Method and device for jointly dispatching data center and power grid

Abstract

The invention provides a method and device for jointly dispatching a data center and a power grid, and relates to the technical field of power transmission. The method comprises the following steps: establishing a first joint scheduling cost function of a data center i and the power grid, wherein the constraint of the first joint scheduling cost function is a first constraint condition; according to a Lyapunov optimization method, the first joint scheduling cost function and the first constraint are converted to obtain a second joint scheduling cost function, and the constraint of the second joint scheduling cost function is a second constraint condition; and determining a minimum value of the second joint scheduling cost function at the current moment, and obtaining a power generation parameter of the power grid required by the data center i, a batch processing calculation load transfer strategy of the data center i and an interactive calculation load scheduling strategy of the data center i. According to the method, the calculation load needing to be processed by the data center is reasonably distributed, and the total operation cost of the data center and the power grid is reduced.

Description

Technical Field

[0001] The present application relates to the field of electric power, and more specifically, to a method and device for jointly scheduling a data center and a power grid. Background Art

[0002] With the development of network technology, the digital age has arrived. Many of the words we are familiar with, such as big data, 5G technology, cloud computing, the Internet of Things, blockchain, artificial intelligence, etc., are all important components of the digital age. Massive amounts of network data may lead to an exponential increase in the computing tasks that data centers need to handle. A single data center can no longer meet user needs, and it has become a trend for multiple data centers to collaboratively handle computing tasks.

[0003] When multiple data centers work together, there is an urgent need to solve the problem of how to dispatch multiple data centers and the power grid to ensure the quality of user experience while fully and reasonably optimizing resource allocation and reducing the total cost of the power grid and data centers. Summary of the invention

[0004] The embodiment of the present application provides a method for jointly scheduling a data center and a power grid, thereby achieving a reasonable distribution of the computing load processed by the data center and reducing the total cost of operating the data center and the power grid.

[0005] In a first aspect, an embodiment of the present application provides a method for jointly scheduling a data center and a power grid, wherein the power grid supplies power to the data center, including:

[0006] Establish a first joint dispatch cost function between data center i and the power grid, i is greater than 1,

[0007] Wherein, the data center i is one of the multiple data centers, the first joint dispatching cost function is the sum of the independent dispatching cost function of the data center i and the independent dispatching cost function of the power grid, and the constraint of the first joint dispatching cost function is the first constraint condition;

[0008] According to the Lyapunov optimization method, converting the first joint scheduling cost function and the first constraint to obtain a second joint scheduling cost function, wherein the constraint of the second joint scheduling cost function is a second constraint condition;

[0009] Determine the minimum value of the second joint scheduling cost function at the current moment, and obtain the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

[0010] In a second aspect, an embodiment of the present application provides a device for jointly scheduling a data center and a power grid, including:

[0011] A processing unit, configured to determine a first joint dispatch cost function between a data center i and the power grid, where i is greater than 1,

[0012] Wherein, the data center i is one of the multiple data centers, the first joint dispatching cost function is the sum of the independent dispatching cost function of the data center i and the independent dispatching cost function of the power grid, and the constraint of the first joint dispatching cost function is the first constraint condition;

[0013] The processing unit is further used to convert the first joint scheduling cost function and the first constraint according to the Lyapunov optimization method to obtain a second joint scheduling cost function, wherein the constraint of the second joint scheduling cost function is a second constraint condition;

[0014] The processing unit is also used to determine the minimum value of the second joint scheduling cost function at the current moment, and obtain the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

[0015] In a third aspect, 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, an embodiment of the present application provides a computer program product or a computer program, the computer program product or the computer program including computer instructions, the computer instructions being stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method of the first aspect above.

[0020] Through the above technical solution, by establishing the joint dispatching cost function of data center i and the power grid, and optimizing the established joint dispatching cost function of data center i and the power grid according to the Lyapunov optimization method, the optimized joint dispatching cost function of data center i and the power grid is obtained. Secondly, by determining the minimum value of the optimized joint dispatching cost function of data center i and the power grid, the power generation parameters of the power grid required by data center i, the batch computing load transfer strategy of data center i and the interactive computing load scheduling strategy of data center i are obtained. This method realizes the reasonable allocation of the computing load to be processed by the data center, reduces the total cost of the operation of the data center and the power grid, and can be decoupled in time through the Lyapunov optimization method. When determining the minimum value of the optimized joint dispatching cost function of data center i and the power grid, only the information at the current moment is required, which makes the calculation simple. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 An optional schematic diagram of a system architecture involved in an embodiment of the present application;

[0022] Figure 2 Another optional schematic diagram of the system architecture involved in the embodiment of the present application;

[0023] Figure 3 A schematic flow chart of a method for jointly scheduling a data center and a power grid provided in an embodiment of the present application;

[0024] Figure 4 A schematic block diagram of a system for independent scheduling of data centers provided in an embodiment of the present application;

[0025] Figure 5 is a schematic block diagram of a system for independent grid dispatching provided in an embodiment of the present application;

[0026] Figure 6 This is a schematic block diagram of a system structure for jointly dispatching a data center and a power grid provided in an embodiment of the present application;

[0027] Figure 7 The real-time queue length of independent scheduling in data centers is shown;

[0028] Figure 8 The real-time queue length of the joint dispatch of data center and power grid is shown;

[0029] Fig. 9 The difference in total cost between joint scheduling and independent scheduling of each subject is shown;

[0030] Fig.10 The cost difference between joint scheduling and independent scheduling of each data center participating in collaborative optimization is shown;

[0031] Fig.11The difference between the cost of each participant after benefit redistribution and the cost during independent scheduling is shown;

[0032] Fig.12 is a schematic block diagram of a device according to an embodiment of the present application;

[0033] Fig.13 A schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0035] Figure 1 FIG. 1 is an optional schematic diagram of a system architecture 100 involved in an embodiment of the present application. Figure 1 As shown, the system architecture 100 includes a power grid dispatch center device 110, a data center batch processing computing load dispatch center 120, a front-end portal server 130, a data center 140, a data center 150 and a joint dispatch device 160. Among them, each device can communicate and interact with each other in a wired or wireless manner.

[0036] The power grid dispatching center 110 is used to receive the power generation information sent by the joint dispatching equipment, and dispatch the thermal generators and wind generators of the power plant to generate electricity according to the power generation information.

[0037] The data center batch computing load scheduling center 120 is used to receive the batch computing load transfer strategies of different data centers sent by the joint scheduling device 160, and schedule data center 140 and data center 150 to transfer or receive batch computing loads according to the batch computing load transfer strategies of different data centers.

[0038] The front-end portal server 130 provides users with Internet access and network access, and the front-end portal server 130 sends the received interactive computing load to the data center 140 and the data center 150 .

[0039] Data center 140 and data center 150 can be independent physical servers, or a server cluster or distributed system composed of multiple physical servers. They can also be cloud servers that provide 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, as well as big data and artificial intelligence platforms.

[0040] The joint dispatching device 160 is used for jointly dispatching the data center 140, the data center 150 and the power grid to reduce the overall operation cost of the multiple data centers and the power grid.

[0041] It should be understood that the above-mentioned joint scheduling device 160 can work independently, or the functions of the above-mentioned joint scheduling device 160 can be implemented in a certain data center or a data center batch processing computing load scheduling center.

[0042] In the system architecture 100, there may also be multiple data centers and multiple front-end portal servers, such as Figure 2 shown.

[0043] Figure 2 FIG. 2 is another optional schematic diagram of the system architecture 200 involved in the embodiment of the present application. Figure 2 As shown, the system architecture 200 includes a front-end portal server 210, a front-end portal server 211, a data center 220, a data center 221, and a power grid. Among them, each device can communicate and interact with each other in a wired or wireless manner, and the power grid provides power to the data center 220 and the data center 221. The front-end portal server 210 and the front-end portal server 211 dispatch the interactive computing load to the data center 220 and the data center 221, and the batch computing load between the data center 220 and the data center 221 is dispatched by the batch computing load dispatching center of the data center.

[0044] It should be understood that Figure 2 Only multiple front-end portal servers and multiple data centers are shown. Figure 2 The system 200 shown also includes the above-mentioned joint dispatching equipment, data center batch processing computing load dispatching center and power grid dispatching center and other equipment.

[0045] In order to more clearly understand the embodiments of the present application, two types of loads processed by the data center in the present application are described below.

[0046] Interactive loads, also known as non-delay-tolerant loads, have high real-time requirements, such as user video chats, voice calls, live broadcasts, and other real-time services.

[0047] Batch computing loads, also known as delay-tolerant loads, do not require real-time performance, such as device system updates and maintenance, and user background downloads of videos, texts, and music.

[0048] With the development of network technology, the digital age has arrived. Many of the words we are familiar with, such as big data, 5G technology, cloud computing, the Internet of Things, blockchain, artificial intelligence, etc., are all important components of the digital age. Massive amounts of network data may lead to an exponential increase in the computing tasks that data centers need to handle. A single data center can no longer meet user needs, and it has become a trend for multiple data centers to collaboratively handle computing tasks.

[0049] Most of the data centers in different regions of my country are operated by different operators, and each data center is usually operated in an independent scheduling mode. However, data centers in different geographical distributions managed by the same data center operator are usually connected by communication networks, and can communicate with each other and transmit data through the communication network. This provides the possibility of spatial transfer of delay-tolerant loads in data centers, and also provides the possibility of joint scheduling of data centers. At the same time, data centers can also be jointly scheduled with the power grid, using the time and space transfer characteristics of delay-tolerant loads in data centers to participate in the demand response of the power grid, helping the power grid achieve system power balance and reduce the operating costs of the power system.

[0050] When multiple data centers work together, how to dispatch multiple data centers and power grids to ensure the quality of user experience while fully and reasonably optimizing resource allocation and reducing the overall cost of power grids and data centers is an urgent problem to be solved.

[0051] Therefore, the present application proposes a method for jointly dispatching a data center and a power grid. The method establishes a joint dispatching cost function of the data center i and the power grid, and optimizes the established joint dispatching cost function of the data center i and the power grid according to the Lyapunov optimization method to obtain the optimized joint dispatching cost function of the data center i and the power grid. Secondly, by determining the minimum value of the optimized joint dispatching cost function of the data center i and the power grid, the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i are obtained. The method adopts the joint dispatching of the data center and the power grid, realizes the reasonable allocation of the computing load to be processed by the data center, reduces the total cost of the operation of the data center and the power grid, and through the Lyapunov optimization method, when determining the minimum value of the joint dispatching cost function of the optimized data center i and the power grid, only the information at the current moment is required, which makes the calculation simple.

[0052] The solution provided by the embodiments of the present application is described below in conjunction with the accompanying drawings.

[0053] Figure 3 The present invention provides a schematic flow chart of a method 300 for jointly dispatching a data center and a power grid. 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 or a computer. The following description takes the electronic device as a joint dispatching device as an example. Figure 3 As shown, method 300 may include steps 310 to 330 .

[0054] S310, the joint dispatching device establishes a first joint dispatching cost function between data center i and the power grid, where i is greater than 1, wherein data center i is one of multiple data centers, and the first joint dispatching cost function is the sum of the independent dispatching cost function of data center i and the independent dispatching cost function of the power grid, and the constraint of the first joint dispatching cost function is the first constraint condition.

[0055] S320, the joint scheduling device converts the first joint scheduling cost function and the first constraint condition according to the Lyapunov optimization method to obtain a second joint scheduling cost function, and the constraint of the second joint scheduling cost function is a second constraint condition.

[0056] S330, the joint scheduling device determines the minimum value of the second joint scheduling cost function at the current moment, and obtains the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

[0057] The joint dispatching device establishes a first joint dispatching cost function of the data center i and the power grid according to the independent dispatching cost function of the data center i and the independent dispatching cost function of the power grid. The constraint condition of the first joint dispatching cost function includes the batch computing load constraint parameter of the data center i. The batch computing load constraint parameter of the data center i can enable the data center i to transfer or receive the batch computing load under the condition of satisfying the computing power, thereby realizing the joint dispatching of the data centers. The joint dispatching device converts the first joint dispatching cost function and the first constraint according to the Lyapunov optimization method to obtain the second joint dispatching cost function. By determining the minimum value of the second joint dispatching cost function at the current moment, the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load dispatching strategy of the data center i are obtained. The method can realize the joint dispatching of multiple data centers and the power grid, and reduce the total cost of the dispatching of multiple data centers and the power grid.

[0058] In order to more clearly understand the embodiment of the present application, the method 300 is described in steps below.

[0059] S310, a joint dispatching device establishes a first joint dispatching cost function between data center i and the power grid.

[0060] Before the joint dispatching device establishes a first joint dispatching cost function between the data center i and the power grid, the joint dispatching device needs to establish an independent dispatching cost function of the data center i and an independent dispatching cost function of the power grid.

[0061] 1. The establishment process of independent scheduling cost function of data center:

[0062] When running independently, data center i does not communicate with the outside world to exchange data, but only processes the computing load that reaches data center i. At the same time, the front-end portal server does not flexibly distribute the interactive computing load, but adopts an average distribution. Figure 4 As shown, Figure 4 A schematic block diagram of a system for independent scheduling of data centers. Figure 4 In the example, the front-end portal server distributes the interactive computing load evenly to data center 1 and data center 2, and there is no load transfer between data center 1 and data center 2.

[0063] The cost function of independent scheduling of data center i is set to the minimum value of the sum of the cost of purchasing electricity from the power grid, the delay cost of interactive computing load, and the cost of discarding batch computing load. At the same time, considering the randomness of the arrival of computing load in the data center, the cost function of independent scheduling of the data center is set as a random programming problem, and the expectation is adopted, as shown in formula (1):

[0064]

[0065] Among them, E{·} is the expectation, C 1, The data center pays the power grid for some costs, C 2, is the latency cost of interactive computing load in the data center, C 3, Computing load dropping costs for batch processing in data centers.

[0066] The cost of data center buying electricity from the grid C 1, The calculation of is shown in formula (2):

[0067]

[0068] Among them, C p,i,t is the electricity price that data center i purchases from the power grid at time t, t i,t is the amount of electricity purchased by data center i from the grid at time t.

[0069] Since there is a certain delay between the front-end portal server and each data center, and the interactive computing load in the data center needs to be processed immediately, if a delay occurs, a certain delay fee must be paid. Therefore, the interactive computing load delay cost C of the data center is 2, The calculation of is shown in formula (3):

[0070]

[0071] Among them, α is the conversion factor between delay time and cost, F is the number of front-end portal servers, N is the number of data centers, and af,i,t is the amount of interactive computing load assigned by the front-end portal server f to data center i at time t, b f,t is the number of interactive computing loads arriving at the front-end portal server at time t, L f,t is the transmission delay between the front-end portal server f and the data center i.

[0072] Since the processing capacity of the data center is limited, when the interactive computing load and batch computing load arriving at the data center exceed the maximum processing capacity, the computing task scheduling center inside the data center will discard part of the batch computing load, and the cost of the discarded part also needs to pay a certain discard fee. Therefore, the batch computing load discarding cost of the data center is C 3, The calculation of is shown in formula (4):

[0073]

[0074] Among them, β i The penalty factor for discarding batch computing load for data center i, g i,t Count the number of batches dropped for data center i.

[0075] The independent scheduling cost function of a data center needs to satisfy the following third constraint:

[0076] (1) The interactive computing load of the data center is evenly distributed by the front-end portal servers, and the distributed interactive computing load is a non-negative value;

[0077] The interactive computing load in the data center is evenly distributed by the front-end portal servers, and the distributed interactive computing load is a non-negative value, as shown in formula (5):

[0078]

[0079] (2) For batch computing loads in data centers, independent scheduling only has the time transfer feature. In terms of time, batch computing loads can be postponed to any time before the deadline, so queues are used to characterize this feature. Since the data center has limited computing load processing capacity, when the number of interactive computing loads and batch computing loads exceeds the maximum capacity that the data center can handle, the computing load scheduling center in the data center will discard part of the batch computing load, as shown in formula (6):

[0080]

[0081] Among them, l i,t is the number of batch processing loads arriving at data center i at time t, e i,tis the number of batch processing loads processed by data center i at time t, g i,t Cap is the number of batch processing loads discarded by data center i at time t, i F is the maximum capacity of the data center to handle the computing load. i,t is the computational load queue length of data center i, The maximum number of batch computing loads that can be processed by data center i at time t.

[0082] (3) In order to ensure that the batch computing load of the data center will not be infinitely accumulated in the queue without being processed, a queue stability constraint is added, as shown in formula (7):

[0083]

[0084] (4) The power consumption of a data center is defined as the sum of the server idle power, the peak cooling power, and the power consumed by processing computing loads, as shown in formula (8):

[0085]

[0086] in, is the idle power of a server in data center i, is the peak power of a server in data center i, PUE i It is the ratio of the total power consumption of the data center to the power consumption of Internet technology (IT) equipment.

[0087] 2. The establishment process of independent grid dispatch cost function:

[0088] like Figure 5 As shown, Figure 5 Schematic block diagram of the system for independent grid dispatch. Figure 5 In the process, thermal power generation and wind power generation in the power grid work simultaneously to supply power to the data center.

[0089] When the power grid is independently dispatched, the data center purchases the required electricity from the power grid, and the power grid controls the power generation of the generator to meet the electricity requirements of each load in the power grid.

[0090] The independent dispatch cost function of the power grid is set as the sum of the power generation cost of the thermal power unit and the wind farm wind curtailment cost minus the minimum value of the power grid’s income from selling electricity to the data center, as shown in formula (9):

[0091]

[0092] in, is the power generation cost of thermal power units, is the wind farm abandonment cost, C ei,Profits from selling electricity to data centers for the power grid.

[0093] The power generation cost of thermal power units in the power grid is expressed by a quadratic polynomial function, as shown in formula (10):

[0094]

[0095] Among them, a 0, is the constant term of the operating cost of the thermal power generator g, a 1, is the first-order coefficient of the operating cost of the thermal power generator g, a 2, is the quadratic coefficient of the operating cost of the thermal power generator g, S TU Subscript for the generating unit of a thermal power plant.

[0096] The power grid also includes wind farm power generation. When the load power consumption is low, the wind farm does not need to use all the wind to generate electricity, and will discard some of the wind. At this time, a certain fee must be paid for the discarded wind. Therefore, the power grid wind abandonment cost is shown in formula (11):

[0097]

[0098] Among them, σ g is the wind turbine generator g wind abandonment penalty factor, The predicted power generation of the wind turbine generator g at time t, is the power generated by the wind turbine generator g at time t, S wind Subscript for the wind turbine generating unit.

[0099] The revenue that the power grid earns by selling electricity to the data center is C ei, , as shown in formula (12):

[0100]

[0101] Among them, C s,i,t The price of electricity sold to the grid.

[0102] The independent dispatch cost function of the power grid needs to meet the following fourth constraint:

[0103] (1) The thermal power unit needs to consider the reserve capacity of the thermal power unit, as shown in formula (13):

[0104]

[0105] Among them, g,t is the upward rotation reserve capacity of the thermal power unit generator g at time t, rd g,t is the downward rotation reserve capacity of the thermal power unit generator g at time t, is the upward climbing rate of the generator g of the thermal power unit, is the upward climbing rate of the generator g of the thermal power unit, is the maximum output of the thermal power generator g, P g is the minimum output of the thermal power generator g, p g,t is the power generation of the thermal power unit generator g at time t, SR up For the power system upward spinning reserve capacity requirement, SR down Spin down reserve capacity requirements for the power system.

[0106] (2) Thermal power units also need to consider the ramp rate constraint, as shown in formula (14):

[0107]

[0108] (3) For the network constraints in the power grid, DC power flow constraints are adopted, as shown in formula (15):

[0109]

[0110] Among them, θ m,t is the phase angle of busbar m at time t, D m,t is the power demand of the external load of busbar m at time t, F (m,n) is the line transmission capacity between busbar m and busbar n, X m,n is the line reactance value between busbar m and busbar n.

[0111] (4) Both the thermal power generators and wind turbine generators in the power grid must meet the output constraints, as shown in formula (16):

[0112]

[0113] It should be understood that in the process of establishing the above-mentioned independent dispatching cost function of the power grid, the situations of wind power generation and thermal power generation are taken into consideration at the same time. Of course, there is also a situation where the power grid only includes wind power generation or thermal power generation. In this case, the independent dispatching cost function of the power grid can be determined according to the existing technology, and the embodiments of this application will not be described again.

[0114] 3. The establishment process of the joint dispatching cost function of data center and power grid:

[0115] The joint dispatching device sets a first joint dispatching cost function between data center i and the power grid, and the first joint dispatching cost function is the sum of the independent dispatching cost function of the data center and the independent dispatching cost function of the power grid.

[0116] However, when the data center is jointly scheduled with the power grid, the interactive computing load in the data center is freely distributed by the front-end portal server, and the batch computing load can be transferred between different data centers. The interactive computing load distribution needs to meet the geographical load balancing constraint. When the data center is independently scheduled, the interactive computing load is evenly distributed by the front-end portal server. Therefore, the interactive computing load distribution constraint needs to be changed to the geographical load balancing constraint in the independent scheduling operation constraint of the data center, as shown in formula (17):

[0117]

[0118] At the same time, during joint operation, since data centers are connected by a communication network, data can be exchanged through the communication network, so batch computing loads can be transferred between data centers. However, during independent scheduling, there is no data exchange between data centers. Therefore, it is necessary to add batch computing load transfer constraints to the independent scheduling operation constraints, as shown in formula (18):

[0119]

[0120] Among them, Cap i,i′ is the data transmission capacity between data center i and data center i', h i→i′,t is the computing load transferred from data center i to data center i', d i,t Calculate the number of total batches that data center i needs to process at time t.

[0121] Therefore, the first joint scheduling cost function is as shown in formula (19):

[0122]

[0123] The joint dispatch constraint is the first constraint, which includes the fifth constraint and the fourth constraint. The fifth constraint is the constraint of the data center during joint dispatch, and the fourth constraint is the constraint of the power grid during joint dispatch. You can refer to the description of the fourth constraint in the above power grid for understanding.

[0124] The fifth constraint is shown in formula (20):

[0125]

[0126] In step S320, the joint scheduling device converts the first joint scheduling cost function and the first constraint condition according to the Lyapunov optimization method to obtain a second joint scheduling cost function, and the constraint of the second joint scheduling cost function is a second constraint condition.

[0127] The following specifically describes how to convert the first joint scheduling cost function and the first constraint condition according to the Lyapunov optimization method to obtain the second joint scheduling cost function.

[0128] First, the first joint scheduling cost function is converted into the Lyapunov standard form, as shown in formula (21):

[0129]

[0130] Its constraints include the sixth constraint and the fourth constraint. The third constraint is shown in formula (22):

[0131]

[0132] Set the Lyapunov function as shown in formula (23):

[0133]

[0134] Define Lyapunov drift as shown in formula (24):

[0135] ΔL(F(t))=E{L(F(t+1))-L(F(t))|F(t)} (24)

[0136] If only ΔL(F(t)) is optimized, it can only ensure that the queue backlog is always pushed to a lower congestion state, and the objective function cannot reach the expected state. Therefore, a drift plus penalty function is defined, as shown in formula (25):

[0137] Δv=ΔL(F(t))+V·E{(C idc,t +C ed, t)|F(t)} (25)

[0138] By controlling the size of parameter V, a smooth trade-off is made between backlog reduction and penalty minimization, thereby obtaining a near-optimal result while ensuring stability.

[0139] By the formula:

[0140] F i,t+1 =max{F i,t -e i,t -g i,t ,0}+d i,t

[0141] Available

[0142]

[0143] Add both sides and multiply have to:

[0144]

[0145]

[0146] Among them, B is a constant, which is The upper bound of

[0147]

[0148] Then the Lyapunov drift plus penalty is as shown in formula (26):

[0149]

[0150] Finally, the second joint scheduling cost function is obtained, as shown in formula (27):

[0151]

[0152] The constraint condition is the second constraint condition, which includes the seventh constraint condition and the fourth constraint condition. The seventh constraint condition is shown in formula (28):

[0153]

[0154] Through the Lyapunov optimization method, the long-term joint scheduling cost function of the data center and the power grid is converted into a single-time joint scheduling cost function that only requires current-time information, and the second joint scheduling cost function does not need to rely on the information of the previous time.

[0155] In step S230, the joint scheduling device determines the minimum value of the second joint scheduling cost function at the current moment, and obtains the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

[0156] Optionally, the joint scheduling device determines the minimum value of the second joint scheduling cost function at the current moment, obtains the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i, including:

[0157] Obtaining the total interactive load received by the front-end server f connected to the data center i at the current moment, the number of batch computing loads received by the data center i, the electricity price at which the data center i purchases electricity from the power grid, and the batch computing load queue length of the data center i, wherein the total interactive load, the number of batch computing loads, the electricity price, and the batch computing load queue length are parameters included in the second constraint condition;

[0158] Based on the total interactive load received by the front-end server f, the number of batch computing loads received by the data center i, the electricity price at which the data center i purchases electricity from the power grid, and the batch computing load queue length of the data center i, the minimum value of the second joint scheduling cost function at the current moment is determined, and the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i are obtained.

[0159] Specifically, at the current time t, the joint scheduling device obtains the total interactive load b received by the front-end server f connected to the data center i at the current time f,t , the number of batch computing loads l received by the data center i i,t , the electricity price C that the data center i purchases from the power grid p,i,t And the batch processing load queue length F of the data center i i,t . The b obtained f,t , l i,t , C p,i,t and F i,t Substitute the second constraint condition and determine the minimum value of the second joint scheduling cost function at the current moment through the solver to obtain the power generation parameters of the power grid required by data center i, the batch computing load transfer strategy of data center i and the interactive computing load scheduling strategy of data center i.

[0160] For example, in a joint dispatching system, including front-end server 1 and front-end server 2, data center 1, data center 2 and data center 3 and a power grid, front-end server 1 is connected to data center 1, data center 2 and data center 3, and front-end server 2 is connected to data centers 1 and 2. At the current moment, if the power generation parameters of the power grid required by data center 1, the batch processing load transfer strategy of data center 1 and the interactive computing load scheduling strategy of data center 1 are determined, the total interactive load b received by the front-end server that needs to be input is f,t Including the interactive load of front-end server 1 and front-end server 2, if the power generation parameters of the power grid required by data center 3, the batch processing load transfer strategy of data center 1 and the interactive computing load scheduling strategy of data center 1 are determined, the total interactive load b received by the front-end server needs to be input f,t Only the interactive load of front-end server 1 is included.

[0161] Optionally, the obtaining the number of batch computing loads received by the data center i, the price of electricity purchased by the data center i from the power grid, and the batch computing load queue length of the data center i includes:

[0162] receiving the number of batch computing loads received by the data center i, the price of electricity purchased by the data center i from the power grid, and the batch computing load queue length of the data center i, which are sent by the data center i; or,

[0163] Receive the batch computing load scheduling device of the data center, which sends the number of batch computing loads received by the data center i, the electricity price at which the data center i purchases electricity from the power grid, and the batch computing load queue length of the data center i.

[0164] Optionally, the method 200 further includes: the joint scheduling device sending the batch computing load transfer strategy of the data center i to the data center i; or,

[0165] The joint scheduling device sends the batch computing load transfer strategy of the data center i to the data center batch computing load scheduling device, so that the data center batch computing load scheduling device can schedule the batch computing load of the data center i according to the batch computing load transfer strategy of the data center i.

[0166] Specifically, when the function of the joint scheduling device is integrated in the data center batch computing load scheduling device, or when the data center batch computing load scheduling device has the function of the joint scheduling device, the joint scheduling device directly sends the batch computing load transfer strategy of the data center i to the data center i, so that the data center i receives or transfers the batch computing load according to the batch computing load transfer strategy. When the joint scheduling device and the data center batch computing load scheduling device are different devices, the joint scheduling device sends the batch computing load transfer strategy of the data center i to the data center batch computing load scheduling device, so that the data center batch computing load scheduling device schedules the batch computing load of the data center i according to the batch computing load transfer strategy of the data center i.

[0167] It should be understood that the function of the joint scheduling device can also be integrated and implemented in a certain data center, such as the first data center. In this case, the first data center sends the batch computing load transfer strategy of other data centers to other data centers.

[0168] Optionally, the method 200 includes:

[0169] The joint dispatching device sends the power generation parameters of the power grid required by the data center i to the power grid dispatching center device, so that the power grid dispatching center device dispatches thermal generators and wind generators to generate electricity according to the power generation parameters of the power grid required by the data center i.

[0170] Optionally, the method 200 includes:

[0171] Determining a power generation strategy of the power grid according to power generation parameters of the power grid required by the plurality of data centers i;

[0172] Send the power generation parameters of the power grid required by the data center i to the communication device of the power grid. Optionally, the batch computing load transfer strategy of the data center i includes the batch computing loads of other data centers received by the data center i and / or the batch computing loads transferred by the data center i to other data centers.

[0173] Optionally, the method 200 further includes:

[0174] The joint scheduling device sends the interactive computing load scheduling strategy of the data center i to the front-end server f.

[0175] Optionally, the method 200 further includes:

[0176] The joint scheduling device updates the computing load queue length of the data center i according to the parameters obtained when determining the minimum value of the second joint scheduling cost function at the current moment.

[0177] Specifically, the joint scheduling device can obtain multiple parameters when determining the minimum value of the second joint scheduling cost function at the current moment, including d i,t , e i,t and g i,t , you can use d i,t , e i,t and g i,t Update the calculation load queue length F of the data center i i,t , to be used to determine the minimum value of the second joint scheduling cost function at the next moment.

[0178] To summarize, after optimizing and calculating the minimum value of the second joint dispatching cost function, the joint dispatching device transmits the optimal power generation decision to the power grid dispatching center device, and the power grid dispatching center device controls the power generation of the thermal power generating set and / or the wind power generating set.

[0179] After optimizing the minimum value of the second joint scheduling cost function, the joint scheduling device transmits the obtained batch computing load transfer scheduling decision of the optimal data center to the data center batch computing load scheduling center, and transmits the obtained optimal data center interactive computing load scheduling decision to the front-end portal server.

[0180] The data center batch computing load dispatching center sends a transfer / receive batch computing load decision to the corresponding data center according to the received dispatching decision. The front-end portal server divides the interactive computing load into each data center according to the received optimal dispatching decision. Each data center executes according to the received batch computing load dispatching decision to realize the spatial transfer of computing power. At this time, the transfer of computing power of the data center is mapped to the power grid as the transfer of electricity, that is, the data center participates in the power grid demand response. Although the joint dispatching device determines the minimum value of the second joint dispatching cost function, that is, the total cost value of the joint dispatching of the data center and the power grid is the minimum, however, since it is very likely that a subject will not participate in the joint dispatching in a single-period collaboration, the final result in this case will lead to dissatisfaction among multiple parties. Therefore, this application adopts long-term joint dispatching.

[0181] Although the total cost of data center-grid joint scheduling is guaranteed to be lower than the total cost during independent operation, the cost of the data center receiving the computing task may be higher than the cost during independent operation, which is not conducive to the data center's participation in the joint scheduling. Therefore, this application redistributes the total benefits obtained from the joint scheduling of data centers and power grids in response to this situation, ensuring that each participating entity will profit in the end, that is, the final operating cost of each participating entity is lower than the cost during independent scheduling, thereby ensuring the enthusiasm of each participating entity to participate in collaborative optimization.

[0182] Optionally, the method 200 further includes:

[0183] According to a plurality of independent dispatching costs of the data center i, the independent dispatching costs of the power grid and the total cost of joint dispatching of the data center i and the power grid, the cost of benefit redistribution of the joint dispatching of the data center i and the cost of benefit redistribution of the joint dispatching of the power grid are determined when the data center i is jointly dispatched with the power grid, so that the cost of benefit redistribution of the joint dispatching of the data center i is less than the independent dispatching cost of the data center i, and the cost of benefit redistribution of the joint dispatching of the power grid is less than the independent dispatching cost of the power grid.

[0184] Optionally, determining, according to a plurality of independent dispatching costs of the data center i, the independent dispatching costs of the power grid, and the total cost of joint dispatching of the data center i and the power grid, a cost of redistribution of benefits of joint dispatching of the data center i and a cost of redistribution of benefits of joint dispatching of the power grid when the data center i is jointly dispatched with the power grid, includes:

[0185] The joint dispatching cost of the data center i and the joint dispatching cost of the power grid are determined according to formula (29):

[0186]

[0187] in, is the cost of independent dispatch of the power grid, is the cost of independent scheduling of the data center i, C e ′ pn,t is the cost of redistribution of the benefits of the joint dispatch of the power grid, C i ′ dc,i,t The cost of redistribution of the joint scheduling benefit for the data center i, The total cost of the joint dispatch of data center i and the power grid.

[0188] Specifically, the above formula (29) is a benefit redistribution model constructed based on the symmetric Nash negotiation model, that is, the cost saved by the joint dispatch of the data center and the power grid is redistributed. It should be understood that the cost saved by the joint dispatch of the data center and the power grid can be called benefit. The benefit generated by the joint dispatch of the data center and the power grid is redistributed to ensure that each participant will profit, that is, the final operating cost of each participant is lower than the cost of independent dispatch, which ensures the enthusiasm of each participant to participate in collaborative optimization.

[0189] It should be understood that the cost of redistributing the benefits of joint dispatching of the power grid and the cost of joint dispatching of data center i can also be achieved through other methods, such as calculating the sum of the cost of independent dispatching of the power grid and the cost of independent dispatching of data center i, determining the difference between the sum of the cost of independent dispatching of the power grid and the cost of independent dispatching of data center i and the total cost of joint dispatching of data center i and the power grid, obtaining the cost saved by the joint dispatching of the data center and the power grid, and allocating the saved cost to each participating entity.

[0190] An embodiment of the present application provides a system for jointly scheduling a data center and a power grid, the system comprising a joint scheduling device, a data center batch processing computing load scheduling device, multiple data centers and multiple front-end servers, wherein the specific functions of the joint scheduling device, the data center batch processing computing load scheduling device, the multiple data centers and the multiple front-end servers can be understood by referring to the above method 300 and will not be repeated here.

[0191] The following is a detailed description of the simulation verification of the method proposed in the embodiment of the present application.

[0192] First, a 6-node 2-data center system was constructed, in which there are two thermal generators and one wind turbine in the power grid, and the data centers are set in node 2 and node 4 respectively. Figure 6 As shown, Figure 6 The following is a schematic diagram of the system structure for jointly dispatching data centers and power grids. Figure 6In the figure, G1 and G2 are thermal generators, G-wind is a wind generator, BUS1~BUS6 are busbars, D1~D3 are the power demands of other users except the data center, such as factory power and ordinary citizens. Data center 1 and data center 2 are set on node 2 and node 4 respectively, the system parameters are set, and one hour is set as a moment. The scheduling of the above function in 24 hours is solved in Matlab.

[0193] Since the load queue constraints are directly removed when the Lyapunov optimization method is used for conversion, it is necessary to verify whether the function can satisfy the original queue constraints, that is, the queue is limited and stable. By solving the data center independent scheduling function mentioned above, the real-time queue length of the data center independent scheduling is obtained, such as Figure 7 As shown, Figure 7 The real-time queue length of independent scheduling in data centers is shown. Figure 7 It can be seen that when the data centers are independently scheduled, the queue of each data center is limited within 24 hours, which means that the batch computing load in the data center will not be infinitely backlogged in the queue without being processed. By solving the joint scheduling function of the data center and the power grid mentioned above, the real-time queue length of the joint scheduling of the data center and the power grid is obtained, as follows: Figure 8 As shown, Figure 8 The real-time queue length of the joint scheduling of data center and power grid is shown by Figure 8 It can be seen that when the data center and the power grid are jointly scheduled, the queue of each data center is limited within 24 hours, which indicates that the batch computing load in the data center will not be infinitely backlogged in the queue and not processed. At the same time, the increase in the queue length of the data center also shows that the time transfer characteristics of the batch computing load of the data center meet the constraints.

[0194] The independent dispatching function of the data center, the independent dispatching function of the power grid, and the joint dispatching function of the data center and the power grid are solved respectively, and the difference diagram between the joint dispatching cost and the independent dispatching cost of each subject is obtained. Fig. 9 As shown, Fig. 9 The total cost difference between joint scheduling and independent scheduling of each subject is shown by Fig. 9 It can be seen that the joint scheduling of data centers and power grids can indeed reduce the total cost of data centers and power grids due to their strong load flexibility. Fig.10 The cost difference between joint scheduling and independent scheduling of each data center participating in collaborative optimization is shown in Fig.10 It can be seen that the cost of joint scheduling increases in Data Center 1 and Data Center 2 at different times.

[0195] Finally, by solving the benefit redistribution model, we can obtain the difference between the cost of each participant after benefit redistribution and the cost during independent scheduling, as shown in Fig.11As shown by Fig.11 It can be seen that after the redistribution of benefits, the cost of each participating entity is lower than the cost during independent scheduling, which ensures the enthusiasm of each participating entity to participate in collaborative optimization. At the same time, due to the adoption of a symmetric Nash negotiation model, each participating entity has the same weight and finally obtains the same benefits.

[0196] The specific implementation modes 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 implementation modes. Within the technical concept of the present application, the technical solution of the present application can be subjected to a variety of simple modifications, and these simple modifications all belong to the protection scope of the present application. For example, the various specific technical features described in the above specific implementation modes can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present application will not further explain various possible combinations. For another example, the various different implementation modes 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.

[0197] It should also be understood that in the various method embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and 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.

[0198] Combination of the above Figures 1 to 10 , describes in detail the method embodiment of the present application, and the following is combined with Fig.11 and Fig.12 , describe in detail the device embodiments of the present application.

[0199] Fig.11 4 is a schematic block diagram of an apparatus 400 according to an embodiment of the present application, and the apparatus 400 can implement the function of the joint scheduling device in the above method. Fig.11 As shown, the apparatus 400 may include a processing unit 410 .

[0200] The processing unit 410 is configured to establish a first joint dispatch cost function between the data center i and the power grid, where i is greater than 1.

[0201] Wherein, the data center i is one of the multiple data centers, the first joint dispatching cost function is the sum of the independent dispatching cost function of the data center i and the independent dispatching cost function of the power grid, the constraint of the first joint dispatching cost function is the first constraint condition, and the constraint condition of the independent dispatching cost function of the data center i includes the batch computing load constraint parameter of the data center i;

[0202] The processing unit 410 is further configured to convert the first joint scheduling cost function and the first constraint according to the Lyapunov optimization method to obtain a second joint scheduling cost function, wherein the constraint of the second joint scheduling cost function is a second constraint condition;

[0203] The processing unit 410 is also used to determine the minimum value of the second joint scheduling cost function at the current moment, and obtain the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

[0204] In some embodiments, the device 400 further includes an acquisition unit 420, which is used to acquire the total interactive load received by the front-end server f, the number of batch computing loads received by the data center i, the electricity price of the electricity purchased by the data center i from the power grid, and the batch computing load queue length of the data center i at the current moment, wherein the total interactive load, the number of batch computing loads, the electricity price, and the batch computing load queue length are parameters included in the second constraint condition;

[0205] The processing unit 410 is also used to determine the minimum value of the second joint scheduling cost function at the current moment based on the total interactive load received by the front-end server f, the number of batch computing loads received by the data center i, the electricity price at which the data center i purchases electricity from the power grid, and the length of the batch computing load queue of the data center i, and obtain the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

[0206] In some embodiments, the processing unit 410 is further used to determine, based on a plurality of independent dispatching costs of the data center i, the independent dispatching costs of the power grid and the total cost of joint dispatching of the data center i and the power grid, the cost of benefit redistribution for the joint dispatching of the data center i and the cost of benefit redistribution for the joint dispatching of the power grid when the data center i is jointly dispatched with the power grid, so that the cost of benefit redistribution for the joint dispatching of the data center i is less than the independent dispatching cost of the data center i, and the cost of benefit redistribution for the joint dispatching of the power grid is less than the independent dispatching cost of the power grid.

[0207] In some embodiments, the processing unit 410 is further configured to determine, when the data center i is jointly dispatched with the power grid, the cost of redistribution of benefits from the joint dispatch of the data center i and the cost of redistribution of benefits from the joint dispatch of the power grid, according to the multiple independent dispatch costs of the data center i, the independent dispatch costs of the power grid, and the total cost of the joint dispatch of the data center i and the power grid, including:

[0208] The joint dispatching cost of the data center i and the joint dispatching cost of the power grid are determined according to formula (30):

[0209]

[0210] in, is the cost of independent dispatch of the power grid, is the cost of independent scheduling of data center i, C e ′ pn,t The cost of redistributing the benefits of the joint dispatch of the grid, C i ′ dc,i,t The cost of redistribution of the benefits of joint scheduling for the data center i, The total cost of the joint dispatch of data center i and the power grid.

[0211] In some embodiments, the apparatus 400 further includes a transceiver unit 430, and the transceiver unit 430 is configured to:

[0212] Sending the batch computing load transfer strategy of the data center i to the data center i; or,

[0213] The batch processing load transfer strategy of the data center i is sent to the data center batch processing load scheduling device.

[0214] In some embodiments, the transceiver unit 430 is further configured to:

[0215] The power generation parameters of the power grid required by the data center i are sent to the communication device of the power grid.

[0216] In some embodiments, the batch processing computing load transfer strategy of the data center i includes the batch processing computing load received by the data center i from other data centers and / or the batch processing computing load transferred by the data center i to other data centers.

[0217] In some embodiments, the transceiver unit 430 is further used to send the interactive computing load scheduling strategy of the data center i to the front-end server f.

[0218] In some embodiments, the independent scheduling cost function of data center i is the minimum value of the sum of the electricity purchase cost, interactive computing load delay cost and batch computing load discard cost of data center i, and the constraint of the independent scheduling cost function of data center i is the third constraint condition.

[0219] In some embodiments, the processing unit 410 is further used to: update the computing load queue length of the data center i according to the parameters obtained when determining the minimum value of the second joint scheduling cost function at the current moment.

[0220] It should be understood that the device embodiment and the method embodiment may correspond to each other, and similar descriptions may refer to the method embodiment. To avoid repetition, no further description is given here. Specifically, when the data processing device 400 in this embodiment may correspond to the execution subject of the method 300 of the embodiment of the present application, the aforementioned and other operations and / or functions of each module in the device 400 are respectively to implement Figure 3 For the sake of brevity, the corresponding processes of each method in are not repeated here.

[0221] The above describes the device and system of the embodiment of the present application from the perspective of the functional module 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 to perform, or a combination of hardware and software modules in the decoding processor to perform. Optionally, the software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, an electrically erasable programmable memory, a register, etc. The storage medium is located in a memory, and the processor reads the information in the memory, and completes the steps in the above method embodiment in conjunction with its hardware.

[0222] like Fig.12 It is a schematic block diagram of an electronic device 500 provided in an embodiment of the present application.

[0223] like Fig.12 As shown, the electronic device 500 may include:

[0224] The memory 510 and the processor 520, the memory 510 is used to store the computer program and transmit the program code to the processor 520. In other words, the processor 520 can call and run the computer program from the memory 510 to implement the method in the embodiment of the present application.

[0225] For example, the processor 520 may be used to execute the steps of each execution subject in the above method 300 according to the instructions in the computer program.

[0226] In some embodiments of the present application, the processor 520 may include but is not limited to:

[0227] General-purpose processor, digital signal processor (DSP), application-specific integrated circuit (ASIC), field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.

[0228] In some embodiments of the present application, the memory 510 includes but is not limited to:

[0229] Volatile memory and / or non-volatile memory. Among them, the non-volatile memory can be read-only memory (ROM), programmable ROM (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or flash memory. The 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).

[0230] In some embodiments of the present application, the computer program may be divided into one or more modules, which are stored in the memory 510 and executed by the processor 520 to complete the method provided by the present application. The one or more modules may be a series of computer program instruction segments capable of completing specific functions, and the instruction segments are used to describe the execution process of the computer program in the electronic device 500.

[0231] Optionally, the electronic device 500 may further include:

[0232] The communication interface 530 may be connected to the processor 520 or the memory 510 .

[0233] The processor 520 may control the communication interface 530 to communicate with other devices, specifically, to send information or data to other devices, or to receive information or data sent by other devices. Exemplarily, the communication interface 530 may include a transmitter and a receiver. The communication interface 530 may further include an antenna, and the number of antennas may be one or more.

[0234] It should be understood that the various components in the electronic device 500 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.

[0235] 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.

[0236] According to one aspect of the present application, a computer storage medium is provided, on which a computer program is stored, and when the computer program is executed by a computer, the computer can perform the method of the above method embodiment. In other words, the present application embodiment also provides a computer program product containing instructions, and when the instructions are executed by a computer, the computer can perform the method of the above method embodiment.

[0237] According to another aspect of the present application, a computer program product or computer program is provided, the computer program product or computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method of the above method embodiment.

[0238] 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 process or function according to the embodiment of the present application is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server, or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or a data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a solid state drive (solid state disk, SSD)), etc.

[0239] It should be understood that in the embodiment 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 only on A, and B can also be determined based on A and / or other information.

[0240] In the description of the present 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 represent: 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 associated objects before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0241] It should also be understood that the first, second, etc. descriptions appearing in the embodiments of the present application are only used for illustration and distinction of the description objects, without any distinction of order, nor do they indicate 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.

[0242] It should also be understood that the specific features, structures or characteristics related to the embodiments in the specification are included in at least one embodiment of the present application. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner.

[0243] In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or device.

[0244] It is understandable 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 relevant laws, regulations and standards of relevant countries and regions.

[0245] Those of ordinary skill 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 to be beyond the scope of this application.

[0246] In the several embodiments provided in the present 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 only schematic. For example, the division of the module is only a logical function division. There may be other division methods in actual implementation, 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.

[0247] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. For example, each functional module in each embodiment 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 one module.

[0248] The above are only specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for jointly scheduling a data center and a power grid, wherein the power grid supplies power to the data center, characterized in that: include: Establish a first joint dispatch cost function between data center i and the power grid, i is greater than 1, Wherein, the data center i is one of the multiple data centers, the first joint dispatching cost function is the sum of the independent dispatching cost function of the data center i and the independent dispatching cost function of the power grid, and the constraint of the first joint dispatching cost function is the first constraint condition; According to the Lyapunov optimization method, converting the first joint scheduling cost function and the first constraint to obtain a second joint scheduling cost function, wherein the constraint of the second joint scheduling cost function is a second constraint condition; Determine the minimum value of the second joint scheduling cost function at the current moment, and obtain the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

2. The method according to claim 1, characterized in that The determining the minimum value of the second joint dispatch cost function at the current moment, obtaining the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load dispatch strategy of the data center i, includes: Obtaining the total interactive load received by the front-end server connected to the data center i at the current moment, the number of batch computing loads received by the data center i, the electricity price at which the data center i purchases electricity from the power grid, and the batch computing load queue length of the data center i, wherein the total interactive load, the number of batch computing loads, the electricity price, and the batch computing load queue length are parameters included in the second constraint condition; Based on the total interactive load received by the front-end server, the number of batch computing loads received by the data center i, the electricity price at which the data center i purchases electricity from the power grid, and the batch computing load queue length of the data center i, the minimum value of the second joint scheduling cost function at the current moment is determined, and the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i are obtained.

3. The method according to claim 1, characterized in that The method further comprises: According to a plurality of independent dispatching costs of the data center i, the independent dispatching costs of the power grid and the total cost of joint dispatching of the data center i and the power grid, the cost of redistribution of benefits from joint dispatching of the data center i and the cost of redistribution of benefits from joint dispatching of the power grid are determined when the data center i is jointly dispatched with the power grid, so that the cost of redistribution of benefits from joint dispatching of the data center i is less than the independent dispatching cost of the data center i, and the cost of redistribution of benefits from joint dispatching of the power grid is less than the independent dispatching cost of the power grid.

4. The method according to claim 3, characterized in that Determining the cost of redistribution of benefits of the joint dispatch of the data center i and the power grid when the data center i is jointly dispatched with the power grid according to the multiple independent dispatch costs of the data center i, the independent dispatch costs of the power grid, and the total cost of the joint dispatch of the data center i and the power grid, includes: The joint dispatching cost of the data center i and the joint dispatching cost of the power grid are determined according to formula (1): in, is the cost of independent dispatch of the power grid, is the cost of independent scheduling of data center i, C e ′ pn,t is the cost of redistribution of the benefits of the joint dispatch of the power grid, C i ′ dc,i,t The cost of redistribution of the joint scheduling benefit for the data center i, The total cost of the joint dispatch of data center i and the power grid.

5. The method according to claim 1, characterized in that The method further comprises: Sending the batch computing load transfer strategy of the data center i to the data center i; or, The batch processing load transfer strategy of the data center i is sent to the data center batch processing load scheduling device.

6. The method according to claim 1, characterized in that The method further comprises: Determining a power generation strategy of the power grid according to power generation parameters of the power grid required by the plurality of data centers i; The power generation parameters of the power grid required by the data center i are sent to the communication device of the power grid.

7. The method according to claim 1, characterized in that The batch processing computing load transfer strategy of the data center i includes the batch processing computing loads of other data centers received by the data center i and / or the batch processing computing loads transferred by the data center i to other data centers.

8. The method according to claim 2, characterized in that: The method further comprises: The interactive computing load scheduling strategy of the data center i is sent to the front-end server f.

9. The method according to claim 1, characterized in that: The independent scheduling cost function of data center i is the minimum value of the sum of the electricity purchase cost, interactive computing load delay cost and batch computing load discard cost of data center i, and the constraint of the independent scheduling cost function of data center i is the third constraint condition.

10. The method according to claim 1, characterized in that The method further comprises: The computing load queue length of the data center i is updated according to the parameters obtained when determining the minimum value of the second joint scheduling cost function at the current moment.

11. A device for jointly dispatching a data center and a power grid, characterized in that: include: A processing unit, configured to determine a first joint dispatch cost function between a data center i and the power grid, where i is greater than 1, Wherein, the data center i is one of the multiple data centers, the first joint dispatching cost function is the sum of the independent dispatching cost function of the data center i and the independent dispatching cost function of the power grid, and the constraint of the first joint dispatching cost function is the first constraint condition; The processing unit is further used to convert the first joint scheduling cost function and the first constraint according to the Lyapunov optimization method to obtain a second joint scheduling cost function, wherein the constraint of the second joint scheduling cost function is a second constraint condition; The processing unit is also used to determine the minimum value of the second joint scheduling cost function at the current moment, and obtain the power generation parameters of the power grid required by the data center i, the batch computing load transfer strategy of the data center i, and the interactive computing load scheduling strategy of the data center i.

12. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores instructions, and when the processor runs the instructions, the processor executes the method according to any one of claims 1 to 10.

13. A computer storage medium, characterized in that: The method comprises instructions which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 10.

Images