CN112994037A - Method, system, medium and device for adjusting power consumption of data center in smart grid environment - Google Patents

Method, system, medium and device for adjusting power consumption of data center in smart grid environment Download PDF

Info

Publication number
CN112994037A
CN112994037A CN202110141041.2A CN202110141041A CN112994037A CN 112994037 A CN112994037 A CN 112994037A CN 202110141041 A CN202110141041 A CN 202110141041A CN 112994037 A CN112994037 A CN 112994037A
Authority
CN
China
Prior art keywords
power
power consumption
data center
cost
task
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110141041.2A
Other languages
Chinese (zh)
Other versions
CN112994037B (en
Inventor
赵梦梦
王晓英
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Qinghai University
Original Assignee
Qinghai University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Qinghai University filed Critical Qinghai University
Priority to CN202110141041.2A priority Critical patent/CN112994037B/en
Publication of CN112994037A publication Critical patent/CN112994037A/en
Application granted granted Critical
Publication of CN112994037B publication Critical patent/CN112994037B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/24—Arrangements for preventing or reducing oscillations of power in networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/04—Circuit arrangements for ac mains or ac distribution networks for connecting networks of the same frequency but supplied from different sources
    • H02J3/06—Controlling transfer of power between connected networks; Controlling sharing of load between connected networks
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/12—Circuit arrangements for ac mains or ac distribution networks for adjusting voltage in ac networks by changing a characteristic of the network load
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/10—Power transmission or distribution systems management focussing at grid-level, e.g. load flow analysis, node profile computation, meshed network optimisation, active network management or spinning reserve management
    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management

Landscapes

  • Engineering & Computer Science (AREA)
  • Power Engineering (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

The invention relates to a method, a system, a storage medium and equipment for adjusting power consumption of a data center in a smart grid environment, wherein the method comprises the following steps: determining the current total power of the data center; determining power consumption adjusting means, and establishing a power adjusting model according to the target power, the current total power and the power to be adjusted of each power adjusting means; determining the operation cost of the cooling system, the operation cost of the energy storage equipment, the adjustment deviation penalty cost and the task delay penalty cost, and normalizing the operation cost and the penalty cost into a total cost; and determining the optimal combination mode of various power consumption adjusting means by using a dynamic optimal scheduling method, realizing accurate regulation and control of the power consumption of the data center and minimizing the total cost. The data center is introduced into the smart power grid as a load of demand response, the self power of the data center is adjusted by various means to participate in the power system energy market, the voltage stabilization effect is achieved on the smart power grid, and meanwhile the voltage stabilization cost of the smart power grid can be greatly reduced.

Description

Method, system, medium and device for adjusting power consumption of data center in smart grid environment
Technical Field
The invention relates to the technical field of power consumption adjustment, in particular to a method, a system, a medium and equipment for adjusting power consumption of a data center in a smart grid environment.
Background
The cloud computing data center generally has hundreds to hundreds of thousands of servers and has the characteristics of large resource amount, strong heterogeneity, high energy consumption, wide distribution and flexible transfer. Data-related Data indicate that the power consumption of Data Centers (DC) is increasing at a rate of doubling every 5 years, and has become a genuine "power consumer".
With the development and utilization of renewable energy sources, the construction of smart power grids is rapidly developed. Due to the intermittent nature of new energy sources (wind, solar, tidal, etc.), the instability of power generation can cause a large impact on the grid structure, and even a mature grid can only accommodate 19% of the load from renewable energy sources. Therefore, a fast peak shaving power supply needs to be added to ensure the safety of the smart grid.
In order to maintain a stable operation of the power grid, passive regulation is generally used in conventional power grids, in which the supply of electrical energy varies as a function of the electrical energy demand. However, the power system is difficult to solve the problem of real-time supply and demand balance after the large-scale fluctuating new energy power generation is accessed only by the adjusting capability of the power generation side.
Disclosure of Invention
The invention aims to solve the technical problem existing in the prior art, and provides a method for adjusting the power consumption of a data center in an intelligent power grid environment, which comprises the following steps:
determining the current total power P of the data center according to the overall power consumption characteristics of the data centerdc；
Determining at least one power consumption adjusting means, and giving a target power P according to the response requirement of the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means;
determining an operating cost Ex of a data center cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskThe operating cost Ex of the cooling systemairThe operation cost OpEx and the regulation deviation penalty cost pe of the energy storage equipmentinaAnd a task delay penalty cost petaskNormalizing to the total Cost of data center power consumption adjustment;
on the premise of meeting the self constraint conditions of each power adjusting means, determining the optimal combination mode of various power consumption adjusting means by using a dynamic optimal scheduling method, adjusting the power consumption of the data center by using the optimal combination mode, and minimizing the total Cost of the power consumption adjustment of the data center.
In order to solve the above technical problem, an embodiment of the present invention provides a system for adjusting power consumption of a data center in a smart grid environment, including:
a total current power determining module for determining the total current power P of the data center according to the overall power consumption characteristics of the data centerdc；
The power regulation model establishing module is used for determining at least one power consumption regulation means and giving a target power P according to the response requirement of the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means;
a total cost determination module for determining an operating cost Ex of the data center cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskThe operating cost Ex of the cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskNormalizing to the total Cost of data center power consumption adjustment;
and the optimal combination determining module is used for determining the optimal combination mode of various power consumption adjusting means by using a dynamic optimal scheduling method on the premise of meeting the self constraint conditions of each power adjusting means, adjusting the power consumption of the data center by using the optimal combination mode and minimizing the total Cost of the power consumption adjustment of the data center.
In order to solve the technical problem, an embodiment of the present invention provides a computer-readable storage medium, which includes instructions, and when the instructions are executed on a computer, the instructions cause the computer to execute the method for adjusting power consumption of a data center in a smart grid environment according to the technical solution described above.
In order to solve the foregoing technical problem, an embodiment of the present invention provides a computer device, which includes a memory, a processor, and a computer program that is stored in the memory and is executable on the processor, where the processor implements the method for adjusting power consumption of a data center in a smart grid environment according to the foregoing technical solution when executing the computer program.
The invention has the beneficial effects that: the data center is introduced into the smart grid as a load of demand response, the self power of the data center is adjusted through one or more means to carry out demand response, and the data center participates in the energy market of the power system, so that the voltage stabilization effect on the smart grid is achieved, and meanwhile, the voltage stabilization cost of the smart grid can be greatly reduced; and in the adjusting process, the optimal combination mode of various power consumption adjusting means is determined by a dynamic optimal scheduling method, so that the power consumption of the data center is accurately regulated and controlled, and the total adjusting cost is minimized.
Additional aspects of the invention and its advantages will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
Fig. 1 is a flowchart of a method for adjusting power consumption of a data center in a smart grid environment according to an embodiment of the present invention;
FIGS. 2(a) -2(c) show the adjustment targets for three experimental cases, respectively;
fig. 3 is a block diagram illustrating power consumption adjustment of a data center in a smart grid environment according to an embodiment of the present invention.
Detailed Description
The principles and features of this invention are described below in conjunction with the following drawings, which are set forth by way of illustration only and are not intended to limit the scope of the invention.
Fig. 1 is a flowchart of a method for adjusting power consumption of a data center in a smart grid environment according to an embodiment of the present invention. As shown in fig. 1, the method includes:
s101, determining the current total power P of the data center according to the overall power consumption characteristics of the data centerdc；
S102, determining at least one power consumption adjusting means, and giving out target power P according to the response requirement of the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means;
s103, determining the operation cost Ex of the cooling system of the data centerairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskThe operating cost Ex of the cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskNormalizing to the total Cost of data center power consumption adjustment;
and S104, determining the optimal combination mode of the multiple power consumption adjusting means by using a dynamic optimal scheduling method on the premise of meeting the self constraint conditions of the power adjusting means, adjusting the power consumption of the data center by using the optimal combination mode, and minimizing the total Cost of the power consumption adjustment of the data center.
In the embodiment, the data center is introduced into the smart grid as a load of demand response, the self power of the data center is adjusted through one or more means to carry out demand response, and the data center participates in the energy market of the power system, so that the voltage stabilization effect on the smart grid is achieved, and meanwhile, the voltage stabilization cost of the smart grid can be greatly reduced; and in the adjusting process, the optimal combination mode of various power consumption adjusting means is determined by a dynamic optimal scheduling method, so that the power consumption of the data center is accurately regulated and controlled, and the total adjusting cost is minimized.
Optionally, the total power model of the data center is established according to the overall power consumption characteristics of the data center, as follows:
Pdc＝Ps+Pc (1)
in the formula, PdcRepresenting the total current power of the DC, PsRepresenting the real-time power consumption, P, of the servercRepresenting the real-time power consumption of the cooling system.
The real-time power consumption model of the server is as follows:
Ps＝α·ucpu+β (2)
in the formula, alpha represents the difference value between the peak power and the idle power of the server, beta represents the idle power of the server, and u represents the idle power of the servercpuRepresenting the current CPU utilization.
The energy consumption of a cooling device is related to its coefficient of Performance, CoP (CoP), where CoP is defined as the ratio of the amount of cooling provided by a cooling system to the amount of power consumed for cooling, and during a period of time in which a server is operating, the water-cooled CoP can be expressed as:
Figure BDA0002928595400000051
where CoP is the ratio of the amount of cooling provided by the cooling system to the amount of power consumed to perform the cooling. The CoP value is not constant and, as can be seen from the model given by hewlett-packard of the water-cooled CoP with the air-conditioning supply temperature of the cooling system, the CoP generally increases with increasing temperature of the air supplied:
Cop＝0.0068Tsup 2+0.0008Tsup+0.458 (4)
where Tsup represents the cooling temperature provided by the cooling system.
In the above embodiment, the real-time power consumption of the server is calculated according to the current utilization rate of the CPU, the idle power of the server, and the peak power, the real-time power consumption of the cooling system is calculated according to the relationship between the energy consumption of the cooling device and the coefficient of performance thereof, and the total power consumption of the data center is calculated according to a specific calculation mode in which the energy consumption of the data center is mainly concentrated on the energy consumption of the IT device itself and the power consumption of the cooling system. Compared with the method only considering the power of the server, the method and the system are combined, so that the evaluation of the power consumption of the data center is more practical, and the potential of power consumption management of different facilities of the data center can be fully exerted.
Optionally, the target power P is given according to the response demand of the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means as follows;
Paim＝Ps+Pc-Pz-ΔPdif (5)
in the formula, PsRepresenting the real-time power consumption, P, of the servercRepresenting the real-time power consumption of the cooling system; pzIs Pair、Pfan、Psc、Pfb、PtdOr PdvfsIs one or the sum of any of a plurality of terms, Pair、PfanRespectively showing the adjusting power to be obtained by the air conditioner cooling power consumption adjusting means and the natural air cooling power consumption adjusting means, Psc、PfbRespectively representing the regulated power to be obtained by the super capacitor power consumption regulating means and the flow battery power consumption regulating means in the energy storage equipment, Ptd、PdvfsRespectively representing the regulated power to be obtained by the task delay power consumption regulating means and the dynamic voltage frequency power consumption regulating means; Δ pdifA deviation value of the power after the power adjustment from the target power is indicated.
Of the above-mentioned powers to be balanced, Psc、PfbAnd respectively representing the regulated power to be obtained by the super capacitor power consumption regulating means and the flow battery power consumption regulating means in the energy storage device.
The flow battery has high energy conversion efficiency and high starting speed, and can ignore self-discharge rate as energy storage equipment. However, Super Capacitors (SC) are not suitable for long-term storage of electrical energy due to a large self-discharge rate, but in combination with Flow Batteries (FB) this disadvantage is avoided. Meanwhile, the SC and the FB are used in a combined mode, so that the effect of the uninterrupted power supply is fully exerted, the speed of self power regulation of the data center is increased, and the demand response is quicker and more accurate. The self-discharge rate and the energy conversion efficiency of the ESD are comprehensively considered, and the conversion relation between the device capacity and the energy needing to be stored is obtained:
Devre＝ESDre÷(ηESD×(1-γESD)) (6)
in the formula, DevreRepresenting a device capacity of the ESD; ESD (electro-static discharge)reIndicating the energy that the energy storage device needs to store; etaESD、γESDConversion efficiency and self-healing for ESDThe discharge rate.
During discharge, to ensure that the data center can operate properly when power is off, this goal can be achieved by limiting the depth of discharge of the ESD:
(1-DoDESD)×Devre≤EESDt≤Devre (7)
in the formula, DoDESDIndicating the depth of discharge of the ESD.
After the energy storage device is charged and discharged for a period of time, the energy stored therein is:
EESDt＝(1-γESD)×EESDt-1±(rESDt-dESDt)×δ (8)
in the formula rESDt、dESDtRespectively representing the charging and discharging power of the energy storage equipment at the moment t; δ represents a time gap. When the energy storage device is discharged, the discharge power of the energy storage device has the following relationship:
dESDt×ηESD＝DESDt (9)
dESDt≤PESDm (10)
Figure BDA0002928595400000061
in the formula DESDtRepresents the useful discharge power at time t; pESDmRepresents the maximum discharge power of the ESD; t isrampRepresenting the ramp rate of the energy storage device. When charging the energy storage device:
rEsDt≤Rt (12)
Figure BDA0002928595400000071
in the formula RtIndicating the chargeable power provided by the load at time t; omegaESDRepresenting the charge rate of the energy storage device.
In the above embodiment, the energy stored in the device ESD in the initial state is calculated by formula (6); equations (7), (9), (10), (11) are the constraints of ESD discharge(7) (12), (13) are constraints on ESD charging; when P is presentdcAnd PaimWhen the difference value is positive, the power of the data center needs to be reduced, at the moment, ESD discharges, the discharge constraint limits the regulated power to be obtained, and the energy stored by the energy storage equipment after discharging can be obtained through a formula (8); if PdcAnd PaimThe difference of (d) is negative, at this time, the ESD is charged, the charging constraint limits the regulated power to be obtained, and the energy stored after charging can be obtained by equation (8).
Of the above-mentioned powers to be balanced, Ptd、PdvfsRespectively representing the regulated power to be obtained by the task delay power consumption regulating means and the dynamic voltage frequency power consumption regulating means.
Delay sensitive tasks refer to workloads that need to give immediate response; delay tolerant tasks refer to loads that are submitted to a data center's task queue first and can wait for the data center to respond when sufficient resources are available. For a machine, a plurality of tasks can be executed in one time interval, but one task can be executed only once in one time interval, and if the time interval does not finish the task, the next time is executed continuously. Here, assuming that one task includes one or more basic tasks, and the processing time of each basic task is Δ T, any task is an integer multiple of a unit of the basic task. To ensure quality of service, task processing needs to be completed within a maximum completion time.
In delay sensitive tasks, the task arrival time is assumed to be Ts,arrThen its deadline completion time is:
Tt,d＝Tt,arr+nΔT (14)
in delay tolerant tasks, the arrival time of the task is assumed to be Tt,arrThe task-tolerant delay time interval is TmaxiOn the premise of ensuring the service quality, the latest start time and the ending completion time of the task are respectively as follows:
Tt，maxs＝Tt，arr+Tmaxi (15)
Tt,d＝Tt,maxs+nΔT (16)
in the formula, Tt,maxsRepresents the latest start time; t ist,dIndicates the expiration completion time; n is the number of basic tasks included in the task.
In order to realize the regulation and control of the power of the data center by using a task delay scheduling method, a model of the number of executed tasks and the utilization rate of server resources is defined:
Figure BDA0002928595400000081
in the formula of Ucpu,iRepresenting server resource utilization; t isbusy,iIndicating the busy time of the server, Δ TiRepresenting a time gap; cou (Chinese character)hRepresenting the maximum processing capacity of the server; m represents the number of active servers in Δ T; a Δ T may be divided into time slices, and λ is the initial position of each task in the time sliceiRepresents each DeltaTiInitial average number of tasks, using imi、emiRespectively represents DeltaTiThe number of tasks migrating in and out, then (lambda)i+imi-emi) Represents DeltaTiThe number of tasks actually processed.
The Dynamic Voltage and Frequency Scaling (DVFS) technology is to dynamically adjust the Frequency and Voltage of a host according to the change of the resource utilization rate occupied by a Dynamic load, thereby reducing power consumption. In this embodiment, power consumption is calculated based on the relationship between the current operating frequency of the server and the utilization rate of the CPU.
Calculating power consumption according to the relationship between the current operating frequency of the server and the utilization rate of the CPU:
Figure BDA0002928595400000082
in the formula (f)max(m) represents a maximum operating frequency of the server; f. ofop(m, k) denotes that the current operating frequency of the host m is fk(m); k denotes a DVFS mode of operation.
Of the above-mentioned powers to be balanced, Pair、PfanRespectively representing nullThe adjusting power to be obtained by the adjusting cooling power consumption adjusting means and the direct natural wind cooling power consumption adjusting means.
In a computer room of a data center, servers generate heat continuously during operation, and therefore, when a server cabinet is used as a heat source, the inlet temperature is expressed as:
Tin＝Tsup-D·Ps (19)
D＝(K-AT·K)-1-K-1 (20)
K＝mf·Cp (21)
wherein D represents a heat distribution matrix; m isfRepresenting the mass flow rate of the rack; cpRepresents the specific heat capacity of air; k represents the product of the two; the constant matrix A represents the heat flow interference between the server nodes; t isinRepresents a server inlet temperature; t issupIndicating a supply temperature of the air conditioner; psRepresenting the real-time power consumption of the server.
In order to ensure that the temperature inside the machine room is within a normal range, the supply temperature of the air conditioner needs to be adjusted according to the inlet temperature of the server:
Tsup＝Tsup+Tadj (22)
Figure BDA0002928595400000091
in the formula (I), the compound is shown in the specification,
Figure BDA0002928595400000092
respectively representing the safe temperature and the maximum temperature of the server inlet; t isadjThen the difference between the two is represented; t issupIndicating the supply temperature of the air conditioner.
In the embodiment of the invention, a double-cold-source refrigeration mode is adopted, and the mode is as follows: when the outdoor air temperature is greater than the upper limit T of the temperature thresholdhigh(ThighWhen the value is 20 ℃), all cold energy is generated by the work of the air conditioner compression refrigeration module; (2) when the outdoor air temperature is greater than the lower limit T of the temperature thresholdlow(TlowValues of 5 ℃) and less than ThighIn the time, the cooling system uses a direct natural wind cooling model for cooling. Thus, the complexity of power consumption adjustment caused by air flow when air conditioner cooling and direct natural wind cooling are used together is avoided. Meanwhile, the implementation of the double-cold-source refrigeration mode fundamentally reduces the power consumption of the cooling system of the data center. The expression for the mass flow of cold air at this time is:
Figure BDA0002928595400000093
wherein G represents the cold air mass flow; t is tmaxRepresents the temperature of the air after heating; t is toIndicating the outdoor fresh air temperature. Thus, the fan power PmfanComprises the following steps:
Figure BDA0002928595400000094
wherein Δ p represents a pressure drop; etafanIndicating fan efficiency.
It is worth noting that the energy storage device is influenced by the self energy conversion efficiency and the self discharge rate, and certain operation cost is generated in the use process. The operation cost OpEx of the data center ESD is determined, and the model is as follows:
OpEx＝price×Eloss (26)
in the formula, price represents the price of electricity; elossRepresents an energy loss of the ESD;
Eloss＝(1-ηESD)×dESDt+γESD×EESDt (27)
in the formula (d)ESDtRepresenting the discharge power of the ESD at the time t; eESDtRepresenting the energy stored by the energy storage device at time t; etaESD、γESDRespectively representing the conversion efficiency and the self-discharge rate of the energy storage device;
EESDt＝(1-γESD)×EESDt-1+(rESDt-dESDt)×δ (8)
in the formula, rESDt、dESDtRespectively representing the charging and discharging power of the energy storage equipment at the moment t; δ represents a time gap.
It should be noted that the annual cost consumed by the cooling system (mainly air conditioner) of the data center during operation includes water cost, electricity cost, administrator wages, management cost, pollution discharge cost, equipment depreciation cost, equipment maintenance cost, and the like, and here, in order to simplify the calculation of the operating cost of the air conditioning system, only considering the electric energy cost, the operating cost model during operation of the air conditioner is as follows:
Figure BDA0002928595400000101
in the formula, ExairRepresents an operating cost for the operation of the cooling system; pcRepresenting the real-time power consumption of the cooling system; CoP represents the ratio of the amount of cooling provided by the cooling system to the amount of power consumed to perform the cooling; price represents electricity price; δ represents a time gap.
It should be noted that, when the data center performs power adjustment, some penalty cost is generated for various reasons.
When the DC regulates power using a manner of task delay scheduling and a manner of dynamic voltage frequency regulation, a portion of the tasks need to be deferred. If the task's deferral time exceeds its own deferrable time, the DC violates the user quality of service agreement with the penalty of:
petask＝(texec-tsub)·τtask (28)
in the formula, texecAnd tsubRepresenting the actual start of execution time and the final commit time of the task, τtaskA penalty constant, i.e. a penalty once per violation, is indicated.
When the DC cannot coordinate various means, so that the regulated power is higher or lower, the DC bears the penalty of
peina＝ΔPdif×price×δ (30)
In the formula,. DELTA.PdifRepresenting a deviation value of the adjusted power of the data center and the target power; priceRepresenting a price of electricity; δ represents a time gap.
Based on the above-described regulation model, we assume that a demand response signal is received from the smart grid, and in response to this signal, the data center can adjust its power consumption by the various means described above. The problem here is how to determine the power consumption to be balanced for each regulation method and to minimize the regulation cost brought to the data center by this regulation. Suppose that the demand response given by the smart grid is to adjust the power consumption to PaimThe balanced power consumption of the air conditioner cooling model and the direct natural wind cooling model in the aspect of temperature control is Pair、PfanThe power consumption of SC and FB balance in the energy storage device is respectively Psc、PfbThe method for fundamentally reducing the power consumption of the server, the power consumption balanced by the task delay scheduling method and the dynamic voltage frequency adjusting method are respectively Ptd、PdvfsThe deviation value of the regulated power and the target power is delta PdifAnd then:
Paims＝Ps+Pc-Pair-Pfan-Psc-Pfb-Pdvfs-Ptd-ΔPdif (31)
in this embodiment, the above six adjustment means are adopted to perform adjustment together, and the adjustment Cost of the multiple adjustment means and the adjustment inaccuracy penalty of the data center affect them respectively, and are recorded as a final total Cost function Cost:
Minimize:
Exair+OpEx+petask+peina (32)
Exair、OpEx、petaskd、peinarespectively representing the operation cost of the cooling system, the operation cost of the energy storage equipment, the penalty of task delay and the penalty of inaccurate regulation;
Subjectto:
constraints(7、10(12)、11(13)、31)
tacou≤tasum (33)
in the formula of tacouTask scheduler for indicating actual delaysThe number of services; ta issumRepresenting the total number of tasks.
In order to solve the above technical problems, embodiments of the present invention provide a dynamic optimal scheduling method, which can achieve the objectives of accurately adjusting the power consumption of a data center by using multiple means to perform demand response and minimizing the adjustment cost. In addition, the methods presented herein are compared to other methods.
In the embodiment, six methods for adjusting the self power consumption of the data center are provided from the aspects of temperature control, energy storage equipment and self power consumption management of the server, and the total power consumption of the data center is adjusted by using super capacitors, flow batteries, air-conditioning refrigeration, direct natural air cooling, task delay scheduling and DVFS technologies respectively. For target power PaimThe six methods can have various different combination modes, the power consumption of the data center can be adjusted to the target power, and the purpose is to find the adjusting mode with the minimum adjusting cost in various feasible schemes, so that the dynamic optimal scheduling method is provided. The main principle of the dynamic optimal scheduling method is that after a given demand response target, power needing to be balanced can be reasonably distributed among multiple power regulation methods, so that the cost of demand response is minimized on the premise of meeting different regulation process constraints. Here, we use a linear optimization algorithm to implement a dynamic optimal scheduling method to obtain the minimum power adjustment cost.
For the constraint optimization problem, the direct method and the indirect method can be classified according to the difference of the solving principle. The interior point method is an indirect algorithm for solving the optimization problem, the solving process is to convert the constraint problem into an unconstrained problem by introducing a utility function, and then continuously update the utility function by using the optimization iteration process so as to make the algorithm converge, and then the steps of the dynamic optimal scheduling method are as follows:
(a) selecting a proper penalty factor r(0)Allowable error ξ and a decrementing coefficient c;
(b) choosing an initial point X in the feasible region(0)Let k be 0;
(c) constructing a penalty function
Figure BDA0002928595400000121
From X(k-1)Method for calculating penalty function by using unconstrained optimization method for point sending
Figure BDA0002928595400000122
Extreme point of
Figure BDA0002928595400000123
(d) Using termination criteria
Figure BDA0002928595400000124
Judging whether convergence occurs; if the condition is met, stopping iteration and obtaining the optimal point of the objective function
Figure BDA0002928595400000125
Otherwise, it orders
Figure BDA0002928595400000126
k +1, and go to (c);
when the termination criterion is met, the solution of the final extreme point obtained in the course of the iteration will be the optimal solution of the output. In this algorithm, an initial point X(0)Randomly generated by a computer, an initial value r of a penalty factor(0)The iteration times can be influenced, 1 is generally taken, then the adjustment is carried out through a plurality of tests, the decrement coefficient plays a role in gradually decrementing the penalty factor, the decisive role is not played in the iteration process, and the value range is usually between 0.1 and 0.7.
(1) Six-means Dynamic Optimal Scheduling Method (DOSM)
In the strategy, the data center uses six means of SC, FB, air-conditioning cooling, direct natural air cooling, task delay scheduling and dynamic voltage frequency adjustment to respond to the demand; the most suitable combination of adjustments is determined from the four adjustment modes using an optimization method to minimize the final adjustment cost.
(2) Four-means dynamic optimization method (Baseline)
Three reference methods are established in the strategy, namely Baseline1, Baseline2 and Baseline 3. Baseline1 means that a data center carries out demand response by using four means, namely SC, FB, task delay scheduling and dynamic voltage frequency regulation, and for a demand response target given by a smart grid, an optimal regulation combination is determined from the four regulation modes by using an optimization method so as to minimize the final regulation cost. Similarly, Baseline2 refers to that the data center uses four means of air-conditioning cooling, direct natural air cooling, task delay scheduling and dynamic voltage frequency adjustment to respond to the demand. Baseline3 refers to that the data center uses four methods of air-conditioning refrigeration, direct natural wind cooling and SC and FB to perform demand response.
(3) Three scheduling methods are equally divided (Avg)
The method averagely divides the power consumption required to be regulated given by a demand response signal into three parts, and balances the power consumption management aspects of temperature control, energy storage equipment and a server, namely, SC and FB regulation means obtain one third of the power consumption to be regulated in total, air-conditioning cooling and direct natural air cooling regulation means obtain one third of the power consumption to be regulated in total, and task delay scheduling and dynamic voltage frequency regulation means obtain one third of the power consumption to be regulated in total. For the allocation method, it should be noted that the six adjustment means are not adopted for the equalization, because the adjustment and control means equally divided by the six means basically cannot meet the requirement of the demand response, only the penalty of power consumption adjustment is continuously increased, and the reference significance is not provided.
Three experimental examples, namely case1, case2 and case3, were prepared in the simulation experiment, as shown in fig. 2. FIG. 2(a) shows case1 showing the response demand on the first day, with the two adjustment targets adjusted to 950KW and half an hour at this power and 1050KW for one hour, respectively; FIG. 2(b) shows case2 showing the response demand on the following day, with the two adjustment targets adjusted to 700KW for one hour and 1100KW for half an hour, respectively; fig. 2(c) shows case3 showing the response demand on the third day, and the three adjustment targets are adjusted to 820KW for two hours and 1250KW for one hour, respectively.
For different cases, the performances of the above proposed six methods of air-conditioning refrigeration, direct natural air cooling, super capacitor, flow battery, task delay scheduling and dynamic voltage frequency regulation are compared. Table 1 shows the adjustment costs corresponding to six adjustment modes for different cases, and it can be seen that, compared to the comparative experiment, the model provided herein has significant advantages in accurately adjusting the adjustment cost for minimizing the power consumption of the data center. In the cases 1, 2, and 3, as the demand response signal increases the power consumption management requirements of the data center, the cost of the DOSM is significantly reduced compared to other methods. Table 1 shows the cost comparison results of demand responses for different strategies.
TABLE 1
Figure BDA0002928595400000141
According to the method for adjusting the power consumption of the data center in the smart grid environment, a detailed model for accurately adjusting and controlling the power consumption of the data center to minimize the total cost is constructed on the basis of comprehensively considering the temperature control, the energy storage equipment, the load characteristics and the characteristics of the server. In the aspect of temperature control, the influence of load change on the power consumption of the cooling system is considered, and meanwhile, the refrigeration power consumption of the cooling system of the data center is reduced fundamentally by utilizing a direct natural wind cooling technology. For the power consumption of the server, the execution of the load is a main factor for generating the power consumption, and in the embodiment of the invention, the task scheduling model is formulated by analyzing the type of the load, and the dynamic voltage frequency adjustment technology is combined, so that the power consumption of the server is reduced. Different from the former two, the energy storage equipment adjusts the power consumption dependence of the data center on the power grid according to the self charging and discharging characteristics in the process of accurately adjusting the power consumption of the data center, so as to meet the requirement of demand response. And finally, normalizing the operation cost, the delay penalty cost and the regulation inaccuracy penalty of the regulation models into the total cost of power consumption regulation, and comprehensively utilizing various power regulation means to achieve the aim of accurately regulating the power consumption of the data center to minimize the total operation cost. The result shows that the model for accurately regulating and controlling the power consumption of the data center to minimize the total cost has a good effect in experiments.
As shown in fig. 3, an embodiment of the present invention further provides a system for adjusting power consumption of a data center in a smart grid environment, including: a total current power determination module 301, a power regulation model building module 302, a total cost determination module 303 and an optimal combination determination module 304.
A total current power determining module 301, configured to determine total current power P of the data center according to overall power consumption characteristics of the data centerdc；
A power regulation model establishing module 302 for determining at least one power consumption regulation means, which is a target power P given according to the response requirement of the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means;
a total cost determination module 303 for determining an operating cost Ex of the data center cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskThe operating cost Ex of the cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskNormalizing to the total Cost of data center power consumption adjustment;
the optimal combination determining module 304 is configured to determine an optimal combination manner of multiple power consumption adjusting means by using a dynamic optimal scheduling method on the premise that constraint conditions of each power adjusting means are met, adjust power consumption of the data center by using the optimal combination manner, and minimize a total Cost for adjusting the power consumption of the data center.
The embodiment of the present invention further provides a computer-readable storage medium, which includes instructions, and when the instructions are run on a computer, the instructions cause the computer to execute the method for adjusting the power consumption of the data center in the smart grid environment, provided by the foregoing embodiment.
The embodiment of the present invention further provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the method for adjusting power consumption of a data center in a smart grid environment provided by the foregoing embodiment.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed.
Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment of the present invention.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention essentially or partially contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and other various media capable of storing program codes.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.

Claims (10)

1. A method for adjusting power consumption of a data center in a smart grid environment is characterized by comprising the following steps:
determining the current total power P of the data center according to the overall power consumption characteristics of the data centerdc；
Determining at least one power consumption adjusting means, and giving a target power P according to the response requirement of the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means;
determining an operating cost Ex of a data center cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskThe operating cost Ex of the cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskNormalizing to the total Cost of data center power consumption adjustment;
on the premise of meeting the self constraint conditions of each power adjusting means, determining the optimal combination mode of various power consumption adjusting means by using a dynamic optimal scheduling method, adjusting the power consumption of the data center by using the optimal combination mode, and minimizing the total Cost of the power consumption adjustment of the data center.
2. The method of claim 1, wherein the total power P of the data center is determined according to the overall power consumption characteristics of the data centerdcThe model is as follows:
Pdc＝Ps+Pc
in the formula, PdcRepresenting the total current power, P, of the data centersRepresenting the real-time power consumption, P, of the servercRepresenting the real-time power consumption of the cooling system;
the real-time power consumption model of the server is as follows:
Ps＝α·ucpu+β；
in the formula, alpha represents the difference value between the peak power and the idle power of the server, beta represents the idle power of the server, and u represents the idle power of the servercpuRepresenting the current CPU utilization;
the CoP is the ratio of the cooling capacity provided by the cooling system to the power consumed for cooling, and the real-time power consumption model of the cooling system is as follows:
Figure FDA0002928595390000021
Cop＝0.0068Tsup 2+0.0008Tsup+0.458；
where Tsup represents the cooling temperature provided by the cooling system.
3. Method according to claim 1, characterized in that said target power P given according to the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means as follows;
Paim＝Ps+Pc-Pz-ΔPdif
in the formula, PsRepresenting the real-time power consumption, P, of the servercRepresenting the real-time power consumption of the cooling system; pzIs Pair、Pfan、Psc、Pfb、PtdOr PdvfsIs one or the sum of any of a plurality of terms, Pair、PfanRespectively showing the adjusting power to be obtained by the air conditioner cooling power consumption adjusting means and the direct natural wind cooling power consumption adjusting means, Psc、PfbRespectively representing the regulated power to be obtained by the super capacitor power consumption regulating means and the flow battery power consumption regulating means in the energy storage equipment, Ptd、PdvfsRespectively representing the regulated power to be obtained by the task delay power consumption regulating means and the dynamic voltage frequency power consumption regulating means; Δ pdifIndicating the deviation of the adjusted power from the target power.
4. The method of claim 1, wherein the determining the operating cost Ex of the data center cooling systemairThe model is as follows:
Figure FDA0002928595390000022
in the formula, ExairRepresents an operating cost for the operation of the cooling system; pcRepresenting the real-time power consumption of the cooling system; CoP represents the ratio of the amount of cooling provided by the cooling system to the amount of power consumed to perform the cooling; price represents electricity price; δ represents a time gap;
the operation cost OpEx of the energy storage equipment of the data center is determined, and the model is as follows:
OpEx＝price×Eloss；
in the formula, price represents the price of electricity; elossRepresents the energy loss of the energy storage device;
Eloss＝(1-ηESD)×dESDt+γESD×EESDt
in the formula (d)ESDtRepresenting the discharge power of the energy storage device at the moment t; eESDtRepresenting the energy stored by the energy storage device at time t; etaESD、γESDAre respectively provided withRepresenting the conversion efficiency and self-discharge rate of the energy storage device;
EESDt＝(1-γESD)×EESDt-1+(rESDt-dESDt)×δ
in the formula, rESDt、dESDtRespectively representing the charging power and the discharging power of the energy storage equipment at the moment t; δ represents a time gap.
5. Method according to claim 1, characterized in that said adjusted deviation penalty cost peinaThe model is as follows:
peina＝ΔPdif×price×δ
in the formula,. DELTA.PdifRepresenting a deviation value of the adjusted power of the data center and the target power; price represents electricity price; δ represents a time gap.
6. The method of claim 1, wherein the task delay penalty cost petaskThe model is as follows:
petask＝(texec-tsub)·τtask
in the formula, texecAnd tsubRepresenting the actual starting execution time and the submission time of the task; tau istaskA penalty constant, i.e. a penalty once per violation, is indicated.
7. The method according to claim 1, wherein the determining an optimal combination manner of multiple power consumption adjusting means by using a dynamic optimal scheduling method on the premise of satisfying self-constraint conditions of each power adjusting means, and the adjusting power consumption of the data center by using the optimal combination manner and minimizing a total Cost of power consumption adjustment of the data center comprises:
on the premise of meeting the server inlet temperature constraint, the energy storage equipment discharge constraint, the charging constraint, the delayed task number constraint and the host frequency constraint, a dynamic optimal scheduling method is realized by using a linear optimization algorithm to obtain the minimum total Cost for adjusting the power consumption of the data center.
8. A data center power consumption adjustment system in a smart grid environment, comprising:
a total current power determining module for determining the total current power P of the data center according to the overall power consumption characteristics of the data centerdc；
The power regulation model establishing module is used for determining at least one power consumption regulation means and giving a target power P according to the response requirement of the smart gridaimCurrent total power P of data centerdcEstablishing a power regulation model with the regulated power to be obtained of the at least one power regulation means;
a total cost determination module for determining an operating cost Ex of the data center cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskThe operating cost Ex of the cooling systemairOperation cost OpEx and adjustment deviation penalty cost pe of energy storage equipmentinaAnd a task delay penalty cost petaskNormalizing to the total Cost of data center power consumption adjustment;
and the optimal combination determining module is used for determining the optimal combination mode of various power consumption adjusting means by using a dynamic optimal scheduling method on the premise of meeting the self constraint conditions of each power adjusting means, adjusting the power consumption of the data center by using the optimal combination mode and minimizing the total Cost of the power consumption adjustment of the data center.
9. A computer-readable storage medium comprising instructions that, when executed on a computer, cause the computer to perform the method for adjusting power consumption of a data center in a smart grid environment according to any one of claims 1 to 7.
10. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method for adjusting power consumption of a data center in a smart grid environment according to any one of claims 1 to 7.
CN202110141041.2A 2021-02-02 2021-02-02 Method, system, medium and device for adjusting power consumption of data center in smart grid environment Active CN112994037B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110141041.2A CN112994037B (en) 2021-02-02 2021-02-02 Method, system, medium and device for adjusting power consumption of data center in smart grid environment

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110141041.2A CN112994037B (en) 2021-02-02 2021-02-02 Method, system, medium and device for adjusting power consumption of data center in smart grid environment

Publications (2)

Publication Number Publication Date
CN112994037A true CN112994037A (en) 2021-06-18
CN112994037B CN112994037B (en) 2022-08-09

Family

ID=76346156

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110141041.2A Active CN112994037B (en) 2021-02-02 2021-02-02 Method, system, medium and device for adjusting power consumption of data center in smart grid environment

Country Status (1)

Country Link
CN (1) CN112994037B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090187782A1 (en) * 2008-01-23 2009-07-23 Palo Alto Research Center Incorporated Integrated energy savings and business operations in data centers
US20130086404A1 (en) * 2011-10-03 2013-04-04 Microsoft Corporation Power regulation of power grid via datacenter
CN107329811A (en) * 2017-06-09 2017-11-07 北京云集智造科技有限公司 A kind of power consumption of data center adjusting method and device
US20180189899A1 (en) * 2015-07-01 2018-07-05 University Of Florida Research Foundation, Inc. Using loads with discrete finite states of power to provide ancillary services for a power grid
CN109034557A (en) * 2018-07-06 2018-12-18 东南大学 Consider the electric system congestion management method of data network Load Regulation potentiality
CN111461920A (en) * 2020-03-23 2020-07-28 青海大学 Smart grid-oriented data center demand response method and system

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20090187782A1 (en) * 2008-01-23 2009-07-23 Palo Alto Research Center Incorporated Integrated energy savings and business operations in data centers
US20130086404A1 (en) * 2011-10-03 2013-04-04 Microsoft Corporation Power regulation of power grid via datacenter
US20180189899A1 (en) * 2015-07-01 2018-07-05 University Of Florida Research Foundation, Inc. Using loads with discrete finite states of power to provide ancillary services for a power grid
CN107329811A (en) * 2017-06-09 2017-11-07 北京云集智造科技有限公司 A kind of power consumption of data center adjusting method and device
CN109034557A (en) * 2018-07-06 2018-12-18 东南大学 Consider the electric system congestion management method of data network Load Regulation potentiality
CN111461920A (en) * 2020-03-23 2020-07-28 青海大学 Smart grid-oriented data center demand response method and system

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
黄焱等: "智能电网下数据中心能耗费用优化综述", 《浙江大学学报（工学版）》 *
黄焱等: "智能电网下数据中心能耗费用优化综述", 《浙江大学学报（工学版）》, vol. 50, no. 12, 31 December 2016 (2016-12-31), pages 2386 - 2399 *

Also Published As

Publication number Publication date
CN112994037B (en) 2022-08-09

Similar Documents

Publication Publication Date Title
Song et al. Energy storage modeling of inverter air conditioning for output optimizing of wind generation in the electricity market
Li et al. Enabling distributed generation powered sustainable high-performance data center
Li et al. Managing green datacenters powered by hybrid renewable energy systems
WO2022127652A1 (en) Air conditioning aggregation control method, system, and regulation apparatus
Li et al. Managing server clusters on renewable energy mix
CN113991655A (en) Method, device and medium for evaluating load aggregation demand response potential of fixed-frequency air conditioner
WO2017149617A1 (en) Control device, supply/demand adjustment control device, power storage device, output control device, supply/demand adjustment system, control method, supply/demand adjustment method, and program
CN112994037B (en) Method, system, medium and device for adjusting power consumption of data center in smart grid environment
CN117559526A (en) Router-simulated energy regulation and control method based on optical storage and charging integrated charging station
CN110008515B (en) Renewable energy data center management method and device
CN115238992B (en) Coordination optimization method and device for power system source load storage and electronic equipment
Wu et al. Real-time energy management of large-scale data centers: A model predictive control approach
CN115330015A (en) Data center participation comprehensive demand response distributed optimization method considering waste heat
CN111861278B (en) Power load peak-valley period division method and system for power system
Teo et al. Energy management controls for chiller system: A review
Mengmeng et al. Datacenter Power Consumption Regulation towards Specific Targets in Smart Grid Environment
CN109753400B (en) Energy consumption adjusting method and edge data center system
CN113852093A (en) Energy storage device and polysilicon load aggregation frequency modulation method and system
CN116700404A (en) Control method, device and system of virtual power plant
CN117267910B (en) Method, device, equipment and medium for flexibly adjusting load of central air-conditioning refrigeration system
Chen et al. Leaky bucket-inspired power output smoothing with load-adaptive algorithm
WO2023228749A1 (en) Electric power adjustment device
Zhang et al. Research on data center computing resources and energy load Co-optimization considering spatial-temporal allocation
CN115164361B (en) Data center control method and device, electronic equipment and storage medium
Che et al. Control Strategy of Constant-Speed Air-Conditioners Based on Virtual State of Charging Priority

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant