Base station energy consumption prediction and modeling method, electronic equipment and computer readable medium
By constructing a target base station energy consumption prediction model, and using data from base stations that have adopted energy-saving strategies and those that have not, combined with a multiple regression model, the problem of inaccurate prediction of energy consumption of base stations that have not adopted energy-saving strategies in traditional methods is solved, thus achieving accurate prediction of future energy consumption and optimized selection of energy-saving strategies.
Patent Information
- Application Number
- CN202410852931.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-27
- Publication Date
- 2025-12-30
AI Technical Summary
Traditional base station energy consumption prediction methods are difficult to accurately predict the energy consumption of areas or base stations that have not yet adopted energy-saving strategies, especially when it is impossible to obtain the original data of remote base stations.
By constructing a target base station energy consumption prediction model, using base station data that has adopted energy-saving strategies as part of the training samples, and combining it with base station data that has not adopted energy-saving strategies, the trained model can predict the base station energy consumption when no energy-saving strategies are adopted. By inputting variables such as RRU model, load parameters and energy-saving strategy parameters, a multivariate regression model such as a random forest regression model is used for prediction.
It enables accurate prediction of future base station energy consumption without adopting energy-saving strategies, and can select the optimal strategy from multiple energy-saving strategies, improving the accuracy and practicality of prediction and avoiding obstacles to cross-regional data transmission.
Smart Images
Figure CN121240178A_ABST
Abstract
Description
Technical Field
[0001] This disclosure relates to the field of communications, and in particular to a method for predicting base station energy consumption. Background Technology
[0002] Base station energy saving is a basic requirement in the field of wireless communication. In order to determine which energy saving strategy is the most reasonable, it is necessary to predict the energy consumption after adopting a certain energy saving strategy.
[0003] Traditional base station energy consumption prediction methods use historical energy consumption characteristics of base stations to perform regression processing to obtain energy consumption models, which makes it difficult to accurately predict the energy consumption of areas or base stations that have not yet adopted energy-saving strategies.
[0004] Therefore, for areas or base stations that have not yet adopted energy-saving strategies, it is urgent to accurately predict the energy consumption of those areas or base stations after adopting certain energy-saving strategies. Summary of the Invention
[0005] This disclosure provides a method for predicting base station energy consumption, a method for establishing a base station energy consumption prediction model, an electronic device, a computer-readable medium, and a computer program product.
[0006] In a first aspect, embodiments of this disclosure provide a base station energy consumption prediction method, comprising: obtaining energy-saving strategy parameter values corresponding to a desired energy-saving strategy and actual base station parameter values of a target base station; inputting the energy-saving strategy parameter values corresponding to the desired energy-saving strategy and the actual base station parameter values into a target base station energy consumption prediction model to obtain the energy consumption of the target base station, wherein the training samples of the target base station energy consumption prediction model include a first energy consumption value and a second energy consumption value, the first energy consumption value is obtained by inputting preset base station parameter values and preset energy-saving strategy parameter values into a specified base station energy consumption prediction model, the specified base station energy consumption prediction model being an energy consumption model employing an energy-saving strategy, and the second energy consumption value being the energy consumption value of the target base station without employing an energy-saving strategy.
[0007] Secondly, embodiments of this disclosure provide a method for establishing a base station energy consumption prediction model, comprising: obtaining a specified base station energy consumption prediction model, wherein the specified base station energy consumption prediction model is trained based on energy consumption sample values of the specified base station under the condition that an energy-saving strategy has been adopted; inputting preset values of base station parameters and preset values of energy-saving strategy parameters corresponding to the energy-saving strategy into the specified base station energy consumption prediction model, and outputting an energy consumption prediction value as a first energy consumption value; and retraining the specified base station energy consumption prediction model using a first training sample including the first energy consumption value and a second training sample including the second energy consumption value to obtain a target base station energy consumption prediction model, wherein the second energy consumption value is the energy consumption sample value of the target base station under the condition that no energy-saving strategy has been adopted.
[0008] Thirdly, embodiments of this disclosure provide an electronic device, which includes a memory and a processor; the memory stores a computer program that can be executed by the processor, and when the computer program is executed by the processor, it implements the above-described base station energy consumption prediction method.
[0009] Fourthly, embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the above-described base station energy consumption prediction method.
[0010] Fifthly, embodiments of this disclosure provide a computer program product, which includes a computer program that, when executed by a processor, implements the above-described base station energy consumption prediction method.
[0011] In a sixth aspect, embodiments of this disclosure provide an electronic device, which includes a memory and a processor; the memory stores a computer program that can be executed by the processor, and when the computer program is executed by the processor, it implements the above-described method for establishing a base station energy consumption prediction model.
[0012] In a seventh aspect, embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the above-described method for establishing a base station energy consumption prediction model.
[0013] Eighthly, this disclosure provides a computer program product including a computer program that, when executed by a processor, implements the above-described method for establishing a base station energy consumption prediction model.
[0014] The base station energy consumption prediction method in this embodiment can utilize a target base station energy consumption prediction model, which is trained using a portion of training samples obtained by using a base station energy consumption prediction model determined based on a specified base station that has adopted an energy-saving strategy, and energy consumption training samples of a target base station that has not yet adopted an energy-saving strategy. Thus, even when the target base station has not yet adopted an energy-saving strategy, it can accurately predict the energy consumption of the target base station when it adopts an energy-saving strategy in the future. Attached Figure Description
[0015] In the accompanying drawings of the embodiments disclosed herein:
[0016] Figure 1 A flowchart illustrating a base station energy consumption prediction method provided in this embodiment of the disclosure;
[0017] Figure 2 A schematic diagram illustrating an example of a base station energy consumption prediction method provided in an embodiment of this disclosure;
[0018] Figure 3 A schematic diagram illustrating another example of the base station energy consumption prediction method provided in this disclosure embodiment;
[0019] Figure 4 A block diagram of an electronic device provided in an embodiment of this disclosure;
[0020] Figure 5 A block diagram of a computer-readable medium provided for embodiments of this disclosure;
[0021] Figure 6 A block diagram of a computer program product provided in an embodiment of this disclosure;
[0022] Figure 7 This is a flowchart illustrating an embodiment of the present disclosure for establishing a base station energy consumption prediction model. Detailed Implementation
[0023] To enable those skilled in the art to better understand the technical solutions of this disclosure, the embodiments of this disclosure will be described in detail below with reference to the accompanying drawings.
[0024] The present disclosure will be described more fully below with reference to the accompanying drawings; however, the embodiments shown may be embodied in different forms, and the present disclosure should not be construed as limited to the embodiments set forth below. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will enable those skilled in the art to fully understand the scope of the disclosure.
[0025] The accompanying drawings of the embodiments disclosed herein are provided to further illustrate the embodiments of this disclosure and form part of the specification. They are used together with the detailed embodiments to explain this disclosure and do not constitute a limitation thereof. The above and other features and advantages will become more apparent to those skilled in the art from the description of the detailed embodiments with reference to the accompanying drawings.
[0026] This disclosure may be described with reference to plan and / or cross-sectional views using the ideal schematic diagrams of this disclosure. Therefore, the example illustrations may be modified according to manufacturing techniques and / or tolerances.
[0027] Where there is no conflict, the various embodiments of this disclosure and the features thereof in the embodiments may be combined with each other.
[0028] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to limit the disclosure. The term "and / or" as used in this disclosure includes any and all combinations of one or more of the associated enumerated entries. The singular forms "a" and "the" as used in this disclosure are also intended to include the plural forms, unless the context clearly indicates otherwise. The terms "comprising," "made of," etc., as used in this disclosure specify the presence of the stated feature, integral, step, operation, element, and / or component, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof.
[0029] Unless otherwise specified, all terms used in this disclosure (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant art and this disclosure, and will not be interpreted as having an idealized or overly formal meaning, unless expressly so defined in this disclosure.
[0030] This disclosure is not limited to the embodiments shown in the accompanying drawings, but includes modifications to the configuration based on the manufacturing process. Therefore, the areas illustrated in the drawings are schematic, and the shapes of the areas shown illustrate specific shapes of the areas of an element, but are not intended to be limiting.
[0031] In some related technologies, when predicting base station energy consumption, a regression model is obtained by using the historical energy consumption characteristics of the base station. Such an energy consumption model needs to be constructed using energy consumption data after the energy-saving strategy has taken effect in order to accurately predict the energy consumption of base stations that have adopted this energy-saving strategy. However, such data is not available before the energy-saving function is promoted (i.e., before the energy-saving strategy has been adopted).
[0032] On the other hand, while the local base station may not yet be employing energy-saving strategies, the remote operator may have already adopted them. Furthermore, equipment of the same model from the same manufacturer has highly similar energy consumption characteristics. If the local and remote base stations use the same model of equipment, data from the remote base station that has already adopted an energy-saving strategy can be used to construct an energy consumption prediction model for the local base station to estimate its energy consumption when employing a particular energy-saving strategy. However, due to data protection mechanisms, it is difficult to directly obtain the specific raw data from the remote base station, making it difficult to construct a local energy consumption prediction model using remote data.
[0033] To address this, the present disclosure provides a base station energy consumption prediction method, electronic device, computer-readable medium, and computer program product for solving at least one aspect of the aforementioned problems.
[0034] In a first aspect, embodiments of this disclosure provide a method for predicting base station energy consumption. Figure 1 This is a flowchart illustrating a base station energy consumption prediction method according to an embodiment of the present disclosure.
[0035] Reference Figure 1 The base station energy consumption prediction method according to the embodiments of this disclosure may include steps S11 and S12.
[0036] In step S11, the values of the energy-saving strategy parameters corresponding to the desired energy-saving strategy and the actual values of the base station parameters of the target base station are obtained.
[0037] In step S12, the energy-saving strategy parameter values corresponding to the desired energy-saving strategy and the actual values of the base station parameters of the target base station are input into the target base station energy consumption prediction model to obtain the energy consumption of the target base station.
[0038] The training samples of the target base station energy consumption prediction model include a first energy consumption value and a second energy consumption value. The first energy consumption value is obtained by inputting the preset values of base station parameters and energy-saving strategy parameters into the specified base station energy consumption prediction model. The specified base station energy consumption prediction model is an energy consumption model that adopts an energy-saving strategy. The second energy consumption value is the energy consumption value of the target base station without adopting an energy-saving strategy.
[0039] Among them, the desired energy-saving strategy is the energy-saving strategy that is expected to be adopted on the target base station.
[0040] In the base station energy consumption prediction method according to embodiments of this disclosure, a target base station energy consumption prediction model is used for prediction. This model is trained using energy consumption values obtained from an energy consumption prediction model based on data reflecting the adoption of energy-saving strategies as part of the training samples, and energy consumption values of target base stations that have not adopted energy-saving strategies as another part of the training samples. This allows for accurate prediction of the energy consumption of the target base station when it adopts energy-saving strategies in the future, even before such strategies are implemented. Furthermore, since accurate prediction of the energy consumption of the target base station when it adopts energy-saving strategies in the future is possible, energy consumption prediction values for the target base station can be obtained for various energy-saving strategies, and the optimal energy-saving strategy configuration can be found based on an optimization algorithm.
[0041] In some embodiments, the first energy consumption value comes from a first training sample, which also includes preset values for base station parameters and preset values for energy-saving strategy parameters.
[0042] In some embodiments, the second energy consumption value is derived from a second training sample, which further includes sample values of base station parameters of the target base station and sample values of energy-saving strategy parameters corresponding to the absence of an energy-saving strategy.
[0043] In some embodiments, the energy consumption prediction model for a specified base station is trained based on sample values of the base station parameters of the specified base station, sample values of energy-saving strategy parameters corresponding to the energy-saving strategy adopted by the specified base station, and sample values of the energy consumption of the specified base station when the energy-saving strategy has been adopted.
[0044] In other words, the specified base station energy consumption prediction model is an energy consumption prediction model trained using data from base stations that have adopted energy-saving strategies, and it can reflect the impact of energy-saving strategies on base station energy consumption.
[0045] In some embodiments, the designated base station is a base station that has adopted an energy-saving strategy.
[0046] In some embodiments, the target base station may be a base station that has not yet adopted an energy-saving strategy.
[0047] In some embodiments, base station power consumption refers to the power consumption of the base station's remote radio units (RRUs). Base station parameters may include the RRU model and RRU load parameters.
[0048] For example, the energy consumption of the target base station is the energy consumption of its remote radio unit (RF unit). The sample value of the energy consumption of the specified base station is the energy consumption value of the RF unit of the target base station. The second energy consumption value is the energy consumption value of the RF unit of the target base station without adopting energy-saving strategies. The sample value of the base station parameters of the specified base station includes the model of the RF unit and the load parameter value of that RF unit. The actual value of the base station parameters of the target base station includes the model of the RF unit and the actual value of the load parameter of that RF unit.
[0049] In some embodiments, the RRU of the designated base station and the RRU of the target base station have the same model number. Since the RRUs of the designated base station and the target base station have the same model number, their energy consumption characteristics are highly similar, thereby improving the accuracy of energy consumption prediction for the target base station when obtaining the energy consumption prediction model for the target base station using the energy consumption prediction model of the designated base station.
[0050] In some embodiments, the RRU load parameters include at least one of the following: number of carriers of the first communication type, number of carriers of the second communication type, total number of users of the first communication type, total number of users of the second communication type, total physical resource block utilization of the first communication type, and total physical resource block utilization of the second communication type.
[0051] For example, RRU load parameters may include at least one of the following: Long Term Evolution (LTE) carrier count, New Radio (NR) carrier count, total LTE users, total NR users, total LTE physical resource block (PRB) utilization, and total NR PRB utilization. The total number of users can refer to the sum of the number of users across all carriers. The total PRB utilization can refer to the sum of the PRB utilization rates across all carriers.
[0052] In some embodiments, a designated base station is one or more base stations located within a designated area, which is an area where energy-saving strategies have been adopted. A target base station is one or more base stations located within a target area.
[0053] In some embodiments, the target area is an area where energy-saving strategies have not yet been adopted.
[0054] For example, the target base station is a base station located locally (e.g., in Beijing), and the designated base station is a base station located in another location. Energy conservation has not yet been promoted locally (i.e., energy conservation strategies have not yet been adopted), while energy conservation has been promoted in the other location (i.e., energy conservation strategies have already been adopted).
[0055] In some embodiments, the base station energy consumption prediction model is a multiple regression model, which includes one of the following: random forest regression model, support vector regression model, K-nearest neighbor regression model, gradient boosting regression model, and neural network regression model.
[0056] For example, the energy consumption prediction model for a specified base station can be determined based on a multiple regression model. The input variables of this multiple regression model can be the aforementioned base station parameters and energy-saving strategy parameters, and the output variable can be base station energy consumption (especially RRU energy consumption). The training samples are composed of the base station parameters of the specified base station, the values of the energy-saving strategy parameters corresponding to the energy-saving strategy adopted by the specified base station, and the energy consumption of the specified base station under the condition that the energy-saving strategy has been adopted. The multiple regression model is trained with multiple such training samples to obtain the energy consumption prediction model for the specified base station.
[0057] Because the base station energy consumption prediction model is determined using energy-saving strategy parameters, it can more accurately reflect the impact of energy-saving strategies on energy consumption compared to techniques that only use base station load parameters. Furthermore, when using multiple regression models, especially random forest regression models, to construct base station energy consumption prediction models, the modeling is more accurate compared to techniques using non-multiple regression models or linear regression models.
[0058] In some embodiments, the power-saving strategy parameters include at least one of a first communication type (e.g., LTE) shutdown duration and a second communication type (e.g., NR) shutdown duration, as well as an RRU-level shutdown duration.
[0059] For example, LTE shutdown duration and NR shutdown duration can be LTE carrier shutdown duration and NR carrier shutdown duration, respectively.
[0060] For example, the RRU-level shutdown duration may include at least one of the following: RRU-level carrier shutdown duration, RRU-level channel shutdown duration, RRU-level sleep duration, and RRU radio frequency remote unit-level power-down duration. Among them, the sleep duration may refer to the time when some modules of the RRU with long power-on recovery times (e.g., FPGA chips) are shut down, and the power-down duration may refer to the entire RRU being shut down.
[0061] In some embodiments, the preset values for base station parameters are multiple values selected at even intervals within the range of values for each of the base station parameters. In some embodiments, the preset values for base station parameters are values selected from the actual values of the base station parameters of the target base station.
[0062] In some embodiments, the load parameters include a first load parameter corresponding to the shutdown duration of a first communication type and / or a second load parameter corresponding to the shutdown duration of a second communication type. Based on preset values of the shutdown duration of the first communication type and / or the shutdown duration of the second communication type, the preset values of the first load parameter and / or the second load parameter are updated. The updated preset values of the load parameters and the preset values of the at least one shutdown duration are input into a designated base station energy consumption prediction model to obtain the first energy consumption value of the first training sample.
[0063] In some embodiments, base station parameters may also include ambient temperature and traffic volume for one or more service types. For example, traffic volume may include voice traffic volume, video traffic volume, WeChat traffic volume, etc.
[0064] In this case, the impact of different factors on base station energy consumption can be reflected more comprehensively, resulting in a more accurate energy consumption prediction model.
[0065] In some embodiments, the designated area includes multiple designated areas that employ different energy-saving strategies, and the designated base station energy consumption prediction model includes multiple sub-designated base station energy consumption prediction models corresponding to the multiple designated areas. The first energy consumption value includes multiple energy consumption prediction values corresponding to the multiple designated areas. The energy consumption prediction value for each designated area is obtained by inputting preset values of base station parameters for that designated area and preset values of energy-saving strategy parameters corresponding to the energy-saving strategy adopted by that designated area into the corresponding sub-designated base station energy consumption prediction model. When predicting the energy consumption of the target base station, based on the energy-saving strategy to be adopted by the target base station, the corresponding sub-target base station energy consumption prediction model is used to predict the energy consumption of the target base station under the condition of adopting that energy-saving strategy.
[0066] In some embodiments, the multiple sub-designated base station energy consumption prediction models are trained based on sample values of base station parameters of base stations in corresponding designated areas, sample values of energy-saving strategy parameters corresponding to the energy-saving strategies adopted in the corresponding designated areas, and sample values of energy consumption of base stations in the corresponding designated areas under the condition that the corresponding energy-saving strategies have been adopted. The first training sample includes multiple sub-sample sets corresponding to multiple designated areas, and each sub-sample set consists of preset values of base station parameters, preset values of energy-saving strategy parameters, and predicted energy consumption values for the corresponding designated area.
[0067] Figure 2 An example of a base station energy consumption prediction method according to an embodiment of this disclosure is shown. Figure 2 In the example, Region 1 is the designated region that has adopted an energy-saving strategy, and Region 2 is the target region that has not adopted an energy-saving strategy.
[0068] The following is combined Figure 2The base station energy consumption prediction method of the present disclosure embodiments will be further described.
[0069] The base station energy consumption prediction model 100 for region 1 is determined based on the sample values of base station parameters in region 1, the sample values of energy-saving strategy parameters corresponding to the energy-saving strategy adopted in region 1, and the sample values of energy consumption obtained in region 1 under the condition that the energy-saving strategy has been adopted.
[0070] For example, the input variables for the base station energy consumption prediction model 100 in Region 1 could be: RRU type, number of LTE carriers, number of NR carriers, total number of LTE users, total number of NR users, total LTE PRB utilization rate, total NR PRB utilization rate, LTE carrier shutdown duration, NR carrier shutdown duration, RRU-level carrier shutdown duration, RRU-level sleep duration, RRU-level power-off duration, RRU-level channel shutdown duration, RRU symbol shutdown switch, ambient temperature, and traffic volume of different types of services. The base station energy consumption prediction model 100 could be a random forest regression model.
[0071] The first training sample 210 can be composed of preset values for base station parameters, preset values for energy-saving strategy parameters, and predicted energy consumption values. The predicted energy consumption values are obtained by inputting the preset values for base station parameters and energy-saving strategy parameters into the base station energy consumption prediction model 100 for region 1. In other words, training sample 210 under energy-saving conditions is constructed based on the base station energy consumption prediction model 100 for region 1.
[0072] In some embodiments, the preset values of base station parameters can be obtained as follows: for each base station parameter, multiple values are selected at uniform intervals within its value range, and the selected multiple values are the preset values of the base station parameter. For example, the base station parameter includes the number of LTE carriers. Within the normal value range of the number of LTE carriers (e.g., 0-3), values are selected at intervals of 1 to obtain multiple values of the number of LTE carriers (e.g., 0, 1, 2, 3), which are used as its preset values.
[0073] In some embodiments, the preset values of the base station parameters can be obtained by selecting some values from the actual values of the base station parameters in region 2 as the preset values of the base station parameters. For example, if the base station parameters include the number of LTE carriers, the actual number of LTE carriers in region 2 (e.g., 1, 2) can be selected as the preset value of the number of LTE carriers.
[0074] In some embodiments, the preset values of energy-saving strategy parameters can be obtained as follows: For various energy-saving strategies (e.g., including energy-saving switches, energy-saving load thresholds, etc.), the preset values of corresponding energy-saving strategy parameters are simulated and generated. For example, different combinations of energy-saving switches and load thresholds can be randomly generated as various energy-saving strategies, and then the corresponding energy-saving duration (e.g., LTE / NR carrier-level shutdown duration, RRU-level shutdown duration, etc.) can be determined as the preset values of the energy-saving strategy parameters.
[0075] For example, for a specific type of RRU for a base station in region 2, the base station communication type is LTE and NR. For a certain load state of the RRU (e.g., the total number of LTE users is 10, the total LTE PRB utilization rate is 6%, the total number of NR users is 15, and the total NR PRB utilization rate is 3%), examples of preset values of energy-saving strategy parameters when using different energy-saving strategies can be shown in Table 1.
[0076] Table 1
[0077]
[0078]
[0079] In some embodiments, the preset values of the corresponding base station parameters can be updated according to the preset values of the energy-saving strategy parameters.
[0080] For example, power-saving strategy parameters may include the LTE carrier shutdown duration, and base station parameters may include the number of LTE carriers in the RRU. When the power-saving strategy is LTE carrier shutdown, the LTE carrier shutdown duration is a non-zero value (e.g., 900 seconds), and the preset value of the number of LTE carriers should be updated to 0. This reflects the actual load situation under carrier shutdown conditions.
[0081] After the base station parameter preset values are updated as described above, in step S2, the preset values of the energy-saving strategy parameters and the updated base station parameter preset values are input into the base station energy consumption prediction model of region 1, and the predicted energy consumption data is output.
[0082] The second training sample 220 may consist of sample values of base station parameters of region 2, sample values of energy-saving strategy parameters corresponding to the absence of an energy-saving strategy (e.g., all energy-saving strategy parameters are 0), and sample values of energy consumption obtained in region 2 without the energy-saving strategy. In other words, the second training sample 220 is a training sample of region 2 in the non-energy-saving state.
[0083] In some embodiments, the target base station energy consumption prediction model 300 for region 2 can be obtained by retraining the base station energy consumption prediction model 100 based on the first training sample 210 and the second training sample 220.
[0084] exist Figure 2 In the example, in step S11, the actual values of the base station parameters of the target base station in region 2 and the values of the energy-saving strategy parameters corresponding to the desired energy-saving strategy can be obtained. In step S12, the values of the energy-saving strategy parameters corresponding to the desired energy-saving strategy and the actual values of the base station parameters of the target base station are input into the target base station energy consumption prediction model 300 to obtain the energy consumption of the target base station.
[0085] Figure 3 Another example of a base station energy consumption prediction method according to embodiments of this disclosure is shown. The following mainly focuses on... Figure 3 Examples and Figure 2 The differences between the examples are explained in detail.
[0086] exist Figure 3 In the example, the designated areas that have adopted energy-saving strategies include Region 1 and Region 2, and the energy-saving strategy 1 adopted by Region 1 is different from the energy-saving strategy 2 adopted by Region 2. Region 3 is the target area that has not adopted an energy-saving strategy.
[0087] The designated base station energy consumption prediction model may include a first sub-base station energy consumption prediction model 101 for region 1 and a second sub-base station energy consumption prediction model 102 for region 2. For example, the energy-saving strategy for region 1 is a carrier shutdown energy-saving strategy, and the energy-saving strategy for region 2 is a channel shutdown energy-saving strategy. The first sub-base station energy consumption prediction model 101 is trained using sample values of base station parameters for region 1, sample values of energy-saving strategy parameters corresponding to the energy-saving strategy adopted by region 1, and sample values of energy consumption obtained by region 1 under the condition that the energy-saving strategy has been adopted. The second sub-base station energy consumption prediction model 102 is trained using sample values of base station parameters for region 2, sample values of energy-saving strategy parameters corresponding to the energy-saving strategy adopted by region 2, and sample values of energy consumption obtained by region 2 under the condition that the energy-saving strategy has been adopted.
[0088] The input variables of the first sub-base station energy consumption prediction model 101 and the second sub-base station energy consumption prediction model 102 can be compared with the reference. Figure 2 The input variables of the base station energy consumption prediction model 100 described are the same.
[0089] The first training samples may include a first subset 211 and a second subset 212. The first subset 211 consists of preset values for base station parameters, preset values for energy-saving strategy parameters corresponding to energy-saving strategy 1, and predicted values for first energy consumption data. The predicted values for first energy consumption data are obtained by inputting the preset values for base station parameters and the preset values for energy-saving strategy parameters corresponding to energy-saving strategy 1 into the first sub-base station energy consumption prediction model 101. The second subset 212 consists of preset values for base station parameters, preset values for energy-saving strategy parameters corresponding to energy-saving strategy 2, and predicted values for second energy consumption data. The predicted values for second energy consumption data are obtained by inputting the preset values for base station parameters and the preset values for energy-saving strategy parameters corresponding to energy-saving strategy 2 into the second sub-base station energy consumption prediction model 102.
[0090] The methods for obtaining the preset values of base station parameters and energy-saving strategy parameters can be the same as those for... Figure 2 The examples are the same.
[0091] For example, for a specific type of RRU in region 1, the base station communication type is LTE and NR. For a certain load state of the RRU (e.g., the total number of LTE users is 10, the total LTE PRB utilization rate is 6%, the total number of NR users is 15, and the total NR PRB utilization rate is 3%), the preset values of the energy-saving strategy parameters when adopting the carrier shutdown energy-saving strategy can be shown in Table 2.
[0092] Table 2
[0093]
[0094] For example, for a specific type of RRU for a base station in region 2, the base station communication type is LTE and NR. For a certain load state of the RRU (e.g., the total number of LTE users is 10, the total LTE PRB utilization rate is 6%, the total number of NR users is 15, and the total NR PRB utilization rate is 3%), the preset values of the energy-saving strategy parameters when adopting the channel shutdown energy-saving strategy can be shown in Table 3.
[0095] Table 3
[0096]
[0097]
[0098] and Figure 2 Similar to the example, the preset values of the corresponding base station parameters can be updated based on the preset values of the energy-saving strategy parameters. The specific method is the same as... Figure 2 The examples are the same, so they will not be repeated here.
[0099] In some embodiments, the target base station energy consumption prediction model 300 for region 3 can be obtained by retraining the first sub-base station energy consumption prediction model 101 and the second sub-base station energy consumption prediction model 102 based on the first sub-sample set 211 and the second sub-sample set 212 of the first training sample and the second training sample 220.
[0100] Specifically, the first sub-base station energy consumption prediction model 101 is retrained using the first sub-sample set 211 and the second training sample 220 to obtain the first sub-target base station energy consumption prediction model 300 corresponding to energy-saving strategy 1; the second sub-base station energy consumption prediction model 102 is retrained using the second sub-sample set 212 and the second training sample 220 to obtain the second sub-target base station energy consumption prediction model 300 corresponding to energy-saving strategy 2.
[0101] exist Figure 3In the example, in step S11, the actual values of the base station parameters of the target base station in region 3 and the values of the energy-saving strategy parameters corresponding to the desired energy-saving strategy can be obtained. In step S12, the values of the energy-saving strategy parameters corresponding to the desired energy-saving strategy and the actual values of the base station parameters of the target base station are input into the target base station energy consumption prediction model 300 to obtain the energy consumption of the target base station.
[0102] Specifically, when region 3 adopts energy-saving strategy 1, the first sub-target base station energy consumption prediction model corresponding to energy-saving strategy 1 is used to predict the base station energy consumption of region 3 under this energy-saving strategy. When region 3 adopts energy-saving strategy 2, the second sub-target base station energy consumption prediction model corresponding to energy-saving strategy 2 is used to predict the base station energy consumption of region 3 under this energy-saving strategy.
[0103] Secondly, embodiments of this disclosure provide a method for establishing a base station energy consumption prediction model. Figure 7 This is a flowchart illustrating a method for establishing a base station energy consumption prediction model according to an embodiment of the present disclosure.
[0104] Reference Figure 7 The method for establishing a base station energy consumption prediction model according to the embodiments of this disclosure may include steps S21-S23.
[0105] In step S21, the energy consumption prediction model of the specified base station is obtained. The energy consumption prediction model of the specified base station is trained based on the sample values of the energy consumption of the specified base station under the condition that energy-saving strategies have been adopted.
[0106] The designated base stations are those that have already adopted energy-saving strategies. In other words, the energy consumption prediction model for the designated base stations is an energy consumption prediction model trained using data from base stations that have adopted energy-saving strategies, and it can reflect the impact of energy-saving strategies on base station energy consumption.
[0107] For example, the energy consumption prediction model for a specified base station is trained based on sample values of the base station parameters of the specified base station, sample values of the energy-saving strategy parameters corresponding to the energy-saving strategy adopted by the specified base station, and the energy consumption sample values of the specified base station.
[0108] In step S22, the preset values of base station parameters and the preset values of energy-saving strategy parameters corresponding to the energy-saving strategy are input into the specified base station energy consumption prediction model, and the output energy consumption prediction value is used as the first energy consumption value.
[0109] In step S23, the specified base station energy consumption prediction model is retrained using a first training sample including a first energy consumption value and a second training sample including a second energy consumption value to obtain the target base station energy consumption prediction model.
[0110] The second energy consumption value is the energy consumption sample value of the target base station without adopting energy-saving strategies.
[0111] For example, the target base station is a base station that has not yet adopted energy-saving strategies.
[0112] The first training sample may also include preset values for base station parameters and preset values for energy-saving strategy parameters. The second training sample may also include sample values for the base station parameters of the target base station and sample values for energy-saving strategy parameters corresponding to not using an energy-saving strategy.
[0113] In the method for establishing a base station energy consumption prediction model according to embodiments of this disclosure, an energy consumption prediction model obtained from data of a designated base station that has adopted energy-saving strategies is used to obtain a portion of training samples for training the energy consumption prediction model of the target base station. Based on this portion of training samples and training samples of the target base station that has not adopted energy-saving strategies, an energy consumption prediction model more suitable for the target base station is obtained. Furthermore, since the energy consumption prediction model of the designated base station is obtained rather than the specific original data of the designated base station, obstacles encountered when transmitting original communication data across regions and operators can be avoided, improving data transmission efficiency and the practicality of the prediction method.
[0114] In some embodiments, base station power consumption refers to the power consumption of the base station's remote radio units (RRUs). Base station parameters may include the RRU model and RRU load parameters.
[0115] In some embodiments, the RRU of the designated base station and the RRU of the target base station have the same model.
[0116] Since the RRUs of the designated base station and the RRUs of the target base station are of the same model, their energy consumption characteristics are highly similar. This improves the accuracy of energy consumption prediction for the target base station when the energy consumption prediction model of the designated base station is used to obtain the energy consumption prediction model of the target base station.
[0117] In some embodiments, the RRU load parameters include at least one of the following: number of carriers of the first communication type, number of carriers of the second communication type, total number of users of the first communication type, total number of users of the second communication type, total physical resource block utilization of the first communication type, and total physical resource block utilization of the second communication type.
[0118] For example, RRU load parameters may include at least one of the following: Long Term Evolution (LTE) carrier count, New Radio (NR) carrier count, total LTE users, total NR users, total LTE physical resource block (PRB) utilization, and total NR PRB utilization. The total number of users can refer to the sum of the number of users across all carriers. The total PRB utilization can refer to the sum of the PRB utilization rates across all carriers.
[0119] In some embodiments, a designated base station is one or more base stations located within a designated area, which is an area where energy-saving strategies have been adopted. A target base station is one or more base stations located within a target area.
[0120] In some embodiments, the target area is an area where energy-saving strategies have not yet been adopted.
[0121] For example, the target base station is a base station located locally (e.g., in Beijing), and the designated base station is a base station located in another location. Energy conservation has not yet been promoted locally (i.e., energy conservation strategies have not yet been adopted), while energy conservation has been promoted in the other location (i.e., energy conservation strategies have already been adopted).
[0122] In some embodiments, the base station energy consumption prediction model is a multiple regression model, which includes one of the following: random forest regression model, support vector regression model, K-nearest neighbor regression model, gradient boosting regression model, and neural network regression model.
[0123] For example, the energy consumption prediction model for a specified base station can be determined based on a multiple regression model. The input variables of this multiple regression model can be the aforementioned base station parameters and energy-saving strategy parameters, and the output variable can be base station energy consumption (especially RRU energy consumption). The training samples are composed of the base station parameters of the specified base station, the values of the energy-saving strategy parameters corresponding to the energy-saving strategy adopted by the specified base station, and the energy consumption of the specified base station under the condition that the energy-saving strategy has been adopted. The multiple regression model is trained with multiple such training samples to obtain the energy consumption prediction model for the specified base station.
[0124] Because the base station energy consumption prediction model is determined using energy-saving strategy parameters, it can more accurately reflect the impact of energy-saving strategies on energy consumption compared to techniques that only use base station load parameters. Furthermore, when using multiple regression models, especially random forest regression models, to construct base station energy consumption prediction models, the modeling is more accurate compared to techniques using non-multiple regression models or linear regression models.
[0125] In some embodiments, the power-saving strategy parameters include at least one of a first communication type (e.g., LTE) shutdown duration and a second communication type (e.g., NR) shutdown duration, as well as an RRU-level shutdown duration.
[0126] For example, LTE shutdown duration and NR shutdown duration can be LTE carrier shutdown duration and NR carrier shutdown duration, respectively.
[0127] For example, the RRU-level shutdown duration may include at least one of the following: RRU-level carrier shutdown duration, RRU-level channel shutdown duration, RRU-level sleep duration, and RRU radio frequency remote unit-level power-down duration. Among them, the sleep duration may refer to the time when some modules of the RRU with long power-on recovery times (e.g., FPGA chips) are shut down, and the power-down duration may refer to the entire RRU being shut down.
[0128] In some embodiments, the preset values for base station parameters are multiple values selected at even intervals within the range of values for each of the base station parameters. In some embodiments, the preset values for base station parameters are values selected from the actual values of the base station parameters of the target base station.
[0129] In some embodiments, the load parameters include a first load parameter corresponding to the shutdown duration of a first communication type and / or a second load parameter corresponding to the shutdown duration of a second communication type. In step S22, the preset values of the first load parameter and / or the second load parameter are updated based on preset values of the shutdown duration of the first communication type and / or the shutdown duration of the second communication type, and the updated preset values of the load parameters are input into the specified base station energy consumption prediction model.
[0130] In some embodiments, base station parameters may also include ambient temperature and traffic volume for one or more service types. For example, traffic volume may include voice traffic volume, video traffic volume, WeChat traffic volume, etc.
[0131] In this case, the impact of different factors on base station energy consumption can be reflected more comprehensively, resulting in a more accurate energy consumption prediction model.
[0132] In some embodiments, the designated area includes multiple designated areas that employ different energy-saving strategies.
[0133] In this case, in step S21, multiple sub-designated base station energy consumption prediction models corresponding to multiple designated areas are obtained. These models are trained based on sample values of base station parameters for the corresponding designated area, sample values of energy-saving strategy parameters corresponding to the energy-saving strategy adopted in the corresponding designated area, and sample values of energy consumption of the base station in the corresponding designated area under the adopted energy-saving strategy. In step S22, for each designated area, preset values of base station parameters and preset values of energy-saving strategy parameters corresponding to the energy-saving strategy adopted in that area are input into the corresponding sub-designated base station energy consumption prediction model, and the corresponding energy consumption prediction value is output. The preset values of base station parameters, preset values of energy-saving strategy parameters, and the energy consumption prediction value for each designated area constitute a corresponding subset of the first training sample. In step S23, the multiple sub-designated base station energy consumption prediction models are retrained using the multiple subsets of the first training sample and the second training sample, respectively, to obtain multiple sub-target base station energy consumption prediction models corresponding to different energy-saving strategies.
[0134] Thirdly, embodiments of this disclosure provide an electronic device.
[0135] Figure 4 This is a block diagram of an electronic device according to an embodiment of the present disclosure.
[0136] Reference Figure 4 The electronic device 1000 includes a memory 1100 and a processor 1200. The memory 1100 stores a computer program PGM that can be executed by the processor 1200. When the computer program PGM is executed by the processor 1200, it implements any of the base station energy consumption prediction methods of the present disclosure embodiments.
[0137] The processor 1200 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 1100 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read-write interface) is connected between the processor 1200 and the memory 1100, enabling information exchange between the memory 1100 and the processor 1200, including but not limited to a data bus (Bus).
[0138] Fourthly, embodiments of this disclosure provide a computer-readable medium.
[0139] Figure 5 This is a block diagram of a computer-readable medium according to embodiments of the present disclosure.
[0140] Reference Figure 5The computer-readable medium 2000 stores a computer program PGM, which, when executed by a processor, implements any of the base station energy consumption prediction methods of the present disclosure embodiments.
[0141] Among them, the processor is a device with data processing capabilities, including but not limited to the central processing unit (CPU).
[0142] Fifthly, embodiments of this disclosure provide a computer program product.
[0143] Figure 6 This is a block diagram of a computer program product according to embodiments of the present disclosure.
[0144] Reference Figure 6 The computer program product 3000 includes a computer program PGM, which, when executed by a processor, implements any of the base station energy consumption prediction methods of the present disclosure embodiments.
[0145] Among them, the processor is a device with data processing capabilities, including but not limited to the central processing unit (CPU).
[0146] Sixthly, embodiments of this disclosure provide an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program executable by the processor. When the computer program is executed by the processor, it implements any of the methods for establishing a base station energy consumption prediction model according to embodiments of this disclosure.
[0147] For specific examples of electronic devices, please refer to Figure 4 And the relevant description above, it is related to Figure 4 The only difference between the electronic devices shown is the method by which the computer program is implemented when it is executed by the processor.
[0148] In a seventh aspect, embodiments of this disclosure provide a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements any of the methods for establishing a base station energy consumption prediction model according to embodiments of this disclosure.
[0149] For specific examples of computer-readable media, please refer to Figure 5 And the relevant description above, it is related to Figure 5 The only difference between the computer-readable media shown is the method by which the computer program is implemented when it is executed by the processor.
[0150] Eighthly, embodiments of this disclosure provide a computer program product. The computer program product includes a computer program that, when executed by a processor, implements any of the methods for establishing a base station energy consumption prediction model according to embodiments of this disclosure.
[0151] For specific examples of computer program products, please refer to Figure 5And the relevant description above, it is related to Figure 5 The only difference between the computer program products shown is the method by which the computer program is implemented when it is executed by the processor.
[0152] Those skilled in the art will understand that all or some of the steps, systems, and devices disclosed above, as functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof.
[0153] In hardware implementations, the division between functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be executed by several physical components working together.
[0154] Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit (CPU), digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software may be distributed on a computer-readable medium, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technique for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory (FLASH) or other disk storage; read-only optical disc (CD-ROM), digital versatile disc (DVD) or other optical disc storage; magnetic cartridges, magnetic tapes, disk storage or other magnetic storage; and any other media that can be used to store desired information and can be accessed by a computer. Furthermore, as is known to those skilled in the art, communication media typically contain computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.
[0155] This disclosure has disclosed exemplary embodiments, and although specific terminology has been used, it is for general illustrative purposes only and should not be construed as limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of this disclosure as set forth by the appended claims.
Claims
1. A method for predicting base station energy consumption, comprising: Obtain the values of the energy-saving strategy parameters corresponding to the desired energy-saving strategy and the actual values of the base station parameters of the target base station; The energy-saving strategy parameter values corresponding to the desired energy-saving strategy and the actual values of the base station parameters are input into the target base station energy consumption prediction model to obtain the energy consumption of the target base station. The training samples of the target base station energy consumption prediction model include a first energy consumption value and a second energy consumption value. The first energy consumption value is obtained by inputting the preset values of base station parameters and energy-saving strategy parameters into the specified base station energy consumption prediction model. The specified base station energy consumption prediction model is an energy consumption model that adopts an energy-saving strategy. The second energy consumption value is the energy consumption value of the target base station without adopting an energy-saving strategy.
2. The method according to claim 1, wherein, The first energy consumption value comes from the first training sample, which also includes the preset values of the base station parameters and the preset values of the energy-saving strategy parameters.
3. The method according to claim 1, wherein, The second energy consumption value comes from the second training sample, which also includes the base station parameter sample values of the target base station and the energy-saving strategy parameters corresponding to the absence of an energy-saving strategy.
4. The method according to claim 1, wherein, The specified base station energy consumption prediction model is trained based on the base station parameter sample values of the specified base station, the energy-saving strategy parameter sample values corresponding to the energy-saving strategy adopted by the specified base station, and the energy consumption sample values of the specified base station under the condition that the energy-saving strategy has been adopted.
5. The method according to claim 1, wherein, The energy consumption of the target base station is the energy consumption of its radio frequency remote unit, and the second energy consumption value is the energy consumption of the radio frequency remote unit of the target base station without employing energy-saving strategies. The base station parameters include the model of the radio frequency remote unit and the load parameters of the radio frequency remote unit.
6. The method according to claim 1, wherein, The designated base station is one or more base stations located within a designated area, which is an area where energy-saving strategies have been adopted. The target base station is one or more base stations located within the target area.
7. The method according to claim 1, wherein, The energy consumption prediction model for the designated base station is a multiple regression model, which includes one of the following: random forest regression model, support vector regression model, K-nearest neighbor regression model, gradient boosting regression model, and neural network regression model.
8. The method according to claim 5, wherein, The energy-saving strategy parameters include at least one of the first communication type shutdown duration and the second communication type shutdown duration, as well as the radio frequency remote unit level shutdown duration.
9. The method according to claim 1, wherein, The preset values for the base station parameters are multiple values selected at even intervals within the range of values for each of the base station parameters.
10. The method according to claim 1, wherein, The preset values for the base station parameters are selected from the actual values of the base station parameters of the target base station.
11. The method according to claim 8, wherein, The load parameters include at least one of a first load parameter corresponding to the shutdown duration of the first communication type and a second load parameter corresponding to the shutdown duration of the second communication type. Based on a preset value of the at least one shutdown duration, the preset value of the load parameter corresponding to the at least one shutdown duration in the preset values of the base station parameters is updated, and The first energy consumption value is obtained by inputting the updated load parameter preset value and the preset value of the at least one shutdown duration into the specified base station energy consumption prediction model.
12. The method according to claim 6, wherein, The designated area includes multiple designated areas that employ different energy-saving strategies. The specified base station energy consumption prediction model includes multiple sub-specified base station energy consumption prediction models corresponding to the multiple specified regions. The first energy consumption value includes multiple energy consumption prediction values corresponding to the multiple specified regions. The energy consumption prediction value for each specified region is obtained by inputting the preset values of the base station parameters for that specified region and the preset values of the energy-saving strategy parameters corresponding to the energy-saving strategy adopted in that specified region into the corresponding sub-specified base station energy consumption prediction model. The step of inputting the actual values of the energy-saving strategy parameters and the actual values of the base station parameters into the target base station energy consumption prediction model to obtain the energy consumption of the target base station includes: based on the energy-saving strategy to be adopted by the target base station, using the corresponding sub-target base station energy consumption prediction model to predict the energy consumption of the target base station under the condition of adopting the energy-saving strategy.
13. The method according to claim 6, wherein, The load parameters include at least one of the following: number of carriers of the first communication type, number of carriers of the second communication type, total number of users of the first communication type, total number of users of the second communication type, total physical resource block utilization rate of the first communication type, and total physical resource block utilization rate of the second communication type.
14. The method according to claim 8, wherein, The radio frequency remote unit level shutdown duration includes at least one of the following: radio frequency remote unit level carrier shutdown duration, radio frequency remote unit level channel shutdown duration, radio frequency remote unit level sleep duration, and radio frequency remote unit level power-off duration.
15. The method according to claim 5, wherein, The base station parameters also include ambient temperature and the traffic volume of one or more service types.
16. The method according to claim 8 or 13, wherein, The first communication type is Long Term Evolution (LTE), and the second communication type is New Radio (NR).
17. A method for establishing a base station energy consumption prediction model, comprising: Obtain a specified base station energy consumption prediction model, which is trained based on the energy consumption sample values of the specified base station under the condition that energy-saving strategies have been adopted. Input the preset values of base station parameters and the preset values of energy-saving strategy parameters corresponding to the energy-saving strategy into the specified base station energy consumption prediction model, and output the energy consumption prediction value as the first energy consumption value. as well as The specified base station energy consumption prediction model is retrained using a first training sample including the first energy consumption value and a second training sample including the second energy consumption value to obtain a target base station energy consumption prediction model. The second energy consumption value is the energy consumption sample value of the target base station without adopting an energy-saving strategy.
18. The method according to claim 17, wherein, The specified base station energy consumption prediction model is trained based on the base station parameter sample values of the specified base station, the energy-saving strategy parameter sample values corresponding to the energy-saving strategy adopted by the specified base station, and the energy consumption sample values of the specified base station. The first training sample also includes the preset values of the base station parameters and the preset values of the energy-saving strategy parameters. The second training sample also includes the base station parameter sample values of the target base station and the sample values of the energy-saving strategy parameters corresponding to those without the energy-saving strategy.
19. An electronic device comprising a memory and a processor; the memory storing a computer program executable by the processor, the computer program, when executed by the processor, implementing the method of any one of claims 1 to 16.
20. A computer-readable medium having a computer program stored thereon, which, when executed by a processor, implements the method of any one of claims 1 to 16.
21. A computer program product comprising a computer program that, when executed by a processor, implements the method of any one of claims 1 to 16.