CN115776813A - Efficient machine room control method and device - Google Patents
Efficient machine room control method and device Download PDFInfo
- Publication number
- CN115776813A CN115776813A CN202310109502.7A CN202310109502A CN115776813A CN 115776813 A CN115776813 A CN 115776813A CN 202310109502 A CN202310109502 A CN 202310109502A CN 115776813 A CN115776813 A CN 115776813A
- Authority
- CN
- China
- Prior art keywords
- temperature
- machine room
- grid
- calculating
- air conditioner
- 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
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
- Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The invention provides a high-efficiency machine room control method and a device, and the method comprises the following steps: according to the boundary conditions and the initial conditions of the machine room, a temperature field and a speed field of air in the machine room are established by CFD software; dividing the interior of the machine room into a plurality of grids, and calculating a first temperature in each grid according to a temperature field; a plurality of temperature sensors are arranged in the machine room, the detected temperature is obtained, and a second temperature in each grid is calculated by combining a speed field; calculating the average value of the corresponding first temperature and the second temperature in each grid, and then calculating the temperature average value of all grids, namely the temperature average value of the machine room; obtaining a power input ratio by adopting a PID algorithm based on a BP neural network according to the temperature mean value and the set temperature of the machine room; and controlling the starting time and the starting sequence of the air conditioner according to the power input ratio and the hot spot area. According to the invention, the correction of the temperature field data can be realized only by installing a small number of temperature sensors, so that the accuracy of the temperature mean value evaluation in the machine room is improved.
Description
Technical Field
The invention relates to the technical field of control of air conditioners in machine rooms, in particular to a high-efficiency machine room control method and device.
Background
In recent years, data centers are rapidly developed domestically, and in order to realize stable operation of equipment in a data center, power consumption generated by refrigeration and air conditioning is very large and accounts for more than 35% of total power consumption required by the data center equipment room, so that the data center has important significance in energy-saving transformation of the existing data equipment rooms.
At present, most of air conditioners installed in a data center machine room are directly set for air conditioner temperature and work without power failure, and the air conditioners operate independently and do not have the capacity of cooperatively controlling room temperature, so that the overall operation efficiency is low, the temperature of the machine room area is uneven, and the indoor environment temperature fluctuates greatly. In some machine rooms, in order to realize cooperative work of air conditioners, a large number of hardware facilities such as sensors need to be arranged indoors, so that the problems of operation and maintenance cost and difficulty in implementation are caused.
Disclosure of Invention
In view of the above problems, the invention provides an efficient machine room control method and device, which solve the problems of low overall operation efficiency, uneven cooling and heating of a machine room area and large indoor environment temperature fluctuation caused by the fact that air conditioners in the prior art do not have the capacity of cooperatively controlling room temperature.
In order to solve the technical problems, the invention adopts the technical scheme that: a high-efficiency machine room control method comprises the following steps: according to the boundary conditions and the initial conditions of the machine room, a temperature field and a speed field of air in the machine room are established by CFD software; dividing the interior of the machine room into a plurality of grids, and calculating a first temperature in each grid according to the temperature field; a plurality of temperature sensors are arranged in the machine room, the detected temperature is obtained, and a second temperature in each grid is calculated by combining a speed field; calculating the average value of the corresponding first temperature and the second temperature in each grid, and then calculating the temperature average value of all the grids, namely the temperature average value of the machine room; obtaining a power input ratio by adopting a PID algorithm based on a BP neural network according to the temperature mean value and the set temperature of the machine room; and controlling the starting time and the starting sequence of the air conditioner according to the power input ratio and the hot spot area.
Preferably, the boundary conditions include air conditioner outlet temperature, air conditioner air flow velocity, rack heat flux and wall heat flux, and the initial conditions include air conditioner outlet initial temperature, air conditioner initial air flow velocity and cabinet power parameters.
Preferably, the establishing a temperature field and a speed field of air in the machine room by using CFD software includes:
constructing a physical model of the machine room according to the structures and the sizes of the machine cabinet and the air conditioner in the machine room; importing the machine room physical model into CFD software, determining a simulation calculation domain, and processing the calculation domain in blocks to obtain a plurality of sub-regions; performing meshing on the sub-regions, and integrating and exporting a mesh file after the meshing is completed; importing the grid file into a solver, selecting a calculation model and setting model parameters, wherein the calculation model is a zero equation turbulence model; and initializing a calculation model and carrying out iterative solution according to the cabinet power parameter and the boundary condition to obtain a temperature field and a speed field.
Preferably, said calculating a second temperature in each of said grids using said binding velocity field comprises: selecting a temperature sensor closest to the central point of the grid to be detected as a selected sensor; acquiring a temperature gradient line between the central point of the grid to be detected and the selected sensor according to the speed field; and obtaining the temperature of the central point of the grid to be measured, namely the second temperature according to the current temperature value and the temperature gradient line of the selected sensor.
Preferably, the calculation formula of the temperature gradient line is as follows:
in the above formula, the first and second carbon atoms are,
the temperature of the first liquid is the first temperature,
in order to select the current temperature value of the sensor,
in order to have a high heat dissipation coefficient,
to select the distance of the sensor from the center point of the grid to be measured,
the average flow velocity between the selected sensor and the center point of the grid under test is determined.
As a preferred scheme, the obtaining of the power input ratio by using the PID algorithm based on the BP neural network includes: step 1: determining the number of input nodes and the number of hidden layer nodes of the BP neural network, and selecting a learning rate and an inertia coefficient; and 2, step: giving an input vector and a target output of a BP neural network, and obtaining node outputs of a hidden layer and an output layer, wherein the output of the output layer is a proportional coefficient, an integral coefficient and a differential coefficient of a PID controller; and step 3: obtaining an input value and an output value through sampling, and calculating a time error; and 4, step 4: and calculating PID output based on a PID algorithm according to the output of the output layer and the time error, wherein the PID output is power input ratio: and 5: and performing iterative training on the BP neural network according to the PID output, and simultaneously adjusting a weighting coefficient.
Preferably, the hot spot area is a grid area in which the average temperature value in the grid is higher than the temperature value of the machine room plus the variation.
Preferably, the controlling the turn-on time and turn-on sequence of the air conditioner includes: calculating the total refrigerating capacity of the currently running air conditioner, determining the currently required refrigerating capacity according to the power input ratio, and determining the starting time of the air conditioner according to the required refrigerating capacity; and if the hot spot area exists, adjusting the air conditioner in the area to be at the first priority, immediately starting the air conditioner, and rapidly cooling.
The invention also discloses a high-efficiency machine room control device, which comprises: the field establishing module is used for establishing a temperature field and a speed field of air in the machine room by using CFD software according to the boundary condition and the initial condition of the machine room; the first temperature calculation module is used for dividing the interior of the machine room into a plurality of grids and calculating a first temperature in each grid according to the temperature field; the second temperature calculation module is used for installing a plurality of temperature sensors in the machine room, acquiring the detected temperature and calculating a second temperature in each grid by combining the speed field; the temperature mean value calculation module is used for calculating the mean value of the corresponding first temperature and the second temperature in each grid and then calculating the temperature mean value of all the grids, namely the temperature mean value of the machine room; the PID algorithm module is used for obtaining a power input ratio by adopting a PID algorithm based on a BP neural network according to the temperature mean value and the set temperature of the machine room; and the air conditioner control module is used for controlling the starting time and the starting sequence of the air conditioner according to the power input ratio and the hot spot area.
Compared with the prior art, the invention has the beneficial effects that: based on the boundary conditions and the initial conditions of the machine room, CFD software is adopted to realize the simulation of an air temperature field and a speed field in the machine room, a first temperature is calculated according to the temperature field, a second temperature is calculated according to the speed field and the detection temperature, the temperature mean value is determined by combining the first temperature and the second temperature, the correction of the temperature field data can be realized only by installing a small number of temperature sensors, and the accuracy of the evaluation of the temperature mean value in the machine room is improved. In addition, the traditional PID algorithm is optimized by using the BP neural network control rule, the stability of the PID control effect is improved, the uniform temperature distribution of a machine room can be ensured, and the energy-saving effect is obvious.
Drawings
The disclosure of the present invention is illustrated with reference to the accompanying drawings. It is to be understood that the drawings are designed solely for the purposes of illustration and not as a definition of the limits of the invention. In the drawings, like reference numerals are used to refer to like parts. Wherein:
fig. 1 is a schematic flow chart of a high-efficiency machine room control method according to an embodiment of the present invention;
FIG. 2 is a layout diagram of a machine room according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a PID algorithm based on a BP neural network according to an embodiment of the present invention;
FIG. 4 is a simulation diagram of a temperature field according to an embodiment of the present invention;
FIG. 5 is a simulation plot of a velocity field according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an efficient machine room control device according to an embodiment of the present invention.
Detailed Description
It is easily understood that according to the technical solution of the present invention, a person skilled in the art can propose various alternative structures and implementation ways without changing the spirit of the present invention. Therefore, the following detailed description and the accompanying drawings are merely illustrative of the technical aspects of the present invention, and should not be construed as limiting or restricting the technical aspects of the present invention.
An embodiment according to the present invention is shown in connection with fig. 1. A high-efficiency machine room control method comprises the following steps:
s101, establishing a temperature field and a speed field of air in the machine room by CFD software according to boundary conditions and initial conditions of the machine room. The boundary conditions comprise air conditioner air outlet temperature, air conditioner flow speed, rack heat flux, wall heat flux and the like, and the initial conditions comprise air conditioner air outlet initial temperature, air conditioner initial flow speed, cabinet power parameters and the like. The simulated graphs of the temperature field and the velocity field are shown in fig. 4 and 5, respectively.
It should be understood that Computational Fluid Dynamics (CFD) is a systematic analysis of physical and chemical phenomena through computer numerical calculation and image display, and the distribution of basic physical quantities (such as velocity, pressure, temperature, concentration, etc.) at various positions in a flow field can be obtained through CFD simulation, and the obtained results have universality and provide a good guiding function for experimental research and industrial design. The common use is ANSYS Fluent, ANSYS CFX, STAR-CCM etc. adopt ANSYS Fluent software to carry out analog simulation to the thermal environment of computer lab in this application.
Specifically, the above-mentioned temperature field and speed field of the air in the machine room established by using the CFD software includes:
1) And constructing a physical model of the machine room according to the structures and the sizes of the machine cabinet and the air conditioner in the machine room.
2) And importing the physical model of the machine room into CFD software, determining a simulated calculation domain, and processing the calculation domain in blocks to obtain a plurality of sub-regions.
3) And carrying out meshing on the sub-regions, and integrating and exporting the mesh files after the meshing is finished.
4) And importing the grid file into a solver, selecting a calculation model and setting model parameters, wherein the calculation model is a zero equation turbulence model.
5) And initializing a calculation model and carrying out iterative solution according to the cabinet power parameter and the boundary condition to obtain a temperature field and a speed field.
S102, dividing the interior of the machine room into a plurality of grids, and calculating a first temperature in each grid according to a temperature field. Collecting a plurality of data points in the grid, wherein each data point has a temperature value, adding the temperature values of all the data points in the grid, and dividing the temperature values by the number of the data points to obtain a first temperature in each grid.
And S103, installing a plurality of temperature sensors in the machine room, acquiring the detected temperature, and calculating a second temperature in each grid by combining the speed field. As shown in fig. 2, one temperature sensor is installed at a grid intersection of a machine room, the temperature sensor detects temperature in real time, the number of grids can be divided according to the area of the machine room, generally 16 grids are arranged for each 100 squares, and 9 temperature sensors are correspondingly installed, so that the number of temperature sensors is greatly saved.
The above-described combined velocity field calculating a second temperature in each grid includes: and selecting the temperature sensor closest to the central point of the grid to be detected as the selected sensor. And acquiring a temperature gradient line between the central point of the grid to be detected and the selected sensor according to the speed field. And obtaining the temperature of the central point of the grid to be measured, namely the second temperature according to the current temperature value and the temperature gradient line of the selected sensor.
When the central point of the grid to be measured is adjacent to the plurality of temperature sensors around and the distances between the central point of the grid to be measured and the plurality of temperature sensors around are equal, all the adjacent temperature sensors are used as selected sensors, the temperatures of the central points of the plurality of grids to be measured are obtained through calculation, and then the average value of the temperatures of the central points of the plurality of grids to be measured is used as the second temperature, so that the evaluation precision of the second temperature can be further improved. When abnormal points exist in the temperatures of the central points of the grids to be measured, the abnormal points need to be removed and then calculated.
The calculation formula of the temperature gradient line is as follows:
in the above formula, the first and second carbon atoms are,
is at a second temperature, and is,
in order to select the current temperature value of the sensor,
in order to have a high heat dissipation coefficient,
to select the distance of the sensor from the center point of the grid to be measured,
the average flow velocity between the selected sensor and the center point of the grid under test is determined.
And S104, calculating the average value of the corresponding first temperature and the second temperature in each grid, and then calculating the temperature average value of all grids, namely the temperature average value of the machine room.
And S105, obtaining the power input ratio by adopting a PID algorithm based on the BP neural network according to the temperature mean value and the set temperature of the machine room. A schematic diagram of a PID algorithm based on a BP neural network is shown in fig. 3.
Specifically, the method for obtaining the power input ratio by adopting the PID algorithm based on the BP neural network comprises the following steps:
step 1: and determining the number of input nodes and the number of hidden layer nodes of the BP neural network, and selecting a learning rate and an inertia coefficient.
Step 2: and giving an input vector and a target output of the BP neural network, and obtaining node outputs of the hidden layer and the output layer, wherein the output of the output layer is a proportional coefficient, an integral coefficient and a differential coefficient of the PID controller. Because all the three coefficients can not be negative, the transformation function of the output layer can adopt a non-negative sigmoid function, and the neuron transformation function of the hidden layer can adopt a positive-negative symmetric sigmoid function.
And step 3: the input value and the output value are obtained by sampling, and the time error is calculated.
And 4, step 4: and calculating PID output based on a PID algorithm according to the output of the output layer and the time error, wherein the PID output is a power input ratio:
and 5: and performing iterative training on the BP neural network according to the PID output, and simultaneously adjusting the weighting coefficient of the learning algorithm, namely increasing or reducing the learning rate according to the total error function of the previous generation.
And S106, controlling the starting time and the starting sequence of the air conditioner according to the power input ratio and the hot spot area. The hot spot area is a grid area with the average temperature value in the grid higher than the average temperature value of the machine room plus variation, and the variation is the maximum value allowed to float on the average temperature value of the machine room. And if the hot spot area exists, adjusting the air conditioner in the area to be at the first priority, immediately starting the air conditioner, and rapidly cooling.
In the embodiment of the invention, the control of the starting time and the starting sequence of the air conditioner comprises the following steps: and calculating the total refrigerating capacity of the currently running air conditioner, determining the currently required refrigerating capacity according to the power input ratio, and determining the starting time of the air conditioner according to the required refrigerating capacity. The power input ratio is a ratio of an output power of the operating air conditioner to a sum of rated powers of all the air conditioners. The required refrigerating capacity is the product of the power input ratio and the refrigerating capacity generated by the power input ratio per unit.
For example: the number of the current running air conditioners is 8, the output power is 80%, the number of the standby air conditioners is 2, the output power is 0%, and then the current power input ratio is 64%; if the power input ratio output by the algorithm model is 75%, the required cooling capacity is (75% -64%) the cooling capacity generated per unit power input ratio, at this time, the standby air conditioner is started, the output power is set to be 11%, and the starting time is the required time of output power input increment per unit output power input increment, namely 11 × 10s. When the output power of the operating air conditioner is lower than the set ratio (such as 80%), the output power of the operating air conditioner is preferentially adjusted, namely, the output power input increment is averagely distributed to the operating air conditioner, so that the temperature in a machine room can be prevented from being greatly fluctuated, and the smoothness of the system is improved.
Referring to fig. 6, the present invention also discloses an efficient machine room control device, comprising:
and the field establishing module 101 is used for establishing a temperature field and a speed field of air in the machine room by using CFD software according to the boundary condition and the initial condition of the machine room.
The first temperature calculating module 102 is configured to divide the interior of the computer room into a plurality of grids, and calculate a first temperature in each grid according to the temperature field.
And the second temperature calculation module 103 is used for installing a plurality of temperature sensors in the machine room, acquiring the detected temperature and calculating a second temperature in each grid by combining the speed field.
And the temperature mean value calculating module 104 is configured to calculate a mean value of the first temperature and the second temperature corresponding to each grid in each grid, and then calculate a temperature mean value of all grids, that is, a temperature mean value of the machine room.
And the PID algorithm module 105 is used for obtaining the power input ratio by adopting a PID algorithm based on the BP neural network according to the temperature mean value and the set temperature of the machine room.
And the air conditioner control module 106 is used for controlling the starting time and the starting sequence of the air conditioner according to the power input ratio and the hot spot area.
In summary, the beneficial effects of the invention include: based on the boundary conditions and the initial conditions of the machine room, CFD software is adopted to realize the simulation of an air temperature field and a speed field in the machine room, a first temperature is calculated according to the temperature field, a second temperature is calculated according to the speed field and the detection temperature, the temperature mean value is determined by combining the first temperature and the second temperature, the correction of the temperature field data can be realized only by installing a small number of temperature sensors, and the accuracy of the evaluation of the temperature mean value in the machine room is improved. In addition, the traditional PID algorithm is optimized by using the BP neural network control rule, the stability of the PID control effect is improved, the uniform temperature distribution of a machine room can be ensured, and the energy-saving effect is obvious.
It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
It should be understood that the integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention essentially or partially contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product stored in a storage medium and including instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk, and various media capable of storing program codes.
The technical scope of the present invention is not limited to the above description, and those skilled in the art can make various changes and modifications to the above-described embodiments without departing from the technical spirit of the present invention, and such changes and modifications should fall within the protective scope of the present invention.
Claims (9)
1. A high-efficiency machine room control method is characterized by comprising the following steps:
according to the boundary conditions and the initial conditions of the machine room, a temperature field and a speed field of air in the machine room are established by CFD software;
dividing the interior of the machine room into a plurality of grids, and calculating a first temperature in each grid according to the temperature field;
a plurality of temperature sensors are arranged in the machine room, the detected temperature is obtained, and a second temperature in each grid is calculated by combining a speed field;
calculating the average value of the corresponding first temperature and the second temperature in each grid, and then calculating the temperature average value of all the grids, namely the temperature average value of the machine room;
obtaining a power input ratio by adopting a PID algorithm based on a BP neural network according to the temperature mean value and the set temperature of the machine room;
and controlling the starting time and the starting sequence of the air conditioner according to the power input ratio and the hot spot area.
2. The efficient machine room control method as claimed in claim 1, wherein the boundary conditions comprise air conditioner outlet temperature, air conditioner flow velocity, rack heat flux, wall heat flux, and the initial conditions comprise air conditioner outlet initial temperature, air conditioner initial flow velocity and cabinet power parameters.
3. The efficient machine room control method according to claim 1, wherein the establishing the temperature field and the speed field of the air in the machine room by using the CFD software comprises the following steps:
constructing a physical model of the machine room according to the structures and the sizes of the machine cabinet and the air conditioner in the machine room;
importing the physical model of the machine room into CFD software, determining a simulation calculation domain, and processing the calculation domain in blocks to obtain a plurality of sub-regions;
carrying out grid division on the sub-regions, and integrating and exporting grid files after the grid division is finished;
importing the grid file into a solver, selecting a calculation model and setting model parameters, wherein the calculation model is a zero equation turbulence model;
and initializing a calculation model and carrying out iterative solution according to the cabinet power parameter and the boundary condition to obtain a temperature field and a speed field.
4. The efficient room control method of claim 1 wherein the calculating a second temperature in each grid in conjunction with the velocity field comprises:
selecting the temperature sensor closest to the central point of the grid to be measured as a selected sensor;
acquiring a temperature gradient line between the central point of the grid to be detected and the selected sensor according to the speed field;
and obtaining the temperature of the central point of the grid to be detected, namely the second temperature according to the current temperature value and the temperature gradient line of the selected sensor.
5. The efficient machine room control method according to claim 4, wherein the calculation formula of the temperature gradient line is as follows:
in the above formula, the first and second carbon atoms are,
the temperature of the first liquid is the first temperature,
in order to select the current temperature value of the sensor,
in order to have a high heat dissipation coefficient,
to select the distance of the sensor from the center point of the grid to be measured,
the average flow velocity between the selected sensor and the center point of the grid under test is determined.
6. The efficient machine room control method according to claim 1, wherein the obtaining of the power input ratio by using the PID algorithm based on the BP neural network comprises:
step 1: determining the number of input nodes and the number of hidden layer nodes of the BP neural network, and selecting a learning rate and an inertia coefficient;
and 2, step: giving an input vector and a target output of a BP neural network, and obtaining node outputs of a hidden layer and an output layer, wherein the output of the output layer is a proportional coefficient, an integral coefficient and a differential coefficient of a PID controller;
and step 3: obtaining an input value and an output value through sampling, and calculating a time error;
and 4, step 4: and calculating PID output based on a PID algorithm according to the output of the output layer and the time error, wherein the PID output is the power input ratio:
and 5: and performing iterative training on the BP neural network according to the PID output, and simultaneously adjusting a weighting coefficient.
7. The efficient machine room control method according to claim 1, wherein the hot spot area is a grid area in which the average temperature value in the grid is higher than the average temperature value of the machine room plus variation.
8. The efficient machine room control method according to claim 1, wherein the controlling of the turn-on time and turn-on sequence of the air conditioners comprises:
calculating the total refrigerating capacity of the currently running air conditioner, determining the currently required refrigerating capacity according to the power input ratio, and determining the starting time of the air conditioner according to the required refrigerating capacity;
and if the hot spot area exists, adjusting the air conditioner in the area to be at the first priority, immediately starting the air conditioner, and rapidly cooling.
9. An efficient machine room control device, comprising:
the field establishing module is used for establishing a temperature field and a speed field of air in the machine room by using CFD software according to the boundary condition and the initial condition of the machine room;
the first temperature calculation module is used for dividing the interior of the machine room into a plurality of grids and calculating a first temperature in each grid according to the temperature field;
the second temperature calculation module is used for installing a plurality of temperature sensors in the machine room, acquiring the detected temperature and calculating a second temperature in each grid by combining the speed field;
the temperature mean value calculation module is used for calculating the mean value of the corresponding first temperature and the second temperature in each grid and then calculating the temperature mean value of all the grids, namely the temperature mean value of the machine room;
the PID algorithm module is used for obtaining a power input ratio by adopting a PID algorithm based on a BP neural network according to the temperature mean value and the set temperature of the machine room;
and the air conditioner control module is used for controlling the starting time and the starting sequence of the air conditioner according to the power input ratio and the hot spot area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310109502.7A CN115776813B (en) | 2023-02-14 | 2023-02-14 | Efficient machine room control method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310109502.7A CN115776813B (en) | 2023-02-14 | 2023-02-14 | Efficient machine room control method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115776813A true CN115776813A (en) | 2023-03-10 |
| CN115776813B CN115776813B (en) | 2023-04-28 |
Family
ID=85393699
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310109502.7A Active CN115776813B (en) | 2023-02-14 | 2023-02-14 | Efficient machine room control method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115776813B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2996980A1 (en) * | 2012-10-12 | 2014-04-18 | Apis Engineering | Cooling system for cooling computer room in data center, has air-conditioning cabinet with refrigeration production integrated direct expansion group coupled to battery for gas/air exchange, and to another battery for cooling water exchange |
| CN107506516A (en) * | 2017-07-03 | 2017-12-22 | 华中科技大学 | A kind of communications equipment room flow field model is established and analysis method and system |
| CN107632524A (en) * | 2017-10-25 | 2018-01-26 | 华中科技大学 | A kind of communication machine room temperature model predictive control method and system |
| CN108386972A (en) * | 2018-02-12 | 2018-08-10 | 南京佳力图机房环境技术股份有限公司 | A kind of machine room air-conditioning energy-saving control system and method |
| CN112616292A (en) * | 2020-11-27 | 2021-04-06 | 湖南大学 | Data center energy efficiency optimization control method based on neural network model |
| CN112888268A (en) * | 2021-02-04 | 2021-06-01 | 中国工商银行股份有限公司 | Energy-saving control method, device and equipment for data center machine room and storage medium |
| CN114364217A (en) * | 2021-12-28 | 2022-04-15 | 广东海悟科技有限公司 | Data center and control method of machine room air conditioner |
| CN115292650A (en) * | 2022-08-08 | 2022-11-04 | 山东新一代信息产业技术研究院有限公司 | Data center machine room local overheating space positioning method based on multipoint temperature data |
-
2023
- 2023-02-14 CN CN202310109502.7A patent/CN115776813B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| FR2996980A1 (en) * | 2012-10-12 | 2014-04-18 | Apis Engineering | Cooling system for cooling computer room in data center, has air-conditioning cabinet with refrigeration production integrated direct expansion group coupled to battery for gas/air exchange, and to another battery for cooling water exchange |
| CN107506516A (en) * | 2017-07-03 | 2017-12-22 | 华中科技大学 | A kind of communications equipment room flow field model is established and analysis method and system |
| CN107632524A (en) * | 2017-10-25 | 2018-01-26 | 华中科技大学 | A kind of communication machine room temperature model predictive control method and system |
| CN108386972A (en) * | 2018-02-12 | 2018-08-10 | 南京佳力图机房环境技术股份有限公司 | A kind of machine room air-conditioning energy-saving control system and method |
| CN112616292A (en) * | 2020-11-27 | 2021-04-06 | 湖南大学 | Data center energy efficiency optimization control method based on neural network model |
| CN112888268A (en) * | 2021-02-04 | 2021-06-01 | 中国工商银行股份有限公司 | Energy-saving control method, device and equipment for data center machine room and storage medium |
| CN114364217A (en) * | 2021-12-28 | 2022-04-15 | 广东海悟科技有限公司 | Data center and control method of machine room air conditioner |
| CN115292650A (en) * | 2022-08-08 | 2022-11-04 | 山东新一代信息产业技术研究院有限公司 | Data center machine room local overheating space positioning method based on multipoint temperature data |
Non-Patent Citations (1)
| Title |
|---|
| 苏杨: ""基于CFD的数据机房热环境模拟研究"" * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115776813B (en) | 2023-04-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP2820488B1 (en) | Multi-dimensional optimization for controlling environmental maintenance modules | |
| CN106920006B (en) | Subway station air conditioning system energy consumption prediction method based on ISOA-LSSVM | |
| Nassif | Modeling and optimization of HVAC systems using artificial neural network and genetic algorithm | |
| CN104566868B (en) | A kind of central air conditioning system and its control method | |
| JP4134781B2 (en) | Air conditioning equipment | |
| Vakiloroaya et al. | Energy-efficient HVAC systems: Simulation–empirical modelling and gradient optimization | |
| US8996193B2 (en) | Computer room cooling control | |
| CN106133462A (en) | Controller and method is found for controlling the extreme value of vapor compression system | |
| Cheng et al. | A robust air balancing method for dedicated outdoor air system | |
| CN113449390B (en) | Air conditioner selection method, system and device | |
| Wright | HVAC optimisation studies: Sizing by genetic algorithm | |
| Jing et al. | An energy-saving control strategy for multi-zone demand controlled ventilation system with data-driven model and air balancing control | |
| Ono et al. | Optimal operation of heat source and air conditioning system with thermal storage tank using nonlinear programming | |
| Hou et al. | Real-time optimal control of HVAC systems: Model accuracy and optimization reward | |
| CN110726219A (en) | Control method, device and system of air conditioner, storage medium and processor | |
| CN115776813A (en) | Efficient machine room control method and device | |
| JP2015114768A (en) | Method for determining position of equipment to be installed in server room | |
| US10458672B2 (en) | Optimized energy usage in an air handling unit | |
| Wang et al. | A global optimization method for data center air conditioning water systems based on predictive optimization control | |
| CN114543273B (en) | Self-adaptive deep learning optimization energy-saving control algorithm for central air conditioner cooling system | |
| Ou et al. | Performance analysis of a liquid desiccant air conditioning system based on a data-driven model | |
| Park et al. | On the Feasibility of Model-Based Design and Optimal Control of Industrial Air-Conditioning System | |
| CN117492495A (en) | Temperature control method, computer device and computer storage medium | |
| Nomura et al. | Design–operation gap caused by parameter variance in HVAC system control sequences: A simulation-based study on energy efficiency and temperature controllability | |
| Zhou et al. | Modelling air-to-air plate-fin heat exchanger without condensation |
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 |