CN110084444A - A kind of cloud data center power load dispatching method considering natural resources randomness - Google Patents
A kind of cloud data center power load dispatching method considering natural resources randomness Download PDFInfo
- Publication number
- CN110084444A CN110084444A CN201910445540.3A CN201910445540A CN110084444A CN 110084444 A CN110084444 A CN 110084444A CN 201910445540 A CN201910445540 A CN 201910445540A CN 110084444 A CN110084444 A CN 110084444A
- Authority
- CN
- China
- Prior art keywords
- power load
- data center
- cloud data
- time shift
- load
- 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
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06315—Needs-based resource requirements planning or analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses a kind of cloud data center power load dispatching method for considering natural resources randomness, which includes the following steps;Predict total power load and renewable energy power output to be scheduled;It can time shift power load from time, space scale scheduling, if the power load treating capacity under current scheduling strategy is less than bearing capacity, then calculate the electrical energy demands amount and Waste Heat Reuse amount of each cloud data center, judge simultaneously electric system and therrmodynamic system whether and meanwhile balance, and power load scheduling scheme and operation totle drilling cost are recorded at equilibrium, to obtain all possible scheduling scheme, using the scheduling scheme of minimum operation totle drilling cost as final scheduling scheme;Power load dispatching method of the invention can guarantee the electric system balance and therrmodynamic system balance of each cloud data center simultaneously, hence it is evident that reduce cloud service provider operating cost, profit margin has been significantly greatly increased, provide effective foundation for Sustainable Development of Enterprises.
Description
Technical field
The present invention relates to cloud data center power load dispatching technique fields, and more specifically, the present invention is a kind of consideration
The cloud data center power load dispatching method of natural resources randomness.
Background technique
It is well known that cloud computing all keeps high speed development in various countries worldwide, data can be under the same cloud
Row transmission, this has also pushed the expansion and development of cloud data center.Under normal conditions, have in a cloud data center hundreds of thousands of
A server, each server require 24 × 7 ground and guarantee steady and continuous operation, this is necessarily required to a large amount of energy consumption.According to
Statistics, the electricity ratio that Global Internet technology industry in 2016 uses account for the 3% of global total electricity, and estimate that this ratio will
It doubles within every 4 years.A large amount of power consumption is that each Internet company under Internet technology industry brings huge power cost,
Excessively high electric cost will affect the performance of enterprises.According to statistics, use energy cost of the large-scale data center at 2016 to 2018
Totally 1,850 hundred million dollars, account for 40% of its totle drilling cost or more.So how to reduce cloud data center with can cost to Guan Chong
It wants.
Although the prior art provides the scheme of some reduction cloud data centers energy costs, for example, BVT (Borrowed
Virtual Time) dispatching algorithm, Credit dispatching algorithm etc., these schemes are possible to improve the energy to a certain extent
Utilization efficiency, but all cloud data center electric system balances can not be met simultaneously, so often there is electricity consumption and bear in the prior art
Lotus scheduling mode Consideration is single, it is subjective, can not consider that carrying out load scheduling etc. ask from the entire electric system overall situation
Topic.
Therefore, how cloud data center power load scheduling is objectively and comprehensively carried out, to improve the energy benefit of data center
With efficiency, the use energy cost of each cloud data center of reduction, become those skilled in the art's technical problem urgently to be resolved and beginning
The emphasis studied eventually.
Summary of the invention
For solve existing for existing cloud data center power load scheduling strategy it is subjective, can not be complete from electric system
The problems such as office's angle carries out power load scheduling, present invention innovation provide in a kind of cloud data of consideration natural resources randomness
Heart power load dispatching method, the cloud data center power load dispatching method can not only guarantee the supply and demand of entire electric system
Balance, and can guarantee simultaneously the equilibrium of supply and demand of entire therrmodynamic system, to reach comprehensive, objectively carry out cloud data center
The scheduling of power load, thoroughly to solve problems existing for existing power load dispatching method.
To realize the above-mentioned technical purpose, the invention discloses a kind of cloud data center electricity consumptions for considering natural resources randomness
Load scheduling method, the dispatching method include the following steps;
Step 1, the following preset time period is predicted using the history power load data of each cloud data center obtained in advance
Interior total power load to be scheduled, and it is pre- to utilize the renewable energy history of each cloud data center obtained in advance to go out force data
Survey the renewable energy power output in the following preset time period;
Step 2, filtering out from total power load to be scheduled can time shift power load;It wherein, can time shift power load
It indicates to postpone the power load that duration processed is greater than the first preset duration;
Step 3, on respectively can time shift power load handled in respective delay duration processed completion premised on, adjustment exist
It is described future preset time period in respectively can time shift power load Annual distribution and Regional Distribution, thus enable can time shift power load
Treating capacity is directly proportional to the power output in real time of the renewable energy of each cloud data center, and enable can time shift power load treating capacity with not
The Spot Price of region is inversely proportional where each cloud data center come in preset time period;
Step 4, judge in the case that currently respectively can time shift power load Annual distribution and Regional Distribution each cloud data in
Whether the power load treating capacity of the heart is less than respective bearing capacity;If so, thening follow the steps 5;If it is not, then return step 3;
Step 5, electrical energy demands amount and waste heat benefit of each cloud data center in the following preset time period are calculated separately
Dosage;
Step 6, at the same judge the power supply volume of the electric system of each cloud data center whether meet respective electrical energy demands amount,
Whether the Waste Heat Reuse amount of each cloud data center meets the calorific value of the therrmodynamic system at respective place;If the two conditions are simultaneously
It sets up, thens follow the steps 7, otherwise return step 3;
Step 7, current power load scheduling scheme is recorded, and calculates the operation totle drilling cost under the scheduling scheme, then
Current power load scheduling scheme and its operation totle drilling cost are saved into scheduling strategy set again;
Step 8, judge whether it is exhaustive in the following preset time period respectively can time shift power load distribution tune
Perfect square formula;If so, thening follow the steps 9;If it is not, then return step 3;
Step 9, using in scheduling strategy set with it is minimum operation totle drilling cost power load scheduling scheme as finally
Cloud data center power load scheduling scheme.
Based on above-mentioned technical solution, the present invention innovatively carries out the power load of cloud data center on spatial and temporal scales
It redistributes, i.e., the present invention can meet the distribution of the power load in time scale and the power load on space scale simultaneously
Distribution, and with kurtosis renewable energy and electricity price effectively cooperated, thus in guarantee electric system and heating power
Efficiency of energy utilization is greatly improved under the premise of the system equilibrium of supply and demand.
Further, in step 3, adjustment respectively can time shift power load Annual distribution when, can time shift power load
It is transferred to the period that renewable energy power output is lower than the second preset value greater than the period of the first preset value and/or electricity price;?
Adjustment respectively can time shift power load Regional Distribution when, can time shift power load be transferred to renewable energy power output be greater than third
The cloud data center and/or electricity price of preset value are lower than the cloud data center of the 4th preset value.
Further, in step 3, adjustment can time shift power load Annual distribution when, further include enable respectively can Shi Yiyong
Electric load handles the step of completion in respective delay duration front half section processed.
It further, further include that filter out from total power load to be scheduled can not time shift power load in step 2
Step;
In step 3, adjust in the following preset time period respectively can not time shift power load Regional Distribution, to enable not
Can time shift power load treating capacity it is directly proportional to the power output in real time of the renewable energy of each cloud data center, and enable can not time shift
Power load treating capacity and the Spot Price of region where each cloud data center in the following preset time period are inversely proportional;Wherein,
It is described can not time shift power load indicate power load of the duration processed less than the second preset duration.
Further, in step 3, adjustment respectively can not time shift power load Regional Distribution when, can not time shift electricity consumption
Load is transferred to the cloud that renewable energy power output is lower than the 6th preset value greater than the cloud data center and/or electricity price of the 5th preset value
Data center.
Further, in step 3, transfer can time shift power load and/or can not be before time shift power load, will be to be transferred
Each power load resolve into multiple sub- loads with same load scale, be that minimum thread carries out with a sub- load
Power load transfer.
Further, in step 6, the electric system include power grid subsystem, renewable energy power generation subsystem, can not
Renewable source of energy generation subsystem and electric energy storage subsystem.
Further, in step 6, the therrmodynamic system includes cloud data center heat release subsystem, generating set heat release
System and electric energy storage heat release subsystem.
Further, in step 7, the operation totle drilling cost includes that cost of electricity-generating, purchases strategies, power load are scheduled to
Originally, cloud data center cost of serving and the punishment cost of removal of load.
It further, further include obtaining going through for each cloud data center in past preset time period simultaneously before step 1
The step of history power load data and renewable energy history go out force data.
The invention has the benefit that the cloud data center electricity consumption provided by the present invention for considering natural resources randomness is negative
Lotus dispatching method can guarantee the electric system balance and therrmodynamic system balance of each cloud data center simultaneously, significantly reduce
The operating cost of cloud service provider has been significantly greatly increased profit margin, has significantly improved the service quality of cloud service provider, has
The strength of enterprise for helping enhance cloud service provider, provides effective foundation for Sustainable Development of Enterprises.
Detailed description of the invention
Fig. 1 be consider natural resources randomness cloud data center can time shift power load dispatching method process signal
Figure.
Fig. 2 is to consider that the process of cloud data center total power load dispatching method to be scheduled of natural resources randomness is shown
It is intended to.
Specific embodiment
With reference to the accompanying drawings of the specification to a kind of cloud data center use for considering natural resources randomness provided by the invention
Electric load dispatching method carries out detailed explanation and illustration.
One cloud service provider (such as Google, Baidu, Tencent etc.) can usually manage multiple in diverse geographic location
Cloud data center, each cloud data center can be carried out data transmission by a proxy server with other cloud data centers;This reality
It applies example to be based on being able to carry out data exchange and transmission principle between different cloud data centers, specifically provides a kind of consideration nature money
The cloud data center power load dispatching method of source randomness, the present invention are innovatively negative to electricity consumption in time, two, space level
Lotus is redistributed, to obtain global optimum as a result, such as farthest reducing cost;As shown in Figure 1, 2, the dispatching party
Method specifically includes following step.
Step 1, the following preset time period is predicted using the history power load data of each cloud data center obtained in advance
Interior total power load to be scheduled, and it is pre- to utilize the renewable energy history of each cloud data center obtained in advance to go out force data
The renewable energy power output in the following preset time period is surveyed, in the present embodiment, passes through Monte Carlo and Latin Hypercube Sampling
Method realize the prediction of total power load in the following preset time period and renewable energy power output.In order to improve cloud data
The accuracy of center power load scheduling, before step 1, the present embodiment further includes being obtained in past preset time period simultaneously
The step of history power load data and renewable energy history of each cloud data center go out force data.
Step 2, filtering out from total power load to be scheduled can time shift power load;It wherein, can time shift power load
Indicate to postpone the power load that duration processed is greater than the first preset duration, in the present embodiment, the first preset duration can be several
Hour or one day.Further including the steps that filtering out from total power load to be scheduled in step 2 can not time shift power load;
Wherein, can not time shift power load indicate power load of the duration processed less than the second preset duration, in the present embodiment,
Two preset durations can be several milliseconds or several seconds.
Step 3, on respectively can time shift power load handled in respective delay duration processed completion premised on, adjustment exist
In the following preset time period respectively can time shift power load Annual distribution and Regional Distribution, thus enable can time shift power load processing
Measure it is directly proportional to the power output in real time of the renewable energy of each cloud data center, and order can time shift power load treating capacity and future it is pre-
If the Spot Price of region is inversely proportional where each cloud data center in the period.Above-mentioned adjustment can time shift electricity consumption Annual distribution
Accordingly indicated with the process of Regional Distribution can time shift load time transfer process and space transfer process, in space scale
Upper transfer means that power load can be handled in the data center for receiving its proxy server locality, can also be by proxy server
It is handled, transfer power load can should be generated accordingly in the process by cloud data centers that network connection is transferred to other areas
Power load cost of transfer;Can shift and mean in time scale, due to can time shift load handle the time compared with
It is long, it is possible to be shifted that can handle in the period.It should be appreciated that the present embodiment can time shift power load in time
Transfer and spatially shift and can occur simultaneously completely, can time shift power load can be both transferred to other times section enterprising
It can also be handled in the cloud data center in other areas while row processing.
For specifically, in step 3, adjustment respectively can time shift power load Annual distribution when, can time shift electricity consumption
Load is transferred to the time that renewable energy power output is lower than the second preset value greater than the period of the first preset value and/or electricity price
Section, i.e. renewable energy are contributed the higher and/or electricity price lower period;For example, special using renewable energy supply time-varying
Property transfer power load (or referred to as workload), it is assumed that can time that handles of time shift power load be one day, if can
Time shift power load reaches data center in morning 0:00, it can be transferred to processing in intraday any time period, is
It is better using renewable energy, reduce cost, the present embodiment by can time shift power load to be transferred to wind power output more
In time (period in morning and late night hours section) or photovoltaic power output more periods (day time period, such as 8:00~16:
00) it is handled;For another example, spot market electricity price time-varying characteristics are utilized to shift power load, it is assumed that can time shift power load
It is one day that the time, which can be handled, if can time shift load morning 0:00 reach data center, it can be transferred in one day
Any time period in processing, for preferably using electricity price in one day the characteristic of real-time change, reduce cost, the present invention can
Time shift power load is transferred in the electricity charge relatively low period and is handled.
Adjustment respectively can time shift power load Regional Distribution when, can time shift power load be transferred to renewable energy and go out
Power is lower than the cloud data center of the 4th preset value greater than the cloud data center and/or electricity price of third preset value, for example, can utilizing
Variation characteristic shifts power load between renewable sources of energy supply region, preferably to utilize renewable energy, reducing cost, this reality
Apply example by can time shift power load be transferred to the more data center region of wind power output or photovoltaic is contributed more number
It is handled according to center region;For another example, power load is shifted using spot market electricity price region variation characteristics, in order to
Preferably utilize different geographical variation characteristic, reduce cost, the present embodiment can time shift power load be transferred to electricity charge phase
Data center in lower region is handled.
As the technical solution of optimization, in step 3, adjustment can time shift power load Annual distribution when, the present embodiment
Further include the steps that enable respectively can time shift power load completion is handled in respective delay duration front half section processed, preceding 50%
The significant reduction data center services cost of processing in time, and which can be met the needs of users as early as possible,
The processing of corresponding task is completed as early as possible.
As shown in Fig. 2, the present embodiment is also adjusted respectively can not time shift power load in the following preset time period in step 3
Regional Distribution, with enable can not the real-time of renewable energy of time shift power load treating capacity and each cloud data center contribute at just
Than, and enable can not each cloud data center place region in time shift power load treating capacity and the following preset time period it is real-time
Electricity price is inversely proportional, in the present embodiment step 3, adjust it is each can not time shift power load Regional Distribution when, can not time shift
Power load is transferred to the cloud data center that renewable energy power output is greater than the 5th preset value and/or electricity price lower than the 6th preset value
Cloud data center;For example, power load is shifted using variation characteristic between renewable energy supply region, in order to preferably sharp
With renewable energy, reduce cost, the present embodiment by can not time shift power load be transferred to the more data center of wind power output
The data center region that region or photovoltaic are contributed more is handled;For another example, using spot market electricity price
Domain variation characteristics shift power load, in order to preferably utilize different geographical variation characteristic and reduce cost, this reality
Apply example by can not time shift power load be transferred to the data center in the relatively low region of the electricity charge and handle.
As the technical solution of optimization, in the present embodiment step 3, transfer can time shift power load and/or can not time shift
Before power load, each power load to be transferred is resolved into multiple sub- loads with same load scale, it is negative with a son
Lotus is that minimum thread carries out power load transfer, i.e., the present embodiment can time shift load equally resolve into small-scale work
Industry, which can not only make full use of computing resource on time and space aspects, reduce computing resource waste, and can
Accurately computation bandwidth occupancy, and then accurately calculating for load cost of transfer is realized according to accurate bandwidth usage.
In addition, in specific implementation, after user submits carry calculation to request, the present embodiment passes through control user region
Proxy server mode complete power load scheduling, power load is collected by proxy server, then is sent out by each proxy server
Corresponding cloud data center under toward scheduling strategy, a proxy server in the present embodiment can be responsible for a ground on geographical location
The power load in area (such as a city) is collected and is summarized.
It should be appreciated that on the basis of present disclosure, first preset value of the present embodiment, the second preset value,
Third preset value, the 4th preset value, the first preset duration, second preset duration etc. can carry out rationally wise as the case may be
Ground adjustment.
Step 4, judge in the case that currently respectively can time shift power load Annual distribution and Regional Distribution each cloud data in
Whether the power load treating capacity of the heart is less than respective bearing capacity, that is, judges under the conditions of current power load scheduling in each cloud data
Whether the heart can normally handle assigned power load amount;If so, it is negative normally to have handled assigned electricity consumption
Lotus amount, thens follow the steps 5;If not, can not normally handle assigned power load amount, then return step 3.
Step 5, electrical energy demands amount and Waste Heat Reuse amount of each cloud data center in the following preset time period are calculated separately.
Step 6, at the same judge the power supply volume of the electric system of each cloud data center whether meet respective electrical energy demands amount,
Whether the Waste Heat Reuse amount of each cloud data center meets the calorific value of the therrmodynamic system at respective place;If the two conditions are simultaneously
It sets up, thens follow the steps 7, otherwise return step 3.The essence of this step deterministic process is: should guarantee the electric energy of data center
Supply, meets in micro-capacitance sensor region again and uses heat demand;Confession of the present embodiment in the electric system for judging each cloud data center
When whether electricity can satisfy respective electrical energy demands amount, further include the steps that using non-IT power load as Consideration, tool
Body be will be that the electric power that refrigeration system, lighting system and other switchgears for preventing server overheat set need also considers
Inside, so the present invention has outstanding advantages of influence factor considers more comprehensively;In addition, when specifically being calculated, thus it is ensured that
The scheme of the electric system equilibrium of supply and demand is known as feasible solution, and the point for guaranteeing the therrmodynamic system equilibrium of supply and demand is made to constitute feasible zone.
Electric system includes power grid subsystem, renewable energy power generation subsystem, non-renewable energy resources power generation sub-system (packet
Include conventional electric power generation unit) and electric energy storage subsystem, electricity stabilization, uninterruptedly supply for guaranteeing cloud data center.
Therrmodynamic system includes cloud data center heat release subsystem, generating set heat release subsystem, electric energy storage heat release subsystem
And electric boiler etc., generating set heat release subsystem may include CHP unit etc., cloud data center heat release subsystem is used for data
Center generate waste heat again Collection utilization, improve capacity usage ratio, specifically, by the heat of collection and other heat release subsystems
Heat be jointly the heat supply of surrounding resident area.
The electric system, therrmodynamic system and the information transmission system of the present embodiment have collectively constituted in the cloud data of geographical distribution
Heart interacted system, the present invention can guarantee the energy utilization situation global optimum of entire cloud data center interacted system.
Step 7, current power load scheduling scheme is recorded, and calculates the operation totle drilling cost under the scheduling scheme, then
Current power load scheduling scheme and its operation totle drilling cost are saved into scheduling strategy set again.Wherein, totle drilling cost packet is runed
Include cost of electricity-generating, (from power grid) purchases strategies, power load scheduling cost, cloud data center cost of serving and the punishment of removal of load
Cost, the punishment cost of removal of load may include therrmodynamic system removal of load punishment cost and therrmodynamic system removal of load punishment at
This.
Step 8, judge whether it is exhaustive in the following preset time period respectively can time shift power load distribution adjustment side
Formula;If so, having obtained all possible power load scheduling scheme, 9 are thened follow the steps;If not, explanation also can be with
The power load scheduling scheme of realization, then return step 3.
Step 9, feasible solution (the i.e. electric power of the lowest cost is determined under feasible zone (i.e. the point of the therrmodynamic system equilibrium of supply and demand)
System equilibrium of supply and demand scheme), will in scheduling strategy set with it is minimum operation totle drilling cost power load scheduling scheme as
Final cloud data center power load scheduling scheme.
In specific implementation, the present embodiment is realized by way of model partition, and model one is Demand-side model, model two
For energy supply side form type, model three is energy balance model, and model four is that whole system calculates operating cost.Model two is mould
One supplying energy of type, model three guarantee energy (including electric energy and thermal energy) Real-time Balancing of model one and model two, and model four is about
Beam model one and model two reach Optimum cost, when the present invention can seek handle according to the time of electricity price, spatial variability
Between cost in section it is minimum, sought that the electric system in the period can be handled according to the time of renewable energy, spatial variability
Balance and therrmodynamic system balance, it is optimal to be finally reached each cloud data center micro-capacitance sensor whole system energy utilization.
In the description of this specification, reference term " the present embodiment ", " one embodiment ", " some embodiments ", " show
The description of example ", " specific example " or " some examples " etc. mean specific features described in conjunction with this embodiment or example, structure,
Material or feature are included at least one embodiment or example of the invention.In the present specification, above-mentioned term is shown
The statement of meaning property is necessarily directed to identical embodiment or example.Moreover, specific features, structure, material or the spy of description
Point may be combined in any suitable manner in any one or more of the embodiments or examples.In addition, without conflicting with each other,
Those skilled in the art can be by different embodiments or examples described in this specification and different embodiments or examples
Feature is combined.
In addition, term " first ", " second " are used for descriptive purposes only and cannot be understood as indicating or suggesting relative importance
Or implicitly indicate the quantity of indicated technical characteristic.Define " first " as a result, the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the present invention, the meaning of " plurality " is at least two, such as two, three
It is a etc., unless otherwise specifically defined.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modification, equivalent replacement and simple modifications etc., should all be included in the protection scope of the present invention in content.
Claims (10)
1. a kind of cloud data center power load dispatching method for considering natural resources randomness, it is characterised in that: the dispatching party
Method includes the following steps;
Step 1, it is predicted in the following preset time period using the history power load data of each cloud data center obtained in advance
Total power load to be scheduled, and go out force data prediction institute using the renewable energy history of each cloud data center obtained in advance
State the renewable energy power output in the following preset time period;
Step 2, filtering out from total power load to be scheduled can time shift power load;It wherein, can the expression of time shift power load
Postpone the power load that duration processed is greater than the first preset duration;
Step 3, on respectively can time shift power load handled in respective delay duration processed completion premised on, adjust described
In the following preset time period respectively can time shift power load Annual distribution and Regional Distribution, thus enable can time shift power load processing
Measure it is directly proportional to the power output in real time of the renewable energy of each cloud data center, and order can time shift power load treating capacity and future it is pre-
If the Spot Price of region is inversely proportional where each cloud data center in the period;
Step 4, judge in the case that currently respectively can time shift power load Annual distribution and Regional Distribution each cloud data center
Whether power load treating capacity is less than respective bearing capacity;If so, thening follow the steps 5;If it is not, then return step 3;
Step 5, electrical energy demands amount and Waste Heat Reuse amount of each cloud data center in the following preset time period are calculated separately;
Step 6, while judging whether the power supply volume of the electric system of each cloud data center meets respective electrical energy demands amount, each cloud
Whether the Waste Heat Reuse amount of data center meets the calorific value of the therrmodynamic system at respective place;If the two conditions simultaneously at
It is vertical, 7 are thened follow the steps, otherwise return step 3;
Step 7, current power load scheduling scheme is recorded, and calculates the operation totle drilling cost under the scheduling scheme, then again will
Current power load scheduling scheme and its operation totle drilling cost are saved into scheduling strategy set;
Step 8, judge whether it is exhaustive in the following preset time period respectively can time shift power load distribution adjustment side
Formula;If so, thening follow the steps 9;If it is not, then return step 3;
Step 9, using the power load scheduling scheme with minimum operation totle drilling cost in scheduling strategy set as final cloud number
According to center power load scheduling scheme.
2. the cloud data center power load dispatching method according to claim 1 for considering natural resources randomness, special
Sign is:
In step 3, adjustment respectively can time shift power load Annual distribution when, can time shift power load be transferred to renewable energy
Source power output is lower than the period of the second preset value greater than the period of the first preset value and/or electricity price;It respectively can Shi Yiyong adjusting
When the Regional Distribution of electric load, by can time shift power load be transferred to renewable energy power output be greater than third preset value cloud data
Center and/or electricity price are lower than the cloud data center of the 4th preset value.
3. the cloud data center power load dispatching method according to claim 2 for considering natural resources randomness, special
Sign is:
In step 3, adjustment can time shift power load Annual distribution when, further include enable respectively can time shift power load respective
Postpone the step of processing is completed in duration front half section processed.
4. the cloud data center power load dispatching method according to claim 3 for considering natural resources randomness, special
Sign is:
In step 2, further including the steps that filtering out from total power load to be scheduled can not time shift power load;
In step 3, adjust in the following preset time period respectively can not time shift power load Regional Distribution, with enable can not when
It is directly proportional to the power output in real time of the renewable energy of each cloud data center to divert from one use to another electric load treating capacity, and enabling can not time shift electricity consumption
Load treating capacity and the Spot Price of region where each cloud data center in the following preset time period are inversely proportional;Wherein, described
Can not time shift power load indicate power load of the duration processed less than the second preset duration.
5. the cloud data center power load dispatching method according to claim 4 for considering natural resources randomness, special
Sign is:
In step 3, adjustment respectively can not time shift power load Regional Distribution when, can not time shift power load be transferred to can be again
Raw energy power output is lower than the cloud data center of the 6th preset value greater than the cloud data center and/or electricity price of the 5th preset value.
6. the cloud data center power load dispatching method according to claim 5 for considering natural resources randomness, special
Sign is:
In step 3, transfer can time shift power load and/or can not be before time shift power load, by each power load to be transferred
Multiple sub- loads with same load scale are resolved into, are that minimum thread carries out power load turn with a sub- load
It moves.
7. the cloud data center power load dispatching method according to claim 6 for considering natural resources randomness, special
Sign is:
In step 6, the electric system includes power grid subsystem, renewable energy power generation subsystem, non-renewable energy resources power generation
Subsystem and electric energy storage subsystem.
8. the cloud data center power load dispatching method according to claim 7 for considering natural resources randomness, special
Sign is:
In step 6, the therrmodynamic system includes that cloud data center heat release subsystem, generating set heat release subsystem and electric energy storage are put
Thermal sub-system.
9. the cloud data center power load dispatching method according to claim 8 for considering natural resources randomness, special
Sign is:
In step 7, the operation totle drilling cost includes cost of electricity-generating, purchases strategies, power load scheduling cost, cloud data center clothes
Cost of being engaged in and the punishment cost of removal of load.
10. according to claim 1, considering the cloud data center power load dispatching party of natural resources randomness described in 2 or 9
Method, it is characterised in that:
It further include the history power load number for obtaining each cloud data center in past preset time period simultaneously before step 1
According to the step of going out force data with renewable energy history.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910445540.3A CN110084444B (en) | 2019-05-27 | 2019-05-27 | Cloud data center power load scheduling method considering randomness of natural resources |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910445540.3A CN110084444B (en) | 2019-05-27 | 2019-05-27 | Cloud data center power load scheduling method considering randomness of natural resources |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110084444A true CN110084444A (en) | 2019-08-02 |
CN110084444B CN110084444B (en) | 2021-05-14 |
Family
ID=67421939
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910445540.3A Active CN110084444B (en) | 2019-05-27 | 2019-05-27 | Cloud data center power load scheduling method considering randomness of natural resources |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110084444B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111550915A (en) * | 2020-05-14 | 2020-08-18 | 郑瀚韬 | Air conditioner intelligent control system based on non-invasive measurement |
CN112103997A (en) * | 2020-09-04 | 2020-12-18 | 天津大学 | Active power distribution network operation flexibility improving method considering data center adjustment potential |
CN114967900A (en) * | 2022-06-06 | 2022-08-30 | 北京亿安天下科技股份有限公司 | Method, system, device and medium for reducing power consumption of data center |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106228258A (en) * | 2016-07-11 | 2016-12-14 | 浙江工业大学 | A kind of meter and the home energy source LAN energy optimal control method of dsm |
CN108183500A (en) * | 2017-11-24 | 2018-06-19 | 国网甘肃省电力公司电力科学研究院 | A kind of rural area provided multiple forms of energy to complement each other is micro- can net capacity configuration optimizing method and device |
CN108898282A (en) * | 2018-06-06 | 2018-11-27 | 华北电力大学 | Data center resource Optimization Scheduling and computer storage medium |
CN109742812A (en) * | 2019-03-11 | 2019-05-10 | 长沙理工大学 | A kind of source-lotus-storage coordinated scheduling method improving new energy consumption |
-
2019
- 2019-05-27 CN CN201910445540.3A patent/CN110084444B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106228258A (en) * | 2016-07-11 | 2016-12-14 | 浙江工业大学 | A kind of meter and the home energy source LAN energy optimal control method of dsm |
CN108183500A (en) * | 2017-11-24 | 2018-06-19 | 国网甘肃省电力公司电力科学研究院 | A kind of rural area provided multiple forms of energy to complement each other is micro- can net capacity configuration optimizing method and device |
CN108898282A (en) * | 2018-06-06 | 2018-11-27 | 华北电力大学 | Data center resource Optimization Scheduling and computer storage medium |
CN109742812A (en) * | 2019-03-11 | 2019-05-10 | 长沙理工大学 | A kind of source-lotus-storage coordinated scheduling method improving new energy consumption |
Non-Patent Citations (1)
Title |
---|
ZHAOHAO DING等: ""Integrated Stochastic Energy Management for Data Center Microgrid Considering Waste Heat Recovery"", 《2018 IEEE INDUSTRY APPLICATION SOCIETY ANNUAL MEETING(IAS)》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111550915A (en) * | 2020-05-14 | 2020-08-18 | 郑瀚韬 | Air conditioner intelligent control system based on non-invasive measurement |
CN112103997A (en) * | 2020-09-04 | 2020-12-18 | 天津大学 | Active power distribution network operation flexibility improving method considering data center adjustment potential |
CN112103997B (en) * | 2020-09-04 | 2022-11-04 | 天津大学 | Active power distribution network operation flexibility improving method considering data center adjustment potential |
CN114967900A (en) * | 2022-06-06 | 2022-08-30 | 北京亿安天下科技股份有限公司 | Method, system, device and medium for reducing power consumption of data center |
Also Published As
Publication number | Publication date |
---|---|
CN110084444B (en) | 2021-05-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Wang et al. | Distributed energy and microgrids (DEM) | |
Pierson et al. | Datazero: Datacenter with zero emission and robust management using renewable energy | |
CN107480847B (en) | Energy source block chain network and virtual power plant operation and scheduling method based on network | |
Liu et al. | A scalable and robust approach to demand side management for smart grids with uncertain renewable power generation and bi-directional energy trading | |
CN110084444A (en) | A kind of cloud data center power load dispatching method considering natural resources randomness | |
Hogade et al. | Minimizing energy costs for geographically distributed heterogeneous data centers | |
CN103489045A (en) | Demand response load optimization potential evaluation method based on multi-scene design | |
Liu et al. | Game-theoretic market-driven smart home scheduling considering energy balancing | |
Robert et al. | The critical role of anchor customers in rural microgrids: Impact of load factor on energy cost | |
Du et al. | RETRACTED ARTICLE: Optimal scheduling of integrated energy system based on improved grey wolf optimization algorithm | |
Amir et al. | Reliability‐constrained optimal design of multicarrier microgrid | |
Han et al. | Waste heat reutilization and integrated demand response for decentralized optimization of data centers | |
Yu et al. | Carbon emission reduction analysis for cloud computing industry: can carbon emissions trading and technology innovation help? | |
Chen et al. | Utility-driven renewable energy sharing systems for community microgrid | |
Chen et al. | Proliferation of small data networks for aggregated demand response in electricity markets | |
Mammoli et al. | Distributed control strategies for high-penetration commercial-building-scale thermal storage | |
Yang et al. | Demand responsive market decision-makings and electricity pricing scheme design in low-carbon energy system environment | |
JP2022188498A (en) | aggregator system | |
Huang et al. | A Day-Ahead Renewables-Based Power Scheduling System for Internet of Energy | |
Yu et al. | Dual-layer optimization design method for collaborative benefits of renewable energy systems in building clusters: Case study of campus buildings | |
CN117639113B (en) | Intelligent micro-grid intelligent power distribution method, device and storage medium | |
Han et al. | A novel 4-level joint optimal dispatch for demand response of data centers with district autonomy realization | |
Munir et al. | Big data of home energy management in cloud computing | |
Butt et al. | Efficient Utilization of Energy using Fog and Cloud based Integrated Environment in Smart Grid | |
Wang et al. | Sharing Economy in Energy Markets: Review and Prospects |
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 |