction set implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The functions specified in one or more boxes; these computer program instructions may also be loaded onto a computer or other programmable data processing terminal equipment to cause a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0113] Although preferred embodiments of the present application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present application.

[0114] Finally, it should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0115] The technical solutions provided in the embodiments of this application have been described in detail above. Specific examples have been used in the embodiments of this application to illustrate the principles and implementation methods of the embodiments of this application. The description of the above embodiments is only for the purpose of helping to understand the methods and core ideas of the embodiments of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the embodiments of this application. Therefore, the content of this specification should not be construed as a limitation on the embodiments of this application.

Claims

1. A multi-computing-center collaborative scheduling method, characterized in that, The method is applied to a multi-computing center collaborative scheduling system, wherein the multi-computing center collaborative scheduling system is equipped with a centralized scheduling platform, and the method includes: The centralized scheduling platform receives photovoltaic output data and computing load data reported by each computing center in the pre-set green electricity sharing alliance; the pre-set green electricity sharing alliance is organized and generated by multiple computing centers with local renewable energy power generation capabilities; In each scheduling cycle, the green electricity surplus and shortage status of each alliance computing center is calculated based on the photovoltaic power output data and the computing load data, and the green electricity surplus and shortage status is integrated into an alliance-level green electricity supply and demand panoramic view. Based on the alliance-level green electricity supply and demand panoramic view, the green electricity supply and demand matching relationship of the current scheduling cycle is identified; A mutual assistance scheduling scheme is generated based on the green electricity supply and demand matching relationship. The mutual assistance scheduling scheme includes a computing power task migration mutual assistance scheme and a green electricity rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme is used to instruct the computing power load to be transferred from the green electricity shortage party to the green electricity surplus party. The green electricity rights settlement mutual assistance scheme is used to instruct the surplus green electricity and its corresponding carbon emission reduction rights of the green electricity surplus party to be converted into green electricity shortage party according to the settlement rules agreed in the alliance. The computing power task migration mutual assistance scheme or the green electricity rights settlement mutual assistance scheme is executed accordingly to obtain the mutual assistance execution results for each scheduling cycle; Based on the mutual assistance execution results, the overall green electricity utilization rate of the alliance and the mutual assistance contribution of each alliance computing center are calculated, and feedback information is generated. The feedback information is used to continuously optimize the scheduling strategy and serves as the basis for the allocation of rights and interests and settlement reconciliation within the alliance.

2. The method according to claim 1, characterized in that, The method further includes: The centralized scheduling platform receives task queue information reported by each computing power center within the preset green electricity sharing alliance; the task queue information includes the determination and labeling of each task to be executed as having a portability identifier.

3. The method according to claim 1, characterized in that, The green energy surplus / shortage status includes a green energy surplus status, a green energy shortage status, and a balanced status; the calculation of the green energy surplus / shortage status of each alliance computing center based on the photovoltaic output data and the computing load data, and the integration of the green energy surplus / shortage status into an alliance-level green energy supply and demand panoramic view, includes: Read the photovoltaic output data and computing load data of each alliance computing center one by one, and calculate the green electricity surplus / deficit difference of each alliance computing center. Alliance computing centers whose photovoltaic output exceeds computing power load power consumption are marked as having a green electricity surplus state and the surplus amount is recorded. Alliance computing centers whose photovoltaic output is less than computing power load power consumption are marked as having a green electricity shortage state and the shortage amount is recorded. Alliance computing centers whose photovoltaic output equals computing power load power consumption are marked as being in a balanced state. This provides a comprehensive overview of the supply and demand of green electricity at the alliance level for the current scheduling period.

4. The method according to claim 1 or 3, characterized in that, The process of identifying the green energy supply and demand matching relationship for the current scheduling cycle based on the alliance-level green energy supply and demand panoramic view includes: Extract the list of members with surplus green electricity and the list of members with shortage green electricity from the alliance-level green electricity supply and demand panoramic view respectively; The green electricity surplus members are sorted in descending order according to the surplus amount of each green electricity surplus member in the green electricity surplus member list, and the green electricity shortage members are sorted in descending order according to the shortage amount of each green electricity shortage member in the green electricity shortage member list. The surplus green electricity members are paired sequentially with the shortage green electricity members in descending order to obtain the green electricity supply and demand matching relationship for the current scheduling cycle.

5. The method according to claim 2, characterized in that, The generation of a mutual assistance dispatch scheme based on the green electricity supply and demand matching relationship includes: Check whether the total power consumption of the transferable tasks with transferability identifiers in the task queue information reported by the green electricity shortage party meets the load transfer requirements corresponding to the mutual assistance amount of this pairing. If the migrateable task meets the load transfer requirements and the network transmission conditions permit, then the computing power task migration method is selected, and the target task to be migrated is selected from the migrateable tasks according to the preset priority to generate a computing power task migration mutual assistance scheme. If the migrated task does not meet the load transfer requirements or the network transmission conditions do not allow it, then the green electricity rights settlement method is selected, and a green electricity rights settlement mutual assistance scheme is generated.

6. The method according to claim 5, characterized in that, The aforementioned computing power task migration and mutual assistance scheme is executed accordingly to obtain the mutual assistance execution results for each scheduling cycle, including: A task pause command is sent to the party with a shortage of green electricity. The input data and execution environment configuration information required for the target task to be migrated are packaged into a data packet and transmitted to the party with a surplus of green electricity via the network. The party with a surplus of green electricity allocates computing resources locally and loads the task execution environment to execute the target task to be migrated, thereby obtaining the mutual assistance execution result.

7. The method according to claim 5, characterized in that, The green electricity rights settlement and mutual assistance scheme is executed accordingly to obtain the mutual assistance execution results for each scheduling cycle, including: Read the current account information of the green electricity surplus party and the green electricity shortage party from the preset green electricity rights ledger, and transfer the agreed settlement green electricity from the account of the green electricity surplus party to the account of the green electricity shortage party; In addition, in accordance with the settlement rules agreed upon within the alliance, the amount of carbon emission reduction rights involved in this settlement of green electricity is calculated, and the corresponding carbon emission reduction rights are transferred and recorded in the preset green electricity rights ledger, thereby obtaining the mutual assistance execution result.

8. A multi-computing-center collaborative scheduling device, characterized in that, An apparatus for use in a multi-computing-power-center collaborative scheduling system, wherein the multi-computing-power-center collaborative scheduling system is equipped with a centralized scheduling platform, the apparatus comprising: The data receiving module is used to receive photovoltaic output data and computing load data reported by each computing center in the preset green electricity sharing alliance through the centralized scheduling platform; the preset green electricity sharing alliance is organized and generated by multiple computing centers with local renewable energy power generation capabilities; The supply and demand panoramic view generation module is used to calculate the green electricity surplus and shortage status of each alliance computing center based on the photovoltaic power output data and the computing load data in each scheduling cycle, and integrate the green electricity surplus and shortage status into an alliance-level green electricity supply and demand panoramic view. The matching relationship generation module is used to identify the green electricity supply and demand matching relationship for the current scheduling period based on the alliance-level green electricity supply and demand panoramic view; The mutual assistance scheduling scheme generation module is used to generate a mutual assistance scheduling scheme based on the green electricity supply and demand matching relationship. The mutual assistance scheduling scheme includes a computing power task migration mutual assistance scheme and a green electricity rights settlement mutual assistance scheme. The computing power task migration mutual assistance scheme is used to instruct the computing power load to be transferred from the green electricity shortage party to the green electricity surplus party. The green electricity rights settlement mutual assistance scheme is used to instruct the surplus green electricity and its corresponding carbon emission reduction rights of the green electricity surplus party to be converted into green electricity shortage party according to the settlement rules agreed upon within the alliance. The mutual assistance execution module is used to execute the computing power task migration mutual assistance scheme or the green electricity rights settlement mutual assistance scheme accordingly, and obtain the mutual assistance execution results for each scheduling cycle; The feedback information generation module is used to calculate the overall green electricity utilization rate of the alliance and the mutual contribution of each alliance computing center based on the mutual assistance execution results, and generate feedback information; the feedback information is used to continuously optimize the scheduling strategy and serve as the basis for the allocation of rights and interests and settlement reconciliation within the alliance.

9. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the multi-computing center collaborative scheduling method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the multi-computing center collaborative scheduling method as described in any one of claims 1 to 7.

Images