CN108990383A - A kind of data center's air-conditioning system forecast Control Algorithm - Google Patents
A kind of data center's air-conditioning system forecast Control Algorithm Download PDFInfo
- Publication number
- CN108990383A CN108990383A CN201810927800.6A CN201810927800A CN108990383A CN 108990383 A CN108990383 A CN 108990383A CN 201810927800 A CN201810927800 A CN 201810927800A CN 108990383 A CN108990383 A CN 108990383A
- Authority
- CN
- China
- Prior art keywords
- air
- data center
- controller
- pue
- parameter
- 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
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05K—PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
- H05K7/00—Constructional details common to different types of electric apparatus
- H05K7/20—Modifications to facilitate cooling, ventilating, or heating
- H05K7/20709—Modifications to facilitate cooling, ventilating, or heating for server racks or cabinets; for data centers, e.g. 19-inch computer racks
- H05K7/208—Liquid cooling with phase change
- H05K7/20827—Liquid cooling with phase change within rooms for removing heat from cabinets, e.g. air conditioning devices
-
- H—ELECTRICITY
- H05—ELECTRIC TECHNIQUES NOT OTHERWISE PROVIDED FOR
- H05K—PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS
- H05K7/00—Constructional details common to different types of electric apparatus
- H05K7/20—Modifications to facilitate cooling, ventilating, or heating
- H05K7/20709—Modifications to facilitate cooling, ventilating, or heating for server racks or cabinets; for data centers, e.g. 19-inch computer racks
- H05K7/20836—Thermal management, e.g. server temperature control
Landscapes
- Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Thermal Sciences (AREA)
- Microelectronics & Electronic Packaging (AREA)
- Air Conditioning Control Device (AREA)
Abstract
The invention discloses a kind of data center's air-conditioning system forecast Control Algorithms, online rolling optimization optimizing is carried out using each parameter of the PREDICTIVE CONTROL optimization object function to field controller including predictive controller, the field controller of field control layer is transmitted to using the resulting parameter of optimizing as controlled variable setting value, field controller controls refrigerating plant based on the setting value of the controlled parameter after optimization.Predictive controller carries out the optimal setting that rolling optimization calculates each sampling instant based on prediction result, field controller makes controlled parameter track optimal setting rapidly using pid algorithm, the large dead time and close coupling characteristic for overcoming data center's air conditioner refrigerating station system, improve control precision；Optimization object function comprehensively considers energy efficiency indexes PUE and composite demand, allows system under the premise of meeting computer room refrigeration demand, reduces air conditioning energy consumption, reduces PUE, realizes energy-efficient purpose.
Description
Technical field
The invention belongs to data center's air conditioner controlling technology fields, pre- more particularly to a kind of data center's air-conditioning system
Survey control method.
Background technique
Formula development is constantly strided forward with computer internet new and high technology, data center's industry has welcome the height of construction
The peak phase.Therewith, the high energy consumption issues of data center enter everybody visual field.Data show that consumption of data center accounts for entirely
The ratio of building trade energy consumption about 20%, and wherein the energy consumption of air-conditioning system is as high as 40%~50%, high energy consumption has changed into
The maximum for restricting the high-efficient development of data center green is kept in check.Face data center's PUE value that the country at present nearly has 80% or more
Greater than 2.0, the low situation of energy efficiency, Optimizing Reconstruction and Energy Saving Control strategy based on data center's air-conditioning system
Research, receives the common concern of insider.However the internal structure design of transformation air-conditioning system is depended merely on, on the one hand it can draw
Carry out the increase of infrastructure investment, compared to for economic benefit, energy-saving benefit is not obvious；On the other hand consider weather, region and ring
Requirement of the border to data center is different, not high to the applicability of Practical Project.Therefore, it has only using Optimal Control Strategy,
The cooling demand of computer room information technoloy equipment could not only be met, but also farthest reduce data center's air conditioning energy consumption, realized real
In meaning data center energy conservation, and the prior art in the Optimizing Reconstruction based on data center's air-conditioning system with
Energy-saving control system and method, for example, Chinese Patent Application No. be 201611079041.X in, disclose a kind of data center
The control method of air-conditioning system, " characterized by further comprising monitoring module, software filtering module and output module, monitoring modules
For the signal monitored to be transferred to controller after software filtering module filtered；Software filtering module be used for signal into
Row filtering；After controller is used to carry out identifying processing to signal, sent out through output module to each module of data center's air-conditioning system
Signal is controlled out；Output module is used for the control signal of o controller." be only to data center carry out effective monitoring and
It adjusts, can not achieve energy-efficient purpose.
Therefore, the emphasis how to solve the above problems as those skilled in the art's research.
Summary of the invention
It is an object of the invention to provide a kind of data center's air-conditioning system forecast Control Algorithms, can effectively solve above-mentioned
Data center's air-conditioning system can not energy-efficient shortcoming.
The purpose of the present invention is realized by following technical proposals:
A kind of data center's air-conditioning system forecast Control Algorithm, including predictive controller utilize PREDICTIVE CONTROL optimization aim
Function carries out online optimizing to each parameter of controller, using the resulting parameter of optimizing as the setting value of field controller, scene
The field controller of control layer controls refrigerating plant based on the setting value of the controlled parameter after optimization.
Preferably, the optimization object function is
Wherein M is prediction time domain, and t is the initial time predicted in time domain, PUEset(k) setting value for being k-th of sampling period PUE,
PUE (k) is the actual value of the PUE in k-th of sampling period；For data center's supply air temperature in k-th of sampling period
Setting value, TASIt (k) is data center's supply air temperature average value in k-th of sampling period, JaFor optimization object function, α is
PUE penalty term weight coefficient, β are supply air temperature penalty term weight coefficient.
Preferably, the predictive controller uses multilayer neural network, the separate refrigeration device and the first joint system
When device for cooling works, current time chilled water supply water temperature, chilled water are discharged by predictive controller for return water pressure difference, cooling tower
Temperature, cooling water supply backwater temperature difference, the air blow and return temperature difference send setting value of the field controller as controlled parameter to；It will be current
It is moment outdoor wet-bulb temperature, data center machine room load, PUE value, supply air temperature and subsequent time PUE setting value, next
Moment supply air temperature setting value and the input to threshold value relevant -1 as field controller；When the second associated refrigerating plant list
Solely when work, identical, the PREDICTIVE CONTROL when input of predictive controller works with separate refrigeration device and the first associated refrigerating plant
The output of device is chilled water supply water temperature, chilled water for return water pressure difference, cooling tower leaving water temperature, the air blow and return temperature difference.
Preferably, the field controller controls refrigerating plant with pid control law.
Preferably, the predictive controller is using predictive control strategy to the controlled pre-set parameter of field controller
It is set.
Compared with prior art, the beneficial effects of the present invention are:
The present invention is directed to data center's chilled water cooling type air-conditioning system, and PREDICTIVE CONTROL optimality criterion takes computer room to send
Air temperature and energy efficiency indexes PUE (Power Usage Effectiveness) are between the two parameter actual values and desired value
Sum of square of deviations.Control system uses two layers of control structure, and upper layer optimization layer is using predictive control algorithm to bottom field control
The setting value of device carries out optimizing, and field controller is calculated using established prediction model, and using Rolling optimal strategy
The optimal setting of controlled parameter, bottom controller make controlled parameter track optimal setting rapidly, make using pid algorithm
System equipment operates in energy-saving run state, realizes data center's air-conditioning system to meet time-varying refrigeration requirement and energy conservation is
The global optimization of target controls.
Detailed description of the invention
Fig. 1 is data center's air-conditioning system neural network prediction model structure chart under summer and conditioning in Transition Season operating condition；
Fig. 2 is the structure chart of data center's air-conditioning system neural network prediction model under winter condition；
Fig. 3 is the neural network topology structure figure of predictive controller under summer operating mode and conditioning in Transition Season operating condition；
Fig. 4 is network response surface device topology diagram under winter condition；
Fig. 5 is Predictive Control System block diagram.
Specific embodiment
The present invention is further illustrated with attached drawing combined with specific embodiments below.
Embodiment one
As shown in Figures 1 to 5, a kind of data center's air-conditioning system forecast Control Algorithm, including control system and controlled system
System, the control system include the optimization layer on upper layer and the field control layer of lower layer, and the optimization layer is equipped with PREDICTIVE CONTROL system
System, the Predictive Control System includes predictive controller, and the field control layer includes field controller, the field control
Device connects refrigerating plant, and the refrigerating plant includes separate refrigeration device；Refrigerating plant include refrigeration unit+plate heat exchanger+
Cooling tower associated refrigerating plant；Refrigerating plant includes cooling tower+plate heat exchanger refrigerating plant, and three kinds of refrigerating plants are equal
Energy isolated operation, the optimization layer connect field control layer, and optimization layer is using predictive control strategy to field control layer controller
Controlled pre-set parameter carry out optimizing, optimization aim be so that system PUE value and supply air temperature is reached setting value, reach energy conservation
With meet cooling capacity requirement purpose, the controlled parameter of field control layer include chilled water supply water temperature, cooling water supply temperature,
Refrigerating water pump flow control parameter, cooling pump flow control parameter, air flow rate control parameter, wherein flow control parameter according to
Different variables may be selected in the control method difference of field control layer, can be pressure difference, the temperature difference or frequency control, preferably freeze
Water is for return water pressure difference, cooling water supply backwater temperature difference and for the return air temperature difference, and the controller of field control layer is based on controlled after optimization
Pre-set parameter controls refrigeration machine, cooling tower, chilled water pump and cooling water pump using PID control method.
In the present embodiment, energy-efficient purpose is realized by controlling different refrigerating plant switchover operations.
Embodiment two
As shown in Figures 1 to 5, a kind of data center's air-conditioning system forecast Control Algorithm, including control system and controlled system
System, the control system include the optimization layer on upper layer and the field control layer of lower layer, and the optimization layer is equipped with PREDICTIVE CONTROL system
System, the Predictive Control System includes predictive controller, and the field control layer includes field controller, the field control
Device connects refrigerating plant, and the refrigerating plant includes separate refrigeration device；First associated refrigerating plant includes that refrigerating plant includes
Refrigeration unit+plate heat exchanger+cooling tower associated refrigerating plant；Second associated refrigerating plant includes cooling tower+plate heat exchanger
Refrigerating plant, three kinds of refrigerating plants energy isolated operation, the optimization layer connect field control layer, and optimization layer is using in advance
Survey control strategy and optimizing carried out to the controlled pre-set parameter of field control layer controller, optimization aim be make system PUE value and
Supply air temperature reaches setting value, achievees the purpose that energy conservation and meets cooling capacity requirement, the controlled parameter of field control layer includes cold
Freeze water supply water temperature, cooling water supply temperature, refrigerating water pump flow control parameter, cooling pump flow control parameter, air flow rate control
Parameter processed, wherein different variables may be selected according to the control method difference of field control layer in flow control parameter, can be pressure
Difference, the temperature difference or frequency control, preferably chilled water are for return water pressure difference, cooling water supply backwater temperature difference and for the return air temperature difference, scene control
The controller of preparative layer is based on the controlled pre-set parameter after optimization, using PID control method to refrigeration machine, cooling tower, chilled water
Pump and cooling water pump are controlled, and the present invention is integrated with plate heat exchanger in water side, makes system when outdoor wet-bulb temperature is lower
It can make full use of natural cooling source cooling supply, reduce the operation power consumption of cold.The present invention independently transports three kinds of operating conditions of controlled system point
Row, respectively summer operating mode, conditioning in Transition Season operating condition and winter condition.Refrigeration unit separate refrigeration under summer operating mode；Conditioning in Transition Season work
Refrigeration unit+plate heat exchanger (water side economizer)+cooling tower joint refrigeration under condition；Cooling tower+plate-type heat-exchange under winter condition
Device (water side economizer) complete natural cooling.Latter two operating condition can be natural in the lower conditioning in Transition Season of outside air temperature and use in winter
Cold source realizes partially or completely natural cooling, to reduce the runing time of water cooler, reduces the operation of air-conditioning system whole year
Power consumption.
There are within 1 year 8760 hours, according to the weather conditions of Beijing area, the runing time of three kinds of operating conditions is divided into it
In 6~August (3624~5832h) be summer operating mode, 12~2 months (8016~1416h) be winter condition, be incorporated in other months
Conditioning in Transition Season operating condition.
In the present embodiment, in different seasons, different refrigerating plants, and energy are selected according to the difference of outside air temperature
Refrigeration enough, which is carried out, using natural cooling source saves energy.
Embodiment three
As shown in Figures 1 to 5, a kind of data center's air-conditioning system forecast Control Algorithm, including control system and controlled system
System, the control system include the optimization layer on upper layer and the field control layer of lower layer, and the optimization layer is equipped with PREDICTIVE CONTROL system
System, the Predictive Control System includes predictive controller, and the field control layer includes field controller, the field control
Device connects refrigerating plant, and the refrigerating plant includes separate refrigeration device；Refrigerating plant include refrigeration unit+plate heat exchanger+
Cooling tower associated refrigerating plant；Refrigerating plant includes cooling tower+plate heat exchanger refrigerating plant, and three kinds of refrigerating plants are equal
Energy isolated operation, the optimization layer connect field control layer, and optimization layer is using predictive control strategy to field control layer controller
Controlled pre-set parameter carry out optimizing, optimization aim be so that system PUE value and supply air temperature is reached setting value, reach energy conservation
With meet cooling capacity requirement purpose, the controlled parameter of field control layer include chilled water supply water temperature, cooling water supply temperature,
Refrigerating water pump flow control parameter, cooling pump flow control parameter, air flow rate control parameter, wherein flow control parameter according to
Different variables may be selected in the control method difference of field control layer, can be pressure difference, the temperature difference or frequency control, preferably freeze
Water is for return water pressure difference, cooling water supply backwater temperature difference and for the return air temperature difference, and the controller of field control layer is based on controlled after optimization
Pre-set parameter controls refrigeration machine, cooling tower, chilled water pump and cooling water pump using PID control method, the present invention
It is integrated with plate heat exchanger in water side, so that system is can make full use of natural cooling source cooling supply when outdoor wet-bulb temperature is lower, subtracts
The operation power consumption of few cold.The present invention divides controlled system to three kinds of operating condition independent operatings, respectively summer operating mode, conditioning in Transition Season work
Condition and winter condition.Refrigeration unit separate refrigeration under summer operating mode；Refrigeration unit+plate heat exchanger (water under conditioning in Transition Season operating condition
Side economizer)+cooling tower joint refrigeration；Cooling tower+plate heat exchanger (water side economizer) is completely free under winter condition supplies
It is cold.Latter two operating condition can be realized partially or completely naturally cold in the lower conditioning in Transition Season of outside air temperature and use in winter natural cooling source
But, to reduce the runing time of water cooler, the operation power consumption of air-conditioning system whole year is reduced, according to the control of field control layer
Method, the flow control parameter that prediction model inputs in parameter may be selected different variables, can be pressure difference, the temperature difference or frequency
Control, preferably chilled water is for return water pressure difference, cooling water supply backwater temperature difference and for the return air temperature difference.At this point, summer and conditioning in Transition Season operating condition
Under, the input parameter of prediction model has current time chilled water supply water temperature, chilled water to go out water temperature for return water pressure difference, cooling tower
Degree, cooling water supply backwater temperature difference, supply air temperature, the air blow and return temperature difference, load (rate), outdoor wet-bulb temperature and system are current
The PUE value at moment and relevant to threshold value -1, the output parameter of prediction model have the PUE value and supply air temperature of subsequent time,
Data center's air-conditioning system neural network prediction model structure under summer and conditioning in Transition Season operating condition is as shown in Figure 1, winter condition
Under, the input parameter of prediction model can reduce cooling water supply backwater temperature difference control variable, and the data center under winter condition is empty
The structure of adjusting system neural network prediction model is as shown in Fig. 2.
In the present embodiment, different variables, setting scene are selected by the flow control parameter that prediction model inputs in parameter
Each parameter of controller is the parameter optimized.
Example IV
As shown in Figures 1 to 5, a kind of data center's air-conditioning system forecast Control Algorithm, including control system and controlled system
System, the control system include the optimization layer on upper layer and the field control layer of lower layer, and the optimization layer is equipped with PREDICTIVE CONTROL system
System, the Predictive Control System includes predictive controller, and the field control layer includes field controller, the field control
Device connects refrigerating plant, and the refrigerating plant includes separate refrigeration device；Refrigerating plant include refrigeration unit+plate heat exchanger+
Cooling tower associated refrigerating plant；Refrigerating plant includes cooling tower+plate heat exchanger refrigerating plant, and three kinds of refrigerating plants are equal
Energy isolated operation, the optimization layer connect field control layer, and optimization layer is using predictive control strategy to field control layer controller
Controlled pre-set parameter carry out optimizing, optimization aim be so that system PUE value and supply air temperature is reached setting value, reach energy conservation
With meet cooling capacity requirement purpose, the controlled parameter of field control layer include chilled water supply water temperature, cooling water supply temperature,
Refrigerating water pump flow control parameter, cooling pump flow control parameter, air flow rate control parameter, wherein flow control parameter according to
Different variables may be selected in the control method difference of field control layer, can be pressure difference, the temperature difference or frequency control, preferably freeze
Water is for return water pressure difference, cooling water supply backwater temperature difference and for the return air temperature difference, and the controller of field control layer is based on controlled after optimization
Pre-set parameter controls refrigeration machine, cooling tower, chilled water pump and cooling water pump using PID control method, the present invention
It is integrated with plate heat exchanger in water side, so that system is can make full use of natural cooling source cooling supply when outdoor wet-bulb temperature is lower, subtracts
The operation power consumption of few cold.The present invention divides controlled system to three kinds of operating condition independent operatings, respectively summer operating mode, conditioning in Transition Season work
Condition and winter condition.Refrigeration unit separate refrigeration under summer operating mode；Refrigeration unit+plate heat exchanger (water under conditioning in Transition Season operating condition
Side economizer)+open cooling tower joint refrigeration；Cooling tower+plate heat exchanger (water side economizer) is completely free under winter condition
Cooling supply.Latter two operating condition can be realized partially or completely natural in the lower conditioning in Transition Season of outside air temperature and use in winter natural cooling source
It is cooling, to reduce the runing time of water cooler, the operation power consumption of air-conditioning system whole year is reduced, data center's air-conditioning system is pre-
Observing and controlling optimization object function are as follows:
Wherein k represents current time, and M is prediction time domain, and t is the initial time predicted in time domain, PUEsetIt (k) is kth
The setting value of a sampling period PUE, PUE (k) are the actual value of the PUE in k-th of sampling period；It is sampled for k-th
The setting value of the data center machine room supply air temperature in period, TAS(k) it blows for the data center machine room in k-th of sampling period warm
Average value is spent, predictive controller selects multilayer neural network, and input layer number is 5, and output layer number of nodes is also 5, hidden
Node layer rule of thumb method and trial and error procedure determines and takes 8, and hidden layer excitation function uses tansig function, and output layer motivates letter
Number is linear function, and predictive controller output variable is current time chilled water supply water temperature, cold under summer and conditioning in Transition Season operating condition
Freeze water and exported for return water pressure difference, cooling tower leaving water temperature, cooling water supply backwater temperature difference, the air blow and return temperature difference as controller, is passed
Give setting value of the field controller as controlled parameter；By wet-bulb temperature, Air Conditioning Facilities cooling load outside current time room
(rate), PUE value, supply air temperature and subsequent time PUE setting value, subsequent time supply air temperature setting value and related to threshold value
- 1 as controller input.Fig. 3 is the neural network topology structure of predictive controller under summer operating mode and conditioning in Transition Season operating condition.
Predictive controller input under winter condition is identical as summer and conditioning in Transition Season, exports and supplies back for chilled water supply water temperature, chilled water
Differential water pressures, cooling tower leaving water temperature, the air blow and return temperature difference, network response surface device topological structure such as Fig. 4 institute under winter condition
Show, Predictive Control System block diagram is as shown in figure 5, wherein x [k] is that the relevant state variables of k time data center air-conditioner system are joined
Number, i.e., current time chilled water supply water temperature, chilled water are for return water pressure difference, cooling tower leaving water temperature, cooling water for return water temperature
Difference, outdoor wet-bulb temperature, supply air temperature, the air blow and return temperature difference, load (rate) and PUE value, wherein load (rate) and PUE value are root
The estimated value of cooling capacity is provided according to electricity consumption and refrigeration plant, surplus can be surveyed and be obtained；It is defeated for k+1 moment prediction model
Out, i.e. subsequent time PUE value and supply air temperature；X* [k+1] is the setting value and the air-supply of subsequent time computer room of subsequent time PUE
The setting value of temperature；U [k] is the optimum control amount at k moment after network response surface device optimizing, i.e. field control
Layer controller is controlled the setting value of parameter；U ' [k+i-1] is rolling optimization in the process according to last moment predictive controller weight
The control amount extrapolated, the control amount are also optimized not according to current time system running state variate-value.
The optimizing step of nerve network controller can be summarized as follows:
1. each connection weight of initialization data center air-conditioner system neural network predictive controller, be assigned a value of [- 1,
1] lesser random number and threshold value (this explanation takes -1) in range, then computing controller exports.Predict that time domain M is 6, in advance
The survey period takes 5 minutes；
2. x [k], x* [k+1] and -1 are acted on network response surface device, controlled using neural network optimizing
Variable processed exports u [k]；
3. control amount output u [k] and state primary quantity x [k] are passed to control target, controlled system reality is obtained
It exports x [k+1], while passing to prediction model, obtain prediction model output
4. keeping the weight W of the neural network structure of predictive controllerjiIt is constant, x [k+1], x* [k+2], -1 are sent to
Predictive controller obtains new dominant vectorIt willNeural network prediction model is passed to x [k+2], is counted
It calculatesAnd so on, above-mentioned steps are recycled, until control output isIt is with prediction output(i=
M-1)；
5. as k < N-M, λ [k]=λ [t1+ M], as k >=N-M, λ [k]=λ [N], recycling 3.~4. walk, retrodict out
Lagrange multiplier vector λ [k] and γ [k], i.e., k=t in order1+M-1,…t1+2,t1+ 1 derives forward, obtains:
Wherein L () indicates that optimality criterion, f () indicate that prediction model, g () indicate nerve network controller.
U [k] and x [k] in above-mentioned derivation process are usedWithSubstitution；
6. finally, correcting the neural network structure of predictive controller by following formula using a calculated γ [k]
Weight:
W=W+ Δ W
Wherein W is the weight battle array of air-conditioning system network response surface device, and μ is right value update rate, μ selection 0.05；
7. recycle 4.~6. walk, the weight of network response surface device is constantly modified, until Δ W=0
8. enable k=k+1, circulation 2.~7. walk, find out the control variable u (k) at controlled system each moment.
Next sampling period then repeats aforesaid operations, the value of each moment control amount after calculating separately out, until
Control process terminates.
In the present embodiment, using optimization object function, track controlled parameter rapidly by repetition optimizing processing optimal
Setting value makes system equipment operate in energy-saving run state, realizes data center's air-conditioning system to meet time-varying refrigeration requirement
And energy conservation is the global optimization control of target.
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 modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (5)
1. a kind of data center's air-conditioning system forecast Control Algorithm, it is characterised in that: utilize PREDICTIVE CONTROL including predictive controller
Optimization object function carries out online optimizing to each parameter of field controller, using the resulting parameter of optimizing as field controller
The field controller of setting value, field control layer controls refrigerating plant based on the setting value of the controlled parameter after optimization.
2. a kind of data center's air-conditioning system forecast Control Algorithm according to claim 1, it is characterised in that: the optimization
Objective function isWherein M is prediction time domain, and t is pre-
Survey the initial time in time domain, PUEset(k) setting value for being k-th of sampling period PUE, PUE (k) are k-th of sampling period
The actual value of PUE；For the setting value of data center's supply air temperature in k-th of sampling period, TASIt (k) is k-th of sampling
Data center's supply air temperature average value in period, JaFor optimization object function, α is PUE penalty term weight coefficient, and β is air-supply temperature
Spend penalty term weight coefficient.
3. a kind of data center's air-conditioning system forecast Control Algorithm according to claim 1, it is characterised in that: the prediction
Controller uses multilayer neural network, and when the separate refrigeration device and the first associated refrigerating plant work, predictive controller will
Current time chilled water supply water temperature, chilled water for return water pressure difference, cooling tower leaving water temperature, cooling water supply backwater temperature difference, send back to
Wind-warm syndrome difference sends setting value of the field controller as controlled parameter to；By wet-bulb temperature, data center's machine outside current time room
Room load, PUE value, supply air temperature and subsequent time PUE setting value, subsequent time supply air temperature setting value and with threshold value phase
- 1 input as field controller closed；When the second associated refrigerating plant works independently, the input of predictive controller and list
Only refrigerating plant and the first associated refrigerating plant are identical when working, and the output of predictive controller is chilled water supply water temperature, freezing
Water is for return water pressure difference, cooling tower leaving water temperature, the air blow and return temperature difference.
4. a kind of data center's air-conditioning system forecast Control Algorithm according to claim 1, it is characterised in that: the scene
Controller controls refrigerating plant with pid control law.
5. a kind of data center's air-conditioning system forecast Control Algorithm according to claim 1, it is characterised in that: the prediction
Controller is set using controlled pre-set parameter of the predictive control strategy to field controller.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810927800.6A CN108990383B (en) | 2018-08-15 | 2018-08-15 | Predictive control method for air conditioning system of data center |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810927800.6A CN108990383B (en) | 2018-08-15 | 2018-08-15 | Predictive control method for air conditioning system of data center |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108990383A true CN108990383A (en) | 2018-12-11 |
| CN108990383B CN108990383B (en) | 2020-08-04 |
Family
ID=64553480
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810927800.6A Active CN108990383B (en) | 2018-08-15 | 2018-08-15 | Predictive control method for air conditioning system of data center |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108990383B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109890176A (en) * | 2019-03-01 | 2019-06-14 | 北京慧辰资道资讯股份有限公司 | A kind of method and device based on artificial intelligence optimization's energy consumption of machine room efficiency |
| CN109917656A (en) * | 2019-03-29 | 2019-06-21 | 重庆大学 | Recirculated cooling water minimum differntial pressure energy-saving control system and method based on processing medium multi-temperature target |
| CN110008575A (en) * | 2019-03-29 | 2019-07-12 | 重庆大学 | Recirculating cooling water system processing medium multi-temperature target set point switches multi-parameter prediction control algolithm |
| CN110594983A (en) * | 2019-09-20 | 2019-12-20 | 东北大学 | Temperature control method suitable for small data center |
| CN111144543A (en) * | 2019-12-30 | 2020-05-12 | 中国移动通信集团内蒙古有限公司 | Data center air conditioner tail end temperature control method, device and medium |
| CN111649465A (en) * | 2020-06-05 | 2020-09-11 | 哈尔滨工业大学 | Automatic control method and system for air conditioning equipment |
| CN112084661A (en) * | 2020-09-10 | 2020-12-15 | 中国船舶重工集团海装风电股份有限公司 | Wind turbine converter water cooling system cooling state assessment early warning method |
| CN112393390A (en) * | 2020-07-15 | 2021-02-23 | 上海有孚智数云创数字科技有限公司 | Cloud computing data center precision air conditioner energy-saving control method based on data analysis |
| CN112487353A (en) * | 2020-11-30 | 2021-03-12 | 广东美的制冷设备有限公司 | Data processing method and device, dehumidifier and storage medium |
| CN113283156A (en) * | 2021-03-29 | 2021-08-20 | 北京建筑大学 | Subway station air conditioning system energy-saving control method based on deep reinforcement learning |
| CN114001443A (en) * | 2021-09-27 | 2022-02-01 | 中盈优创资讯科技有限公司 | Machine room energy saving method and device based on AI |
| CN114167964A (en) * | 2021-12-01 | 2022-03-11 | 上海德衡数据科技有限公司 | Wind side natural cooling method and system based on data center |
| CN114322209A (en) * | 2021-12-27 | 2022-04-12 | 上海美控智慧建筑有限公司 | Data processing method and device of ice storage air conditioning system and electronic equipment |
| CN114462171A (en) * | 2022-01-05 | 2022-05-10 | 浙江力德节能科技有限公司 | Method for calculating predicted value of temperature difference between supply water and return water at demand side of central air conditioner |
| CN115581056A (en) * | 2022-11-09 | 2023-01-06 | 宁波亮控信息科技有限公司 | Energy-saving prediction control method and system suitable for water cooling system of data center |
| CN115623753A (en) * | 2022-10-27 | 2023-01-17 | 南京群顶科技股份有限公司 | Water-cooling energy-saving timing regulation and control method based on dynamic time table |
| CN115696871A (en) * | 2022-11-04 | 2023-02-03 | 中国电子工程设计院有限公司 | Machine learning-based data center water cooling system regulation and control method and device |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130190930A1 (en) * | 2011-01-27 | 2013-07-25 | International Business Machines Corporation | Energy Saving Control for Data Center |
| US20160187397A1 (en) * | 2014-12-30 | 2016-06-30 | Schneider Electric It Corporation | Power consumption model for cooling equipment |
| CN106483843A (en) * | 2015-08-31 | 2017-03-08 | 中兴通讯股份有限公司 | A kind of method and system realizing IDC power consumption control |
| CN107169562A (en) * | 2017-06-15 | 2017-09-15 | 郑州云海信息技术有限公司 | A kind of energy efficiency controlling method and device for data center |
| CN108089440A (en) * | 2017-12-06 | 2018-05-29 | 北京百度网讯科技有限公司 | Energy-saving control method and device |
-
2018
- 2018-08-15 CN CN201810927800.6A patent/CN108990383B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130190930A1 (en) * | 2011-01-27 | 2013-07-25 | International Business Machines Corporation | Energy Saving Control for Data Center |
| US20160187397A1 (en) * | 2014-12-30 | 2016-06-30 | Schneider Electric It Corporation | Power consumption model for cooling equipment |
| CN106483843A (en) * | 2015-08-31 | 2017-03-08 | 中兴通讯股份有限公司 | A kind of method and system realizing IDC power consumption control |
| CN107169562A (en) * | 2017-06-15 | 2017-09-15 | 郑州云海信息技术有限公司 | A kind of energy efficiency controlling method and device for data center |
| CN108089440A (en) * | 2017-12-06 | 2018-05-29 | 北京百度网讯科技有限公司 | Energy-saving control method and device |
Non-Patent Citations (2)
| Title |
|---|
| 冉义兵: "基于神经网络的VAV系统预测控制策略研究及实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 安册册: "数据中心空调系统预测模型研究", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109890176A (en) * | 2019-03-01 | 2019-06-14 | 北京慧辰资道资讯股份有限公司 | A kind of method and device based on artificial intelligence optimization's energy consumption of machine room efficiency |
| CN109917656B (en) * | 2019-03-29 | 2022-03-01 | 重庆大学 | Circulating cooling water minimum pressure difference energy-saving control system and method based on process medium multi-temperature target |
| CN109917656A (en) * | 2019-03-29 | 2019-06-21 | 重庆大学 | Recirculated cooling water minimum differntial pressure energy-saving control system and method based on processing medium multi-temperature target |
| CN110008575A (en) * | 2019-03-29 | 2019-07-12 | 重庆大学 | Recirculating cooling water system processing medium multi-temperature target set point switches multi-parameter prediction control algolithm |
| CN110008575B (en) * | 2019-03-29 | 2023-01-31 | 重庆大学 | Multi-parameter predictive control algorithm for switching process medium multi-temperature target set values of circulating cooling water system |
| CN110594983A (en) * | 2019-09-20 | 2019-12-20 | 东北大学 | Temperature control method suitable for small data center |
| CN111144543A (en) * | 2019-12-30 | 2020-05-12 | 中国移动通信集团内蒙古有限公司 | Data center air conditioner tail end temperature control method, device and medium |
| CN111649465A (en) * | 2020-06-05 | 2020-09-11 | 哈尔滨工业大学 | Automatic control method and system for air conditioning equipment |
| CN112393390A (en) * | 2020-07-15 | 2021-02-23 | 上海有孚智数云创数字科技有限公司 | Cloud computing data center precision air conditioner energy-saving control method based on data analysis |
| CN112084661A (en) * | 2020-09-10 | 2020-12-15 | 中国船舶重工集团海装风电股份有限公司 | Wind turbine converter water cooling system cooling state assessment early warning method |
| CN112084661B (en) * | 2020-09-10 | 2022-07-05 | 中国船舶重工集团海装风电股份有限公司 | Wind turbine converter water cooling system cooling state assessment early warning method |
| CN112487353A (en) * | 2020-11-30 | 2021-03-12 | 广东美的制冷设备有限公司 | Data processing method and device, dehumidifier and storage medium |
| CN112487353B (en) * | 2020-11-30 | 2022-08-09 | 广东美的制冷设备有限公司 | Data processing method and device, dehumidifier and storage medium |
| CN113283156A (en) * | 2021-03-29 | 2021-08-20 | 北京建筑大学 | Subway station air conditioning system energy-saving control method based on deep reinforcement learning |
| CN113283156B (en) * | 2021-03-29 | 2023-09-15 | 北京建筑大学 | Energy-saving control method for subway station air conditioning system based on deep reinforcement learning |
| CN114001443A (en) * | 2021-09-27 | 2022-02-01 | 中盈优创资讯科技有限公司 | Machine room energy saving method and device based on AI |
| CN114167964A (en) * | 2021-12-01 | 2022-03-11 | 上海德衡数据科技有限公司 | Wind side natural cooling method and system based on data center |
| CN114167964B (en) * | 2021-12-01 | 2024-05-24 | 上海德衡数据科技有限公司 | Wind side natural cooling method and system based on data center |
| CN114322209A (en) * | 2021-12-27 | 2022-04-12 | 上海美控智慧建筑有限公司 | Data processing method and device of ice storage air conditioning system and electronic equipment |
| CN114462171A (en) * | 2022-01-05 | 2022-05-10 | 浙江力德节能科技有限公司 | Method for calculating predicted value of temperature difference between supply water and return water at demand side of central air conditioner |
| CN114462171B (en) * | 2022-01-05 | 2024-05-24 | 浙江力德节能科技有限公司 | Calculation method for supply and return water temperature difference predicted value of central air conditioner on demand side |
| CN115623753A (en) * | 2022-10-27 | 2023-01-17 | 南京群顶科技股份有限公司 | Water-cooling energy-saving timing regulation and control method based on dynamic time table |
| CN115623753B (en) * | 2022-10-27 | 2024-03-15 | 南京群顶科技股份有限公司 | Water-cooling energy-saving timing regulation and control method based on dynamic time table |
| CN115696871A (en) * | 2022-11-04 | 2023-02-03 | 中国电子工程设计院有限公司 | Machine learning-based data center water cooling system regulation and control method and device |
| CN115696871B (en) * | 2022-11-04 | 2023-06-13 | 中国电子工程设计院有限公司 | Data center water cooling system regulation and control method and device based on machine learning |
| CN115581056A (en) * | 2022-11-09 | 2023-01-06 | 宁波亮控信息科技有限公司 | Energy-saving prediction control method and system suitable for water cooling system of data center |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108990383B (en) | 2020-08-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108990383A (en) | A kind of data center's air-conditioning system forecast Control Algorithm | |
| CN110288164B (en) | Predictive control method for building air-conditioning refrigeration station system | |
| CN108413567B (en) | Central air conditioner cost-saving optimization method and system based on Internet of things | |
| CN103064285B (en) | A kind of heat pump heating multiobjective optimization control method based on model | |
| CN110392515B (en) | Historical data-based cold and heat source machine room energy-saving control method and system | |
| CN103062861B (en) | Energy-saving method and system for central air conditioner | |
| CN103020481B (en) | A kind of method based on energy-conservation determination air source heat pump floor heating optimal operating condition | |
| CN110410942B (en) | Energy-saving optimization control method and system for cold and heat source machine room | |
| CN102367980B (en) | Central air-conditioning multi-dimension integrating optimization control system and method | |
| CN101737899B (en) | Wireless sensor network-based central air-conditioning control system and method | |
| CN107192097A (en) | The control method and device of air conditioning water system | |
| CN109945402A (en) | A kind of central air conditioning water system power-economizing method | |
| CN108489012A (en) | Cold source of air conditioning energy efficiency model control method based on load prediction and constraint | |
| CN107421029B (en) | Tail end cold quantity balance control method | |
| CN110107989A (en) | Small-sized based on chilled water return water temperature optimum set point determines frequency water cooler and becomes temperature control method of water | |
| CN105004002A (en) | Energy saving control system and energy saving control method used for central air conditioner cooling water system | |
| CN103994553A (en) | Refrigerating system cooling water energy-saving control method, system and device | |
| CN113446661B (en) | Intelligent and efficient heat supply network operation adjusting method | |
| CN104965409A (en) | Industrial circulating water system energy consumption self-learning optimization control method | |
| CN110940061A (en) | Central air conditioner control method and system | |
| CN111649464B (en) | Control method and system of intelligent refrigeration system, server and computer storage medium | |
| CN212720195U (en) | Cooling water system control device based on system overall energy efficiency ratio COP is best | |
| CN110895029A (en) | Building load prediction method based on temperature of chilled water | |
| CN113218040A (en) | Energy efficiency improvement control method for central air-conditioning system | |
| CN108224632A (en) | Hotel Buildings central air conditioner room chilled water system comprehensive energy efficiency lifting control method |
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 |