Management method, device, equipment and storage medium for multi-data center microgrid

Abstract

The present application provides a multi-data center management method, device, equipment, and storage medium, which belongs to the field of data center microgrid scheduling. The management method includes: obtaining a first predicted value of renewable energy, a second predicted value of thermal energy demand, and a third predicted value of workload of each data center microgrid in the multi-data center microgrid; determining the operating cost of the multi-data center microgrid based on the first predicted value and the second predicted value; determining the data migration constraint based on the third predicted value; optimizing the operating cost of the multi-data center microgrid based on the data migration constraint, and obtaining the data migration status between each data center microgrid and the workload response status of each data center microgrid. The embodiment of the present application can effectively reduce the operating cost of the multi-data center microgrid architecture while maintaining the stability of the multi-data center microgrid system.

Description

Technical Field

[0001] The present application relates to the field of data center microgrid scheduling, and more specifically, to a management method, apparatus, device, and storage medium for multi-data center microgrids. Background Art

[0002] A data center microgrid is a microgrid primarily focused on data centers, typically consisting of a data center and a power system. Data centers are used to transmit, accelerate, display, compute, and store data on the internet infrastructure, while the power system provides energy to the data center. This power system can primarily be supplied by local renewable resources. Data center microgrids can also include a thermal system that recycles heat generated by the data center, meeting the heating needs of nearby residents while reducing environmental pollution.

[0003] Multiple data center microgrids can be connected via a distributed network to form a multi-data center microgrid architecture. For individual data center microgrids, optimizing energy scheduling on the supply side or workload scheduling on the demand side can be used to reduce operating costs. However, there is currently no effective optimization solution to reduce the operating costs of a multi-data center microgrid architecture. Summary of the Invention

[0004] The embodiments of the present application provide a management method, apparatus, device, and storage medium for a multi-data center microgrid, which can effectively reduce the operating costs of a multi-data center microgrid architecture.

[0005] In a first aspect, an embodiment of the present application provides a method for managing a multi-data center microgrid, including:

[0006] Obtaining a first predicted value of renewable energy, a second predicted value of thermal energy demand, and a third predicted value of workload for each data center microgrid in the multi-data center microgrid, wherein the workload includes a deferrable task;

[0007] Determining an operating cost of the multi-data center microgrid based on the first predicted value and the second predicted value;

[0008] Determining a data migration constraint based on the third predicted value; wherein the data migration constraint includes that the deferrable task is completed within a specified time;

[0009] According to the data migration constraints, the operating cost of the multi-data center microgrid is optimized, and the data migration status between the data center microgrids and the workload response status of the data center microgrids are obtained.

[0010] In a second aspect, an embodiment of the present application provides a management device for a multi-data center microgrid, including:

[0011] an acquisition unit, configured to acquire a first predicted value of renewable energy, a second predicted value of thermal energy demand, and a third predicted value of workload of each data center microgrid in the multi-data center microgrid; wherein the workload includes a deferrable task;

[0012] a determining unit, configured to determine an operating cost of the multi-data center microgrid based on the first predicted value and the second predicted value;

[0013] The determining unit is further configured to determine a data migration constraint based on the third predicted value; wherein the data migration constraint includes that the deferrable task is completed within a specified time;

[0014] The optimization unit is used to optimize the operating cost of the multi-data center microgrid according to the data migration constraint, and obtain the data migration status between the data center microgrids and the workload response status of the data center microgrids.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, including:

[0016] a processor adapted to implement computer instructions; and,

[0017] The memory stores computer instructions, where the computer instructions are suitable for being loaded by the processor and executing the method of the first aspect.

[0018] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, which stores computer instructions. When the computer instructions are read and executed by a processor of a computer device, the computer device executes the method of the first aspect above.

[0019] In a fifth aspect, embodiments of the present application provide a computer program product or computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method of the first aspect described above.

[0020] Through the above technical solution, the embodiment of the present application can use the operating cost of the multi-data center microgrid as the objective function and the data migration constraint as the constraint condition to construct a stochastic optimization model for the joint scheduling of the multi-data center microgrid, and further optimize the stochastic optimization model to obtain a scheduling plan for the multi-data center microgrid with the optimal operating cost, thereby effectively reducing the operating cost of the multi-data center microgrid architecture. At the same time, the present application can also help maintain the stability of the multi-data center microgrid system. In addition, the embodiment of the present application performs data migration between each data center microgrid, so that each data center microgrid does not need to be constrained by geographical distance, and can achieve a wider range of cooperation. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is an optional schematic diagram of a system architecture to which the solution provided in the embodiments of the present application can be applied;

[0022] Figure 2 This is a schematic flow chart of a multi-data center microgrid management method provided in an embodiment of the present application;

[0023] Figure 3 This is a schematic flow chart of determining the operating cost of a multi-data center microgrid provided in an embodiment of the present application;

[0024] Figure 4 This is an example of the task arrival curve, the latest task completion curve, and the task completion curve provided in the embodiment of the present application;

[0025] Figure 5 This is a schematic flowchart of determining data migration constraints provided by an embodiment of the present application;

[0026] Figure 6A This is a schematic diagram of a nonlinear primitive model of a server provided in an embodiment of the present application;

[0027] Figure 6B is a schematic diagram of an improved model of a server provided in an embodiment of the present application;

[0028] Figure 7 is a schematic flow chart of another method for managing a multi-data center microgrid provided in an embodiment of the present application;

[0029] Figure 8 This is a schematic diagram of an example of a task arrival curve, a task latest completion curve, and a task completion curve provided in an embodiment of the present application;

[0030] Figure 9 A schematic diagram showing an example of electricity generated by renewable energy, residential heat load values, and real-time electricity prices;

[0031] Figure 10AIt is a schematic diagram of the operating status of a standalone data center microgrid;

[0032] Figure 10B yes Figure 10A A schematic diagram of the operating status of the data center microgrid after joining the multi-data center microgrid alliance;

[0033] Figure 11A This is a schematic diagram of the operating status of another independently operated data center microgrid;

[0034] Figure 11B yes Figure 11A A schematic diagram of the operating status of the data center microgrid after joining the multi-data center microgrid alliance;

[0035] Figure 12A This is a schematic diagram of the operating status of another independently operated data center microgrid;

[0036] Figure 12B yes Figure 12A A schematic diagram of the operating status of the data center microgrid after joining the multi-data center microgrid alliance;

[0037] Figure 13 A schematic block diagram of a management device for a multi-data center microgrid provided in an embodiment of the present application;

[0038] Figure 14 It is a schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0039] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments of this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0040] It should be understood that in the embodiments of the present application, "B corresponding to A" means that B is associated with A. In one implementation, B can be determined based on A. However, it should also be understood that determining B based on A does not mean determining B based solely on A; B can also be determined based on A and / or other information.

[0041] In the description of this application, unless otherwise specified, "at least one" means one or more, and "plurality" means two or more than two. In addition, "and / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following items" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0042] It should also be understood that the first, second, etc. descriptions appearing in the embodiments of the present application are only for illustration and distinction of the description objects, and there is no order. They do not represent any special limitation on the number of devices in the embodiments of the present application, and cannot constitute any limitation on the embodiments of the present application.

[0043] It should also be understood that the specific features, structures, or characteristics associated with the embodiments in the specification are included in at least one embodiment of the present application. In addition, these specific features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.

[0044] In addition, the terms "include" and "have" and any variations thereof are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, product, or device.

[0045] The solution provided in this application may involve artificial intelligence technology.

[0046] Artificial Intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that seeks to understand the essence of intelligence and produce new intelligent machines that can respond in a manner similar to human intelligence. AI also involves studying the design principles and implementation methods of various intelligent machines, enabling them to possess the capabilities of perception, reasoning, and decision-making.

[0047] It should be understood that artificial intelligence (AI) technology is a comprehensive discipline encompassing a wide range of fields, encompassing both hardware and software technologies. Foundational AI technologies generally include sensors, specialized AI chips, cloud computing, distributed storage, big data processing, operating / interaction systems, and mechatronics. AI software technologies primarily encompass computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0048] With the research and advancement of artificial intelligence technology, artificial intelligence technology has been studied and applied in many fields, such as common smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, unmanned driving, autonomous driving, drones, robots, smart medical care, smart customer service, etc. It is believed that with the development of technology, artificial intelligence technology will be applied in more fields and play an increasingly important role.

[0049] The solution provided in this application may also involve cloud technology.

[0050] Cloud technology refers to a hosting technology that unifies hardware, software, network and other resources within a wide area network or local area network to achieve data computing, storage, processing and sharing.

[0051] Cloud technology is a general term for network technology, information technology, integration technology, management platform technology, and application technology based on the cloud computing business model. It can form a resource pool, providing flexible and convenient on-demand use. Cloud computing technology will become a key support. Backend services of technical network systems, such as video websites, image websites, and more portals, require extensive computing and storage resources. With the rapid development and application of the internet industry, every item will likely have its own unique identification, requiring transmission to backend systems for logical processing. Different levels of data will be processed separately. Data from various industries requires strong system support, which can be achieved through cloud computing.

[0052] Cloud computing refers to the delivery and usage model of IT infrastructure, enabling on-demand, scalable access to resources over the internet. In a broader sense, cloud computing refers to the delivery and usage model of services, enabling on-demand, scalable access to services over the internet. These services can be IT-related, software-related, internet-related, or other services. Driven by the growth of the internet, real-time data streams, and the diversity of connected devices, as well as the demand for search services, social networks, mobile commerce, and open collaboration, cloud computing has rapidly developed. Unlike previous parallel and distributed computing approaches, the emergence of cloud computing will fundamentally revolutionize the entire internet model and enterprise management models.

[0053] As a provider of cloud computing infrastructure, a cloud computing resource pool (referred to as a cloud platform, commonly referred to as an IaaS (Infrastructure as a Service) platform) is established. Various types of virtual resources are deployed in the resource pool for external customers to choose from. The cloud computing resource pool mainly includes computing devices (virtualized machines, including operating systems), storage devices, and network devices.

[0054] Based on logical functional divisions, the Platform as a Service (PaaS) layer can be deployed on top of the Infrastructure as a Service (IaaS) layer, and the Software as a Service (SaaS) layer can be deployed on top of the PaaS layer. SaaS can also be deployed directly on IaaS. PaaS is a platform for software execution, such as databases and web containers. SaaS is a variety of business software, such as web portals and text messaging apps. Generally speaking, SaaS and PaaS are layers above IaaS.

[0055] The embodiments of the present application can use cloud computing to manage the energy of multi-data center microgrids. The solutions provided by the embodiments of the present application can be implemented or deployed on network elements or functions of a cloud platform.

[0056] Currently, multiple data center microgrids can be connected via a distributed network, forming a multi-data center microgrid architecture. These microgrids can share power across this distributed network. However, power sharing requires each microgrid to be connected to the distribution network. This process increases the operating costs of the multi-data center microgrid architecture. Furthermore, power sharing requires transmission lines, which necessitates close proximity between the microgrids.

[0057] In light of this, embodiments of the present application provide a multi-data center microgrid architecture in which individual data center microgrids can be coupled (connected) through data sharing, eliminating the need for connection via the distribution network. Specifically, data sharing among the data center microgrids can be achieved through the transfer of work tasks between data centers, i.e., data migration.

[0058] On the one hand, the multi-data center microgrid architecture utilizes the characteristics of coupling of each data center microgrid based on data sharing, so that each data center microgrid can cooperate with each other.

[0059] For example, when a data center microgrid needs to spend a lot of money to process all the work tasks it receives, it can consider transferring part of the work tasks to other data center microgrids in the multi-data center microgrid architecture for processing, which helps reduce operating costs.

[0060] For example, when a data center microgrid is unable to process all the work tasks it receives, it can consider transferring some of the work tasks to other data center microgrids in the multi-data center microgrid architecture for processing, which helps maintain the stability of the system.

[0061] As an example, the multi-data center microgrid architecture may include a dispatching center that can schedule the number and time period of work tasks processed by the data centers in each data center microgrid, as well as the migration status of work task data in each data center microgrid.

[0062] On the other hand, under this multi-data center microgrid architecture, each data center microgrid does not need to be constrained by geographical distance. Even data center microgrids in different provinces can be connected based on data sharing, thereby achieving cooperation in a larger geographical range.

[0063] Figure 1 It is an optional schematic diagram of a system architecture 10 to which the solution provided in the embodiment of the present application can be applied.

[0064] like Figure 1 As shown, the system architecture 10 in the left block diagram includes K data center microgrids (such as data center microgrids 1 to K) and a dispatch center 12. Wherein, K is a positive integer greater than 1.

[0065] Based on the data sharing characteristics of the K data center microgrids, the dispatch center 12 can schedule the number and time periods for each data center to process tasks (such as IT tasks), as well as the migration status of task data within each data center microgrid. For example, if the cost of processing all tasks within the specified timeframe for a data center in data center microgrid 2 is too high, the dispatch center 12 can choose to migrate some of the tasks in that data center to data centers in other data center microgrids for processing, thereby reducing costs.

[0066] Figure 1 The left block diagram shows a schematic diagram of a single data center microgrid 10 , which may be, for example, the microgrid 2 .

[0067] For the operation and scheduling of a single data center microgrid, please refer to Figure 1 The data center microgrid 10 in the following is divided into three parts:

[0068] (1) Power section 110

[0069] In the power system 110, power demand is provided by data centers 130 and electric boilers 201, while power is primarily supplied by local renewable energy sources, such as wind power plants 101 and photovoltaic power plants 102. When renewable energy sources are insufficient to meet power demand, a certain amount of power can be purchased from the main power grid 140. The power supply and demand of the power system 110 are managed by the power dispatch module 103.

[0070] In some examples, power section 110 integrates a power storage system 104. On the one hand, power storage system 104 can store excess power during certain periods to meet demand during other periods. On the other hand, power storage system 104 can also purchase and store power from main grid 140 during periods of low electricity prices and use this power during periods of high electricity prices, thereby avoiding high electricity purchase prices.

[0071] (2) Thermal section 120

[0072] The thermal system 120 primarily focuses on reusing the heat generated by the data center 130, meeting the heating needs of nearby residents while also reducing environmental pollution. The demand side of the thermal system is the residential heat load 204, while the supply of thermal energy primarily comes from the heat generated by the data center 130 and the electric furnace 201. The thermal system 120's supply and demand are managed by the thermal scheduling module 203.

[0073] Since the heat source of the thermal power unit 120 and the demand of residents are uncertain, the thermal power unit 120 needs to include an electric boiler device 201. When the heat generated by the data center 130 is insufficient to meet the residents' needs, the electric boiler 201 can be used to generate heat to maintain stable operation of the system.

[0074] Similar to the power component 110, the thermal component 120 also integrates a thermal storage system 202. This system can store excess heat generated by the data center 130 during certain periods for use in other periods. Furthermore, it can store heat generated by the electric boiler 201 during periods of low electricity prices.

[0075] (3) Data Center 130

[0076] The data center connects the power supply 110 and the thermal supply 120. Generally speaking, a data center 130 typically includes three components: servers for processing loads, air conditioning equipment for cooling, and other electronic devices such as storage and disks. The ratio of the sum of the power consumption of all components in data center 130 to the power consumption of the servers, or the PUE value, is typically used to measure the energy efficiency of data center 130. In the embodiments of this application, this PUE value is used to measure data center power consumption.

[0077] Since the energy consumption of servers in the data center accounts for the largest proportion and has the largest range of variation, dynamic voltage and frequency adjustment technology can be used to model and calculate the overall power consumption of the data center 130. This technology assumes that the power consumption of the core circuit of the microprocessor is proportional to its operating frequency and is also proportional to the square of its operating voltage, and the changes in operating frequency and operating voltage have the same direction and rhythm. Based on this, the operating frequency of the server can be used to calculate the energy consumption of the CPU, thereby calculating the total energy consumption of the data center. In the embodiment of this application, the model modeled by this dynamic voltage and frequency adjustment technology will be used to calculate the power of the server.

[0078] As a possible implementation method, the dispatch center 12 can be deployed separately on equipment outside the microgrid of each data center.

[0079] As another possible implementation, the dispatch center 12 may be deployed in a data center microgrid, for example, deployed on the same device as the data center in the data center microgrid.

[0080] The data center 130 or the dispatch center 12 can be deployed on a server. The server can be one or more servers. When there are multiple servers, at least two servers are used to provide different services, and / or at least two servers are used to provide the same service, such as providing the same service in a load balancing manner. This embodiment of the present application is not limited to this.

[0081] The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. Servers can also become nodes in the blockchain.

[0082] It should be noted that some constraints need to be considered when migrating work task data between microgrids in various data centers.

[0083] For example, data migration incurs certain economic costs, primarily consisting of hardware installation and maintenance costs, as well as the energy consumption of receiving and sending tasks. However, compared to the economic costs of data center 130 itself processing the workload, the economic costs of data migration are relatively small. For example, hardware maintenance is often performed on a monthly or annual basis, while a multi-data center microgrid architecture system can be scheduled daily, eliminating the need to consider the economic costs of data migration.

[0084] For another example, data migration is also subject to response time constraints. For real-time tasks, data center 130 should process them immediately upon arrival. However, if such tasks are migrated to other data center microgrids, the response time during data migration may prevent them from being processed in a timely manner. Therefore, this embodiment of the application only considers the migration of delay-capable tasks, which have more ample processing time, between multiple data center microgrids.

[0085] Hereinafter, the embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0086] Figure 2 The schematic flow chart of a multi-data center microgrid management method 200 provided in an embodiment of the present application. The execution subject of the method 200 can be any electronic device with data processing capability, such as a server, which includes Figure 1 The method 200 can be used to manage energy in a multi-data center microgrid. Figure 1 The scenario in is used as an example to illustrate, but the application is not limited to this.

[0087] like Figure 2 As shown, method 200 may include steps 210 to 250 .

[0088] 210 , obtaining a first predicted value of renewable energy, a second predicted value of thermal energy demand, and a third predicted value of workload of each data center microgrid in the multi-data center microgrid.

[0089] Among them, renewable energy can include wind energy, solar energy, etc., without limitation.

[0090] In some embodiments, the historical real values of wind power generation, photovoltaic power generation, and thermal energy demand of residents and apartments near the data center microgrid can be obtained, and the predicted values of wind power generation, photovoltaic power generation, and thermal energy demand at a certain time in the future can be estimated based on the historical real values, that is, the above-mentioned first predicted value and second predicted value.

[0091] For example, the first predicted value may be the amount of wind power generation and / or photovoltaic power generation per unit time (eg, a certain day), and the second predicted value may be the amount of heat energy demand per unit time.

[0092] In some embodiments, a historical real value of the workload of the data center microgrid can be obtained, and a predicted value of the workload of the data center microgrid at a certain time in the future, that is, the third predicted value, can be estimated based on the historical real value.

[0093] The workload of each data center microgrid includes delayable tasks. In the embodiment of the present application, since the processing time of delayable tasks is more sufficient, data migration of delayable tasks can be performed between multiple data center microgrids.

[0094] The workload of each data center microgrid may also include real-time tasks. In the embodiment of the present application, real-time tasks should be processed immediately upon arrival, and data migration of real-time tasks may not be performed between multiple data center microgrids.

[0095] Exemplarily, the third prediction value may be the workload size and time distribution per unit time (eg, a certain day), such as the amount of real-time tasks, the amount of deferrable tasks, and the delay time of deferrable tasks per hour in a certain day.

[0096] As an example, after obtaining the above-mentioned first prediction value and second prediction value, a large number of possible real values can be generated for the first prediction value and the second prediction value using Gaussian distribution, and a random scenario set can be constructed by combining the possible real values and the real-time electricity price.

[0097] 220. Determine an operating cost of the multi-data center microgrid based on the first predicted value and the second predicted value.

[0098] As an example, the set of random scenarios in 210 may be used as input to determine the operating cost of a multi-data center microgrid.

[0099] In this embodiment of the present application, the operating cost of a multi-data center microgrid can be used as the objective function of a stochastic optimization model for the joint scheduling of the multi-data center microgrid. The goal of the joint scheduling in this embodiment of the present application is to optimize (i.e., minimize) the operating cost of the multi-data center microgrid.

[0100] In some embodiments, the operating cost of the multi-data center microgrid may include an economic cost, a first penalty cost for not utilizing thermal energy, and a second penalty cost for not utilizing renewable energy.

[0101] As a possible implementation, see Figure 3 , the operating cost of the multi-data center microgrid can be determined through the process shown in 221 to 223.

[0102] 221. Determine a first penalty cost for non-utilization of renewable energy in each data center microgrid based on the first predicted value, the second predicted value, the power consumption of each data center microgrid, the charge and discharge amount of the power storage system of each data center microgrid, and at least one of the power consumption of the electric boiler, and utilizing the power system supply and demand balance condition.

[0103] Specifically, for power dispatch, the power on the demand side should be less than the power on the supply side. Because the power dispatch module prioritizes renewable energy for system power and purchases power from the main grid when renewable energy is depleted, any excess power on the supply side at a given moment is unused renewable energy.

[0104] For example, the power supply side includes not only wind power generation, photovoltaic power generation, and power purchased from the main grid, but also some power released by the power storage system. The power demand side primarily includes data centers, electric boilers, and power stored in the power storage system. Therefore, the power supply and demand balance constraint is that the sum of the renewable energy, power purchases, and power storage system discharge of each data center microgrid is greater than or equal to the sum of the power consumption of each data center microgrid, the charge of the power storage system, and the power consumption of the electric boiler.

[0105] For example, the power supply and demand balance constraint can be expressed as follows:

[0106] + (1)

[0107] in, and Represents data center microgrid k At the moment t The amount of wind power and photovoltaic power generated, Represents a data center microgrid k At the moment t Electricity purchased from the main grid, Represents a data center microgrid k At the moment t Power consumption, Indicates that the electric boiler is at time t Power consumption, and .

[0108] At the same time, data center microgrid can be obtained k At the moment t Unused renewable energy The expression is as follows:

[0109] + (2)

[0110] Data Center Microgrid k At the moment t The first penalty cost of not using renewable energy can be calculated based on the data center microgrid k At the moment t Unused renewable energy Penalty cost coefficient for non-use of renewable energy The result can be, for example, the product of the two.

[0111] 222. Determine a second penalty cost for unused thermal energy of each data center microgrid based on the second predicted value, the heat generated by each data center microgrid, the heat generated by the electric boiler, and at least one of the thermal energy released and absorbed by the thermal storage system of each data center microgrid, and utilizing the supply and demand balance condition of the thermal system.

[0112] Specifically, thermal dispatch also needs to meet the constraint that heat supply exceeds demand. During the dispatch process, the dispatch center can prioritize the use of heat generated by the data center, and only activate the electric boiler equipment when heat production is insufficient.

[0113] For example, the supply of thermal energy mainly comes from the heat production of the data center, the heat production of the electric boiler and the heat energy released by the thermal storage system, while the demand part is mainly the heat load and the heat energy released by the thermal storage system. Therefore, it can be obtained that the heat supply and demand balance constraint is that the sum of the heat production of each data center microgrid, the heat production of the electric boiler and the heat energy released by the thermal storage system is greater than or equal to the sum of the heat load of each data center microgrid and the heat energy absorbed by the thermal storage system.

[0114] For example, the heat supply and demand balance constraint can be expressed as follows:

[0115] + 3)

[0116] in, and Represent the efficiency of heat generation of data center and electric boiler respectively, For the moment t The heat load, and They represent the thermal energy absorbed and released by the thermal storage system at time t.

[0117] At the same time, data center microgrid can be obtained k At the moment t Unused heat energy The expression is as follows:

[0118] + (4)

[0119] Data Center Microgrid k At the moment t The second penalty cost of unused thermal energy can be calculated based on the data center microgrid k At the moment t Unused heat energy Unused penalty cost coefficient The result can be, for example, the product of the two.

[0120] 223. Determine the operating cost of the multi-data center microgrid based on the power purchase amount of each data center microgrid, the first penalty cost, and the second penalty cost.

[0121] For example, Represents a data center microgrid k The operating costs are:

[0122] (5)

[0123] Where T is the set of time nodes.

[0124] Furthermore, the total operating cost of the multi-data center microgrid system architecture is the sum of the operating costs of all data center microgrids, which can be expressed as F Indicates as follows:

[0125] (6)

[0126] Among them, K is the number of data center microgrids in the multi-data center microgrid system architecture, which can also be called the set of data center microgrids.

[0127] 230. Determine data migration constraints based on the third prediction value.

[0128] The data migration constraint includes that the delayed task must be completed within a specified time.

[0129] It can be understood that the data migration constraint, as a constraint condition of the stochastic optimization model of multi-data center microgrid joint scheduling, can ensure that the deferrable tasks are completed within the specified time during the optimization process of the objective function.

[0130] In some embodiments, data migration constraints may be determined based on a task arrival curve, a task latest completion curve, and a task completion curve of each data center microgrid.

[0131] First, we will introduce the task arrival curve, the task latest completion curve, and the task completion curve for a single data center. The following description uses IT tasks as an example.

[0132] See also Figure 4, shows an example of a task arrival curve, a task latest completion curve, and a task completion curve.

[0133] Among them, the broken line Arr t It represents the task arrival curve, which means that the data center processes the tasks as they arrive, regardless of whether they are real-time tasks or delayed tasks. The formula can be as follows:

[0134] (7)

[0135] in, For the moment Arrived deferrable tasks, For the moment Arrival of real-time tasks.

[0136] Polyline Com t It represents the latest completion curve of the task, which means that for all delayable tasks, the data center chooses to complete the task at the deadline. Its formula can be as follows:

[0137] (8)

[0138] in, Indicates time t The deadline of the deferrable task that was reached.

[0139] Polyline is a possible task completion curve, indicating the accumulation to the moment t The number of IT tasks that the data center has processed so far. Figure 4 The shaded area in the graph is consistent with the task arrival curve Arr t And the task latest completion curve Com t The following relationship is satisfied:

[0140] (9)

[0141] In addition, the number of IT tasks being processed by the data center at time t and Has the following relationship:

[0142] (10)

[0143] In addition, the amount The CPU processing capacity of the data center server should not be exceeded, so:

[0144] (11)

[0145] in, represents the number of servers running at time t, R is a collection of server types, Indicates the frequency The server at time t work efficiency, Indicates the upper limit of the number of tasks that the server can process. For example, It can be 0.9.

[0146] Then, aiming at the scheduling and operation problem of multi-data center microgrid, the relevant task arrival curve, task latest completion curve and task completion curve are introduced.

[0147] See also Figure 5 , the data migration constraints can be determined through the process shown in 231 to 233.

[0148] 231 , determining a task arrival curve of the delayable tasks of each data center microgrid based on the number of delayable tasks arriving at the first time, the number of delayable tasks received, and the number of delayable tasks transferred out of each data center microgrid.

[0149] Specifically, considering that only delayable tasks can be migrated between data center microgrids during the joint scheduling of multiple data center microgrids, the embodiment of the present application can design a task arrival curve for delayable tasks. For example, the task arrival curve is as follows:

[0150] (12)

[0151] in, Represents a data center microgrid k At the moment The number of IT work tasks received, Indicates the data center microgrid k At the moment The number of IT workloads transferred to other data center microgrids.

[0152] 232. Determine the latest completion curve of the delayable task of each data center microgrid based on the deadline of the delayable task and the task arrival curve.

[0153] For example, the latest completion curve of a deferrable task is as follows:

[0154] (13)

[0155] 233. Determine the data migration constraint based on the task completion curve, the task arrival curve, and the latest completion curve of the processed deferrable tasks of the multi-data center microgrid, wherein the data migration constraint includes that the task completion curve is between the task arrival curve and the latest completion curve.

[0156] Exemplary, moment t The number of deferrable tasks processed by all data centers Should meet the following requirements:

[0157] (14)

[0158] in, Represents a data center microgrid k At the moment t The number of deferrable tasks processed.

[0159] Formula (14) shows that as long as the delayable task is completed within the specified time, any data center microgrid in the multi-data center microgrid system architecture can process it.

[0160] Therefore, the embodiment of the present application can ensure that the deferred tasks are completed within the specified time during the optimization process of the objective function through the data migration constraint.

[0161] In some embodiments, data migration constraints may further include:

[0162] At the moment t When is the starting time, the number of deferrable tasks being processed by each data center microgrid at time t is the same as the number of deferrable tasks it has processed at time t;

[0163] At the moment t When the time is a moment after the start time, each data center microgrid is at time t The number of deferrable tasks being processed, and their t The number of deferrable tasks processed, at time ( t- 1) Related to the number of deferrable tasks processed.

[0164] For example, a data center microgrid k The number of deferrable tasks being processed at time t Satisfies the following relationship:

[0165] (15)

[0166] in, Represents a data center microgrid k At the moment ( t -1) The number of deferrable tasks processed.

[0167] Therefore, the embodiment of the present application can ensure that the number of delayable tasks being processed by the data center microgrid meets the objective operating laws of the system through the data migration constraint during the optimization process of the objective function, thereby expanding the flexible space required for system scheduling.

[0168] In some embodiments, data migration constraints may further include:

[0169] The sum of the number of deferrable tasks and the number of real-time tasks being processed by each data center microgrid exceeds the working capacity of the CPU of each data center microgrid.

[0170] For example, the number of deferrable tasks being processed by data center microgrid k at time t is and the number of real-time tasks Satisfies the following formula:

[0171] (16)

[0172] Therefore, the embodiment of the present application can ensure that the total amount of tasks processed by the data center microgrid does not exceed the working capacity of the CPU of the data center microgrid during the optimization process of the objective function through the data migration constraint.

[0173] 240 , optimizing the operating cost of the multi-data center microgrid according to the data migration constraint, and obtaining the data migration status between each data center microgrid and the workload response status of each data center microgrid.

[0174] Specifically, a stochastic optimization model for the joint scheduling of multi-DC microgrids can be constructed based on the data migration constraints and the operating costs of the multi-DC microgrids. The operating costs of the multi-DC microgrids (e.g., Formula (6)) can be used as the objective function of the stochastic optimization model for the joint scheduling of multi-DC microgrids, and the data migration constraints (e.g., Formulas (12)-(16)) can be used as the constraints of the stochastic optimization model.

[0175] Furthermore, by optimizing the stochastic optimization model, a scheduling scheme for a multi-data center microgrid with optimal operating costs can be obtained. This scheduling scheme can include data migration between each data center microgrid and the workload response status of each data center microgrid.

[0176] Therefore, the embodiment of the present application takes the operating cost of the multi-data center microgrid as the objective function and the data migration constraint as the constraint condition, constructs a random optimization model for the joint scheduling of the multi-data center microgrid, and further optimizes the random optimization model. It can obtain a scheduling plan for the multi-data center microgrid with the optimal operating cost, thereby effectively reducing the operating cost of the multi-data center microgrid architecture.

[0177] In addition, by scheduling the work tasks of each data center microgrid, when a data center microgrid is unable to handle all the work tasks it receives, part of the work tasks can be migrated to other idle data center microgrids for processing. The embodiment of this application can help maintain the stability of the multi-data center microgrid system.

[0178] In addition, the embodiment of the present application connects the data center microgrids through data migration (i.e., data sharing), which can achieve the goal that the data center microgrids do not need to be connected through the distribution network, so that the data center microgrids do not need to be constrained by geographical distance, thereby achieving cooperation on a larger scale.

[0179] In some embodiments, as a possible implementation method, the above-mentioned operating costs can be optimized based on data migration constraints, and at least one of the constraints of each data center server, power storage system constraints, thermal storage system constraints, power supply and demand balance constraints, and thermal supply and demand balance constraints.

[0180] That is to say, in the above-mentioned random optimization model of joint scheduling of multi-data center microgrids, in addition to data migration constraints, at least one of the constraints of each data center server, power storage system constraints, thermal storage system constraints, power supply and demand balance constraints, and thermal supply and demand balance constraints can be used as constraints of the random optimization model.

[0181] Because each constraint effectively serves at least one component of the multi-DC microgrid, multiple constraints enable decentralized management of each component, preventing interference between components and improving overall system scheduling. This approach offers the advantage that, when a module, such as the power storage system, experiences a problem, the corresponding constraints can be directly corrected, significantly facilitating the stable operation of the multi-DC microgrid. More importantly, when modules need to be added or removed, the corresponding constraints can be added or removed directly, facilitating system management and adjustments.

[0182] The power supply and demand balance constraint can be described in Formula (1) above, and the heat supply and demand balance constraint can be described in Formula (3) above, which will not be repeated here.

[0183] In some embodiments, each data center server constraint may include:

[0184] The working efficiency of each data center server does not exceed a first preset value, wherein the working efficiency of each data center server is related to the number of work tasks currently processed by each data center server and the upper limit of the number of processed tasks.

[0185] Specifically, the power consumption of a server CPU is approximately proportional to its work efficiency. The work efficiency of a server can be defined as the ratio of the number of tasks the CPU is currently processing to the upper limit of the number of tasks the CPU can process, which can be expressed as:

[0186] (17)

[0187] in, Indicates the frequency The server at time t work efficiency, Indicates the number of tasks currently being processed by the server. Indicates the upper limit of the number of tasks that the server can handle. Generally speaking, the server's work efficiency does not exceed 0.9.

[0188] In some optional embodiments, the power consumption of the multi-data center microgrid is related to the power consumption and energy utilization efficiency of the servers in each data center microgrid. Here, the power consumption of the multi-data center microgrid can also be used as part of the data center server constraints.

[0189] Among them, the server power consumption of each data center microgrid is related to the CPU power consumption and static power consumption of each data center microgrid, and the CPU power consumption of each data center microgrid is related to the operating frequency, chip model and working efficiency of the CPU of each data center microgrid.

[0190] For example, based on the working efficiency of the server's CPU, the dynamic voltage and frequency adjustment technology can be used to model the CPU's power consumption. For example, the power consumption of the CPU at time t is It can be expressed as:

[0191] (18)

[0192] in, It is a constant coefficient and is related to the CPU chip model.

[0193] The power consumption of CPU accounts for the largest part of the server power consumption. The rest of the power consumption mainly comes from other devices such as memory. This power consumption can be called static power consumption and can be regarded as a constant. Therefore, the power consumption of the server can be determined as follows:

[0194] (19)

[0195] Furthermore, you can use represents the number of servers running at time t, then the power consumption of the multi-data center microgrid can be obtained by summing the power consumption of all servers:

[0196] (20)

[0197] in, is the total power consumption of data center microgrid k at time t, R is the set of server types, and PUE represents energy utilization efficiency.

[0198] In some embodiments, formula (20) needs to calculate the product of the number of each type of server and the power of such server to obtain the power consumption of the entire multi-data center microgrid system. Figure 6A A schematic diagram of the original nonlinear model of the server is shown. Since the number and power of the servers are decision variables, the model is nonlinear and its solution is relatively complex.

[0199] In order to solve the above problems, we can make the following assumptions:

[0200] (twenty one)

[0201] Figure 6B A schematic diagram of the improved linear model is shown. Figure 6B As shown in Equation (22), assuming that servers are fully loaded once powered on, the system only needs to count the number of such servers powered on, thus linearizing the model. In this model, the power consumption of the data center is greater than the power consumption of actually processing all workloads, as shown in Equation (22). Here, this error does not exceed the power consumption of the server with the lowest power consumption when running at full load.

[0202] (twenty two)

[0203] In some embodiments, the power storage system and the thermal storage system can be widely used in the electric heat scheduling model, and both can provide greater flexibility for the electric heat scheduling model. The constraints of the power storage system and the thermal storage system are described below.

[0204] Exemplarily, the power storage system constraints may include at least one of the following:

[0205] The initial storage capacity of the power storage system per unit time remains constant;

[0206] The amount of electricity stored in the power storage system is between an upper limit and a lower limit of the amount of electricity stored in the power storage system;

[0207] The charge amount of the power storage system at time t does not exceed the upper limit of the charge amount of the power storage system at time t;

[0208] The discharge amount of the power storage system at time t does not exceed the upper limit of the discharge amount of the power storage system at time t;

[0209] The amount of power stored in the power storage system at a time point next to time t is related to the amount of power stored, the charge amount, the charge efficiency, the discharge amount, and the discharge efficiency of the power storage system at time t.

[0210] As an example, taking the daily dispatch model as an example, in order to ensure that the initial operating state of the power storage system is the same every day, the initial storage capacity of the power storage system should remain constant, that is:

[0211] (twenty three)

[0212] in, Represents a data center microgrid k The initial storage capacity of the corresponding power storage system on day T is, Represents a data center microgrid k The corresponding initial storage capacity of the power storage system on the (T+1) day.

[0213] In addition, the power stored in the power storage system should have upper and lower boundary constraints, and its charging and discharging amounts at each moment should also have ramping constraints, as follows:

[0214] (twenty four)

[0215] (25)

[0216] (26)

[0217] in, Represents a data center microgrid k The corresponding power storage system is at time t The amount of stored electricity, and They represent the lower and upper limits of the amount of electricity that can be stored in the power storage system, and denote the charge and discharge amounts of the power storage system at time t, and Indicates the upper limit of the charge and discharge amount at each moment.

[0218] It is worth noting that the power storage system has a certain amount of energy loss during the storage and release process, so the charging efficiency and discharge efficiency The entry constraints should be considered. k The relationship between the amount of electricity stored in the power storage system at the next moment and the amount of electricity stored in the power storage system at time t is as follows:

[0219] (27)

[0220] Exemplarily, the thermal storage system constraints may include at least one of the following:

[0221] The initial stored thermal energy per unit time of the thermal storage system remains constant;

[0222] The stored thermal energy of the thermal storage system is between an upper limit and a lower limit of the stored thermal energy of the thermal storage system;

[0223] The thermal energy absorbed by the thermal storage system at time t does not exceed the upper limit of the thermal energy absorbed by the thermal storage system at time t;

[0224] The thermal energy released by the thermal storage system at time t does not exceed the upper limit of the thermal energy released by the thermal storage system at time t;

[0225] The stored thermal energy of the thermal storage system at the next moment after time t is related to the stored thermal energy, absorbed thermal energy, thermal energy absorption efficiency, released thermal energy, and thermal energy release efficiency of the thermal storage system at time t.

[0226] Similar to the electricity storage system, the thermal storage system should also ensure that the initial state is the same every day:

[0227] (28)

[0228] in, Data center microgrid k Thermal storage system at the moment t thermal energy storage capacity.

[0229] For the boundary conditions and ramp constraints of the thermal storage system, there is the following relationship:

[0230] (29)

[0231] (30)

[0232] (31)

[0233] in, and Represents data center microgrid k The lower and upper limits of thermal storage in thermal storage systems, and They represent the thermal energy absorbed and released by the thermal storage system at time t, T and T Indicates the upper limit of heat energy absorbed and released at each moment.

[0234] The relationship between the thermal energy of the thermal storage system at the next moment and the thermal energy stored at time t can be expressed as follows:

[0235] (32)

[0236] in, and They represent the efficiency of the thermal storage system in absorbing and releasing thermal energy, respectively.

[0237] Therefore, the embodiments of the present application can ensure that data center servers, power storage systems, thermal storage systems, etc. meet objective laws during the optimization process of the objective function through data center server constraints, power storage system constraints, thermal storage system constraints, power supply and demand balance constraints, and thermal supply and demand balance constraints.

[0238] Therefore, the embodiment of the present application takes the operating cost of the multi-data center microgrid as the objective function, and takes the data migration constraints, as well as at least one of the constraints of each data center server, the power storage system constraints, the thermal storage system constraints, the power supply and demand balance constraints, and the thermal supply and demand balance constraints as constraints, to construct a random optimization model for the joint scheduling of multi-data center microgrids, and further optimizes the random optimization model to obtain a scheduling plan for the multi-data center microgrid with the optimal operating cost, thereby effectively reducing the operating cost of the multi-data center microgrid architecture.

[0239] While effectively reducing the operating costs of a multi-data center microgrid architecture, the present embodiment also helps maintain the stability of the multi-data center microgrid system. Furthermore, by connecting data center microgrids through data migration (i.e., data sharing), the present embodiment enables collaboration across a wide range of data center microgrids, eliminating the need for geographical distance constraints.

[0240] Figure 7 The schematic flow chart of another method 300 for managing a multi-data center microgrid provided in an embodiment of the present application is shown. The method 300 can be executed by any electronic device with data processing capabilities, for example, the electronic device can be implemented as a server. For another example, the electronic device can be implemented as Figure 1 The dispatch center 12 is not limited in this application.

[0241] The method 300 may be used to perform energy management on a multi-data center microgrid. As an example, a dispatch center may perform energy management on a multi-data center microgrid.

[0242] For the convenience of description, the following embodiments are Figure 1 The scenario in is used as an example to illustrate, but the application is not limited to this.

[0243] It should be understood that Figure 7 The steps or operations of the multi-data center microgrid management method are shown, but these steps or operations are only examples. The embodiment of the present application may also perform other operations or Figure 7 In addition, Figure 7 The steps in Figure 7 are executed in a different order than the ones presented, and may not be executed Figure 7 All operations in .

[0244] like Figure 7 As shown, the management method 300 may include steps 301 to 320 .

[0245] 301, obtaining the actual value of wind power generation.

[0246] 302, obtaining the actual value of photovoltaic power generation.

[0247] 303, obtain the actual value of residents' heat demand.

[0248] After obtaining the actual values of wind power generation, photovoltaic power generation, and residential heat demand, process 31 may be performed to construct a random scenario set. Process 31 may specifically include the following steps 304 to 308 .

[0249] 304, obtaining a wind power generation prediction value.

[0250] For example, a large number of possible wind power generation real values may be generated as the wind power generation prediction value using Gaussian distribution based on the wind power generation real value.

[0251] 305, obtaining a photovoltaic power generation prediction value.

[0252] For example, based on the actual value of photovoltaic power generation, a large number of actual values of photovoltaic power generation may be generated by Gaussian distribution as the wind power generation prediction value.

[0253] 306, obtain the predicted value of residents' heat demand.

[0254] For example, based on the actual value of the heat demand of the residents, a large number of actual values of the heat demand of the residents that may exist can be generated using a Gaussian distribution as the predicted value of the heat demand of the residents.

[0255] 307, obtain real-time electricity prices.

[0256] 308 , constructing a random scenario set. For example, a random scenario set can be constructed based on the wind power generation forecast value, photovoltaic power generation forecast value, residential heat demand forecast value, and real-time electricity price obtained in steps 304 to 307 .

[0257] After the random scenario set is constructed, process 32 can be performed to schedule power supply and power demand for each time period. The power storage system can be used to flexibly adjust the power supply, and the main grid's power purchase mechanism can be used as a backup to obtain penalty costs for unused renewable energy.

[0258] Continue to see Figure 7 , process 32 may include steps 309 to 312.

[0259] 309, main network electricity purchase.

[0260] 310, power supply.

[0261] For example, the power supply side purchases electricity for the main grid and discharges electricity to the power storage system, and the power demand side charges the power storage system and meets power demand (such as the power consumption of the data center microgrid).

[0262] 311, charging and discharging of power storage systems.

[0263] 312, power demand.

[0264] Exemplarily, the dispatch center may obtain the power supply and power demand in steps 309 to 312 .

[0265] 313. Obtaining the penalty cost of not utilizing renewable energy. Specifically, the penalty cost of not utilizing renewable energy can be calculated based on the power supply and power demand in process 32.

[0266] After constructing the randomized scenario set, process 33 can be performed to schedule heat supply and demand for each time period. The thermal storage system can be used to flexibly adjust the supplied heat energy, while the electric boiler heat generation mechanism can be used as a backup to obtain penalty costs for unused heat energy.

[0267] Continue to see Figure 7 , process 33 may include steps 314 to 317.

[0268] 314, heat energy demand.

[0269] 315, heat supply.

[0270] For example, the heat supply side includes electric boilers supplying heat, data center microgrids generating heat, and heat storage systems releasing heat, while the heat demand side includes the heat storage system absorbing heat energy and residents' heat demand.

[0271] 316. Thermal storage system charging and discharging heat.

[0272] 317, heating provided by electric furnace.

[0273] For example, the dispatch center may obtain the heat supply and heat demand in steps 314 to 317 .

[0274] 318. Obtaining the penalty cost of unused heat energy. Specifically, the penalty cost of unused heat energy can be calculated based on the heat energy supply and heat demand in process 33.

[0275] 319,Workload Migration and Response in Multi-data Center Microgrids.

[0276] 320, obtain the economic cost (i.e., operating cost) of a multi-data center microgrid.

[0277] Exemplarily, in steps 319 to 320, according to the method in method 200, the operating cost of the multi-data center microgrid can be used as the objective function, and the data migration constraint, as well as at least one of the constraints of each data center server, the power storage system constraint, the thermal storage system constraint, the power supply and demand balance constraint, and the thermal supply and demand balance constraint can be used as constraints to construct a random optimization model for the joint scheduling of multiple data center microgrids, and further optimize the random optimization model to obtain a scheduling plan for the multi-data center microgrid with the optimal operating cost, determine the data migration status between each data center microgrid and the workload response status of each data center microgrid.

[0278] For example, the dispatch center may determine during this process whether to migrate part of the work tasks of a certain data center microgrid to other data center microgrids for processing.

[0279] Therefore, the embodiments of the present application utilize a dispatch center to manage the energy of multiple data center microgrids, enabling cooperative operation among the data center microgrids, effectively reducing the operating costs of the multi-data center microgrid architecture. Furthermore, the embodiments of the present application also help maintain the stability of the multi-data center microgrid system. Furthermore, by connecting the data center microgrids through data migration (i.e., data sharing), the embodiments of the present application eliminate the need for geographical distance constraints among the data center microgrids, enabling wide-scale collaboration.

[0280] The beneficial effects of the embodiments of the present application are described below with reference to specific simulation examples.

[0281] To investigate the effectiveness of the technical solutions provided in the embodiments of this application, simulation parameters were set. The data center parameters are shown in Table 1. The server CPU can be an Intel Pentium 950, with parameters shown in Table 2. The electric boiler's heat production efficiency was set to 80%, while the parameters of the power storage system and thermal storage system are shown in Table 3.

[0282] Table 1

[0283]

[0284] Table 2

[0285]

[0286] Table 3

[0287]

[0288] The predicted values of IT workload in the data center are shown in Table 4. The corresponding task arrival curve and task latest completion curve are shown in Table 4. Figure 8 shown.

[0289] Table 4

[0290]

[0291] The electricity generated by renewable energy, the heat load value of residents, and the real-time electricity price are Figure 9 In display.

[0292] At the same time, in order to study the cooperation between different data center microgrids, the embodiment of the present application sets the uncertainty scale of the three data center microgrids as shown in Table 5.

[0293] Table 5

[0294]

[0295] The following describes the results of the electric and heat scheduling after the cooperation of three data center microgrids with the help of the attached figures. Figure 10A 、 Figure 11A and Figure 12A Respectively represent the operating status of the three data center microgrids when running independently, 10B, Figure 11B and Figure 12B It indicates the operating status of these three data center microgrids after joining the multi-data center microgrid alliance.

[0296] 1) For a standalone data center microgrid 1

[0297] See also Figure 10A Data Center Microgrid 1 has abundant renewable energy (RE) to meet both electricity and heat needs. Furthermore, both the power and heat storage systems help shift peak demand and fill valleys during scheduling, further maximizing the utilization of renewable energy. Therefore, this data center microgrid no longer needs to purchase electricity from the main grid, resulting in zero economic cost, with operating costs consisting solely of two penalty costs.

[0298] Renewable energy is primarily unused during the early morning and midday hours. The data center microgrid receives fewer IT tasks during the early morning hours, making it unable to absorb more energy, thus underutilizing renewable energy. Meanwhile, photovoltaic (PV) power generation is extremely abundant during midday hours. Even if the dispatch center concentrates IT tasks during this time, significant amounts of renewable energy will still remain unused. Therefore, the renewable energy penalty cost for data center microgrid 1 is relatively high.

[0299] Penalty costs for unused heat arise because the data center's heat production periods don't coincide with residents' heat demand periods. The thermal storage system stores and releases the data center's heat during this process, ensuring a stable heat supply while maximizing heat utilization efficiency. However, since both heat production and demand are uncertain, some penalty costs for unused heat are inevitable.

[0300] 2) For a standalone data center microgrid 2

[0301] See also Figure 11A , Data Center Microgrid 2 needs to handle more IT workloads, and the local renewable energy is not as abundant as that of Data Center Microgrid 1. Therefore, Data Center Microgrid 2 needs to purchase a large amount of electricity from the main grid, which results in higher economic costs.

[0302] In this environment, the dispatch center fully utilizes renewable energy. During midday, when renewable energy is most abundant and grid electricity prices are highest, the dispatch center avoids purchasing electricity from the grid. Instead, it uses all renewable energy combined with energy released by the power storage system to complete as much IT workload as possible. Ultimately, the renewable energy penalty cost for Data Center Microgrid 2 is zero. However, due to the large amount of IT workload it handles, the data center generates a significant amount of heat, far exceeding the heat needs of local residents. This results in a high penalty cost for unused heat in Data Center Microgrid 2.

[0303] 3) For independently operated data center microgrids

[0304] See also Figure 12A Data Center Microgrid 3 needs to handle a greater amount of heat load from nearby residents. Therefore, Data Center Microgrid 3 also needs to purchase a large amount of electricity from the main grid. However, this electricity not only provides the data center with IT load but also generates heat for the electric boiler system to meet the heating needs of residents. The electric boiler system can be activated when the data center's heat output is insufficient to meet the heat load, thus maintaining system security and stability. Note that during midday, when electricity prices are higher, the dispatch center will activate the electric boiler to meet the necessary heat demand during this period, delaying the processing of IT tasks and thus avoiding purchasing electricity during high-price periods.

[0305] In environments with high heat demand, the data center's heat production can be fully utilized. At the same time, the large investment in electric boilers also increases electricity demand, fully utilizing renewable energy. Therefore, both renewable energy and data center heat production are fully utilized in Data Center Microgrid 3, and its operating costs only include the economic cost of purchasing electricity from the main grid.

[0306] 4) Data center microgrid after joint scheduling

[0307] Figure 10B 、 Figure 11B and Figure 12B The following table shows the power and heat scheduling of the three data center microgrids after cooperation. Figure 10A 、 Figure 11A and Figure 12A It can be seen that the dispatch center mainly arranges data center microgrid 1 to receive and process IT work tasks, data center microgrid 2 migrates a large number of IT work tasks outward, and data center microgrid 3 both receives and migrates tasks, depending on the specific time period.

[0308] See also Figure 10B Although Data Center Microgrid 1 handles a large number of IT tasks from Data Center Microgrid 2, it still does not purchase electricity from the main grid. Instead, it handles IT tasks to the greatest extent possible while making full use of renewable energy. Figure 10A , Figure 10B The cost of renewable energy penalty has dropped significantly. On the contrary, since a lot of heat energy will be generated by processing IT workloads, Figure 10B The penalty cost for unused thermal energy has increased significantly.

[0309] See also Figure 11B Since the bulk of IT workloads have been offloaded to Data Center Microgrids 1 and 3, Data Center Microgrid 2 only needs to schedule IT workloads based on the distribution and scale of renewable energy. This eliminates the need to purchase large amounts of electricity from the main grid, significantly reducing economic costs. Furthermore, thanks to the reduced IT workload, the data center's heat output has also dropped significantly, making it more suitable for heating needs of local residents. Consequently, the cost of unused heat from Data Center Microgrid 2 has also been significantly reduced.

[0310] See also Figure 12BData Center Microgrid 3 is prone to accepting and processing more IT workloads, a characteristic of its own. Since Data Center Microgrid 3 needs to meet the significant heating needs of nearby residents, it must generate sufficient heat. When IT workloads are high, it can generate significant heat through its CPU processing. When IT workloads are low, CPU heat generation is insufficient, and the system can activate its electric boiler to generate heat. However, both processing a large number of IT workloads and activating the electric boiler to generate heat require Data Center Microgrid 3 to purchase a significant amount of electricity from the main grid. Therefore, its economic cost is insensitive to the number of IT workloads it processes. After joint scheduling, Data Center Microgrid 3 accepts a large number of IT workloads from Data Center Microgrid 2 and processes these workloads to generate heat to meet the heat load, resulting in a reduction in the use of its electric boiler. Notably, at certain times, Data Center Microgrid 3 chooses not to accept IT workloads, instead transferring them to Data Center Microgrid 1. This is because electricity prices are high during these periods, making it uneconomical for Data Center Microgrid 3 to handle the large workload.

[0311] In summary, data center microgrid 1 is suitable for accepting a large number of IT workloads to maximize the utilization rate of renewable energy; data center microgrid 2 is suitable for migrating a large number of IT workloads to avoid excessive electricity purchases and excessive heat generation; the cost of data center microgrid 3 is not sensitive to the number of IT workloads processed, and it is suitable for accepting more IT workloads when electricity prices are not high to help alleviate the task processing pressure of data center microgrids 1 and 2.

[0312] 5) Cost structure of each data center microgrid before and after cooperation

[0313] Because the model's objective function is composed of multiple costs, it is necessary to analyze the changes in various costs for different types of data center microgrids before and after joint scheduling. The cost components before and after joint scheduling of data center microgrids are calculated, and the values are shown in Table 6 below.

[0314] Table 6

[0315]

[0316] Data Center Microgrid 1 did not need to purchase electricity from the main grid before and after the partnership, so its costs only included two penalty costs. After the partnership, Data Center Microgrid 1 took on a large amount of IT workload from Data Center Microgrid 2, which improved its renewable energy utilization efficiency. However, it also generated more heat that could not be used by nearby residents. Ultimately, its renewable energy penalty costs decreased, while its heat penalty costs increased, resulting in a 3.48% increase in operating costs.

[0317] Data Center Microgrid 2 fully utilized renewable energy before and after the partnership, eliminating the penalty. However, since it migrated a significant portion of its IT workload to Data Center Microgrids 1 and 3, it no longer needed to purchase significant amounts of electricity from the grid after the partnership. The heat generated by processing IT workloads was also significantly reduced, ultimately resulting in a 97.78% reduction in operating costs.

[0318] For data center microgrid 3, its cost is mainly the electricity purchase cost and will not change much. After joint scheduling, its operating cost increases by 3.94%.

[0319] Although two of the three data center microgrids experienced increased costs after joint scheduling, the significant cost reduction for data center microgrid 2 ultimately resulted in a 52.48% reduction in the total operating costs of the three data center microgrids. This cost allocation ensures that each data center microgrid's costs are reduced. Furthermore, the electricity purchase cost, which accounts for the largest portion of the cost structure and is of primary concern to operators, decreased by 53.24%, demonstrating the superiority of the model in this application for improving economic efficiency.

[0320] The specific embodiments of the present application are described in detail above in conjunction with the accompanying drawings. However, the present application is not limited to the specific details in the above embodiments. Within the technical concept of the present application, a variety of simple modifications can be made to the technical solution of the present application, and these simple modifications all fall within the scope of protection of the present application. For example, the various specific technical features described in the above specific embodiments can be combined in any suitable manner unless there is any contradiction. In order to avoid unnecessary repetition, the present application will not further explain various possible combinations. For another example, the various different embodiments of the present application can also be arbitrarily combined, and as long as they do not violate the ideas of the present application, they should also be regarded as the contents disclosed in the present application.

[0321] It should also be understood that in the various method embodiments of the present application, the order of the sequence numbers of the above-mentioned processes does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. It should be understood that these sequence numbers can be interchanged where appropriate, so that the embodiments of the present application described can be implemented in an order other than those shown or described.

[0322] The above describes the method embodiment of the present application in detail. Figures 13 and 14 , describe in detail the device embodiments of the present application.

[0323] Figure 13 FIG is a schematic block diagram of a management device 700 for a multi-data center microgrid according to an embodiment of the present application. Figure 13 As shown, the apparatus 700 may include an acquiring unit 710 , a determining unit 720 , and an optimizing unit 730 .

[0324] An acquisition unit 710 is configured to acquire a first predicted value of renewable energy, a second predicted value of thermal energy demand, and a third predicted value of workload of each data center microgrid in the multi-data center microgrid; wherein the workload includes a deferrable task;

[0325] a determining unit 720, configured to determine an operating cost of the multi-data center microgrid based on the first predicted value and the second predicted value;

[0326] The determining unit 720 is further configured to determine a data migration constraint based on the third predicted value; wherein the data migration constraint includes that the deferrable task is completed within a specified time;

[0327] The optimization unit 730 is configured to optimize the operating cost of the multi-data center microgrid according to the data migration constraint, and obtain the data migration status between the data center microgrids and the workload response status of the data center microgrids.

[0328] Optionally, the third predicted value includes the number and deadline of the deferrable tasks at the first time;

[0329] The determining unit 720 is specifically configured to:

[0330] Determining a task arrival curve of the delayable tasks of each data center microgrid according to the number of delayable tasks arriving at, the number of delayable tasks received by, and the number of delayable tasks transferred out of each data center microgrid at the first time;

[0331] Determining the latest completion curve of the delayable task of each data center microgrid according to the deadline of the delayable task and the task arrival curve;

[0332] The data migration constraint is determined based on a task completion curve of processed deferrable tasks of the multi-data center microgrid, the task arrival curve, and the latest completion curve, wherein the data migration constraint includes the task completion curve being between the task arrival curve and the latest completion curve.

[0333] Optionally, the data migration constraints further include:

[0334] At the moment t When the starting time is t The number of deferrable tasks being processed, and the number of each data center microgrid at time t The number of deferrable tasks processed is the same;

[0335] At the moment t When the time is after the start time, the time t The number of delayable tasks being processed by each data center microgrid is the number of delayable tasks processed by each data center microgrid at time t, time ( t -1) The number of deferrable tasks processed.

[0336] Optionally, the data migration constraints further include:

[0337] The sum of the number of deferrable tasks and the number of real-time tasks being processed by each data center microgrid does not exceed the working capacity of the CPU of each data center microgrid.

[0338] Optionally, the determining unit 720 is specifically configured to:

[0339] determining, based on the first predicted value, the second predicted value, at least one of the power consumption of each data center microgrid, the charge and discharge amounts of the power storage systems of each data center microgrid, and the power consumption of electric boilers, and utilizing a power system supply and demand balance condition, a first penalty cost for not utilizing renewable energy in each data center microgrid;

[0340] determining a second penalty cost for unused thermal energy of each data center microgrid based on at least one of the second predicted value, heat generated by each data center microgrid, heat generated by the electric boiler, and thermal energy released and absorbed by a thermal storage system of each data center microgrid, and utilizing a supply and demand balance condition of the thermal system;

[0341] An operating cost of the multi-data center microgrid is determined based on the amount of electricity purchased by each data center microgrid, the first penalty cost, and the second penalty cost.

[0342] Optionally, the optimization unit 730 is specifically configured to:

[0343] The operating cost of the multi-data center microgrid is optimized based on the data migration constraints and at least one of the data center server constraints, power storage system constraints, thermal storage system constraints, power supply and demand balance constraints, and thermal supply and demand balance constraints.

[0344] Optionally, the constraints on the servers in each data center include:

[0345] The working efficiency of each data center server does not exceed a first preset value, wherein the working efficiency of each data center server is related to the number of work tasks currently processed by each data center server and the upper limit of the number of processing tasks.

[0346] Optionally, the power storage system constraint includes at least one of the following:

[0347] The initial storage capacity of the power storage system per unit time remains constant;

[0348] The amount of electricity stored in the power storage system is between an upper limit and a lower limit of the amount of electricity stored in the power storage system;

[0349] The power storage system is at t The charge capacity of the power storage system does not exceed the t The upper limit of the charging capacity;

[0350] The power storage system is at t The discharge amount does not exceed the power storage system at the time t The upper limit of the discharge capacity;

[0351] The power storage system is at t The stored power of the power storage system at the next moment is t It is related to the storage capacity, charging capacity, charging efficiency, discharge capacity and discharge efficiency.

[0352] Optionally, the thermal storage system constraint includes at least one of the following:

[0353] The initial stored thermal energy per unit time of the thermal storage system remains constant;

[0354] The stored thermal energy of the thermal storage system is between an upper limit and a lower limit of the stored thermal energy of the thermal storage system;

[0355] The thermal storage system is at t The absorbed heat energy does not exceed the upper limit of the absorbed heat energy of the thermal storage system at time t;

[0356] The thermal storage system is at t The released thermal energy does not exceed the thermal storage system at the moment t The upper limit of heat energy released;

[0357] The thermal storage system is at t The stored thermal energy of the next moment is the same as the thermal storage system at the moment t It is related to the storage of thermal energy, absorption of thermal energy, thermal energy absorption efficiency, release of thermal energy, and thermal energy release efficiency.

[0358] Optionally, the power supply and demand balance constraint includes:

[0359] The sum of the renewable energy, purchased electricity and the discharge capacity of the power storage system of each data center microgrid is greater than or equal to the sum of the power consumption of each data center microgrid, the charge capacity of the power storage system and the power consumption of the electric boiler.

[0360] Optionally, the heat supply and demand balance constraint includes:

[0361] The sum of the heat generated by the microgrids of each data center, the heat generated by the electric boiler, and the heat energy released by the thermal storage system is greater than or equal to the sum of the heat load of the microgrids of each data center and the heat energy absorbed by the thermal storage system.

[0362] Optionally, the power consumption of the multi-data center microgrid is related to the power consumption and energy utilization efficiency of the servers of each data center microgrid;

[0363] Among them, the server power consumption of each data center microgrid is related to the CPU power consumption and static power consumption of each data center microgrid, and the CPU power consumption of each data center microgrid is related to the operating frequency, chip model and working efficiency of the CPU of each data center microgrid.

[0364] It should be understood that the device embodiments and method embodiments may correspond to each other, and similar descriptions may refer to the method embodiments. To avoid repetition, they will not be described in detail here. Specifically, in this embodiment, the management device 700 of the multi-data center microgrid may correspond to the corresponding subject that performs the method 200 or 300 of the embodiment of the present application, and the aforementioned and other operations and / or functions of the various modules in the device 700 are respectively for implementing the various methods described above, or the corresponding processes in each method. For the sake of brevity, they will not be described in detail here.

[0365] The above describes the apparatus and system of the embodiment of the present application from the perspective of functional modules in conjunction with the accompanying drawings. It should be understood that the functional module can be implemented in hardware form, can be implemented by instructions in software form, and can also be implemented by a combination of hardware and software modules. Specifically, the steps of the method embodiment in the embodiment of the present application can be completed by the hardware integrated logic circuit and / or software form instructions in the processor, and the steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor for execution, or can be completed by a combination of hardware and software modules in the decoding processor. Optionally, the software module can be located in a mature storage medium in the art such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, etc. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps in the above method embodiment in conjunction with its hardware.

[0366] like Figure 14 It is a schematic block diagram of an electronic device 800 provided in an embodiment of the present application.

[0367] like Figure 14 As shown, the electronic device 800 may include:

[0368] The memory 810 and the processor 820 are configured to store computer programs and transmit the program code to the processor 820. In other words, the processor 820 can call and run the computer program from the memory 810 to implement the method in the embodiment of the present application.

[0369] For example, the processor 820 may be configured to execute the steps in the method 200 according to the instructions in the computer program.

[0370] In some embodiments of the present application, the processor 820 may include but is not limited to:

[0371] General-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0372] In some embodiments of the present application, the memory 810 includes but is not limited to:

[0373] Volatile memory and / or non-volatile memory. Non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can be random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct RAM bus random access memory (DR RAM).

[0374] In some embodiments of the present application, the computer program may be divided into one or more modules, which are stored in the memory 810 and executed by the processor 820 to implement the encoding method provided by the present application. The one or more modules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device 800.

[0375] Optional, such as Figure 14 As shown, the electronic device 800 may further include:

[0376] The transceiver 830 may be connected to the processor 820 or the memory 810 .

[0377] The processor 820 may control the transceiver 830 to communicate with other devices. Specifically, the processor 820 may send information or data to other devices or receive information or data sent by other devices. The transceiver 830 may include a transmitter and a receiver. The transceiver 830 may further include one or more antennas.

[0378] It should be understood that the various components in the electronic device 800 are connected via a bus system, wherein the bus system includes not only a data bus but also a power bus, a control bus and a status signal bus.

[0379] According to one aspect of the present application, a communication device is provided, including a processor and a memory, wherein the memory is used to store a computer program, and the processor is used to call and run the computer program stored in the memory, so that the encoder executes the method of the above method embodiment.

[0380] According to one aspect of the present application, a computer storage medium is provided, on which a computer program is stored. When the computer program is executed by a computer, the computer is enabled to perform the method of the above-mentioned method embodiment. Alternatively, the present application also provides a computer program product containing instructions. When the computer is executed by the instructions, the computer is enabled to perform the method of the above-mentioned method embodiment.

[0381] According to another aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the method of the above-described method embodiment.

[0382] In other words, when implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the embodiments of the present application are fully or partially generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that integrates one or more available media. The available medium can be magnetic media (e.g., floppy disk, hard disk, magnetic tape), optical media (e.g., digital video disc (DVD)), or semiconductor media (e.g., solid-state drive (SSD)).

[0383] It is understood that in the specific implementation of this application, user information and other related data may be involved. When the above embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0384] Those skilled in the art will appreciate that the modules and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0385] In the several embodiments provided in this application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0386] Modules described as separate components may or may not be physically separate, and components displayed as modules may or may not be physical modules, i.e., they may be located in one place or distributed across multiple network elements. Some or all of the modules may be selected based on actual needs to achieve the purpose of the present embodiment. For example, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module.

[0387] The above are only specific embodiments of the present application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A method for managing a multi-data center microgrid, characterized in that: include: Obtaining a first predicted value of renewable energy, a second predicted value of thermal energy demand, and a third predicted value of workload for each data center microgrid in the multi-data center microgrid, wherein the workload includes a deferrable task; Determining an operating cost of the multi-data center microgrid based on the first predicted value and the second predicted value; Determining a data migration constraint based on the third predicted value; wherein the data migration constraint includes that the deferrable task is completed within a specified time; Optimizing the operating cost of the multi-data center microgrid according to the data migration constraints, and obtaining data migration status between the data center microgrids and workload response status of the data center microgrids; The third predicted value includes the number and deadline of the deferrable tasks at the first time; The determining of the data migration constraint according to the third prediction value includes: Determining a task arrival curve of the delayable tasks of each data center microgrid according to the number of delayable tasks arriving at, the number of delayable tasks received by, and the number of delayable tasks transferred out of each data center microgrid at the first time; Determining the latest completion curve of the delayable task of each data center microgrid according to the deadline of the delayable task and the task arrival curve; The data migration constraint is determined based on a task completion curve of the processed deferrable tasks of the multi-data center microgrid, the task arrival curve, and the latest completion curve, wherein the data migration constraint includes the task completion curve being between the task arrival curve and the latest completion curve.

2. The method according to claim 1, characterized in that The data migration constraints also include: At the moment t When the starting time is t The number of delayable tasks being processed is related to the number of delays in each data center microgrid at time t The number of deferrable tasks processed is the same; At the moment t When the time is after the start time, each data center microgrid is at time t The number of delayable tasks being processed is related to the number of delayable tasks processed by each data center microgrid at time t, time ( t -1) The number of delayed tasks that have been processed.

3. The method according to claim 1, characterized in that The data migration constraints also include: The sum of the number of deferrable tasks and the number of real-time tasks being processed by each data center microgrid does not exceed the working capacity of the CPU of each data center microgrid.

4. The method according to any one of claims 1 to 3, characterized in that The determining the operating cost of the multi-data center microgrid according to the first predicted value and the second predicted value includes: determining, based on the first predicted value, the second predicted value, at least one of the power consumption of each data center microgrid, the charge and discharge amounts of the power storage systems of each data center microgrid, and the power consumption of electric boilers, and utilizing a power system supply and demand balance condition, a first penalty cost for not utilizing renewable energy in each data center microgrid; determining a second penalty cost for unused thermal energy of each data center microgrid based on at least one of the second predicted value, heat generated by each data center microgrid, heat generated by the electric boiler, and thermal energy released and absorbed by a thermal storage system of each data center microgrid, and utilizing a supply and demand balance condition of the thermal system; An operating cost of the multi-data center microgrid is determined based on the amount of electricity purchased by each data center microgrid, the first penalty cost, and the second penalty cost.

5. The method according to any one of claims 1 to 3, characterized in that Optimizing the operating cost of the multi-data center microgrid according to the data migration constraint includes: The operating cost of the multi-data center microgrid is optimized based on the data migration constraints and at least one of the server constraints, power storage system constraints, thermal storage system constraints, power supply and demand balance constraints, and thermal supply and demand balance constraints of each data center microgrid.

6. The method according to claim 5, characterized in that The server constraints of each data center microgrid include: The working efficiency of the server of each data center microgrid does not exceed a first preset value, wherein the working efficiency of each server of the data center microgrid is related to the number of work tasks currently processed by the server of each data center microgrid and the upper limit of the number of processed tasks.

7. The method according to claim 5, characterized in that The power storage system constraint includes at least one of the following: The initial storage capacity of the power storage system per unit time remains constant; The amount of electricity stored in the power storage system is between an upper limit and a lower limit of the amount of electricity stored in the power storage system; The power storage system is at t The charge capacity of the power storage system does not exceed the t The upper limit of the charging capacity; The power storage system is at t The discharge amount does not exceed the power storage system at the time t The upper limit of the discharge capacity; The power storage system is at t The stored power of the power storage system at the next moment is t It is related to the storage capacity, charging capacity, charging efficiency, discharge capacity and discharge efficiency.

8. The method according to claim 5, characterized in that The thermal storage system constraint includes at least one of the following: The initial stored thermal energy per unit time of the thermal storage system remains constant; The stored thermal energy of the thermal storage system is between an upper limit and a lower limit of the stored thermal energy of the thermal storage system; The thermal storage system is at t The absorbed heat energy does not exceed the upper limit of the absorbed heat energy of the thermal storage system at time t; The thermal storage system is at t The released thermal energy does not exceed the thermal storage system at the moment t The upper limit of heat energy released; The thermal storage system is at t The stored thermal energy of the next moment is the same as the thermal storage system at the moment t It is related to the storage of thermal energy, absorption of thermal energy, thermal energy absorption efficiency, release of thermal energy, and thermal energy release efficiency.

9. The method according to claim 5, characterized in that The power supply and demand balance constraints include: The sum of the renewable energy, purchased electricity and the discharge capacity of the power storage system of each data center microgrid is greater than or equal to the sum of the power consumption of each data center microgrid, the charge capacity of the power storage system and the power consumption of the electric boiler.

10. The method according to claim 5, characterized in that The heat supply and demand balance constraints include: The sum of the heat generated by the microgrids of each data center, the heat generated by the electric boilers and the heat energy released by the thermal storage system is greater than or equal to the sum of the heat load of the microgrids of each data center and the heat energy absorbed by the thermal storage system.

11. The method according to any one of claims 1 to 3, characterized in that The power consumption of the multi-data center microgrid is related to the server power consumption and energy utilization efficiency of each data center microgrid; Among them, the server power consumption of each data center microgrid is related to the CPU power consumption and static power consumption of each data center microgrid, and the CPU power consumption of each data center microgrid is related to the operating frequency, chip model and working efficiency of the CPU of each data center microgrid.

12. A management device for a multi-data center microgrid, characterized in that: include: an acquisition unit, configured to acquire a first predicted value of renewable energy, a second predicted value of thermal energy demand, and a third predicted value of workload of each data center microgrid in the multi-data center microgrid; wherein the workload includes a deferrable task; a determining unit, configured to determine an operating cost of the multi-data center microgrid based on the first predicted value and the second predicted value; The determining unit is further configured to determine a data migration constraint based on the third predicted value; wherein the data migration constraint includes that the deferrable task is completed within a specified time; an optimization unit, configured to optimize the operating cost of the multi-data center microgrid according to the data migration constraint, and obtain the data migration status between the data center microgrids and the workload response status of the data center microgrids; The third predicted value includes the number and deadline of the deferrable tasks at the first time; The determining unit is specifically configured to: Determining a task arrival curve of the delayable tasks of each data center microgrid according to the number of delayable tasks arriving at, the number of delayable tasks received by, and the number of delayable tasks transferred out of each data center microgrid at the first time; Determining the latest completion curve of the delayable task of each data center microgrid according to the deadline of the delayable task and the task arrival curve; The data migration constraint is determined based on a task completion curve of the processed deferrable tasks of the multi-data center microgrid, the task arrival curve, and the latest completion curve, wherein the data migration constraint includes the task completion curve being between the task arrival curve and the latest completion curve.

13. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores instructions, and when the processor executes the instructions, the processor executes the method according to any one of claims 1 to 11.

14. A computer storage medium, characterized in that Used to store a computer program, the computer program comprising instructions for executing the method according to any one of claims 1 to 11.

15. A computer program product, characterized in that The method comprises a computer program code, which, when executed by an electronic device, causes the electronic device to perform the method according to any one of claims 1 to 11.

Images