CN115907246A - Data center carbon-benefit model construction method based on electric power data - Google Patents
Data center carbon-benefit model construction method based on electric power data Download PDFInfo
- Publication number
- CN115907246A CN115907246A CN202211308911.1A CN202211308911A CN115907246A CN 115907246 A CN115907246 A CN 115907246A CN 202211308911 A CN202211308911 A CN 202211308911A CN 115907246 A CN115907246 A CN 115907246A
- Authority
- CN
- China
- Prior art keywords
- carbon
- data
- data center
- index
- power consumption
- 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.)
- Pending
Links
- 238000010276 construction Methods 0.000 title claims abstract description 10
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims abstract description 140
- 229910052799 carbon Inorganic materials 0.000 claims abstract description 140
- 238000004458 analytical method Methods 0.000 claims abstract description 20
- 238000000034 method Methods 0.000 claims abstract description 18
- 230000000694 effects Effects 0.000 claims abstract description 15
- 239000011159 matrix material Substances 0.000 claims abstract description 10
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 230000005611 electricity Effects 0.000 claims description 45
- 238000012360 testing method Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 10
- 230000008901 benefit Effects 0.000 claims description 8
- 238000011156 evaluation Methods 0.000 claims description 4
- 101100297732 Mus musculus Plekhm3 gene Proteins 0.000 claims description 3
- 230000001174 ascending effect Effects 0.000 claims 2
- 230000000630 rising effect Effects 0.000 claims 2
- 239000002699 waste material Substances 0.000 claims 2
- 230000005540 biological transmission Effects 0.000 claims 1
- 239000003245 coal Substances 0.000 claims 1
- 230000003247 decreasing effect Effects 0.000 claims 1
- 238000011038 discontinuous diafiltration by volume reduction Methods 0.000 claims 1
- 230000002349 favourable effect Effects 0.000 abstract description 2
- 238000010835 comparative analysis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 238000004134 energy conservation Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
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
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/84—Greenhouse gas [GHG] management systems
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a data center carbon-benefit model construction method based on electric power data, which comprises the following steps of: (1) constructing an index system; the method mainly comprises four aspects of power consumption analysis, peak valley analysis, load analysis and variable loss analysis; (2) acquiring power consumption related data; (3) preprocessing data; (4) calculating the carbon emission index; summarizing and calculating a carbon emission index according to each index, and quantifying the carbon emission level of the data center; the larger the index, the higher the carbon emission level, and vice versa; (5) calculating a carbon emission reduction potential index; (6) establishing a carbon preference matrix; (7) evaluating carbon preference grade; and evaluating the carbon preference grade of the data center according to the quadrant in which the data center is positioned, determining the carbon reduction priority of the data center according to different grade characteristics, and taking corresponding measures to implement carbon reduction. The method can evaluate the carbon emission level and the emission reduction potential of the data center, screen the data center which is most easy to generate carbon reduction effect, and help management departments to position the most favorable path for carbon reduction.
Description
Technical Field
The invention relates to energy conservation and carbon reduction of a data center, in particular to a data center carbon-benefit model construction method based on electric power data.
Background
The data center is used as a base of the digital economic era, the total electricity consumption reaches 870 hundred million kilowatt hours in 2020, the total electricity consumption exceeds 1.2% of the electricity consumption of the whole society of China, the electricity consumption still keeps a rapid growth potential in the future, and the data center has a huge carbon reduction space.
The carbon reduction potential is different due to different objective conditions of different data centers, and the situation that the carbon reduction amount is not ideal but costs a lot may occur in a one-time cutting mode. At present, carbon reduction research of a data center is mostly aimed at a single data center, starting from carbon reduction measures, means for scientific assessment and comparative analysis of carbon reduction potential of the data center are provided, so that the most favorable carbon reduction path cannot be found, and the result is more than half.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to provide a data center carbon-benefit model construction method based on electric power data, so that the carbon emission level and the emission reduction potential of a data center are evaluated, the data center which is most prone to generating carbon reduction effect is screened, and management departments are helped to position the most carbon-reduction optimal path.
The technical scheme is as follows: the invention relates to a data center carbon-benefit model construction method based on electric power data, which comprises the following steps of:
(1) Constructing an index system: a carbon-benefit model index system is constructed based on electricity utilization data, and the method mainly comprises four aspects of electricity consumption analysis, peak valley analysis, load analysis and variable loss analysis, and is shown in a table 1.
TABLE 1 carbon Hui model index System
(2) Acquiring electricity related data: the method comprises the steps of obtaining electricity consumption data, line loss data and load data of a data center based on a national power grid electricity consumption information acquisition system.
The electricity consumption related data in the step (2) comprise daily electricity consumption data, monthly line loss electricity quantity data, daily peak electricity consumption data, daily valley electricity consumption data, daily 96-point load and monthly variable loss electricity quantity data.
(3) Preprocessing data; the preprocessing comprises removing null data and calculating an intermediate index.
The step (3) is specifically as follows:
and (3.1) removing the null data.
(3.2) calculating an intermediate index; the indicators include:
peak power consumption mean: in the statistical period, the average value of the peak power consumption of each day of the data center;
mean value of valley power consumption: in the statistical period, the average value of the electricity consumption of each valley of the data center;
daily load rate: the ratio of the maximum load value of the data center on the day to the contract capacity is specifically as follows:
average load factor: in the statistical period, the average value of the daily load rate of the data center;
days of light load: when the daily load rate is in the interval of 0-60), defining the current day as light load; the number of light load days is the number of days in the statistical period that the daily load rate is in the interval of 0-60.
(4) Calculating the carbon emission index: summarizing and calculating a carbon emission index according to each index, and quantifying the carbon emission level of the data center; the larger the index, the higher the carbon emission level and vice versa.
The step (4) is specifically as follows:
assuming that M data centers are shared in a certain area, the power consumption is PQ = { PQ = 1 ,pq 2 ,…,pq m Line loss capacity AP _ LL = { AP _ LL 1 ,ap_ll 2 ,…,ap_ll m }。
(4.1) calculating the power consumption index score:
wherein,
the maximum value of the electricity consumption of the M data centers.
(4.2) calculating the line loss electric quantity index score:
wherein,
the maximum value of the line power loss of the M data centers.
(4.3) calculating the power consumption trend index score:
because the power consumption data is time series data, a Mann-Kendall trend test is adopted, and the principle is as follows:
for time series X, the statistics of the Mann-Kendall trend test are as follows:
wherein x is j Is the j data value of the time sequence; n is the length of the data sample; sgn is a sign function defined as follows:
when n is more than or equal to 8, the statistic S approximately follows normal distribution, the mean value E (S) =0, and the variance is as follows:
the normalized test statistic Z was calculated as follows:
the index for measuring the magnitude of the trend is represented by the gradient beta as:
median represents median, β is positive represents "upward trend", and β is negative represents "downward trend".
The Mann-Kendall trend testing method comprises the following steps: for a given confidence level α, if | Z | ≧ Z 1-α/2 That is, at the confidence level α (significance test level), there is a significant trend of the time series data up or down.
Take α =0.1 as an example, Z 1-α/2 ＝Z 0.95 Query of the normal distribution table Z 0.95 =1645, so when a certain data center Z is more than or equal to 1.645, the significance test of 90% is passed, namely the power consumption of the data center has a remarkable trend.
Finally, the power consumption trend index is divided into:
(4.4) the carbon index of a certain data center is:
w_pq i +w_apll i +w_pqt i ＝1
wherein, w _ pq i ，w_apll i ，w_pqt i The weight of the power consumption index, the line loss index and the power consumption trend index are respectively.
(5) Calculating a carbon emission reduction potential index: summarizing and calculating a carbon emission reduction potential index according to each index, and quantifying the carbon emission reduction potential of the data center; the larger the index, the higher the carbon reduction potential.
The step (5) is specifically as follows:
assuming that M data centers are in total in a certain area, the average value of peak electricity consumption is TPQ = { TPQ in a statistical period 1 ,tpq 2 ,…,tpq m Mean value of valley power consumption VPQ = { VPQ = 1 ,vpq 2 ,…,vpq m Mean load rate APR = { APR = } 1 ,apr 2 ,…,apr m } light load duration LAP = { LAP = 1 ,lap 2 ,…,lap m And variable loss capacity AP _ TL = { AP _ TL = } 1 ,ap_tl 2 ,…,ap_tl m }。
(5.1) a calculation mode of the peak-valley fluctuation rate index score:
tvr i ＝(tpq i -vpq i )/tpq i
wherein,
tvr i : peak-to-valley fluctuation rate in a certain data center;
S_tvr i : scoring a peak-to-valley fluctuation rate index of a certain data center;
max (TVR): the maximum value of the peak-to-valley fluctuation rate of the M data centers.
(5.2) calculating the average load rate index score:
wherein,
the maximum value of the average load rate of the M data centers.
(5.3) a light load duration index score calculation mode:
wherein max (LAP) is the maximum value of the light load duration of the M data centers.
(5.4) daily load Rate tendency index score (S _ dapr) i ) The calculation method is the same as the step (4.3).
(5.5) the calculation mode of the variable loss electric quantity index score is as follows:
wherein,
the maximum value of the variable loss electric quantity of the M data centers is obtained.
(5.6) the carbon reduction potential index of a certain data center is as follows:
w_tvr i +w_apr i +w_lap i +w_dapr i +w_aptl i ＝1
wherein, w _ tvr i ，w_apr i ，w_lap i ，w_dapr i ，w_aptl i The weight of the peak-valley fluctuation rate index, the average load rate index, the light load duration index, the load trend index and the variable loss electric quantity index are respectively.
(6) Establishing a carbon preference matrix: constructing carbon-benefit matrix according to index condition and setting threshold value epsilon 1 、ε 2 Dividing the carbon preference matrix into 4 quadrants; when the carbon index is larger than epsilon 1 When the carbon emission level of the data center exceeds the upper limit, emission reduction needs to be implemented; when the carbon reduction potential index is larger than epsilon 2 And considering that the carbon emission reduction potential of the data center is relatively high, and preferably executing carbon reduction measures.
(7) Evaluation of carbon preference grade: and evaluating the carbon preference grade of the data center according to the quadrant in which the data center is positioned, determining the carbon reduction priority of the data center according to different grade characteristics, and taking corresponding measures to implement carbon reduction.
The step (7) is specifically as follows:
the carbon number level of the data center is divided into 4 levels according to the quadrant in which the data center is located.
A first level: the carbon emission level is high, the emission reduction potential is also large, a good effect can be achieved by reducing carbon for the data center, and the whole carbon emission level of the data center is effectively reduced.
A second stage: the carbon emission level is low, but the emission reduction potential is also large, and carbon reduction is performed on the data center, so that although the contribution to the whole carbon emission reduction of the data center is small, a good effect can be achieved on the specific data center.
Third level: the carbon emission level is low, the emission reduction potential is small, the carbon reduction is carried out on the data center, the effect on the whole and single data centers is not obvious, and the emission reduction priority should be properly reduced.
Fourth level: carbon emissions levels are high, but emission reduction potential is small, and for such data centers, carbon reduction is recommended to be implemented in view of other measures.
A computer storage medium having stored thereon a computer program which, when executed by a processor, implements a method of constructing a data center carbon-benefit model based on power data as described above.
A computer device comprises a storage, a processor and a computer program stored on the storage and capable of running on the processor, wherein the processor executes the computer program to realize the above data center carbon benefit model construction method based on the power data.
Has the advantages that: compared with the prior art, the invention has the following advantages:
1. the model gives full play to the advantages of the power data, and the carbon condition of the data center is evaluated by using the power data with multiple dimensions of power consumption, power load, line loss condition and variable loss condition, so that the model is more comprehensive and accurate;
2. the model eliminates the influence of objective conditions of the data center from the performance of power utilization of the data center, and realizes scientific evaluation and comparative analysis of multiple data centers in a specified area;
3. the model carries out quantitative analysis on the carbon reduction potential of the data center, helps management departments to locate the most preferential path of carbon reduction, and achieves the effect of double effect and half effect.
Drawings
FIG. 1 is a flow chart of the steps of the method of the present invention;
FIG. 2 is a schematic diagram of carbon preference grade.
Detailed Description
The technical scheme of the invention is further explained by combining the attached drawings.
As shown in fig. 1, a data center carbon-benefit model construction method based on power data includes the following steps:
(1) Constructing an index system: a carbon-benefit model index system is constructed based on electricity utilization data, and the method mainly comprises four aspects of electricity consumption analysis, peak valley analysis, load analysis and variable loss analysis, and is shown in a table 1.
TABLE 1 carbon Hui model index System
(2) Acquiring electricity consumption related data: the method comprises the steps of obtaining electricity consumption data, line loss data and load data of a data center based on a national power grid electricity consumption information acquisition system.
The electricity consumption related data in the step (2) comprise daily electricity consumption data, monthly line loss electricity quantity data, daily peak electricity consumption data, daily valley electricity consumption data, daily 96-point load and monthly variable loss electricity quantity data.
(3) Preprocessing data; the preprocessing comprises removing null data and calculating an intermediate index.
The step (3) is specifically as follows:
and (3.1) removing the null data.
(3.2) calculating an intermediate index; the indicators include:
peak power consumption mean: in the statistical period, the average value of the peak power consumption of each day of the data center;
mean value of valley power consumption: in the statistical period, the average value of the electricity consumption of each valley of the data center;
daily load rate: the ratio of the maximum load value of the data center on the day to the contract capacity is specifically as follows:
average load factor: in the statistical period, the average value of the daily load rate of the data center;
light load days: when the daily load rate is in the interval of 0-60), defining the current day as light load; the number of light load days is the number of days in the statistical period that the daily load rate is in the interval of 0-60.
(4) Calculating the carbon emission index: summarizing and calculating carbon emission indexes according to the indexes, and quantifying the carbon emission level of the data center; the larger the index, the higher the carbon emission level and vice versa.
The step (4) is specifically as follows:
assuming that M data centers are shared in a certain area, the power consumption is PQ = { PQ = 1 ,pq 2 ,…,pq m Line loss capacity AP _ LL = { AP _ LL 1 ,ap_ll 2 ,…,ap_ll m }。
(4.1) calculating the power consumption index score:
wherein,
the maximum value of the electricity consumption of the M data centers.
(4.2) calculating the line loss electric quantity index score:
wherein,
the maximum value of the power loss of the M data center lines is obtained.
(4.3) calculating the power consumption trend index score:
because the electricity consumption data is time series data, a Mann-Kendall trend test is adopted, and the principle is as follows:
for time series X, the statistics of the Mann-Kendall trend test are as follows:
wherein x is j Is the j data value of the time sequence; n is the length of the data sample; sgn is a sign function, which is defined as follows:
when n is more than or equal to 8, the statistic S approximately follows normal distribution, the mean value E (S) =0, and the variance is as follows:
the normalized test statistic Z was calculated as follows:
the index for measuring the magnitude of the trend is represented by the gradient beta as:
median represents median, β is positive represents "upward trend", and β is negative represents "downward trend".
The Mann-Kendall trend testing method comprises the following steps: for a given confidence level α, if | Z | ≧ Z 1-α/2 That is, at the confidence level α (significance test level), there is a significant trend of the time series data up or down.
Take α =0.1 as an example, Z 1-α/2 ＝Z 0.95 Query of the normal distribution table Z 0.95 =1645, so when a certain data center Z is more than or equal to 1.645, the significance test of 90% is passed, namely the power consumption of the data center has a remarkable trend.
Finally, the power consumption trend index is divided into:
(4.4) the carbon index of a certain data center is:
w_pq i +w_apll i +w_pqt i ＝1
wherein, w _ pq i ，w_apll i ，w_pqt i The weight of the power consumption index, the line loss power index and the power consumption trend index are respectively.
(5) Calculating a carbon emission reduction potential index: summarizing and calculating a carbon emission reduction potential index according to each index, and quantifying the carbon emission reduction potential of the data center; the larger the index, the higher the carbon reduction potential.
The step (5) is specifically as follows:
assuming that M data centers are in total in a certain area, the average value of peak electricity consumption is TPQ = { TPQ in a statistical period 1 ,tpq 2 ,…,tpq m Mean value of valley power consumption VPQ = { VPQ = 1 ,vpq 2 ,…,vpq m Mean load rate APR = { APR = } 1 ,apr 2 ,…,apr m Light load duration LAP = { LAP = } 1 ,lap 2 ,…,lap m And variable loss capacity AP _ TL = { AP _ TL = } 1 ,ap_tl 2 ,…,ap_tl m }。
(5.1) a calculation mode of the peak-valley fluctuation rate index score:
tvr i ＝(tpq i -vpq i )/tpq i
wherein,
tvr i : peak-to-valley fluctuation rate of a certain data center;
S_tvr i : scoring a peak-to-valley fluctuation rate index of a certain data center;
max (TVR): the maximum of the peak-to-valley fluctuation rates for the M data centers.
(5.2) calculating the average load rate index score:
wherein,
the maximum value of the average load rate of the M data centers.
(5.3) the light load duration index score calculation mode is as follows:
wherein max (LAP) is the maximum value of the light load time length of the M data centers.
(5.4) daily load Rate tendency index score (S _ dapr) i ) The calculation method is the same as the step (4.3).
(5.5) the calculation mode of the variable loss electric quantity index score is as follows:
wherein,
the maximum value of the variable loss electric quantity of the M data centers is obtained.
(5.6) the carbon reduction potential index of a certain data center is as follows:
w_tvr i +w_apr i +w_lap i +w_dapr i +w_aptl i ＝1
wherein, w _ tvr i ，w_apr i ，w_lap i ，w_dapr i ，w_aptl i The weight of the peak-valley fluctuation rate index, the average load rate index, the light load duration index, the load trend index and the variable loss electric quantity index are respectively.
(6) Establishing a carbon preference matrix: constructing carbon-benefit matrix according to index condition and setting threshold value epsilon 1 、ε 2 Dividing the carbon-preference matrix into 4 quadrants; when the carbon index is larger than epsilon 1 When the carbon emission level of the data center exceeds the upper limit, emission reduction needs to be implemented; when the carbon reduction potential index is larger than epsilon 2 And considering that the carbon emission reduction potential of the data center is relatively high, and preferably carrying out carbon reduction measures.
(7) Evaluation of carbon preference grade: and evaluating the carbon preference grade of the data center according to the quadrant in which the data center is positioned, determining the carbon reduction priority of the data center according to different grade characteristics, and taking corresponding measures to implement carbon reduction.
The step (7) is specifically as follows:
as shown in FIG. 2, the carbon benefits level of the data center is divided into 4 levels, depending on the quadrant in which the data center is located.
A first grade: the carbon emission level is high, the emission reduction potential is also large, a good effect can be achieved by reducing carbon for the data center, and the overall carbon emission level of the data center is effectively reduced.
A second stage: the carbon emission level is low, but the emission reduction potential is also large, and carbon reduction is performed on the data center, so that although the contribution to the whole carbon emission reduction of the data center is small, a good effect can be achieved on the specific data center.
Third level: the carbon emission level is low, the emission reduction potential is small, the carbon reduction is performed on the data center, the overall effect and the single data center effect are not obvious, and the emission reduction priority should be properly reduced.
Fourth level: carbon emissions levels are high, but emission reduction potential is small, and for such data centers, carbon reduction is recommended to be implemented in view of other measures.
A computer storage medium having stored thereon a computer program which, when executed by a processor, implements a method of constructing a data center carbon-benefit model based on power data as described above.
A computer device comprises a storage, a processor and a computer program stored on the storage and capable of running on the processor, wherein the processor executes the computer program to realize the above data center carbon benefit model construction method based on the power data.
Claims (9)
1. A data center carbon-benefit model construction method based on electric power data is characterized by comprising the following steps:
(1) Constructing an index system: constructing a carbon-benefit model index system based on electricity consumption data, wherein the index system mainly comprises four aspects of electricity consumption analysis, peak valley analysis, load analysis and variable loss analysis;
(2) Acquiring electricity related data: acquiring power consumption data, line loss data and load data of a data center based on a national power grid power consumption information acquisition system;
(3) Preprocessing data; the preprocessing comprises removing null data and calculating an intermediate index;
(4) Calculating the carbon emission index: summarizing and calculating a carbon emission index according to each index, and quantifying the carbon emission level of the data center; the larger the index, the higher the carbon emission level and vice versa;
(5) Calculating a carbon emission reduction potential index: summarizing and calculating a carbon emission reduction potential index according to each index, and quantifying the carbon emission reduction potential of the data center; the larger the index, the higher the carbon reduction potential;
(6) Establishing a carbon preference matrix: constructing carbon-benefit matrix according to index condition and setting threshold value epsilon 1 、ε 2 Dividing the carbon preference matrix into 4 quadrants; when the carbon index is larger than epsilon 1 When the data center carbon emissions are consideredThe level exceeds the upper limit, and emission reduction needs to be implemented; when the carbon reduction potential index is larger than epsilon 2 When the carbon emission reduction potential of the data center is relatively high, carbon reduction measures are preferably implemented;
(7) Evaluation of carbon preference grade: and evaluating the carbon preference grade of the data center according to the quadrant in which the data center is positioned, determining the carbon reduction priority of the data center according to different grade characteristics, and taking corresponding measures to implement carbon reduction.
2. The method for constructing the data center carbon-benefit model based on the electric power data as claimed in claim 1, wherein the analysis of the electric power consumption in the step (1) mainly comprises the following indexes:
electricity consumption: the actual power consumption of the data center is the most direct index for calculating the carbon emission;
line loss capacity: the part of electric quantity refers to electric energy loss and loss generated from the outgoing line of the power plant to the electric energy meter of the client in the electric energy transmission and marketing process, although the electric energy loss and loss are not actual electricity consumption of the data center, carbon emission is generated, and the electric energy loss and the loss need to be considered when the carbon reduction overall planning is carried out;
trend of electricity consumption: and representing the future power consumption rising or falling trend of the data center. If the carbon emission level is in an ascending trend, the carbon emission level is continuously increased, and if the carbon emission level is in a descending trend, the carbon emission level is gradually reduced;
the peak-to-valley analysis in the step (1) mainly comprises the following indexes:
peak-to-valley fluctuation ratio: the electric energy belongs to real-time balance resources, the resource waste can be caused by generating electricity according to the peak, and the requirement can not be met by generating electricity according to the valley, so that if a large peak-valley difference exists, the generator set can deviate from the optimal working condition, the efficiency is reduced, and the unit coal consumption is increased;
the load analysis in the step (1) mainly comprises the following indexes:
average load factor: representing whether the configuration of the transformer is reasonable or not under the current power utilization condition of the data center; if the average load rate is continuously low, adopting a volume reduction means to reduce the power consumption; the lower the average load rate, the larger the carbon emission reduction space;
light load duration: the longer the light load duration is, namely the longer the transformer low-load operation time is, the higher the transformer capacity configuration is, and the resource waste is caused; the higher the index is, the larger the carbon emission reduction space is;
load trend: representing the rising or falling trend of the future load level of the data center; if the carbon emission reduction space is in an ascending trend, the carbon emission reduction space is gradually reduced, otherwise, the carbon emission reduction space is increased;
the loss variation analysis in the step (1) mainly comprises the following indexes:
and (3) variable loss electric quantity: the power consumption of the transformer is referred, and the power consumption does not produce actual value and belongs to the part to be reduced; the larger the variable loss electric quantity is, the larger the carbon emission reduction space is.
3. The method for constructing the data center carbon-benefit model based on the power data according to claim 1, wherein the power consumption related data in the step (2) comprises daily power consumption data, monthly line loss data, daily peak power consumption data, daily valley power consumption data, daily 96-point load, and monthly variable loss data.
4. The method for constructing the data center carbon-benefit model based on the electric power data as claimed in claim 1, wherein the step (3) is specifically as follows:
(3.1) removing null data;
(3.2) calculating an intermediate index; the indexes include:
peak power consumption mean: in the statistical period, the average value of the peak power consumption of each day of the data center;
mean value of valley power consumption: in the statistical period, the average value of the electricity consumption of each valley of the data center;
daily load rate: the ratio of the maximum load value of the data center on the day to the contract capacity is specifically as follows:
average load factor: in the statistical period, the average value of the daily load rate of the data center;
light load days: when the daily load rate is in the range of [ 0-60), defining the daily load as light load; the number of light load days is the number of days in the statistical period that the daily load rate is in the interval of 0-60.
5. The method for constructing the data center carbon-benefit model based on the electric power data as claimed in claim 1, wherein the step (4) is specifically as follows:
assuming that M data centers are shared in a certain area, the power consumption is PQ = { PQ = 1 ,pq 2 ,…,pq m Line loss capacity AP _ LL = { AP _ LL 1 ,ap_ll 2 ,…,ap_ll m }；
(4.1) calculating the power consumption index score:
wherein,
the maximum value of the electricity consumption of the M data centers is obtained;
(4.2) calculating the line loss electric quantity index score:
wherein,
the maximum value of the power loss of the M data center lines;
(4.3) calculating the power consumption trend index score:
because the electricity consumption data is time series data, a Mann-Kendall trend test is adopted, and the principle is as follows:
for time series X, the statistics of the Mann-Kendall trend test are as follows:
wherein x is j Is the j data value of the time sequence; n is the length of the data sample; sgn is a sign function, which is defined as follows:
when n is more than or equal to 8, the statistic S approximately follows normal distribution, the mean value E (S) =0, and the variance is as follows:
the normalized test statistic Z was calculated as follows:
the index for measuring the magnitude of the trend is represented by the gradient beta as:
median represents the median, β is a positive value representing an "upward trend", and β is a negative value representing a "downward trend";
the Mann-Kendall trend testing method comprises the following steps: for a given confidence level α, if | Z | ≧ Z 1-α/2 That is, at the confidence level α, there is a significant trend of increasing or decreasing time series data;
finally, the power consumption trend index is divided into:
(4.4) the carbon index of a certain data center is:
w_pq i +w_apll i +w_pqt i ＝1
wherein, w _ pq i ，w_apll i ，w_pqt i The weight of the power consumption index, the line loss power index and the power consumption trend index are respectively.
6. The method for constructing the data center carbon-benefit model based on the electric power data as claimed in claim 1, wherein the step (5) is specifically as follows:
assuming that M data centers are in total in a certain area, the average value of peak electricity consumption is TPQ = { TPQ in a statistical period 1 ,tpq 2 ,…,tpq m Mean value of valley power consumption VPQ = { VPQ = 1 ,vpq 2 ,…,vpq m }, average load rate APR = { APR = 1 ,apr 2 ,…,apr m Light load duration LAP = { LAP = } 1 ,lap 2 ,…,lap m And variable loss capacity AP _ TL = { AP _ TL = } 1 ,ap_tl 2 ,…,ap_tl m }；
(5.1) a calculation mode of the peak-valley fluctuation rate index score:
tvr i ＝(tpq i -vpq i )/tpq i
wherein,
tvr i : peak-to-valley fluctuation rate of a certain data center;
S_tvr i : scoring a peak-to-valley fluctuation rate index of a certain data center;
max (TVR): the maximum of the peak-to-valley fluctuation rates for the M data centers.
(5.2) calculating the average load rate index score:
wherein,
the maximum value of the average load rate of the M data centers;
(5.3) the light load duration index score calculation mode is as follows:
wherein max (LAP) is the maximum value of the light load duration of the M data centers;
(5.4) daily load Rate tendency index score (S _ dapr) i ) The calculation mode is the same as that of the step (4.3);
(5.5) the calculation mode of the variable loss electric quantity index score is as follows:
wherein,
the maximum value of the variable loss electric quantity of the M data centers is obtained;
(5.6) the carbon reduction potential index of a certain data center is as follows:
w_tvr i +w_apr i +w_lap i +w_dapr i +w_aptl i ＝1
wherein, w _ tvr i ，w_apr i ，w_lap i ，w_dapr i ，w_aptl i Respectively peak-to-valley fluctuation ratioAnd the weight of the standard, the average load rate index, the light load duration index, the load trend index and the variable loss electric quantity index.
7. The method for constructing the data center carbon-benefit model based on the electric power data as claimed in claim 1, wherein the step (7) is specifically as follows:
dividing the carbon benefits level of the data center into 4 levels according to the quadrant in which the data center is located;
a first grade: the carbon emission level is high, the emission reduction potential is also large, a good effect can be obtained by reducing carbon for the data center, and the whole carbon emission level of the data center is effectively reduced;
a second stage: the carbon emission level is low, but the emission reduction potential is also large, carbon reduction is performed on the data center, and although the contribution to the whole carbon emission reduction of the data center is small, a good effect can be achieved on the specific data center;
third level: the carbon emission level is low, the emission reduction potential is small, the carbon reduction is carried out on the data center, the effect on the whole and single data centers is not obvious, and the emission reduction priority should be properly reduced;
fourth level: carbon emissions levels are high, but emission reduction potential is small, and for such data centers, carbon reduction is recommended to be implemented in view of other measures.
8. A computer storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements a method for constructing a data center carbon-benefit model based on power data according to any one of claims 1-7.
9. A computer device comprising a storage, a processor and a computer program stored on the storage and executable on the processor, wherein the processor when executing the computer program implements a method for constructing a data center carbon benefit model based on power data according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211308911.1A CN115907246A (en) | 2022-10-25 | 2022-10-25 | Data center carbon-benefit model construction method based on electric power data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211308911.1A CN115907246A (en) | 2022-10-25 | 2022-10-25 | Data center carbon-benefit model construction method based on electric power data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115907246A true CN115907246A (en) | 2023-04-04 |
Family
ID=86494127
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211308911.1A Pending CN115907246A (en) | 2022-10-25 | 2022-10-25 | Data center carbon-benefit model construction method based on electric power data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115907246A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116775589A (en) * | 2023-08-23 | 2023-09-19 | 湖北华中电力科技开发有限责任公司 | Data security protection method for network information |
-
2022
- 2022-10-25 CN CN202211308911.1A patent/CN115907246A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116775589A (en) * | 2023-08-23 | 2023-09-19 | 湖北华中电力科技开发有限责任公司 | Data security protection method for network information |
| CN116775589B (en) * | 2023-08-23 | 2023-10-27 | 湖北华中电力科技开发有限责任公司 | Data security protection method for network information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021213192A1 (en) | Load prediction method and load prediction system employing general distribution | |
| CN112288141A (en) | Comprehensive benefit evaluation method for fishing light complementary photovoltaic power station | |
| CN107528350A (en) | A kind of wind power output typical scene generation method for adapting to long -- term generation expansion planning | |
| CN115081902B (en) | Comprehensive planning method, device, equipment and medium based on source network load storage cooperation | |
| CN115907246A (en) | Data center carbon-benefit model construction method based on electric power data | |
| CN111724049B (en) | Research and judgment method for potential electric power energy efficiency service clients | |
| CN113657789A (en) | Distributed photovoltaic roof resource multi-dimensional evaluation method | |
| CN114662909A (en) | Rural land operation right circulation transaction price index calculation system | |
| CN115907511A (en) | Method and device for constructing adjustable load resource value evaluation model | |
| CN112561252B (en) | Reactive power combination evaluation method for power grid in new energy-containing region | |
| CN116484166B (en) | Station identification method for salient values based on change rate of atmospheric pollution concentration | |
| CN111191876A (en) | Comprehensive energy system evaluation method for college park | |
| CN114676931B (en) | Electric quantity prediction system based on data center technology | |
| CN115115470A (en) | Green data center carbon emission management method based on emission factor method | |
| CN114066254A (en) | Carbon emission estimation method and system for urban total planning construction land | |
| CN111626509B (en) | Method and system for evaluating effective supply capacity of regional new energy | |
| CN110135619B (en) | Method and system for predicting medium-and-long-term electric heating requirements | |
| CN109149644B (en) | Light-storage integrated online strategy matching and collaborative optimization method based on big data analysis | |
| CN112113316A (en) | Method for extracting air conditioner load | |
| CN112801333B (en) | XGBoost-based power distribution network line summer peak load prediction method | |
| CN117495203B (en) | Multi-target electric energy management method and system based on energy storage system | |
| Fan | The advantages and disadvantages of agricultural economic management information | |
| Wang et al. | A calculation model for wind power consumed by electric heating load based on multi-state probability distribution | |
| CN118137500B (en) | Photovoltaic power generation prediction method and device | |
| CN118195132A (en) | Multi-parameter evaluation decision-making method and system for enterprise |
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 |