Multi-energy-flow cooperative control method and system for edge data center cluster

Abstract

The invention discloses a multi-energy-flow cooperative control method and system for an edge data center cluster, and relates to the technical field of multi-energy-flow cooperative control, and the method comprises the steps: S1, building a power consumption model of an edge data center server according to the data load type of an edge data center and a data load scheduling method in combination with a server power consumption control strategy; s2, constructing an edge data center cluster water cooling system in a water polymerization mode based on the power consumption model; and S3, constructing a multi-energy flow coupling model based on a load scheduling method and a power consumption model of the data of the edge data center in combination with an edge data center cluster water cooling system in an aggregation mode, and performing multi-energy flow cooperative control of the edge data center cluster. The situation that systems of a traditional data center operate independently is broken through, multi-energy-flow efficient interaction and cooperation are achieved, cluster operation of the whole edge data center is more stable and smoother, the data load response time is greatly shortened, and emission of greenhouse gases such as carbon dioxide is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of multi-energy flow collaborative control, and in particular to a multi-energy flow collaborative control method and system for an edge data center cluster. Background Art

[0002] With its unique geographical location close to the user side, the edge data center can quickly respond to various data needs with ultra-low latency, providing solid computing power support for many scenarios such as real-time interactive applications and local data analysis, facilitating intelligent use in the fields of smart Internet of Things, Industrial Internet, etc.

[0003] With the iteration of information technology and the expansion of business needs, the scale of edge data centers continues to expand. Correspondingly, their energy consumption problems are becoming more severe and the phenomenon of waste heat emissions is becoming increasingly prominent.

[0004] Therefore, the existing technology considers the data center park computing power and power resource optimization planning method guided by the diversified electricity market, characterizes the aggregated adjustable characteristics of the data center park and the methods and benefits of participating in the electricity-frequency-capacity market, or simulates the performance of the waste heat recovery system of the lake water source data center, which provides a theoretical basis for the design and energy-saving operation of the waste heat recovery system of the data center to further solve this problem. However, the existing technology is all aimed at large and medium-sized data centers, and fails to fully consider the unique characteristics of edge data center clusters such as small scale, cluster distribution, and proximity to users, making it difficult to directly transplant and apply the traditional information flow-power flow-thermal flow collaborative control technology suitable for large and medium-sized data centers.

[0005] To this end, this application specifically proposes a multi-energy flow collaborative control method for an edge data center cluster to solve the above technical problems. Summary of the invention

[0006] The main purpose of the present invention is to provide a multi-energy flow collaborative control method and system for an edge data center cluster, which is used to construct an information flow-power flow-thermal flow collaborative mechanism for the edge data center cluster, and provide a practical information flow-power flow-thermal flow collaborative operation solution for the efficient and energy-saving operation of the edge data center cluster, so as to improve the overall stability and sustainability of the edge data center, and solve the technical problems raised in the background technology.

[0007] The present invention adopts the following technical solutions to solve the above technical problems: A multi-energy flow collaborative control method for an edge data center cluster comprises the following steps: S1. Energy consumption modeling of edge data centers based on the flexible time-space scheduling of data loads: According to the data load type and data load scheduling method of the edge data center, the load volume of the data processed in each period is determined, and combined with the server power consumption control strategy, the power consumption modeling of auxiliary equipment such as edge data center servers, refrigeration equipment, and power supply and distribution systems is realized to build a power consumption model of edge data center servers; S2. Modeling of water cooling waste heat recovery mechanism for edge data center clusters in aggregation mode: Based on the power consumption model of edge data center servers, a heat transfer model between the water cooling module and the server and a water distribution network model of the edge data center cluster are constructed to achieve refined simulation of the edge data center cluster water cooling system, establish an edge data center cluster waste heat recovery model under aggregation mode, collect and preliminarily process the low-grade waste heat generated by the edge data center, and form an edge data center cluster water cooling system under aggregation mode; S3. Establish a coordinated control mechanism for information flow, power flow and thermal flow in edge data center clusters: The load scheduling method based on edge data center data and the power consumption model of edge data center servers, combined with the edge data center cluster water cooling system in aggregation mode, is used to build a multi-energy flow coupling model to achieve a refined simulation of the coordinated interaction of information flow, power flow, and thermal flow in the edge data center cluster, and then conduct coordinated control of information flow, power flow, and thermal flow in the edge data center cluster; The multi-energy flow coupling model is in the period The overall energy cost of the edge data center cluster It is expressed as:

[0008] in, Indicates Each edge data center purchases electricity from the mains; Indicates the working power of the heat pump; Indicates The time-of-use electricity price in the area where the edge data center is located; represents the carbon emission factor; represents the carbon emission cost coefficient; Indicates the waste heat recovery benefit coefficient; Indicates the amount of cooling water heat obtained after being heated by a heat pump at the waste heat recovery point; Indicates the total number of time periods; Indicates the number of edge data centers; Indicates the time it takes for edge data centers to converge to a unified waste heat recovery point.

[0009] Preferably, in the step S1, the edge data center is deployed at the access network level, in a data center close to the user side, with a single unit size not exceeding 100 standard racks, and is equipped with lithium iron phosphate energy storage batteries for maintaining its uninterrupted operation. Preferably, in step S1, each edge data center is set as a uniform data center, that is, the technical indicators such as the performance and rated power of the servers deployed in each edge data center are the same. Total energy consumption of edge data centers It can be expressed as the sum of IT equipment power consumption, cooling equipment power consumption and auxiliary equipment power consumption, as shown in the following formula:

[0010] in, Indicates Power consumption of IT equipment in edge data centers; Indicates Power consumption of cooling equipment in edge data centers; Indicates The power consumption of auxiliary equipment such as the power supply and distribution system of an edge data center.

[0011] Preferably, Power consumption of IT equipment in edge data centers For IT equipment, energy consumption modeling can be completed based on the server utilization model combined with parameters such as server peak power and idle power. The calculation formula is:

[0012] in Indicates the first A collection of servers in operation in an edge data center; Indicates A collection of dormant servers in edge data centers; Indicates the static power of the server when the IT equipment in the edge data center is running; Indicates the peak power of the server when the IT equipment in the edge data center is running; Indicates Edge Data Center Real-time data load processing capacity of each server; Indicates the rated data load processing capacity of a single server in an edge data center; Indicates the server standby power of IT equipment in the edge data center in sleep mode.

[0013] Preferably, Power consumption of cooling equipment in edge data centers , based on energy utilization efficiency PUE and combined with IT equipment energy consumption And auxiliary equipment Calculation, since PUE defines the time period is the ratio of the total input energy consumption of the data center to the total energy consumption of IT equipment, which is: .

[0014] Preferably, the data load types of the edge data center in step S1 include online load and offline load, wherein: Online loads include real-time interactive loads; Offline loads include time-shiftable data loads and space-shiftable data loads, which are used for service requests including load data analysis and scientific computing.

[0015] Preferably, the data load scheduling method of the edge data center in step S1 includes: a data load migration method based on spatial flexibility and a data load transfer method based on temporal flexibility.

[0016] Preferably, in step S2, the heat loss coefficient of IT equipment in the edge data center is set If the heat generated by IT equipment is completely recovered by cooling water, then The calculation formula for the heat energy output by the water cooling module of an edge data center is:

[0017] in, , are the specific heat capacity and density of water at normal temperature and pressure respectively; , Respectively represent the first Cooling water flow and inlet water temperature per unit time for an edge data center.

[0018] Preferably, in step S2, the heat loss coefficient of the IT equipment server in the edge data center is set Constant, and the heat generated by the equipment can be completely recovered by cooling water, then the cooling water temperature at the outlet of the water cooling module of each edge data center The calculation formula is:

[0019] in, Indicates the basic value of waste heat recovery temperature; Indicates the specific heat capacity of cooling water at normal temperature and pressure; Indicates Cooling water flow at the outlet of the water-cooling module in an edge data center.

[0020] Preferably, the edge data center cluster water cooling system in the aggregation mode in step S3 is used to transport cooling water to each edge data center node through a preset water distribution network by a water pump to achieve cooling of each edge data center.

[0021] Preferably, the heat transfer model in step S3 is used in the edge data center cluster water cooling system to centrally guide the low-grade waste heat of each edge data center to a unified waste heat recovery point, and use a heat pump to reheat the collected low-grade waste heat.

[0022] Preferably, in the step S3, a heat energy recovery threshold is set during the start-stop control process of the heat pump at the unified waste heat recovery point to determine whether to start the heat pump.

[0023] Preferably, a multi-energy flow collaborative control system of an edge data center cluster is constructed based on any of the multi-energy flow collaborative control methods of an edge data center cluster described above, and is used for collaborative control of information flow, power flow and thermal flow of an edge data center cluster. The system is based on coupling acquisition of the power consumption model of the edge data center cluster water cooling system and the edge data center server in an aggregation mode.

[0024] Preferably, the power consumption model is provided with multiple groups, and through data load coordination, the power consumption model includes: energy storage battery, distributed photovoltaic and edge data center, the edge data center is used for power regulation and acquisition of energy consumption of energy storage battery and distributed photovoltaic The edge data center cluster water cooling system includes: a cooling water source, a water distribution network connected to a power consumption model, and a heat pump for recycling and upgrading waste heat of the water distribution network.

[0025] It can be seen from the above technical solution that the present invention provides a multi-energy flow collaborative control method for edge data center clusters. Compared with the prior art, the present invention has the following advantages: 1. The present invention can ensure efficient and smooth data processing by setting up a collaborative control mechanism between the information flow, power flow and heat flow of the edge data center cluster, thereby greatly shortening the data load response time, providing an excellent user experience for real-time interactive applications, and reducing greenhouse gas emissions such as carbon dioxide.

[0026] 2. The present invention establishes a coordinated control mechanism of information flow, power flow and thermal flow in edge data center clusters, which can facilitate efficient and energy-saving operation of edge data center clusters, improve the overall stability and sustainability of edge data centers, and promote the green and healthy development of the edge computing industry.

[0027] 3. The present invention performs energy consumption modeling by comprehensively considering data load type, scheduling method and server power consumption control strategy, and goes deep into the refined management of each time period and each component of the edge data center, and can dynamically adjust power consumption according to actual data processing needs.

[0028] 4. The present invention solves the problem that the existing data center waste heat recovery methods are difficult to directly transplant and apply to edge data centers by further constructing a water-cooled waste heat recovery mechanism in an aggregation mode for edge data center clusters, thereby realizing the accurate collection and utilization of low-grade waste heat generated by edge data centers and improving the overall energy utilization rate of edge data center clusters.

[0029] 5. The present invention breaks the situation of isolated operation of each system in the traditional data center by constructing an edge data center cluster information flow, power flow and thermal flow collaborative control mechanism, realizes efficient interaction and collaboration among the three, and makes the operation of the entire edge data center cluster more stable and smooth.

[0030] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become easy to understand through the following description. Of course, it is not necessary to achieve all of the advantages described above simultaneously for any product implementing the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The drawings constituting a part of the present application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings: Figure 1 It is a schematic diagram of the overall process of an embodiment of the present invention; Figure 2 A schematic diagram of the coupling system structure framework according to an embodiment of the present invention; Figure 3 Schematic diagram of simulation experiment results of an embodiment of the present invention. DETAILED DESCRIPTION

[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. In the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0033] In the embodiment, see Figures 1 to 3 .

[0034] like Figure 1 As shown, a multi-energy flow collaborative control method of an edge data center cluster proposed in an embodiment of the present invention mainly focuses on information flow, power flow and thermal flow in multi-energy flow, including: S1. Energy consumption modeling of edge data centers based on the flexible time-space scheduling of data loads: According to the data load type and data load scheduling method of the edge data center, the load volume of the data processed in each time period is determined, and combined with the server power consumption control strategy, the power consumption modeling of auxiliary equipment such as edge data center servers, refrigeration equipment, and power supply and distribution systems is realized to build a power consumption model of edge data center servers.

[0035] The edge data center is deployed at the access network level, close to the user side. The single data center does not exceed 100 standard racks and is equipped with lithium iron phosphate energy storage batteries to maintain its uninterrupted operation.

[0036] Specifically, each edge data center is set as a uniform data center, that is, the technical indicators such as server performance and rated power deployed in each edge data center are the same. Total energy consumption of edge data centers It can be expressed as the sum of IT equipment power consumption, cooling equipment power consumption and auxiliary equipment power consumption, as shown in the following formula:

[0037] in, Indicates Power consumption of IT equipment in edge data centers; Indicates Power consumption of cooling equipment in edge data centers; Indicates The power consumption of auxiliary equipment such as the power supply and distribution system of an edge data center.

[0038] No. Power consumption of IT equipment in edge data centers For IT equipment, energy consumption modeling can be completed based on the server utilization model combined with parameters such as server peak power and idle power. The calculation formula is:

[0039] in Indicates the first A collection of servers in operation in an edge data center; Indicates A collection of dormant servers in edge data centers; Indicates the static power of the server when the IT equipment in the edge data center is running; Indicates the peak power of the server when the IT equipment in the edge data center is running; Indicates Edge Data Center Real-time data load processing capacity of each server; Indicates the rated data load processing capacity of a single server in an edge data center; Indicates the server standby power when the IT equipment in the edge data center is in sleep mode; At this time, for the Power consumption of cooling equipment in edge data centers , based on energy utilization efficiency PUE and combined with IT equipment energy consumption And auxiliary equipment Calculation, since PUE defines the time period is the ratio of the total input energy consumption of the data center to the total energy consumption of IT equipment, which is:

[0040] In addition, it can be added that the data load types of edge data centers include online loads and offline loads. Among them: online loads include real-time interactive loads; offline loads include time-transferable data loads and spatially transferable data loads, which are used for service requests including load data analysis and scientific computing. At this time, the data load scheduling methods of edge data centers include: migration data load methods based on spatial flexibility and transfer data load methods based on time flexibility.

[0041] Therefore, based on the above edge data center energy consumption modeling analysis, according to the edge data center cluster data load scheduling technology, that is, the migration data load technology based on spatial flexibility and the transfer data load technology based on time flexibility, the time period can be defined First in the interior Real-time data load processing capacity of edge data centers As shown below:

[0042] in, Indicates time period Neidi The online load arrival of each edge data center; Indicates time period Neidi The edge data center can time-shift the data load arrival; Indicates time period Neidi The amount of data load that can be spatially transferred by edge data centers; Indicates time period Neidi The edge data center The amount of load moved into the edge data center; It indicates the time period Neidi The edge data center The amount of load moved out of edge data centers; express Edge data center slots The amount of load shifted to future time periods; express Edge data centers past period to period The amount of load moved in.

[0043] Among them, the period Neidi Real-time data load processing capacity of edge data centers It can be further combined with the data load processing capacity of the edge data center server to express as:

[0044] In addition, the above-mentioned migration data load technology based on spatial flexibility and the transfer data load technology based on time flexibility should satisfy the following constraint formula conditions:

[0045] Among them, c1 represents the time period First Edge Data Center The real-time data load processing capacity of the server in the running state is subject to the rated data load processing capacity as the upper limit; c2 represents the time period No. The edge data center The load volume removed from each edge data center is capped by the amount of data load that can be transferred spatially in the current period; c3 represents the period No. The load amount that the edge data center moves out to the future period is capped by the amount of data load that can be transferred in the current period; the data load transfer flag matrix between edge data centers is constructed in c4 ,but Indicates time period The first The edge data center The edge data center data load transfer flag is 1 when data load transfer occurs, otherwise it is 0. This constraint indicates that if the edge data center receives data load transferred from other edge data centers, it cannot transfer data load to other edge data centers. c5 represents the time period First in the interior Average queuing delay of edge data center data load , average processing delay , communication delay and transmission delay The sum of the maximum tolerable delay time in the current period is the upper limit.

[0046] Furthermore, in the c5 constraint content, Average queuing delay of edge data center data load , average processing delay , communication delay and transmission delay The specific definitions are as follows: (1) Calculate the average queuing delay of data load based on M / M / 1 queuing theory ,have:

[0047] (2) Calculate the average processing delay of the data load based on the inverse of the average service rate of the data center ,have:

[0048] (3) The communication delay can be calculated based on the data network branch communication delay, bandwidth and transmitted data load and transmission delay ,have:

[0049] in Indicates the data network branch number and corresponding set; Indicates time period Whether the internal data load passes through the branch Communication transmission flag, if it flows through the branch, it is 1, otherwise it is 0; , Represents data network branches The corresponding communication delay and bandwidth; Indicates time period Inbound data network branch The amount of data payload transferred.

[0050] In summary, by comprehensively considering the data load type, scheduling method and server power consumption control strategy to carry out energy consumption modeling, we can carry out refined management of each time period and each component of the edge data center, and dynamically adjust the power consumption according to the actual data processing needs.

[0051] S2. Modeling of water cooling waste heat recovery mechanism for edge data center clusters in aggregation mode: Based on the power consumption model of edge data center servers, a heat transfer model between the water cooling module and the server and a water distribution network model of the edge data center cluster are constructed to achieve refined simulation of the edge data center cluster water cooling system, establish an edge data center cluster waste heat recovery model under aggregation mode, collect and preliminarily process the low-grade waste heat generated by the edge data center, and form an edge data center cluster water cooling system under aggregation mode; In the specific use process, the edge data center cluster water cooling waste heat recovery system constructed by this application is as follows: Figure 2 As shown in the figure, cooling water is pumped to each edge data center node through a preset water network. At each node, advanced water cooling technology is used to meet the cooling needs of its server IT equipment, ensuring the stable operation of the edge data center while collecting and preliminarily processing the low-grade waste heat generated. Subsequently, the preliminarily processed low-grade waste heat will be centrally guided to a unified waste heat recovery point, and the collected low-grade waste heat will be upgraded using heat pump technology to convert it into high-grade thermal energy that can meet the heating needs of surrounding communities or industrial facilities, thereby realizing the cascade utilization of waste heat resources.

[0052] Therefore, by further constructing a water-cooled waste heat recovery mechanism under an aggregated mode for edge data center clusters, the problem that the existing data center waste heat recovery methods are difficult to directly transplant and apply to edge data centers is solved, and the accurate collection and utilization of low-grade waste heat generated by edge data centers is achieved, thereby improving the overall energy utilization rate of edge data center clusters.

[0053] Specifically, heat is transferred between the water cooling module and the server's heating components by conduction, so the heat loss coefficient of IT equipment in the edge data center is set If the heat generated by IT equipment is completely recovered by cooling water, then The calculation formula for the heat energy output by the water cooling module of an edge data center is:

[0054] in, , are the specific heat capacity and density of water at normal temperature and pressure respectively; , Respectively represent the first Cooling water flow and inlet water temperature per unit time for an edge data center.

[0055] At this time, set the heat loss coefficient of the IT equipment server in the edge data center Constant, and the heat generated by the equipment can be completely recovered by cooling water, then the cooling water temperature at the outlet of the water cooling module of each edge data center The calculation formula is:

[0056] in, Indicates the basic value of waste heat recovery temperature; Indicates the specific heat capacity of cooling water at normal temperature and pressure; Indicates Cooling water flow at the outlet of the water-cooling module in an edge data center.

[0057] Then, the cooling water of each edge data center is gathered to a unified waste heat recovery point and heated again by the heat pump. The calculation of the recoverable heat after gathering is as follows:

[0058] in, It represents the initial heat collected by each edge data center when it converges to a unified waste heat recovery point; Indicates the heat dissipation coefficient in the convergence process caused by the different distances between each edge data center and the unified waste heat recovery point; Indicates the amount of cooling water heat obtained after being heated by a heat pump at the waste heat recovery point; Indicates the working power of the heat pump; represents the coefficient of performance of the heat pump; Indicates the time it takes for edge data centers to converge to a unified waste heat recovery point.

[0059] In addition, the water cooling waste heat recovery of edge data center clusters in the aggregation mode should meet the following constraint formula conditions:

[0060] Among them, c6 represents the flow balance constraint of the water distribution network node; c7 represents the working power constraint of the water pump; c8 and c9 are the Constraints on water inflow and outflow for edge data centers; Indicates time period Neidi Unit flow rate of cooling water consumed by water cooling modules in edge data centers; is the rated operating power of the heat pump; , Respectively represent the first The maximum water inflow and outflow flow of an edge data center.

[0061] S3. Establish a coordinated control mechanism for information flow, power flow and thermal flow in edge data center clusters: The load scheduling method based on edge data center data and the power consumption model of edge data center servers, combined with the edge data center cluster water cooling system in aggregation mode, are used to construct a multi-energy flow coupling model to achieve refined simulation of the coordinated interaction of information flow, power flow, and thermal flow in the edge data center cluster, and then conduct coordinated control of information flow, power flow, and thermal flow in the edge data center cluster.

[0062] Based on the energy consumption modeling of the edge data center in S1 above, the spatiotemporal redistribution of the edge data center cluster load can be achieved through the migration data load technology based on spatial flexibility and the transfer data load technology based on temporal flexibility.

[0063] Among them, the water-cooled waste heat recovery mechanism of the edge data center cluster under the aggregation mode in S2 above can realize the efficient recovery and cascade utilization of waste heat resources of the edge data center cluster with small scale, cluster distribution and close to users.

[0064] Furthermore, in order to ensure uninterrupted power supply to the edge data center in the event of a power outage in the power grid and to guarantee the quality of user service, the edge data center is equipped with energy storage batteries (here in the present invention, it refers to lithium iron phosphate energy storage batteries) as emergency power supplies. Therefore, based on the flexible charging and discharging characteristics of the energy storage batteries, the coordinated control of the information flow, power flow, and thermal flow of the edge data center cluster can be achieved.

[0065] Furthermore, the present invention application can reduce the electricity expenditure cost of edge data centers and effectively reduce carbon emissions in the operation process. Some edge data centers are equipped with distributed photovoltaic equipment to provide green energy supply for the data centers.

[0066] In summary, the power supply composition of the edge data center at this time is expressed as:

[0067] in, , Respectively indicate time periods Neidi The charging and discharging power of the energy storage batteries equipped in each edge data center; Indicates time period Neidi Output power of distributed photovoltaic equipment in each edge data center; Indicates time period Neidi Each edge data center purchases electricity from the mains.

[0068] Therefore, the overall energy cost of the edge data center cluster can be described as follows:

[0069] in, The overall energy cost of the edge data center cluster; Indicates time period Neidi Each edge data center purchases electricity from the mains; Indicates time period Neidi The time-of-use electricity price in the area where the edge data center is located; represents the carbon emission factor; represents the carbon emission cost coefficient; Indicates the waste heat recovery benefit coefficient; Indicates time period The heat of cooling water obtained after being heated by a heat pump at the internal waste heat recovery point; Indicates the total number of time periods; Indicates the number of edge data centers in the edge data center cluster; Indicates the time it takes for edge data centers to converge to a unified waste heat recovery point.

[0070] In addition, for the energy storage batteries equipped in the edge data center, the following constraint formula conditions should be met:

[0071] Among them, c10 and c11 represent the energy storage battery charging and discharging power constraints respectively; c12 represents the energy storage battery capacity constraint; c13 represents the energy storage battery charging and discharging state constraint; c14 represents the energy storage battery charging and discharging times constraint; , Respectively indicate time periods Neidi The charging and discharging power of the energy storage batteries equipped in each edge data center; , , , Respectively indicate time periods Neidi The minimum and maximum charging and discharging power of the energy storage batteries equipped in each edge data center; Indicates The maximum capacity of the energy storage batteries equipped in each edge data center; , Indicates The number of charge and discharge changes and the maximum number of changes of the energy storage batteries equipped in each edge data center during the total period.

[0072] Among them, The calculation formula for the number of charge and discharge changes of the energy storage batteries equipped in each edge data center during the total period is as follows:

[0073]

[0074] in, Indicates the first The equivalent conversion of the charging and discharging power of the energy storage battery equipped in each edge data center. A value greater than 0 indicates that the energy storage battery is in a charging state, and a value less than 0 indicates that the energy storage battery is in a discharging state; represents an infinite positive integer; Indicates The charging and discharging status switching flag of the energy storage battery equipped in each edge data center.

[0075] In the specific implementation process, the edge data center cluster water cooling system in the aggregation mode is used to transport cooling water to each edge data center node through a preset water distribution network through a water pump to achieve cooling of each edge data center.

[0076] At this time, the heat transfer model is used in the edge data center cluster water cooling system to centrally guide the low-grade waste heat of each edge data center to a unified waste heat recovery point, and use a heat pump to reheat the collected low-grade waste heat.

[0077] Specifically, a heat recovery threshold is set during the start-stop control process of the heat pump at the unified waste heat recovery point to determine whether to start the heat pump, which can be specifically expressed as the following condition:

[0078] in, It represents the initial heat collected by each edge data center when it converges to a unified waste heat recovery point; Indicates the heat recovery threshold; Indicates the start and stop flag of the heat pump at the unified waste heat recovery point. If enabled, it is 1, otherwise it is 0.

[0079] In summary, based on the above S1, S2, and S3, the coordinated control of information flow, power flow, and thermal flow of the edge data center cluster can be achieved. The spatial flexibility-based migration data load technology and the temporal flexibility-based transfer data load technology can be used to realize the spatiotemporal redistribution of the edge data center cluster load. The flexible charging and discharging characteristics of the edge data center energy storage battery can be used to enhance the energy flexibility of the edge data center. At the same time, the water-cooled waste heat recovery mechanism of the edge data center cluster under the aggregation mode can realize the efficient recovery and cascade utilization of the waste heat resources of the edge data center cluster.

[0080] Therefore, by establishing a coordinated control mechanism for information flow, power flow and thermal flow in edge data center clusters, it is possible to facilitate efficient and energy-saving operation of edge data center clusters, improve the overall stability and sustainability of edge data centers, and promote the green and healthy development of the edge computing industry. At this time, coordinated control of information flow, power flow and thermal flow is carried out for edge data center clusters to ensure efficient and smooth data processing, greatly shorten the data load response time, and provide an excellent user experience for real-time interactive applications. It also helps edge data centers improve the efficiency of the use of power and computing resources, and effectively reduce greenhouse gas emissions such as carbon dioxide. In addition, it can inject new impetus into the economic feasibility of the edge data center industry through waste heat recovery and reuse.

[0081] In addition, it should be noted that by building an edge data center cluster information flow-power flow-heat flow collaborative control mechanism, the situation of isolated operation of each system in the traditional data center has been broken, and efficient interaction and collaboration among the three have been achieved, making the operation of the entire edge data center cluster more stable and smooth.

[0082] On the other hand, reference Figure 2 The present invention also discloses a multi-energy flow collaborative control system for an edge data center cluster, which is constructed based on any of the above-mentioned multi-energy flow collaborative control methods for an edge data center cluster and is used for collaborative control of information flow, power flow and thermal flow in an edge data center cluster. The system is based on coupling acquisition of the power consumption model of the edge data center cluster water cooling system and the edge data center server in an aggregation mode.

[0083] There are multiple sets of power consumption models, and through data load coordination, the power consumption models include: energy storage batteries, distributed photovoltaics and edge data centers. The edge data center is used for power regulation and obtains the energy consumption of energy storage batteries and distributed photovoltaics. The edge data center cluster water cooling system includes: a cooling water source, a water distribution network connected to the power consumption model, and a heat pump for waste heat recovery and upgrading of the water distribution network.

[0084] Based on the above method and the constructed system, the effectiveness of the proposed information flow-power flow-thermal flow coordination mechanism of the edge data center cluster is further verified. Therefore, the following edge data center cluster joint scheduling scenario is built: The Gurobi solver is used based on the MATLAB R2020a simulation platform for problem solving and simulation verification.

[0085] In this simulation experiment scenario, it is assumed that three edge data centers are deployed within an area of ​​30 square kilometers and their geographical distribution follows Poisson distribution. The time-of-use electricity price in the region is determined according to the specified value of City A in a specified time period. The key technical parameter information of the edge data center refers to the setting of the prefabricated modular data center in B; the data load situation in each time period refers to the real data set of the cloud computer cluster in C, and the distributed photovoltaic output information refers to the real data of City A on a certain day, which is: refer to Figure 3 Taking the No. 1 edge data center in the edge data center cluster as an example, the simulation analysis of the information flow-power flow-thermal flow coordination mechanism of the edge data center cluster proposed by the present invention is as follows: On the premise of ensuring that the user service quality is not affected, based on the information flow, power flow and heat flow coordination mechanism of the edge data center cluster, in the time dimension, the internal data load of the edge data center can be migrated to the period when the distributed photovoltaic output reaches its peak (11:00-14:00) for processing, thereby making full use of the efficient period of solar power generation and avoiding local server overload problems, reducing dependence on traditional energy; in the spatial dimension, data load redistribution between different edge data centers can be achieved, thereby comprehensively improving the overall energy utilization efficiency and computing utility of the edge data center cluster.

[0086] Based on the flexible charging and discharging characteristics of the energy storage batteries equipped in the edge data center, when the city power supply is sufficient and the electricity price is at a low period (23:00-7:00 the next day), the energy storage battery is in the charging period; when the city power fluctuates, fails, or the electricity price is at a peak period (10:00-13:00, 17:00-22:00), the energy storage battery switches to the discharge period to ensure stable power supply for various devices in the edge data center. At the same time, it can be seen that when the edge data center cannot fully absorb the distributed photovoltaic output period (11:00-14:00), the edge data center increases the data load processing share of this period through data load time-space scheduling, and uses the energy storage battery charging operation to absorb excess power.

[0087] Based on the water-cooled waste heat recovery mechanism of the edge data center cluster in the aggregation mode, the small-scale and low-grade waste heat of each edge data center node can be effectively utilized. The waste heat resources that each node can utilize are linearly related to its computing power load curve. This part of the simulation analysis will not be repeated.

[0088] Based on the above simulation analysis results, the daily operating cost of each edge data center can be calculated. After considering the information flow-power flow-heat flow coordination mechanism of the edge data center cluster proposed in this application, the daily operating cost of the edge data center cluster in the non-cooperative scenario is significantly reduced, thereby achieving the overall daily operating cost reduction of the edge data center cluster.

[0089] In another embodiment provided in the present application, a computer program product including instructions is also provided, which, when executed on a computer, enables the computer to execute the multi-energy flow collaborative control method of any edge data center cluster in the above-mentioned embodiments.

[0090] It is understandable that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and the explanation, examples and beneficial effects of the relevant contents can refer to the corresponding parts in the above method.

[0091] The embodiment of the present application also provides an electronic device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus. Memory, used to store computer programs; The processor is used to implement the multi-energy flow collaborative control method of the above-mentioned edge data center cluster when executing the program stored in the memory.

[0092] The communication bus mentioned in the above electronic device can be a peripheral component interconnect standard bus or an extended industrial standard architecture bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc.

[0093] The communication interface is used for communication between the above electronic device and other devices.

[0094] The memory may include a random access memory, or may include a non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0095] The above-mentioned processor can be a general-purpose processor, including a central processing unit, a network processor, etc.; it can also be a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component.

[0096] It should also be noted that electronic devices also include terminal devices, which can also be called terminals, user equipment, mobile stations, mobile terminals, etc. Terminal devices can be mobile phones, smart TVs, wearable devices, tablet computers, computers with wireless transceiver functions, virtual reality terminal devices, augmented reality terminal devices, wireless terminals in industrial control, wireless terminals in unmanned driving, wireless terminals in remote surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, etc. The embodiments of this application do not limit the specific technology and specific device form used by the terminal devices.

[0097] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. 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 described in 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 one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. 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 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 DVD), or a semiconductor medium (e.g., a solid-state hard disk), etc.

[0098] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

[0099] In addition, it should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back...), the directional indications are only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly.

[0100] In addition, if there are descriptions involving "first", "second", etc. in the embodiments of the present invention, the descriptions of "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, the meaning of "and / or" appearing in the full text includes three parallel schemes. Taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in the field to implement. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

Claims

1. A multi-energy flow collaborative control method for edge data center clusters, characterized in that: include: S1. Build a power consumption model of edge data center servers based on the data load type and data load scheduling method of the edge data center and the server power consumption control strategy; S2. Build a heat transfer model and a water distribution network model based on the power consumption model to form an edge data center cluster water cooling system in an aggregation mode; S3. Load scheduling method and power consumption model based on edge data center data, combined with edge data center cluster water cooling system in aggregation mode, is used to build a multi-energy flow coupling model and conduct coordinated control of information flow, power flow and thermal flow of edge data center cluster. The overall energy cost of the edge data center cluster It is expressed as: Indicates Each edge data center purchases electricity from the mains; Indicates the working power of the heat pump; Indicates The time-of-use electricity price in the area where the edge data center is located; represents the carbon emission factor; represents the carbon emission cost coefficient; Indicates the waste heat recovery benefit coefficient; Indicates the amount of cooling water heat obtained after being heated by a heat pump at the waste heat recovery point; Indicates the total number of time periods; Indicates the number of edge data centers; Indicates the time it takes for edge data centers to converge to a unified waste heat recovery point.

2. The multi-energy flow collaborative control method of the edge data center cluster as claimed in claim 1, characterized in that: In the S1 step, the edge data center is deployed at the access network level, close to the user side of the data center, with a single unit size not exceeding 100 standard racks, and is equipped with lithium iron phosphate energy storage batteries to maintain its uninterrupted operation.

3. The multi-energy flow collaborative control method of the edge data center cluster as claimed in claim 1, characterized in that: The data load types of the edge data center in step S1 include online load and offline load, where: Online loads include real-time interactive loads; Offline loads include time-shiftable data loads and space-shiftable data loads, which are used for service requests including load data analysis and scientific computing.

4. The multi-energy flow collaborative control method of the edge data center cluster as claimed in claim 1, characterized in that: The data load scheduling method of the edge data center in the step S1 includes: a data load migration method based on spatial flexibility and a data load transfer method based on temporal flexibility.

5. The multi-energy flow collaborative control method of the edge data center cluster as claimed in claim 1, characterized in that: In the step S2, a waste heat recovery model for edge data center clusters in an aggregation mode is also established to collect and preliminarily process low-grade waste heat generated by edge data centers.

6. The multi-energy flow collaborative control method of the edge data center cluster as claimed in claim 1, characterized in that: The edge data center cluster water cooling system in the aggregation mode in step S3 is used to transport cooling water to each edge data center node through a preset water distribution network by a water pump to achieve cooling of each edge data center.

7. The multi-energy flow collaborative control method of the edge data center cluster as claimed in claim 6, characterized in that: The heat transfer model in step S3 is used in the edge data center cluster water cooling system to centrally guide the low-grade waste heat of each edge data center to a unified waste heat recovery point, and use a heat pump to reheat the collected low-grade waste heat.

8. The multi-energy flow collaborative control method of the edge data center cluster as claimed in claim 7, characterized in that: In the step S3, a heat energy recovery threshold is set during the start-stop control process of the heat pump at the unified waste heat recovery point to determine whether to start the heat pump.

9. A multi-energy flow collaborative control system for an edge data center cluster, constructed based on the multi-energy flow collaborative control method for an edge data center cluster described in any one of claims 1 to 8, for collaborative control of information flow, power flow, and thermal flow in an edge data center cluster, characterized in that: The system is based on coupling acquisition of power consumption models of edge data center cluster water cooling system and edge data center server in aggregation mode.

10. The multi-energy flow collaborative control system of the edge data center cluster as claimed in claim 9, characterized in that: The power consumption model is configured with multiple groups and coordinated through data load. The power consumption model includes: energy storage battery, distributed photovoltaic and edge data center. The edge data center is used for power regulation and acquisition of energy consumption of energy storage battery and distributed photovoltaic. The edge data center cluster water cooling system includes: a cooling water source, a water distribution network connected to a power consumption model, and a heat pump for recycling and upgrading waste heat of the water distribution network.

Images