A new base station energy consumption prediction method based on digital twinning

Abstract

The present application relates to a kind of new base station energy consumption prediction method based on digital twinning, wherein method includes the following steps: S1 obtains the network and user related data of entity space, and input to the data center of digital twinning space;S2 according to the network base station distribution topology formed by parameter data, and the network base station distribution topology is mapped to digital twinning space;S3 service area gridding, and it is mapped in digital twinning space;S4 user information is geographically distributed, and entity space data is mapped to digital twinning space;S5 according to the power consumption gridding distribution of user distribution of twin space;S6 calculate the grid covered by new base station;S7 calculate the power consumption of all grids covered by base station and, obtain the power consumption of new base station;S8 output new base station power consumption table.The new base station energy consumption prediction method based on digital twinning can accurately predict the power consumption of new base station, and improve the investment benefit of new base station.

Description

Technical Field

[0001] This invention relates to the field of communication technology, and in particular to a method for predicting the energy consumption of newly built base stations based on digital twins. Background Technology

[0002] Traditional energy consumption calculation methods typically rely on the equipment used when building a base station, such as the number of BBUs, RRUs, and transmission equipment, and then calculate the approximate power consumption of each device to determine the energy consumption of the base station. This method can only provide a rough estimate of the base station's power consumption.

[0003] As the deployment of 5G base stations in China expands, the power consumption of 5G base stations is approximately 3-4 times that of 4G base stations. High power consumption is a thorny issue for operators deploying 5G on a large scale. How to accurately predict the energy consumption of 5G base stations and improve the refined management of wireless network investment are urgent problems that operators need to solve.

[0004] Chinese patent document (CN 105682109 A) discloses an energy-saving method and apparatus. The method includes: predicting the traffic volume of each cell in a future time window based on the historical traffic volume of each cell in the stored network; simulating the base station energy consumption of the entire network after shutting down the cells to be saved based on the predicted traffic volume of each cell in the future time window; and determining one or more cells to be saved based on the simulation results that have the lowest base station energy consumption in the entire network after shutdown, as the cells to be saved.

[0005] Chinese patent document (CN 112566226 A) discloses an intelligent energy-saving method for 5G base stations. The method includes: S1, distinguishing the specific characteristics of the wireless base station to determine the initial energy-saving configuration; S2, predicting energy-saving parameter thresholds using a second-order smoothing prediction algorithm; and S3, an energy-saving parameter adjustment mechanism based on real-time KPI monitoring. The advantages of this invention are that it effectively overcomes the problems of rigid application modes, poor flexibility, slow response time, poor energy-saving effect, and inability to effectively integrate with user perception and KPIs in traditional energy-saving methods. The system interface, developed using C++, processes massive amounts of historical performance data at the cell level in the current network. It filters and classifies data according to energy-saving effect, selecting cells suitable for the energy-saving strategy. A second-order smoothing prediction algorithm is used to predict the service volume development trend of the applicable cells, deriving the energy-saving time window for each cell. Then, through an energy-saving parameter adjustment mechanism based on real-time KPI monitoring, the energy-saving parameters are dynamically adjusted according to changes in cell load, achieving the optimal balance between cell energy-saving effect and user perception.

[0006] Chinese patent document (CN 112654077 A) discloses an energy-saving method and apparatus, and a computer-storable medium, relating to the field of network technology. The energy-saving method includes: obtaining a user's energy-saving request, the energy-saving request including multiple physical devices, a target time period, and constraints; based on the constraints, selecting physical devices other than a specified physical device from the multiple physical devices as devices to be processed, and obtaining the real-time traffic volume of each device to be processed; using the real-time traffic volume of each device to be processed, predicting the target traffic volume of each device to be processed within the target time period; and performing energy-saving operations on each device to be processed based on the target traffic volume.

[0007] Chinese patent document (CN 113207162 A) discloses a method for intelligent management and control of base station energy consumption based on service prediction, relating to the field of energy-saving technology for wireless communication base stations. The method includes the following steps: collecting historical service volume data on a cell-by-cell basis and classifying the historical service volume data according to storage duration; constructing a service prediction model using the historical service volume data storage duration as input, and outputting the predicted service volume for the next time period; classifying cell capacity into levels and setting trigger conditions based on the prediction results; calculating the overlap coverage between cells based on cell MR measurement reports and cell location information, and ranking the cells in real time based on the calculation results; and adopting an LTE carrier shutdown energy-saving scheme based on the ranking results, intelligently hibernating, waking up, and monitoring some cells during periods of low service volume, thereby reducing the overall energy consumption of the base station while ensuring normal network coverage, achieving energy saving and emission reduction.

[0008] Chinese patent document (CN 113810878 A) discloses a macro base station placement method based on vehicle-to-everything (V2X) task offloading decision-making. Specifically, the method includes: Step 1: Establishing Y*Y encoding matrices and combining rows and columns; Step 2: Establishing a digital twin network and simulating each combination; Step 3: Calculating the optimal task offloading decision for each macro base station in each combination; Step 4: Calculating the total energy consumption of each macro base station under each combination; Step 5: Establishing a minimum objective function for total energy consumption and solving it using a particle swarm optimization algorithm; thus obtaining the optimal combination. This technical solution aims to reduce energy consumption.

[0009] Chinese patent document (CN 114845323 A) discloses a wireless network optimization platform and method based on digital twins. The system of the present invention constructs a mapping model of real wireless network indicators in the digital space, and finally achieves synchronous operation and two-way interaction between the real network management system and the digital network multi-dimensional system, thereby improving the correlation between data.

[0010] Therefore, it is necessary to develop a method for predicting the energy consumption of newly built base stations based on digital twins, so as to accurately predict the power consumption of newly built base stations and provide investment reference for operation. Summary of the Invention

[0011] The technical problem to be solved by the present invention is to provide a method for predicting the energy consumption of newly built base stations based on digital twins. This method can quickly calculate the power consumption distribution based on user distribution, thereby accurately predicting the power consumption of newly built base stations and improving the investment efficiency of newly built base stations.

[0012] To address the aforementioned problems, the technical solution adopted by this invention is: a method for predicting the energy consumption of newly built base stations based on digital twins, comprising the following steps:

[0013] S1 acquires and inputs data: acquires network and user-related data of the physical space and inputs it into the data center of the digital twin space;

[0014] S2 forms and maps the network base station distribution topology: Based on the engineering parameter data, the network base station distribution topology is formed and mapped to the digital twin space;

[0015] S3 Business Area Rasterization and Mapping: The geographical information of the business area in the physical space is rasterized and mapped to the digital twin space to achieve digital twin space rasterization.

[0016] S4 Geographic Distribution: Geographically distribute user information, extract MDT and MR data, and map them to the digital twin space after geographic distribution;

[0017] S5 power consumption distribution: Calculate the power consumption distribution based on the distribution of user information in the twin space;

[0018] S6 determines the base station coverage area: import the location and related parameters of the new base station from the data center, determine the base station coverage area, and calculate the grid area covered by the new base station;

[0019] S7: Calculate the sum of power consumption of all grids covered by the base station to obtain the power consumption of the newly built base station;

[0020] S8: Output the power consumption table of the newly built base station.

[0021] By adopting the above technical solution and combining user distribution information, the distribution and data are mapped to a digital twin space. Then, the power consumption distribution is calculated in the twin space based on the user information distribution, thereby calculating the grid covered by the newly built base station and finally obtaining the power consumption of the newly built base station, providing an investment reference for operation. This method avoids the inaccuracies of traditional methods and improves the investment efficiency of newly built base stations.

[0022] Preferably, the network and user-related data in the physical space in step S1 includes base station power consumption data, sector operating parameter data, MDT data, MR data, and newly built base station data.

[0023] Preferably, the base station power consumption data in step S1 includes base station identifier, date, time, granularity, and power consumption, with the granularity being 15 minutes, 30 minutes, or 1 hour; the sector engineering parameter data includes base station identifier, sector identifier, longitude, latitude, and azimuth; the MDT data includes user identifier, longitude, latitude, and primary serving sector; the MR data includes user identifier, longitude, latitude, user TA, user AOA, and primary serving sector; and the newly built base station data includes base station identifier, sector identifier, longitude, latitude, azimuth, antenna horizontal beamwidth, and coverage radius.

[0024] Preferably, in step S3, the geographic information of the physical space service area is rasterized to form a raster P(m), where m = 1, 2, ..., n, and mapped in the digital twin space to form a raster V(m), where m = 1, 2, ..., n, n is a natural number, and n ≤ m. MDT and MR user data are location data based on latitude and longitude points, while base station coverage is based on a coverage area. Rasterization can aggregate point data into small-area data with a certain granularity, and then stitch these small-area data together to form the coverage area of ​​the base station, facilitating the statistics and presentation of base station coverage data.

[0025] Preferably, in step S4, the MDT data is geographically distributed according to the user's latitude and longitude, and distributed into a grid P(m), where m = 1, 2, ..., n, n is a natural number, and n ≤ m; the MR data is geographically distributed according to latitude and longitude information or user TA and AOA data, and distributed into a grid P(m), where m = 1, 2, ..., n, n is a natural number, and n ≤ m; then the data in grid P(m) is mapped to grid V(m) in the digital twin space, realizing the mapping of physical space data to the digital twin space.

[0026] Preferably, in step S5, calculating the power consumption distribution based on user distribution specifically involves: firstly, calculating the proportion of user data in the grids covered by each base station in the twin space, using the following formula;

[0027]

[0028] Where x is the base station identifier; V_Point_Ratio(x, y) is the proportion of user data of base station x in grid y, y = 1, 2, ..., z, z is a natural number, z ≤ y; V_Point(x, y) is the number of user data of base station x in grid y;

[0029] Then, the power consumption is calculated based on the proportion of user data in the grid covered by each base station, using the following formula:

[0030] V_Power(x, y)=V_Point_Ratio(x, y)×Site_Power(x);

[0031] Where x is the base station identifier; V_Power(x, y) is the power consumption of base station x in grid y, y = 1, 2, ..., z, z is a natural number, z ≤ y; Site_Power(x) is the power consumption of base station x, with the statistical granularity being hourly, daily, monthly, or yearly.

[0032] Finally, the power consumption is summed to obtain the total power consumption within the grid. The calculation formula is as follows:

[0033] V_Power_Sum(m)=∑V_Power(x, m);

[0034] Where x represents all base station identifiers falling into the grid; m represents the grid identifier, m = 1, 2, ..., n, n is a natural number, n ≤ m; V_Power_Sum(m) is the total power consumption of the V(m) grid.

[0035] Preferably, in step S6, the location and related parameters of the newly built base station are imported from the data center, including the new base station identifier, longitude, latitude, sector identifier, azimuth angle, antenna beamwidth, and coverage radius; then, the grid that the newly built base station can cover is calculated based on the longitude, latitude, antenna azimuth angle, antenna beamwidth, and coverage radius.

[0036] Preferably, in step S7, the power consumption of the newly built base station is calculated based on the sum of the power consumption of all grid cells within the coverage area of ​​the newly built base station, using the following formula:

[0037]

[0038] Where h represents the identifier of the newly built base station; NSite_V_Power(h, y) is the power consumption of the newly built base station h, y represents all the grids covered by the newly built base station, y = 1, 2, ..., z, z is a natural number, z ≤ y.

[0039] Preferably, the power consumption table of the newly built base station output in step S8 includes the base station identifier, longitude, latitude, date, time, granularity, and power consumption, with the statistical granularity being hourly, daily, monthly, or yearly.

[0040] Preferably, in step S2, the network base station distribution topology mapped to the digital twin space includes the distribution in geographic space and digital twin space, the relationship between sectors and base stations, the coverage relationship in digital twin space, and the sector coverage direction; in step S3, the geographic information of the physical space service area is divided into 50m*50m grids to form grid P(m).

[0041] Compared with existing technologies, this digital twin-based method for predicting the energy consumption of newly built base stations can quickly calculate the power consumption distribution based on user distribution, thus providing a more accurate prediction of the power consumption of newly built base stations. This avoids the inaccuracies of traditional methods and improves the investment efficiency of newly built base stations. Attached Figure Description

[0042] The technical solution of the present invention will be further described below with reference to the accompanying drawings:

[0043] Figure 1 This is a schematic diagram of a new base station energy consumption prediction method based on digital twins according to an embodiment of the present invention;

[0044] Figure 2 This is a schematic diagram of the network base station distribution topology for the new base station energy consumption prediction method based on digital twins according to an embodiment of the present invention.

[0045] Figure 3 This is a schematic diagram of the service area gridding of the energy consumption prediction method for newly built base stations based on digital twins according to an embodiment of the present invention;

[0046] Figure 4 This is a schematic diagram illustrating the method for predicting the energy consumption of newly built base stations based on digital twins in this invention, where user location is determined via TA+AOA.

[0047] Figure 5 This is a schematic diagram illustrating the calculation of base station power distribution using the digital twin-based new base station power consumption prediction method according to an embodiment of the present invention.

[0048] Figure 6 This is a schematic diagram illustrating the calculation of grid power consumption in the energy consumption prediction method for newly built base stations based on digital twins, according to an embodiment of the present invention.

[0049] Figure 7 This is a schematic diagram of the antenna horizontal beamwidth and coverage of the energy consumption prediction method for newly built base stations based on digital twins according to an embodiment of the present invention.

[0050] Figure 8 This is a schematic diagram of the coverage grid for a newly built base station using the digital twin-based energy consumption prediction method for newly built base stations, as described in an embodiment of the present invention. Detailed Implementation

[0051] To enhance understanding of the present invention, it will be further described in detail below with reference to the accompanying drawings and embodiments. These embodiments are merely some examples of the present invention and are used to explain the invention, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0052] Example: Figure 1 As shown, the energy consumption prediction method for newly built base stations based on digital twins includes the following steps:

[0053] S1 Data Acquisition and Input: Acquire network and user-related data in the physical space and input it into the data center of the digital twin space; the network and user-related data in the physical space in step S1 includes base station power consumption data, sector parameter data, MDT data, MR data, and newly built base station data; the base station power consumption data in step S1 includes base station identifier, date, time, granularity, and power consumption (as shown in Table 1), with granularity of 15 minutes, 30 minutes, or 1 hour; the sector parameter data (as shown in Table 2) includes base station identifier, sector identifier, longitude, latitude, and azimuth; the MDT data (as shown in Table 3) includes user identifier, longitude, latitude, and primary serving sector; the MR data (as shown in Table 4) includes user identifier, longitude, latitude, user TA, user AOA, and primary serving sector; the newly built base station data (as shown in Table 5) includes base station identifier, sector identifier, longitude, latitude, azimuth, antenna horizontal beamwidth, and coverage radius;

[0054] Table 1 Base Station Power Consumption Data

[0055] 1 SITE0001 2023 / 1 / 10 20:00 60 4.00 2 SITE0002 2023 / 1 / 10 20:00 60 3.90 3 SITE0003 2023 / 1 / 10 20:00 60 4.20 4 SITE0004 2023 / 1 / 10 20:00 60 3.80 5 ……

[0056] Table 2 Sector Engineering Parameter Data

[0057] 1 SITE0001 CELL0001-1 120.319444 31.565 0 2 SITE0001 CELL0001-2 120.319444 31.565 120 3 SITE0001 CELL0001-3 120.319444 31.565 240 4 SITE0002 CELL0002-1 120.3242 31.56402 0 5 SITE0002 CELL0002-2 120.3242 31.56402 120 6 SITE0002 CELL0002-3 120.3242 31.56402 240 7 SITE0003 CELL0003-1 120.322 31.5619 0 8 SITE0003 CELL0003-2 120.322 31.5619 120 9 SITE0003 CELL0003-3 120.322 31.5619 240 10 SITE0004 CELL0004-1 120.3189 31.56175 0 11 SITE0004 CELL0004-2 120.3189 31.56175 120 12 SITE0004 CELL0004-3 120.3189 31.56175 240 13 ……

[0058] Table 3 MDT Data

[0059] 1 USE00100 120.319765 31.565648 CELL0001-1 2 USE00101 120.319773 31.5656864 CELL0001-1 3 ……

[0060] Table 4 MR Data

[0061]

[0062] Table 5 Data on newly built base stations

[0063]

[0064] S2 forms and maps the network base station distribution topology: Based on engineering parameter data, a network base station distribution topology is formed, and then mapped to the digital twin space; for example... Figure 2 As shown, the network base station distribution topology mapped to the digital twin space in step S2 includes the distribution in geographic space and digital twin space, the relationship between sectors and base stations, the coverage relationship in digital twin space, and the sector coverage direction.

[0065] S3 Business Area Rasterization and Mapping: The geographical information of the business area in the physical space is rasterized and mapped to the digital twin space to achieve digital twin space rasterization.

[0066] like Figure 3 As shown, in step S3, the geographic information of the physical space business area is rasterized to form a raster P(m), where m = 1, 2, ..., n, n is a natural number, and n ≤ m; and it is mapped in the digital twin space to form a raster V(m), where m = 1, 2, ..., n; n is a natural number, and n ≤ m; in this embodiment, in step S3, the geographic information of the physical space business area is rasterized into a 50m * 50m raster to form a raster P(m);

[0067] S4 Geographic Distribution: Geographically distribute user information, extract MDT and MR data, and map them to the digital twin space after geographic distribution;

[0068] In step S4, the MDT data is geographically distributed according to the user's latitude and longitude, and distributed into a grid P(m), where m = 1, 2, ..., n; n is a natural number, and n ≤ m; for example... Figure 4 As shown, the MR data is geographically distributed according to latitude and longitude information or user TA and AOA data, and distributed into grid P(m), where m = 1, 2, ..., n; n is a natural number, n ≤ m; then the data of grid P(m) is mapped to grid V(m) of digital twin space to realize the mapping of physical space data to digital twin space;

[0069] S5 power consumption distribution: Calculate the power consumption distribution based on the distribution of user information in the twin space;

[0070] In step S5, calculating the power consumption distribution based on user distribution specifically involves: firstly, calculating the proportion of user data in the grids covered by each base station in the twin space, i.e., calculating the base station power distribution, as shown in the diagram. Figure 5 As shown, the calculation formula is:

[0071]

[0072] Where x is the base station identifier; V_Point_Ratio(x, y) is the proportion of user data of base station x in grid y, y = 1, 2, ..., z, z is a natural number, z ≤ y; V_Point(x, y) is the number of user data of base station x in grid y;

[0073] like Figure 6 As shown, the power consumption is then calculated based on the proportion of user data in the grid covered by each base station, using the following formula:

[0074] V_Power(x, y)=V_Point_Ratio(x, y)×Site_Power(x);

[0075] Where x is the base station identifier; V_Power(x, y) is the power consumption of base station x in grid y, y = 1, 2, ..., z; z is a natural number, z ≤ y; Site_Power(x) is the power consumption of base station x, with a statistical granularity of hour, day, month or year;

[0076] Finally, the power consumption is summed to obtain the total power consumption within the grid. The calculation formula is as follows:

[0077] V_Power_Sum(m)=∑V_Power(x, m);

[0078] Where x represents all base station identifiers falling into the grid; m represents the grid identifier, m = 1, 2, ..., n; n is a natural number, n ≤ m; V_Power_Sum(m) is the total power consumption of the V(m)th grid; when m is 1002, the power consumption of the grid base station is shown in Table 6.

[0079] Table 6 Power Consumption of V(1002) Grid Base Station

[0080] 1 V(1002) SITE0001 2023 / 1 / 10 20:00 60 0.031 2 V(1002) SITE0002 2023 / 1 / 10 20:00 60 0.065 3 V(1002) SITE0003 2023 / 1 / 10 20:00 60 0.047 4 V(1002) ……

[0081] S6 determines the base station coverage area: import the location and related parameters of the new base station from the data center, determine the base station coverage area, and calculate the grid that the new base station may cover;

[0082] In step S6, the location and related parameters of the newly built base station are imported from the data center, including the new base station identifier, longitude, latitude, sector identifier, azimuth angle, antenna beamwidth, and coverage radius, such as... Figure 7 As shown; then, based on longitude, latitude, antenna azimuth angle, antenna beamwidth, and coverage radius, the grid that the newly built base station can cover is calculated; such as Figure 8 As shown;

[0083] S7: Calculate the sum of power consumption of all grids covered by the base station to obtain the power consumption of the newly built base station;

[0084] In step S7, the power consumption of the new base station is calculated based on the sum of the power consumption of all grid cells within the coverage area of ​​the new base station. The formula is as follows:

[0085]

[0086] Where h represents the identifier of the newly built base station; NSite_V_Power(h, y) is the power consumption of the newly built base station h, and y represents all the grids covered by the newly built base station, y = 1, 2, ..., z, z is a natural number, z ≤ y; Table 7 shows the grids covered by the newly built base station and the power consumption of the grids.

[0087] Table 7 Coverage grid of newly built base stations and grid power consumption

[0088]

[0089] S8: Output the power consumption table of the newly built base station; The power consumption table of the newly built base station output in step S8 (as shown in Table 8) includes the base station identifier, longitude, latitude, date, time, granularity and power consumption, and the statistical granularity is hourly, daily, monthly or yearly.

[0090] Table 8 Power Consumption of Newly Built Base Stations

[0091]

[0092] For those skilled in the art, the specific embodiments are merely illustrative descriptions of the present invention. Obviously, the specific implementation of the present invention is not limited to the above-described manner. Any non-substantial improvements made using the inventive concept and technical solution of the present invention, or the direct application of the inventive concept and technical solution to other situations without modification, are all within the protection scope of the present invention.

Claims

1. A method for predicting the energy consumption of newly built base stations based on digital twins, characterized in that, Includes the following steps: S1 acquires and inputs data: acquires network and user-related data of the physical space and inputs it into the data center of the digital twin space; S2 forms and maps the network base station distribution topology: Based on the engineering parameter data, the network base station distribution topology is formed and mapped to the digital twin space; S3 Business Area Rasterization and Mapping: The geographical information of the business area in the physical space is rasterized and mapped to the digital twin space to achieve digital twin space rasterization. S4 Geographic Distribution: Geographically distribute user information, extract MDT and MR data, and map them to the digital twin space after geographic distribution; S5 power consumption distribution: Calculate the power consumption distribution based on the distribution of user information in the twin space; S6 determines the base station coverage area: import the location and related parameters of the new base station from the data center, determine the base station coverage area, and calculate the grid area covered by the new base station; S7: Calculate the sum of power consumption of all grids covered by the base station to obtain the power consumption of the newly built base station; S8: Output the power consumption table of the newly built base station; In step S6, the location and related parameters of the newly built base station are imported from the data center, including the new base station identifier, longitude, latitude, sector identifier, azimuth angle, antenna beamwidth, and coverage radius; then, the grid that the newly built base station can cover is calculated based on the longitude, latitude, antenna azimuth angle, antenna beamwidth, and coverage radius. In step S5, the calculation of power consumption distribution based on user distribution is specifically as follows: First, the proportion of user data in the grids covered by each base station in the twin space is calculated, and the formula is as follows; ； Where x is the base station identifier; V_Point_Ratio(x, y) is the proportion of user data of base station x in grid y, y=1,2,…,z, z is a natural number, y≤z; V_Point(x, y) is the number of user data of base station x in grid y; Then, the power consumption is calculated based on the proportion of user data in the grid covered by each base station, using the following formula: ; Where x is the base station identifier; V_Power(x, y) is the power consumption of base station x in grid y, y=1,2,…,z; Site_Power(x) is the power consumption of base station x, with the statistical granularity being hourly, daily, monthly, or yearly. Finally, the power consumption is summed to obtain the total power consumption within the grid. The calculation formula is as follows: ； Where x represents all base station identifiers falling into the grid; m represents the grid identifier, m=1,2,…,n, n is a natural number, m≤n; V_Power_Sum(m) is the total power consumption of the V(m) grid.

2. The method for predicting the energy consumption of newly built base stations based on digital twins according to claim 1, characterized in that, The network and user-related data in the physical space in step S1 include base station power consumption data, sector operating parameter data, MDT data, MR data, and newly built base station data.

3. The method for predicting the energy consumption of newly built base stations based on digital twins according to claim 2, characterized in that, The base station power consumption data in step S1 includes base station identifier, date, time, granularity, and power consumption, with granularity being 15 minutes, 30 minutes, or 1 hour; the sector engineering parameter data includes base station identifier, sector identifier, longitude, latitude, and azimuth; the MDT data includes user identifier, longitude, latitude, and primary serving sector; the MR data includes user identifier, longitude, latitude, user TA, user AOA, and primary serving sector; the newly built base station data includes base station identifier, sector identifier, longitude, latitude, azimuth, antenna horizontal beamwidth, and coverage radius.

4. The method for predicting the energy consumption of newly built base stations based on digital twins according to claim 2, characterized in that, In step S3, the geographic information of the physical space business area is divided into rasterized grids to form grids P(m), where m = 1, 2, ..., n, n is a natural number, and m ≤ n; It is then mapped in the digital twin space to form a grid V(m), where m = 1, 2, ..., n, n is a natural number, and m ≤ n.

5. The method for predicting the energy consumption of newly built base stations based on digital twins according to claim 4, characterized in that, In step S4, the MDT data is geographically distributed according to the user's latitude and longitude, and distributed into grid P(m), where m = 1, 2, ..., n, n is a natural number, and m ≤ n; the MR data is geographically distributed according to latitude and longitude information or user TA and AOA data, and distributed into grid P(m), where m = 1, 2, ..., n, n is a natural number, and m ≤ n; then the data in grid P(m) is mapped to grid V(m) in the digital twin space, realizing the mapping of physical space data to digital twin space.

6. The method for predicting the energy consumption of newly built base stations based on digital twins according to claim 1, characterized in that, In step S7, the power consumption of the new base station is calculated based on the sum of the power consumption of all grid cells within the coverage area of ​​the new base station. The formula is as follows: ； Where h represents the identifier of the newly built base station; NSite_V_Power(h, y) is the power consumption of the newly built base station h, y represents all the grids covered by the newly built base station, y=1,2,...,z, z is a natural number, and y≤z.

7. The method for predicting the energy consumption of newly built base stations based on digital twins according to claim 6, characterized in that, The power consumption table of the newly built base station output in step S8 includes the base station identifier, longitude, latitude, date, time, granularity, and power consumption. The statistical granularity is hourly, daily, monthly, or yearly.

8. The method for predicting the energy consumption of newly built base stations based on digital twins according to claim 3, characterized in that, In step S2, the network base station distribution topology mapped to the digital twin space includes the distribution in geographic space and digital twin space, the relationship between sectors and base stations, the coverage relationship in digital twin space, and the sector coverage direction; in step S3, the geographic information of the physical space service area is divided into 50m*50m grids to form grid P(m).

Images