Comprehensive energy efficiency evaluation method and system of 5G base station and storage medium

Abstract

The invention provides a comprehensive energy efficiency evaluation method and system of a 5G base station and a storage medium. The comprehensive energy efficiency evaluation method comprises the following steps: establishing a comprehensive energy efficiency evaluation index system of the 5G base station based on carbon emission-energy efficiency coupling; calculating the comprehensive weight of each index based on an entropy weight method and an analytic hierarchy process; calculating an evaluation result based on a fuzzy comprehensive evaluation method of hierarchical recursion; a 5G base station comprehensive energy efficiency reference line considering multiple typical operation scenes is constructed, and an efficient reference and a tolerance threshold are set; and comparing the comprehensive evaluation score of each typical operation scene of the 5G base station with a set efficient reference and a tolerance threshold, and performing comprehensive energy efficiency evaluation on the 5G base station. According to the method, a referenceable standardized reference line is established for comprehensive energy efficiency evaluation of the 5G base station, clear guidance is provided for energy efficiency benchmarking of the 5G base stations of different types and different regions, and a subsequent energy efficiency improvement scheme can be systematically developed.

Description

Technical Field

[0001] The present invention belongs to the technical field of new energy management, and relates to a comprehensive energy efficiency evaluation method, system and storage medium. Background Art

[0002] The full-load power consumption of a single 5G base station is about 3.8 kW, which is more than three times that of a 4G base station. With the continuous construction of 5G base stations, the power consumption of 5G base stations is bound to increase continuously. With the rapid development of the new power system and the 5G communication network, the energy consumption of 5G base stations is increasing day by day. However, the traditional energy efficiency evaluation method only conducts a single-dimensional evaluation from the perspective of the energy consumption of the base station itself or the communication index, and it is difficult to take into account the access of renewable energy, the utilization of energy storage devices and carbon emission factors. In the face of the macro requirements of "carbon peak and carbon neutrality" and the needs of the development of green communication and green power, there is an urgent need for a multi-dimensional, quantifiable 5G base station comprehensive energy efficiency evaluation system that combines carbon efficiency and energy efficiency, and conducts management and optimization in a larger scope including base station clusters and grid interactions. However, at present, there is a lack of a mature 5G base station comprehensive energy efficiency evaluation index system and evaluation method, a lack of a standardized baseline for reference, a single focus dimension, and the coupling relationship between carbon emission and energy efficiency is ignored, resulting in a lack of clear guidance for the energy efficiency benchmarking of 5G base stations of different types and in different regions, and it is also difficult to systematically carry out subsequent energy efficiency improvement plans. Summary of the Invention

[0003] To solve the problems in the comprehensive energy efficiency evaluation of 5G base stations described in the background art, such as the lack of a standardized baseline for reference, a single focus dimension, the coupling relationship between carbon emission and energy efficiency is ignored, resulting in a lack of clear guidance for the energy efficiency benchmarking of 5G base stations of different types and in different regions, and it is difficult to systematically carry out subsequent energy efficiency improvement plans, the present invention provides a comprehensive energy efficiency evaluation method, system and storage medium for 5G base stations.

[0004] The method of the present invention includes:

[0005] Based on the energy consumption, carbon emissions, 5G base station-grid interaction, data flow and load characteristics of 5G base stations, multi-level evaluation indicators are set, and the evaluation indicators are grouped and linked into an evaluation hierarchy according to the dominance relationship, and a comprehensive energy efficiency evaluation index system for 5G base stations is established based on the evaluation hierarchy;

[0006] Based on the entropy weight method and the analytic hierarchy process, the entropy weight and subjective weight of each index in the comprehensive energy efficiency evaluation index system of 5G base stations are calculated respectively, and weighted fusion is carried out to obtain the comprehensive weight of each index;

[0007] According to the comprehensive energy efficiency evaluation index system of 5G base stations, the comprehensive weights of each index, and the typical operation scenario characteristics of 5G base stations, based on the hierarchical recursive fuzzy comprehensive evaluation method, for different typical operation scenarios of 5G base stations, establish the fuzzy evaluation matrix of each index respectively. Through the calculation of the fuzzy evaluation matrix, obtain the comprehensive evaluation scores of each typical operation scenario of 5G base stations and the fuzzy evaluation results of each typical operation scenario of 5G base stations;

[0008] According to the fuzzy evaluation results of each typical operation scenario of 5G base stations, use the dimensionality reduction and discrimination method of principal component analysis to obtain the criterion set of key evaluation indexes for the comprehensive energy efficiency of 5G base stations;

[0009] According to the criterion set of key evaluation indexes for the comprehensive energy efficiency of 5G base stations, based on the label division and clustering grouping including different regions, different load types, and different equipment scales, for each typical operation scenario of 5G base stations, construct the comprehensive energy efficiency baseline of 5G base stations, and set the high-efficiency baseline and tolerance threshold based on the comprehensive energy efficiency baseline of 5G base stations;

[0010] Compare the comprehensive evaluation scores of each typical operation scenario of 5G base stations with the set high-efficiency baseline and tolerance threshold to conduct a comprehensive energy efficiency evaluation of 5G base stations.

[0011] Furthermore, the multi-level evaluation indexes of the comprehensive energy efficiency evaluation index system of the 5G base stations include: primary evaluation indexes and secondary evaluation indexes;

[0012] The primary evaluation indexes include: energy consumption evaluation index of 5G base stations, carbon emission evaluation index, 5G base station - power grid interaction evaluation index, data flow and load characteristic evaluation index; The secondary evaluation indexes are the sub-indexes of each evaluation index in the primary evaluation indexes; The sub-indexes of the energy consumption evaluation index of the 5G base stations include: base station hardware energy consumption, energy storage system energy consumption / efficiency, standby / idle energy consumption ratio, energy consumption per unit traffic flow; The sub-indexes of the carbon emission evaluation index include: direct carbon emission of the base station, indirect carbon emission of the base station, renewable energy utilization rate, carbon emission intensity per unit energy consumption; The sub-indexes of the 5G base station - power grid interaction evaluation index include: regulation load capacity ratio, energy storage system peak shaving capacity, power grid interaction participation degree, auxiliary service revenue / emission reduction amount; The sub-indexes of the data flow and load characteristic evaluation index include: service traffic scale, service traffic bandwidth utilization rate, energy consumption per unit data flow, proportion of delay-sensitive services, load peak-valley ratio, load fluctuation index, load coupling degree, load adaptability.

[0013] Even further, the calculation methods of the sub-indexes of the energy consumption evaluation index of the 5G base stations are as follows:

[0014] Base station hardware energy consumption E HWRefers to the total energy consumption of the core hardware during the operation of a 5G base station and the energy consumption of the corresponding refrigeration / auxiliary equipment. The formula is as follows:

[0015]

[0016] In the formula, P k represents the operating power of the k-th type of hardware device; t k represents the cumulative operating time; N represents the total number of device types to be counted;

[0017] Energy consumption / efficiency η of the energy storage system es Refers to the energy loss or efficiency of the base station energy storage system during the charging and discharging processes. The formula is as follows:

[0018]

[0019] In the formula, E in represents the energy stored during charging of the energy storage system; E out represents the actually available energy during discharging;

[0020] Standby / idle energy consumption ratio α idle Refers to the ratio of the energy consumption of a 5G base station in a low-traffic or idle state to the energy consumption at full load operation. The formula is as follows:

[0021]

[0022] In the formula, P idle represents the average power in the idle state; P full represents the average power at full load operation;

[0023] Energy consumption per unit traffic flow E data Refers to the electric energy consumed by the base station for each unit of data transmission or processing. The formula is as follows:

[0024]

[0025] In the formula, E total represents the total energy consumption of the base station during the statistical period, and V data represents the total data volume processed or transmitted by the base station during the period.

[0026] Furthermore, the calculation methods of the sub-indicators of the carbon emission assessment indicators are as follows:

[0027] Direct carbon emission C of the base station direct Refers to the direct generation of greenhouse gas emissions by the base station using its own power supply in case of power shortage or emergency. The formula is as follows:

[0028]

[0029] In the formula, F kRepresents the usage amount of the k-th type of fuel; EF k Represents the carbon emission factor corresponding to this type of fuel; N represents the total number of fuel types;

[0030] Indirect carbon emissions C of the base station indirect Refers to the indirect carbon emissions generated by the base station when obtaining electricity from the power grid. The formula is:

[0031] C indirect = E grid × EF grid (6),

[0032] In the formula, E grid Represents the total electricity consumption (kWh) obtained by the base station from the power grid; EF grid Represents the average carbon emission factor of the power grid;

[0033] Renewable energy utilization rate γ renew Refers to the proportion of the electricity generated by renewable energy in the total energy consumption of the base station. The formula is:

[0034]

[0035] In the formula, E renew Represents the electricity generated by renewable energy; E total Represents the total energy consumption of the base station;

[0036] Carbon emission intensity per unit energy consumption C eco Refers to the comprehensive carbon emissions corresponding to each 1 kWh of electricity consumed by the base station, including the sum of direct and indirect emissions. The formula is:

[0037]

[0038] Furthermore, the calculation method of the sub-indicators of the 5G base station-power grid interaction evaluation index is as follows:

[0039] Ratio of adjustable load capacity θ adj Refers to the ratio of the load capacity that the base station participates in scheduling or can moderately turn off / reduce power. The formula is:

[0040]

[0041] In the formula, P adj Represents the adjustable load power, P total Represents the normal full-load power of the base station;

[0042] Peak shaving capacity of the energy storage system θ peak Refers to the contribution made by the base station's energy storage to peak shaving the power grid when coordinated with the load. The formula is:

[0043]

[0044] wherein, ΔE peak represents the electricity consumption reduced during peak hours by using energy storage, and E peak,total represents the total electricity consumption of the base station during the original peak hours;

[0045] The grid interaction participation degree θ grid refers to the ratio of the number of times, time, or load regulation amount of the base station participating in grid demand response or price incentive within a cycle to its own adjustable load capacity. The formula is:

[0046]

[0047] wherein, P resp (t) represents the load power of the base station actually responding during the schedulable period t; P adj (t) represents the ideally adjustable load power; T resp represents the set of response periods; T represents the set of schedulable periods;

[0048] Auxiliary service revenue / emission reduction amount ΔC AS refers to the carbon emission reduction amount or economic benefit obtained by the base station and the grid through active peak shaving, valley filling, and energy storage participating in auxiliary services to reduce the grid's call for high-carbon units during peak periods. The formula is:

[0049]

[0050] wherein, ΔEF grid (t) represents the unit power generation carbon emission factor reduced after replacing high-carbon units during period t; Δt represents the period length.

[0051] Furthermore, the calculation method of the sub-indicators of the data stream and the load characteristic evaluation index is as follows:

[0052] Service traffic scale V data : refers to the total amount of data transmitted by the base station within the statistical period. The formula is:

[0053]

[0054] wherein, throughput(t) represents the downlink / uplink throughput at time t; Δt represents the time interval;

[0055] Service traffic bandwidth utilization rate β bandwidth : refers to the ratio of the bandwidth actually used by the base station to the allocable bandwidth. The formula is:

[0056]

[0057] Energy consumption per unit data stream E data refers to the energy consumed by the base station when processing a unit amount of data. The formula is:

[0058]

[0059] Proportion δ of delay-sensitive services delay It refers to the proportion of the traffic of delay-sensitive services that can be statistically counted among the services with different delay levels in the 5G network. The formula is:

[0060]

[0061] In the formula, represents the total data volume of delay-sensitive services;

[0062] Peak-to-valley load ratio ρ peak-valley It refers to the ratio of the highest load to the lowest load within the statistical period of the base station. The formula is:

[0063]

[0064] In the formula, P max and P min represent the maximum / minimum load power within the period respectively;

[0065] Load fluctuation index σ load It refers to quantitatively measuring the degree of fluctuation according to the change curve within the statistical period of the base station load. The formula is:

[0066]

[0067] In the formula, P load (t) represents the base station load power at time t; represents the average load power;

[0068] Load coupling degree φ data It refers to the correlation or temporal overlap degree between the load curve and the service traffic curve. The formula is:

[0069]

[0070] In the formula, Corr(·,·) represents the correlation coefficient;

[0071] Load adaptability α load It refers to the ability of the base station to dynamically adjust its own power supply, energy storage, or baseband unit start / stop strategy according to the real-time service load. The formula is:

[0072]

[0073] In the formula, ΔP config (t) represents the load adjustment amount at the base station configuration end at time t; ΔP demand (t) represents the actual demand change.

[0074] Furthermore, the calculation methods for the comprehensive evaluation scores of each typical operating scenario of the 5G base station and the fuzzy evaluation results of each typical operating scenario of the 5G base station are as follows:

[0075] Divide the typical operating scenarios of the 5G base station in different regions, different load characteristics, different equipment scales, or business type dimensions, and let m s represent the number of typical operating scenarios, and the scenario set is denoted as Collect the actual data of each indicator in each scenario S j and denote it as X j =(x 1,j , x 2,j , …, x n,j ), where x i,j represents the actual observed value of the indicator I j in the scenario S i ; Set the fuzzy comment set V = {v1, v2, …, v m}, and each comment level represents the corresponding performance level, all of which have fuzzy membership intervals; Adopt the triangular fuzzy function to define the membership function of the indicator I i at the evaluation level v k :

[0076]

[0077] In the formula, μ i,k (x) represents the membership degree of x belonging to the indicator I i at the evaluation level v k ;

[0078] Form the fuzzy evaluation matrix of all indicators in the same scenario for each evaluation level as follows:

[0079]

[0080] Multiply the comprehensive weight results of each indicator by the comprehensive weight to obtain the fuzzy comprehensive evaluation vector or matrix from the secondary evaluation indicators to the primary evaluation indicators:

[0081]

[0082] In the formula, represents the fuzzy weight vector of each secondary indicator; B j represents the fuzzy evaluation matrix of each secondary indicator; represents the fuzzy operator, represents the comprehensive membership degree vector of the scenario S j at each evaluation level;

[0083] Perform secondary weighted fuzzy evaluation based on the comprehensive weight results of the first-level indicators to obtain the fuzzy comprehensive membership degree vector of the first-level indicators The expression of the membership degree vector of the 5G base station is as follows:

[0084]

[0085] In the formula, represents the comprehensive membership degree of this scenario at the evaluation level v k ;

[0086] According to the principle of maximum membership degree, the comprehensive energy efficiency level of the 5G base station scenario S j is determined to be the level with the highest membership degree:

[0087]

[0088] Based on the membership degree vector of the 5G base station Calculate the comprehensive evaluation score S i of each typical operating scenario of the 5G base station, assign corresponding score values to the evaluation levels and perform weighted summation to obtain the fuzzy evaluation result Score(S j ) of each typical operating scenario of the 5G base station. The expression is as follows:

[0089]

[0090] In the formula, α k represents the score value corresponding to the preset level v k ;

[0091] Furthermore, the method for obtaining the criterion set of the key evaluation indicators for the comprehensive energy efficiency of the 5G base station is as follows:

[0092] Stack the membership degree vectors obtained for each indicator I i at each language evaluation V = {v1, v2,..., v m} level to form a matrix: i a i1 = [μ i2 , μ im ,..., μ Stack the membership degree vectors of all indicators row by row to form a matrix:

[0093]

[0094] In the formula, M represents the structured expression of the fuzzy membership degrees of each indicator at each evaluation level;

[0095] Perform standardization processing on each column of M. Let the mean of the j-th column be and the standard deviation be σ j , then the standardized matrix Z = [zij is defined as:

[0096]

[0097] Calculate the covariance matrix S for the standardized matrix Z:

[0098]

[0099] where the dimension of S is m×m;

[0100] Perform eigenvalue decomposition on S to obtain the eigenvalues λ1, λ2, …, λ m and the corresponding eigenvectors e1, e2, …, e m :

[0101] Se j =λ j e j , j = 1, 2, …, m. (43),

[0102] Set the cumulative variance contribution rate threshold and select the first k principal components that satisfy ;

[0103] Project the standardized matrix Z onto the selected principal components to obtain the score matrix:

[0104] Y = ZE (44),

[0105] where E = [e1 e2 … e k represents the matrix of the selected k eigenvectors;

[0106] At the same time, the loadings of the indicators on each principal component are:

[0107]

[0108] where L represents the correlation between the indicator characteristics of each evaluation level and each principal component;

[0109] For the indicator I i the communality is the sum of the squares of the loadings of the indicator on the retained principal components, that is:

[0110]

[0111] where represents the proportion of variance that the indicator I i can explain under the selected principal components; Determine a threshold θ for the communality according to the cumulative contribution rate. Usually, if it indicates that the indicator has a high contribution on the principal component, then retain it; if consider that the redundancy of the indicator is high or the contribution is low, and it can be eliminated;

[0112] Finally, the set of retained indicators is the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations.

[0113] Furthermore, the method for setting the high-efficiency benchmark and tolerance threshold is as follows:

[0114] Assume there are N 5G base station samples, and each sample contains the data of each indicator in the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations and the corresponding label information; for the key indicators of the i-th base station, the vector x i =(x i1 , x i2 , …, x ip ), where p represents the number of indicators in the criterion set; each sample contains the label information of the region R i , load type L i and equipment scale E i ; first, standardize each indicator. Let the standardized indicator vector be:

[0115] z i =(z i1 , z i2 , …, z ip ) (47);

[0116] Among them, for the j-th indicator:

[0117]

[0118] In the formula, represents the mean of the j-th indicator, and σ j represents the standard deviation;

[0119] According to the labels including different regions, different load types and different equipment scales, preliminarily group the samples, and define the grouping function:

[0120] G: {1, 2, …, N} → {1, 2,, M} (49),

[0121] Divide the samples into M preliminary categories, and the samples within each category have similar label attributes. For each grouping g (g = 1,, M), denote the samples in this group as:

[0122] D g ={z i |G(i) = g} (50),

[0123] Within each preliminary grouping D g , use the K-means clustering method to cluster the samples with the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations. Assume that K clusters are used within the grouping D g within the sample set Dg Clustering is performed, and its objective function is:

[0124]

[0125] In the formula, C k represents the k-th cluster; μ k represents the central vector of this cluster.

[0126] Each clustering cluster represents a typical operating scenario, and several clusters {C g , C g,1 , C g,2 , …, C g,k} are obtained within the grouping D; for the samples in a certain clustering cluster C g,k , the comprehensive evaluation score set {S i |z i ∈C g,k}, using the quantile method, an efficient baseline and a tolerance threshold are set. Among them, the 75th percentile p1 and the 25th percentile p2

[0127] For the clustering cluster C g,k , the expressions of the efficient baseline Efficient Baseline g,k and the tolerance threshold TolerableThreshol g,k are as follows:

[0128]

[0129] Among them, the median Q 50％ is defined as the overall reference score;

[0130] Summarize the baseline values of each clustering cluster C g,k within all groupings g = 1, …, M, establish a mapping relationship, and obtain the set of efficient baselines and tolerance thresholds for each typical operating scenario of the 5G base station:

[0131] f: (area, load type, equipment scale) →

[0132]

[0133] The present invention proposes a comprehensive energy efficiency evaluation system for 5G base stations, including an evaluation index system establishment module, a comprehensive weight calculation module, an evaluation calculation module, a key evaluation index criterion set acquisition module, an efficient benchmark and tolerance threshold setting module, and a comprehensive energy efficiency evaluation module. The evaluation index system establishment module is used to set multi-level evaluation indexes based on the energy consumption, carbon emissions, 5G base station-grid interaction, data flow and load characteristics of 5G base stations, group and link the evaluation indexes according to the dominance relationship to form an evaluation hierarchy structure, and establish a comprehensive energy efficiency evaluation index system for 5G base stations based on the evaluation hierarchy structure. The comprehensive weight calculation module is used to calculate the entropy weight and subjective weight of each index in the comprehensive energy efficiency evaluation index system of 5G base stations based on the entropy weight method and the analytic hierarchy process respectively, and perform weighted fusion to obtain the comprehensive weight of each index. The evaluation calculation module is used to establish a fuzzy evaluation matrix for each index for different typical operation scenarios of 5G base stations based on the comprehensive energy efficiency evaluation index system of 5G base stations, the comprehensive weight of each index, and the characteristics of typical operation scenarios of 5G base stations, using the hierarchical recursive fuzzy comprehensive evaluation method, and obtain the comprehensive evaluation score and fuzzy evaluation result of each typical operation scenario of 5G base stations through the calculation of the fuzzy evaluation matrix. The key evaluation index criterion set acquisition module is used to obtain the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations by using the dimensionality reduction and discrimination method of principal component analysis based on the fuzzy evaluation results of each typical operation scenario of 5G base stations. The efficient benchmark and tolerance threshold setting module is used to construct a comprehensive energy efficiency baseline for 5G base stations for each typical operation scenario of 5G base stations based on the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations, by clustering and grouping based on label division including different regions, different load types and different equipment scales, and set the efficient benchmark and tolerance threshold based on the comprehensive energy efficiency baseline of 5G base stations. The comprehensive energy efficiency evaluation module is used to compare the comprehensive evaluation scores of each typical operation scenario of 5G base stations with the set efficient benchmark and tolerance threshold to conduct a comprehensive energy efficiency evaluation of 5G base stations.

[0134] The specific implementation manners of each module in this system are the same as those described in the above method, and will not be elaborated here.

[0135] The present invention also proposes a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the comprehensive energy efficiency evaluation method for 5G base stations as described above.

[0136] Compared with the prior art, the present invention has the following beneficial effects: (1) Comprehensive evaluation of comprehensive energy efficiency: By integrating multi-dimensional indicators including energy consumption, carbon emissions, base station-grid interaction, data flow, and load characteristics, a comprehensive energy efficiency evaluation index system based on the coupling of "carbon emissions - energy efficiency" is constructed, breaking through the limitation of traditional energy efficiency evaluation that only focuses on single energy consumption, and realizing a comprehensive quantitative evaluation of the energy utilization, low-carbon attributes, and grid coordination ability of 5G base stations; (2) High scientificity of evaluation: Combining the entropy weight method driven by objective data and the analytic hierarchy process of subjective expert experience, the comprehensive weight of indicators is dynamically calculated, avoiding data deviation and reducing subjective bias, and significantly improving the scientificity of evaluation results; (3) High calculation efficiency: Based on the fuzzy comprehensive evaluation method of hierarchical recursion, differential evaluation is carried out for different typical operation scenarios of 5G base stations, and redundant indicators are removed through principal component analysis to form a key evaluation index criterion set, solving the problems of redundant evaluation indicators and low calculation efficiency in complex scenarios; (4) Precise evaluation benchmark basis: Based on different regions, different load types, and different equipment scales, through the clustering analysis method, a comprehensive energy efficiency baseline of 5G base stations considering multiple typical operation scenarios is constructed, and high-efficiency benchmarks and tolerance thresholds are set through the quantile method, providing a precise evaluation benchmark basis for 5G base stations, so as to intuitively and accurately identify energy efficiency gaps and formulate targeted optimization strategies; (5) Strong reliability: Carbon emissions and base station-grid interaction are simultaneously considered in the comprehensive energy efficiency evaluation index system, taking into account carbon emission control and base station operation reliability, promoting the green and low-carbon development of 5G base stations without sacrificing performance stability; (6) Strong universality: Through the full-chain technical path of "index construction - weight integration - scenario evaluation - dimensionality reduction criterion - baseline generation", a standardized and popularizable energy efficiency and carbon efficiency evaluation system for 5G base stations is formed, providing a systematic energy efficiency improvement methodology and tool support for the industry.

[0137] In summary, the present invention has a wide range of concerned dimensions. Based on the coupling of "carbon emissions - energy efficiency", a comprehensive energy efficiency evaluation index system is established, providing a standardized baseline for the comprehensive energy efficiency evaluation of 5G base stations, and providing clear guidance for the energy efficiency benchmarking of 5G base stations of different types and regions, enabling the subsequent energy efficiency improvement plan to be carried out systematically. Through the integration of multi-disciplinary methods and the collaboration of multi-source data, the present invention realizes the leap of 5G base station energy efficiency evaluation from a single dimension to comprehensiveness, from static benchmarking to dynamic scenario-based, and from experience-driven to data-model joint-driven, providing accurate and reliable decision-making support for the energy conservation and carbon reduction of 5G networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0138] Figure 1 It is a flowchart of the method of the present invention.

[0139] Figure 2 It is a system architecture diagram of the present invention.

[0140] Figure 3 It is the architecture diagram of the comprehensive energy efficiency evaluation index system for 5G base stations. Specific implementation manners

[0141] In order to make the technical problems, technical solutions and beneficial effects to be solved by this application clearer and more understandable, the following further details this application in combination with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0142] Embodiment 1

[0143] The comprehensive energy efficiency evaluation method for 5G base stations, the flowchart is as Figure 1 shown, and the specific steps are as follows.

[0144] Based on the energy consumption, carbon emissions, 5G base station - power grid interaction, data stream and load characteristics of 5G base stations, multi-level evaluation indicators are set, and the evaluation indicators are grouped and linked into an evaluation hierarchy according to the domination relationship, and a comprehensive energy efficiency evaluation index system for 5G base stations is established based on the evaluation hierarchy.

[0145] The architecture diagram of the comprehensive energy efficiency evaluation index system for 5G base stations is as Figure 3 shown, and the multi-level evaluation indicators include: primary evaluation indicators and secondary evaluation indicators. Among them, the primary evaluation indicators include: energy consumption evaluation indicators of 5G base stations, carbon emission evaluation indicators, 5G base station - power grid interaction evaluation indicators, data stream and load characteristic evaluation indicators; the secondary evaluation indicators are sub-indicators of each evaluation indicator in the primary evaluation indicators.

[0146] The sub-indicators of the energy consumption evaluation indicators of 5G base stations include: base station hardware energy consumption, energy storage system energy consumption / efficiency, standby / idle energy consumption ratio, and energy consumption per unit traffic flow. The calculation methods are as follows:

[0147] Base station hardware energy consumption E HW refers to the total energy consumption of the core hardware in the operation of the 5G base station and the energy consumption of the corresponding refrigeration / auxiliary equipment. The formula is:

[0148]

[0149] In the formula, P k represents the operating power of the k-th type of hardware equipment; t k represents the cumulative operating time; N represents the total number of equipment types to be counted;

[0150] Energy storage system energy consumption / efficiency η es refers to the energy loss or efficiency of the base station energy storage system during the charging and discharging processes. The formula is:

[0151]

[0152] Where, E in represents the energy stored in the energy storage system during charging; E out represents the actually available energy during discharging;

[0153] Standby / idle energy consumption ratio α idle refers to the ratio of the energy consumption of a 5G base station in a low-traffic or idle state to the energy consumption at full load operation, and the formula is:

[0154]

[0155] Where, P idle represents the average power in the idle state; P full represents the average power at full load operation;

[0156] Energy consumption per unit traffic flow E data refers to the electric energy consumed by the base station for transmitting or processing each unit of data, and the formula is:

[0157]

[0158] Where, E total represents the total energy consumption of the base station within the statistical period, and V data represents the total data volume processed or transmitted by the base station within the period.

[0159] The sub-indicators of the carbon emission assessment index include: direct carbon emission of the base station, indirect carbon emission of the base station, renewable energy utilization rate, and carbon emission intensity per unit energy consumption. The calculation methods are as follows:

[0160] Direct carbon emission C of the base station direct refers to the direct greenhouse gas emissions generated by the base station using its own power supply in case of power shortage or emergency, and the formula is:

[0161]

[0162] Where, F k represents the usage amount of the kth type of fuel; EF k represents the carbon emission factor corresponding to this type of fuel; N represents the total number of fuel types;

[0163] Indirect carbon emission C of the base station indirect refers to the indirect carbon emission amount generated by the base station obtaining electricity from the power grid, and the formula is:

[0164] C indirect = E grid × EF grid (6),

[0165] Where, E grid represents the total electricity obtained by the base station from the power grid (kWh); EF grid represents the average carbon emission factor of the power grid;

[0166] Renewable energy utilization rate γ renew It refers to the proportion of the electricity generated by renewable energy in the total energy consumption of the base station for power supply. The formula is:

[0167]

[0168] In the formula, E renew represents the electricity generated by renewable energy; E total represents the total energy consumption of the base station;

[0169] Carbon emission intensity per unit energy consumption C eco It refers to the comprehensive carbon emissions corresponding to every 1 kWh of electricity consumed by the base station, including the sum of direct and indirect emissions. The formula is:

[0170]

[0171] The sub - indicators of the 5G base station - power grid interaction evaluation index include: the proportion of adjustable load capacity, the peak - shaving ability of the energy storage system, the power grid interaction participation rate, and the auxiliary service revenue / emission reduction amount. The calculation methods are as follows:

[0172] Proportion of adjustable load capacity θ adj It refers to the proportion of the load capacity that the base station participates in scheduling or can be moderately shut down / power - reduced. The formula is:

[0173]

[0174] In the formula, P adj represents the adjustable load power, P total represents the normal full - load power of the base station;

[0175] Peak - shaving ability of the energy storage system θ peak It refers to the contribution made by the base station's energy storage to power grid peak - shaving when it cooperates with the load. The formula is:

[0176]

[0177] In the formula, ΔE peak represents the electricity consumption reduced by using the energy storage during peak hours, E peak,total represents the total electricity consumption of the base station during the original peak hours;

[0178] Power grid interaction participation rate θ grid It refers to the ratio of the number of times, time, or load regulation amount that the base station participates in power grid demand response or price incentives within a cycle to its own adjustable load capacity. The formula is:

[0179]

[0180] In the formula, P resp(t) represents the load power that the base station actually responds to during the schedulable period t; P adj (t) represents the ideally controllable load power; T resp represents the set of response periods; T represents the set of schedulable periods;

[0181] Auxiliary service revenue / emission reduction ΔC AS It refers to the carbon emissions reduced or economic benefits obtained by the base station and the power grid through active peak shaving, valley filling, and energy storage participating in auxiliary services to reduce the use of high-carbon units by the power grid during peak periods. The formula is:

[0182]

[0183] In the formula, ΔEF grid (t) represents the unit power generation carbon emission factor reduced after replacing the high-carbon unit in period t; Δt represents the period length.

[0184] The sub-indicators of the data stream and load characteristic evaluation index include: service traffic scale, service traffic bandwidth utilization rate, unit data stream energy consumption, proportion of delay-sensitive services, load peak-valley ratio, load fluctuation index, load coupling degree, and load adaptability. The calculation methods are as follows:

[0185] Service traffic scale V data : It refers to the total amount of data transmitted by the base station within the statistical period. The formula is:

[0186]

[0187] In the formula, throughput(t) represents the downlink / uplink throughput at time t; Δt represents the time interval;

[0188] Service traffic bandwidth utilization rate β bandwidth : It refers to the proportion of the bandwidth actually used by the base station to the allocable bandwidth. The formula is:

[0189]

[0190] Unit data stream energy consumption E data It refers to the energy consumed by the base station when processing a unit of data volume. The formula is:

[0191]

[0192] Proportion of delay-sensitive services δ delay It refers to the traffic proportion of delay-sensitive services that can be statistically counted among the services with different delay levels in the 5G network. The formula is:

[0193]

[0194] In the formula, Represents the total data volume of delay-sensitive services;

[0195] Peak-to-valley load ratio ρ peak-valley Refers to the ratio of the highest load to the lowest load within the base station statistical period, and the formula is:

[0196]

[0197] In the formula, P max and P min Respectively represent the maximum / minimum load power within the period;

[0198] Load fluctuation index σ load Refers to quantitatively measuring the degree of fluctuation according to the change curve within the base station load statistical period, and the formula is:

[0199]

[0200] In the formula, P load (t) represents the base station load power at time t; Represents the average load power;

[0201] Load coupling degree φ data Refers to the correlation or time-sequence overlap degree between the load curve and the service traffic curve, and the formula is:

[0202]

[0203] In the formula, Corr(·,·) represents the correlation coefficient;

[0204] Load adaptability degree α load Refers to the ability of the base station to dynamically adjust its own power supply, energy storage or baseband unit start-stop strategy according to the real-time service load, and the formula is:

[0205]

[0206] In the formula, ΔP config (t) represents the load adjustment amount at the base station configuration end at time t; ΔP demand (t) represents the actual demand change.

[0207] Analyze the importance of each index in the comprehensive energy efficiency evaluation index system of 5G base stations, calculate the entropy weight and subjective weight of each index in the comprehensive energy efficiency evaluation index system of 5G base stations respectively based on the entropy weight method and the analytic hierarchy process, and perform weighted fusion to obtain the comprehensive weight of each index.

[0208] Specifically, based on the entropy weight method, assume that there are m evaluation indexes, denoted as X = {X1, X2, …, X m}}, collect the index data, and there are n samples in total, forming an n×m original data matrix R = (rij ), where r ij represents the observed value of the i-th sample on the j-th index; the original matrix R is normalized using range normalization to obtain the normalized matrix Z = (z ij ), where z ij ∈ [0, 1], and the expression is as follows:

[0209] Normalization of positive indicators:

[0210] Normalization of negative indicators:

[0211] Calculate the proportion p ij of the j-th index under the i-th sample:

[0212]

[0213] Calculate the information entropy e j of the j-th index:

[0214]

[0215] In the formula, the larger e j , the smaller the difference between indicators among different samples, and the lower the information content. Conversely, the higher it is; it is agreed that when p ij = 0, the value is taken as 0;

[0216] Calculate the redundancy d j , and the larger the value, the more abundant the information contained in the index in the data. The expression is as follows:

[0217] d j = 1 - e j (25),

[0218] Calculate the entropy weight of the entropy weight method

[0219]

[0220] In the formula, j = 1,..., m, and the weight vector is the relative entropy weight of each index obtained based on the entropy weight method;

[0221] Specifically, based on the analytic hierarchy process, according to the Saaty 1 - 9 scale method: 1 means equally important, 3 means slightly important, 5 means significantly important, 7 means strongly important, 9 means extremely important, and 2, 4, 6, 8 are intermediate values. Each index is compared pairwise and assigned values to obtain the pairwise comparison scores a jk of relative importance, forming the judgment matrix A = (a jk ), where a jkIndicates the importance of index j relative to index k; perform consistency test:

[0222] Maximum eigenvalue:

[0223] Consistency index:

[0224] Random consistency ratio

[0225] Where RI is the average random consistency index, and it is required that CR < 0.1. If not passed, the judgment matrix needs to be adjusted again;

[0226] For the j-th row of the judgment matrix A, calculate its geometric mean:

[0227]

[0228] Normalize to obtain the weight vector and get the subjective weights of each index The formula is as follows:

[0229]

[0230] The calculation method of the comprehensive weight of each index is: construct a minimum deviation optimization model with the goal of minimizing the gap between subjective and objective weights:

[0231]

[0232] The constraint conditions are:

[0233]

[0234] Entropy weight And subjective weight Linearly weighted fusion to obtain the comprehensive weight The formula is as follows:

[0235]

[0236] Where α represents the adjustment coefficient, which is set according to the actual situation.

[0237] Based on the comprehensive energy efficiency evaluation index system of 5G base stations, the comprehensive weights of each index, and the typical operation scenario characteristics of 5G base stations, using the hierarchical recursive fuzzy comprehensive evaluation method, establish the fuzzy evaluation matrix of each index for different typical operation scenarios of 5G base stations respectively. Through the calculation of the fuzzy evaluation matrix, obtain the comprehensive evaluation scores of each typical operation scenario of 5G base stations and the fuzzy evaluation results of each typical operation scenario of 5G base stations.

[0238] Specifically, the typical operating scenarios of 5G base stations are divided according to different regions, different load characteristics, different equipment scales, or business type dimensions, and let m s represent the number of typical operating scenarios, and the scenario set is denoted as Collect the actual data of each indicator in each scenario S j , denoted as X j =(x 1,j ,x 2,j ,…,x n,j ), where x i,j represents the actual observed value of indicator I j in scenario S i ; Set the fuzzy comment set V = {v1, v2,..., v m}, and each comment level represents the corresponding performance level, all of which have fuzzy membership intervals; Adopt the triangular fuzzy function to define the membership function of indicator I i at evaluation level v k :

[0239]

[0240] In the formula, μ i,k (x) represents the membership degree of x belonging to indicator I i at evaluation level v k ;

[0241] Form the fuzzy evaluation matrix of all indicators in the same scenario for each evaluation level as follows:

[0242]

[0243] Weight the comprehensive weight results of each indicator according to the comprehensive weight to obtain the fuzzy comprehensive evaluation vector or matrix from the secondary evaluation indicator to the primary evaluation indicator:

[0244]

[0245] In the formula, represents the fuzzy weight vector of each secondary indicator; B j represents the fuzzy evaluation matrix of each secondary indicator; represents the fuzzy operator, represents the comprehensive membership degree vector of scenario S j at each evaluation level;

[0246] Conduct a secondary weighted fuzzy evaluation based on the comprehensive weight results of the primary indicators to obtain the fuzzy comprehensive membership degree vector The expression of the membership degree vector of the 5G base station is:

[0247]

[0248] In the formula, Indicates that the scene is at the evaluation level v k The comprehensive membership on ;

[0249] According to the maximum membership principle, the 5G base station scene S j The comprehensive energy efficiency level under the following conditions is determined as the level with the highest membership degree:

[0250]

[0251] Membership vector based on 5G base station Calculate the comprehensive evaluation score S of each typical operation scenario of 5G base station i , assign the corresponding score values to the evaluation level and perform weighted summation to obtain the fuzzy evaluation results Score (S) of each typical operation scenario of 5G base stations. j ), the expression is as follows:

[0252]

[0253] In the formula, α k Represents the pre-set level v k The corresponding score.

[0254] According to the fuzzy evaluation results of various typical operating scenarios of 5G base stations, the dimension reduction and discrimination method of principal component analysis was adopted to eliminate indicators with high redundancy and low contribution, and the criterion set of key evaluation indicators for the comprehensive energy efficiency of 5G base stations was obtained.

[0255] Specifically, each indicator I i The obtained evaluation V = {v1, v2, ..., v m The membership vector a on the level i =[μ i1 ,μ i2 ,…,μ im ],satisfy Stack the membership vectors of all indicators row by row to form a matrix:

[0256]

[0257] In the formula, M represents the structured expression of the fuzzy membership of each indicator at each evaluation level;

[0258] Standardize each column of M and set the mean of the jth column to be and the standard deviation is σ j , then the standardized matrix Z = [z ij ] is defined as:

[0259]

[0260] Calculate the covariance matrix S for the standardized matrix Z:

[0261]

[0262] where the dimension of S is m×m;

[0263] Perform eigenvalue decomposition on S to obtain the eigenvalues λ1, λ2, …, λ m and the corresponding eigenvectors e1, e2, …, e m :

[0264] Se j =λ j e j , j = 1, 2, …, m. (43),

[0265] Set the cumulative variance contribution rate threshold and select the first k principal components that satisfy τ = 0.85;

[0266] Project the standardized matrix Z onto the selected principal components to obtain the score matrix:

[0267] Y = ZE (44),

[0268] where E = [e1 e2 … e k represents the matrix of the selected k eigenvectors;

[0269] Meanwhile, the loadings of the indicators on each principal component are:

[0270]

[0271] where L represents the correlation between the indicator characteristics of each evaluation level and the principal components;

[0272] For the indicator I i the common factor variance is the sum of the squares of the loadings of the indicator on the retained principal components, that is:

[0273]

[0274] where represents the proportion of the variance that the indicator I i can explain under the selected principal components; determine a threshold θ for the common factor variance according to the cumulative contribution rate. Usually, if it indicates that the indicator has a high contribution on the principal component, then retain it; if considering that the redundancy of the indicator is high or the contribution is low, it can be excluded;

[0275] Finally, the set of retained metrics is the key evaluation metric criterion set for the comprehensive energy efficiency of 5G base stations.

[0276] Based on the key evaluation metric criterion set for the comprehensive energy efficiency of 5G base stations, clustering groups are divided according to tags including different regions, different load types, and different equipment scales. For each typical operating scenario of 5G base stations, a comprehensive energy efficiency baseline for 5G base stations is constructed, and an efficient baseline and tolerance threshold are set using the quantile method based on the comprehensive energy efficiency baseline of 5G base stations.

[0277] Specifically, assume there are N 5G base station samples, and each sample contains the data of each metric in the key evaluation metric criterion set for the comprehensive energy efficiency of 5G base stations and the corresponding tag information; for the key metrics of the i-th base station, they form a vector x i =(x i1 ,x i2 ,…,x ip ), where p represents the number of metrics in the criterion set; each sample contains the tag information of the region R i , load type L i , and equipment scale E i ; first, standardize each metric. Let the standardized metric vector be:

[0278] z i =(z i1 ,z i2 ,…,z ip ) (47);

[0279] Among them, for the j-th metric:

[0280]

[0281] In the formula, represents the mean of the j-th metric, and σ j represents the standard deviation;

[0282] According to the tags including different regions, different load types, and different equipment scales, the samples are initially grouped. Define the grouping function:

[0283] G:{1,2,…,N}→{1,2,…,M} (49),

[0284] The samples are divided into M initial categories, and the samples within each category have similar tag attributes. For each group g (g = 1,, M), denote the samples in this group as:

[0285] D g ={z i |G(i)=g} (50),

[0286] In each initial group D gThe K-means clustering method is used to cluster the samples using the 5G base station comprehensive energy efficiency key evaluation index criterion set. g K clusters are used in the sample set D g For clustering, the objective function is:

[0287]

[0288] In the formula, C k represents the kth cluster; μ k represents the center vector of the cluster,

[0289] Each cluster represents a typical operation scenario. g Several clusters {C g,1 ,C g,2 ,…,C g,k}; For a cluster C g,k The comprehensive evaluation score set {S i |z i ∈C g,k}, using the quantile method, set the high efficiency benchmark and tolerance threshold, where the p1 = 75% quantile is taken and p2 = 25% quantile

[0290] For cluster C g,k ,,Define Efficient Baseline g,k and TolerableThreshol g,k The expression is as follows:

[0291]

[0292] Among them, the median Q is defined 50％ As an overall reference score;

[0293] All groups g=1,…,M in each cluster C g,k The benchmark values are summarized and a mapping relationship is established to obtain the efficient benchmark and tolerance threshold set for each typical operation scenario of 5G base stations:

[0294] f: (region, load type, equipment size) →

[0295]

[0296] Finally, the comprehensive evaluation scores of each typical operating scenario of the 5G base station are compared with the set high-efficiency benchmark and tolerance threshold to conduct a comprehensive energy efficiency evaluation of the 5G base station.

[0297] Example 2

[0298] The comprehensive energy efficiency evaluation system of a 5G base station, the architecture diagram is as Figure 2 shown, and it consists of an evaluation index system establishment module, a comprehensive weight calculation module, an evaluation calculation module, a key evaluation index criterion set acquisition module, an efficient benchmark and tolerance threshold setting module, and a comprehensive energy efficiency evaluation module.

[0299] The evaluation index system establishment module is used to set multi-level evaluation indexes based on the energy consumption, carbon emissions, 5G base station - power grid interaction, data flow and load characteristics of the 5G base station, group the evaluation indexes according to the dominance relationship and link them into an evaluation hierarchy, and establish a comprehensive energy efficiency evaluation index system for the 5G base station according to the evaluation hierarchy.

[0300] The architecture diagram of the comprehensive energy efficiency evaluation index system of the 5G base station is as Figure 3 shown. The multi-level evaluation indexes include: primary evaluation indexes and secondary evaluation indexes. Among them, the primary evaluation indexes include: energy consumption evaluation index of the 5G base station, carbon emission evaluation index, 5G base station - power grid interaction evaluation index, data flow and load characteristic evaluation index; the secondary evaluation indexes are sub-indexes of each evaluation index in the primary evaluation indexes.

[0301] The sub-indexes of the energy consumption evaluation index of the 5G base station include: base station hardware energy consumption, energy storage system energy consumption / efficiency, standby / idle energy consumption ratio, energy consumption per unit traffic flow. The calculation methods are as follows:

[0302] Base station hardware energy consumption E HW refers to the total energy consumption of the core hardware of the 5G base station during operation and the energy consumption of the corresponding cooling / auxiliary equipment. The formula is:

[0303]

[0304] In the formula, P k represents the operating power of the kth type of hardware device; t k represents the cumulative operating time; N represents the total number of device types to be counted;

[0305] Energy storage system energy consumption / efficiency η es refers to the energy loss or efficiency of the base station energy storage system during the charging and discharging processes. The formula is:

[0306]

[0307] In the formula, E in represents the energy stored in the energy storage system during charging; E out represents the actually available energy during discharging;

[0308] Standby / idle energy consumption ratio α idleRefers to the ratio of the energy consumption of a 5G base station in a low or idle traffic state to its full-load operation energy consumption. The formula is:

[0309]

[0310] In the formula, P idle represents the average power in the idle state; P full represents the average power during full-load operation;

[0311] The energy consumption per unit traffic flow E data refers to the electric energy consumed by the base station for each unit of data transmission or processing. The formula is:

[0312]

[0313] In the formula, E total represents the total energy consumption of the base station within the statistical period, and V data represents the total data volume processed or transmitted by the base station during the period.

[0314] The sub-indicators of the carbon emission assessment index include: direct carbon emissions of the base station, indirect carbon emissions of the base station, renewable energy utilization rate, and carbon emission intensity per unit energy consumption. The calculation methods are as follows:

[0315] Direct carbon emissions C of the base station direct refers to the direct greenhouse gas emissions generated by the base station using its own power supply in case of power shortage or emergency. The formula is:

[0316]

[0317] In the formula, F k represents the usage amount of the k-th type of fuel; EF k represents the carbon emission factor corresponding to this type of fuel; N represents the total number of fuel types;

[0318] Indirect carbon emissions C of the base station indirect refers to the indirect carbon emissions generated by the base station obtaining electricity from the power grid. The formula is:

[0319] C indirect = E grid × EF grid (6),

[0320] In the formula, E grid represents the total electricity obtained by the base station from the power grid (kWh); EF grid represents the average carbon emission factor of the power grid;

[0321] Renewable energy utilization rate γ renew refers to the proportion of the electricity of renewable energy in the total energy consumption of the base station for energy supply. The formula is:

[0322]

[0323] In the formula, E renew represents the electricity provided by renewable energy; E total represents the total energy consumption of the base station;

[0324] The carbon emission intensity C of unit energy consumption eco refers to the comprehensive carbon emissions corresponding to each 1 kWh of electric energy consumed by the base station, including the sum of direct and indirect emissions. The formula is:

[0325]

[0326] The sub - indicators of the 5G base station - power grid interaction evaluation index include: the proportion of adjustable load capacity, the peak - shaving ability of the energy storage system, the power grid interaction participation degree, and the auxiliary service revenue / emission reduction amount. The calculation methods are as follows:

[0327] The proportion of adjustable load capacity θ adj refers to the proportion of the load capacity that the base station participates in scheduling or can be moderately shut down / reduced in power. The formula is:

[0328]

[0329] In the formula, P adj represents the adjustable load power, P total represents the normal full - load power of the base station;

[0330] The peak - shaving ability of the energy storage system θ peak refers to the contribution made by the base station's energy storage to peak - shaving for the power grid when it cooperates with the load. The formula is:

[0331]

[0332] In the formula, ΔE peak represents the electricity consumption reduced by using the energy storage during peak hours, E peak,total represents the total electricity consumption of the base station during the original peak hours;

[0333] The power grid interaction participation degree θ grid refers to the ratio of the number of times, time, or load regulation amount that the base station participates in power grid demand response or electricity price incentives within a cycle to its own adjustable load capacity. The formula is:

[0334]

[0335] In the formula, P resp (t) represents the load power that the base station actually responds to during the schedulable period t; P adj (t) represents the ideally adjustable load power; T resp represents the set of response periods; T represents the set of schedulable periods;

[0336] Auxiliary service revenue / emission reduction volume ΔC AS It refers to the carbon emissions reduced or economic benefits obtained by the base station and the power grid through active peak shaving, valley filling, and energy storage participating in auxiliary services, reducing the high-carbon units called by the power grid during peak periods. The formula is as follows:

[0337]

[0338] In the formula, ΔEF grid (t) represents the unit power generation carbon emission factor reduced after replacing high-carbon units during period t; Δt represents the period length.

[0339] The sub-indicators of the data flow and load characteristic evaluation indicators include: service traffic scale, service traffic bandwidth utilization rate, energy consumption per unit data flow, proportion of delay-sensitive services, load peak-valley ratio, load fluctuation index, load coupling degree, and load adaptability. The calculation methods are as follows:

[0340] Service traffic scale V data : It refers to the total amount of data transmitted by the base station within the statistical period. The formula is:

[0341]

[0342] In the formula, throughput(t) represents the downlink / uplink throughput at time t; Δt represents the time interval;

[0343] Service traffic bandwidth utilization rate β bandwidth : It refers to the ratio of the bandwidth actually used by the base station to the allocable bandwidth. The formula is:

[0344]

[0345] Energy consumption per unit data flow E data It refers to the energy consumed by the base station when processing a unit amount of data. The formula is:

[0346]

[0347] Proportion of delay-sensitive services δ delay It refers to the traffic proportion of delay-sensitive services that can be counted among services with different delay levels in the 5G network. The formula is:

[0348]

[0349] In the formula, represents the total data volume of delay-sensitive services;

[0350] Load peak-valley ratio ρ peak-valley It refers to the ratio of the highest load to the lowest load within the statistical period of the base station. The formula is:

[0351]

[0352] Wherein, P max and P min respectively represent the maximum / minimum load power within a period;

[0353] The load fluctuation index σ load refers to quantitatively measuring the degree of fluctuation according to the change curve within the base station load statistical period. The formula is:

[0354]

[0355] Wherein, P load (t) represents the base station load power at time t; represents the average load power;

[0356] The load coupling degree φ data refers to the correlation degree or time sequence overlap degree between the load curve and the service traffic curve. The formula is:

[0357]

[0358] Wherein, Corr(·,·) represents the correlation coefficient;

[0359] The load adaptability degree α load refers to the ability of the base station to dynamically adjust its own power supply, energy storage or baseband unit start / stop strategy according to the real-time service load. The formula is:

[0360]

[0361] Wherein, ΔP config (t) represents the load adjustment amount at the base station configuration end at time t; ΔP demand (t) represents the actual demand change.

[0362] The comprehensive weight calculation module is used to analyze the importance of each index in the comprehensive energy efficiency evaluation index system of 5G base stations, calculate the entropy weight and subjective weight of each index in the comprehensive energy efficiency evaluation index system of 5G base stations respectively based on the entropy weight method and the analytic hierarchy process, and perform weighted fusion to obtain the comprehensive weight of each index.

[0363] Specifically, based on the entropy weight method, assuming there are m evaluation indexes, denoted as X = {X1, X2,..., X m}), collect index data, and there are n samples, forming an n×m original data matrix R = (r ij ), where r ij represents the observed value of the i-th sample on the j-th index; perform normalization processing on the original matrix R using range normalization to obtain the normalized matrix Z = (z ij), where z ij ∈ [0, 1], and the expression is as follows:

[0364] Positive index normalization:

[0365] Negative index normalization:

[0366] Calculate the proportion p of the j-th index under the i-th sample ij :

[0367]

[0368] Calculate the information entropy e of the j-th index j :

[0369]

[0370] In the formula, the larger e j is, the smaller the difference between the indexes among different samples, the lower the information content. On the contrary, the higher it is; it is agreed that when p ij = 0, the value is taken as 0;

[0371] Calculate the redundancy d j , and the larger the value, the richer the information content contained in the index in the data. The expression is as follows:

[0372] d j = 1 - e j (25),

[0373] Calculate the entropy weight of the entropy weight method

[0374]

[0375] In the formula, j = 1,..., m, and the weight vector is the relative entropy weight of each index obtained based on the entropy weight method;

[0376] Specifically, based on the analytic hierarchy process, according to the Saaty 1-9 scale method: 1 is equally important, 3 is slightly important, 5 is significantly important, 7 is strongly important, 9 is extremely important, and 2, 4, 6, 8 are intermediate values. Compare and assign values to each pair of indexes to obtain the pairwise comparison scores a jk of relative importance, and form a judgment matrix A = (a jk ), where a jk represents the importance degree of index j relative to index k; conduct a consistency test:

[0377] Largest eigenvalue:

[0378] Consistency index:

[0379] Random consistency ratio

[0380] Where RI is the average random consistency index, and it is required that CR < 0.1. If it fails, the judgment matrix needs to be readjusted;

[0381] For the j-th row of the judgment matrix A, calculate its geometric mean:

[0382]

[0383] Normalize to obtain the weight vector and get the subjective weights of each index The formula is as follows:

[0384]

[0385] The calculation method of the comprehensive weight of each index is: construct a minimum deviation optimization model with the goal of minimizing the gap between subjective and objective weights:

[0386] min{||w - w AHP || 2 + ||w - w entropy || 2} (32);

[0387] The constraint conditions are:

[0388]

[0389] Combine the entropy weight with the subjective weight by linear weighted fusion to obtain the comprehensive weight The formula is as follows:

[0390]

[0391] Where α represents the adjustment coefficient, which is set according to the actual situation.

[0392] The evaluation calculation module is used to establish a fuzzy evaluation matrix for each index for different typical operation scenarios of the 5G base station based on the hierarchical recursive fuzzy comprehensive evaluation method according to the comprehensive energy efficiency evaluation index system of the 5G base station, the comprehensive weights of each index, and the typical operation scenario characteristics of the 5G base station. Through the calculation of the fuzzy evaluation matrix, the comprehensive evaluation score of each typical operation scenario of the 5G base station and the fuzzy evaluation result of each typical operation scenario of the 5G base station are obtained.

[0393] Specifically, divide the typical operation scenarios of the 5G base station in different regions, different load characteristics, different equipment scales or service type dimensions, and let m sIndicates the number of typical operating scenarios, and the set of scenarios is denoted as Collect the actual data of each metric in each scenario S j and denote it as X j =(x 1,j , x 2,j , …, x n,j ), where x i,j represents the actual observed value of metric I j in scenario S i ; Set the fuzzy comment set V = {v1, v2, …, v m}, and each comment level represents the corresponding performance level, all of which have fuzzy membership intervals; Adopt the triangular fuzzy function to define the membership function of metric I i at evaluation level v k :

[0394]

[0395] In the formula, μ i,k (x) represents the membership degree of x belonging to metric I i at evaluation level v k ;

[0396] Form the fuzzy evaluation matrix of all metrics in the same scenario for each evaluation level as follows:

[0397]

[0398] Weight the comprehensive weight results of each metric according to the comprehensive weight to obtain the fuzzy comprehensive evaluation vector or matrix from the secondary evaluation index to the primary evaluation index:

[0399]

[0400] In the formula, represents the fuzzy weight vector of each secondary index; B j represents the fuzzy evaluation matrix of each secondary index; represents the fuzzy operator, represents scenario S j at the comprehensive membership degree vector of each evaluation level;

[0401] Perform a secondary weighted fuzzy evaluation based on the comprehensive weight results of the primary index to obtain the fuzzy comprehensive membership degree vector The expression of the membership degree vector of the 5G base station is:

[0402]

[0403] In the formula, Indicates the comprehensive membership degree of this scenario at evaluation level v k ;

[0404] According to the principle of maximum membership degree, the comprehensive energy efficiency level of the 5G base station scenario S j is determined as the level with the highest membership degree:

[0405]

[0406] Based on the membership degree vector of the 5G base station Calculate the comprehensive evaluation score S of each typical operation scenario of the 5G base station i , assign the corresponding score values to the evaluation levels for weighted summation, and obtain the fuzzy evaluation result Score(S j ) of each typical operation scenario of the 5G base station. The expression is as follows:

[0407]

[0408] In the formula, α k represents the score value corresponding to the preset level v k .

[0409] The key evaluation index criterion set acquisition module is used to obtain the key evaluation index criterion set of the comprehensive energy efficiency of the 5G base station by using the dimension reduction and discrimination method of principal component analysis based on the fuzzy evaluation results of each typical operation scenario of the 5G base station, and eliminate the indexes with high redundancy and low contribution degree.

[0410] Specifically, for each index I i The membership degree vector a m obtained at each language evaluation V = {v1, v2,..., v i} level is i1 , μ i2 , …, μ im , satisfying Stack the membership degree vectors of all indexes row by row to form a matrix:

[0411]

[0412] In the formula, M represents the structured expression of the fuzzy membership degree of each index at each evaluation level;

[0413] Perform standardized processing on each column of M. Let the mean of the j-th column be and the standard deviation be σ j , then the standardized matrix

[0414] Z = [z ij is defined as:

[0415]

[0416] Calculate the covariance matrix S for the standardized matrix Z:

[0417]

[0418] Wherein, the dimension of S is m×m;

[0419] Perform eigenvalue decomposition on S to obtain the eigenvalues λ1, λ2, …, λ m and the corresponding eigenvectors e1, e2, …, e m :

[0420] Se j = λ j e j , j = 1, 2, …, m. (43),

[0421] Set the cumulative variance contribution rate threshold and select the first k principal components that satisfy τ = 0.85;

[0422] Project the standardized matrix Z onto the selected principal components to obtain the score matrix:

[0423] Y = ZE (44),

[0424] Wherein, E = [e1 e2 … e k represents the matrix of the selected k eigenvectors;

[0425] Meanwhile, the load of the index on each principal component is:

[0426]

[0427] Wherein, L represents the correlation between the index characteristics of each evaluation level and each principal component;

[0428] For the index I i the common factor variance is the sum of the squares of the loads of the index on the retained principal components, that is:

[0429]

[0430] Wherein, represents the proportion of the variance that the index I i can explain under the selected principal components; determine a threshold θ of the common factor variance according to the cumulative contribution rate. Generally, if it indicates that the index has a high contribution on the principal component, then retain it; if considering that the index has a high redundancy or low contribution, it can be eliminated;

[0431] Finally, the set of retained indexes is the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations.

[0432] An efficient benchmark and tolerance threshold setting module, which is used to construct an integrated energy efficiency baseline for 5G base stations for each typical operating scenario of 5G base stations based on the key evaluation index criterion set of the integrated energy efficiency of 5G base stations, and set the efficient benchmark and tolerance threshold by the quantile method based on the integrated energy efficiency baseline of 5G base stations, through clustering and grouping based on label division including different regions, different load types, and different equipment scales.

[0433] Specifically, assume that there are N 5G base station samples, and each sample contains the data of each index in the key evaluation index criterion set of the integrated energy efficiency of 5G base stations and the corresponding label information; for the key indicators of the i-th base station, they form a vector x i =(x i1 ,x i2 ,…,x ip ), where p represents the number of indicators in the criterion set; each sample contains the label information of the region R i , load type L i , and equipment scale E i ; first, standardize each indicator. Let the standardized indicator vector be:

[0434] z i =(z i1 ,z i2 ,…,z ip ) (47);

[0435] Among them, for the j-th indicator:

[0436]

[0437] In the formula, represents the mean of the j-th indicator, and σ j represents the standard deviation;

[0438] According to the labels including different regions, different load types, and different equipment scales, initially group the samples, and define the grouping function:

[0439] G:{1,2,…,N}→{1,2,…,M} (49),

[0440] Divide the samples into M initial categories, and the samples within each category have similar label attributes. For each group g (g = 1,, M), denote the samples in this group as:

[0441] D g ={z i |G(i)=g} (50),

[0442] In each initial group D gThe K-means clustering method is used to cluster the samples using the 5G base station comprehensive energy efficiency key evaluation index criterion set. g K clusters are used in the sample set D g For clustering, the objective function is:

[0443]

[0444] In the formula, C k represents the kth cluster; μ k represents the center vector of the cluster,

[0445] Each cluster represents a typical operation scenario. g Several clusters {C g,1 ,C g,2 ,…,C g,k}; For a cluster C g,k The comprehensive evaluation score set {S i |z i ∈C g,k}, using the quantile method, set the high efficiency benchmark and tolerance threshold, where the p1 = 75% quantile is taken and p2 = 25% quantile

[0446] For cluster C g,k ,,Define Efficient Baseline g,k and TolerableThreshol g,k The expression is as follows:

[0447]

[0448] Among them, the median Q is defined 50％ As an overall reference score;

[0449] All groups g=1,…,M in each cluster C g,k The benchmark values are summarized and a mapping relationship is established to obtain the efficient benchmark and tolerance threshold set for each typical operation scenario of 5G base stations:

[0450] f: (region, load type, equipment size) →

[0451]

[0452] The comprehensive energy efficiency evaluation module is used to compare the comprehensive evaluation scores of various typical operating scenarios of 5G base stations with the set high-efficiency benchmarks and tolerance thresholds, and conduct a comprehensive energy efficiency evaluation of 5G base stations.

[0453] Embodiment 3

[0454] A computer-readable storage medium stores a computer program. When the computer program is executed by a processor, it implements the comprehensive energy efficiency evaluation method of the 5G base station as described in Embodiment 1 above, and the comprehensive energy efficiency evaluation system of the 5G base station as described in Embodiment 2.

[0455] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present application can be implemented in various computer languages. For example, object-oriented programming languages such as Java, C++, Python, and interpreted scripting languages such as JavaScript.

[0456] The present application is described with reference to the flowcharts and / or block diagrams of methods, electronic devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, and the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing electronic devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing electronic devices generate for implementing in the process Figure 1 one process or multiple processes and / or blocks Figure 1 a device for the functions specified in one block or multiple blocks.

[0457] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing electronic device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements in the process Figure 1 one process or multiple processes and / or blocks Figure 1 a device for the functions specified in one block or multiple blocks.

[0458] These computer program instructions can also be loaded onto a computer or other programmable data processing electronic device, so that a series of operation steps are executed on the computer or other programmable electronic device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable electronic device provide for implementing in the process Figure 1steps of the functions specified in one or more processes and / or blocks Figure 1 steps of the functions specified in one or more blocks

[0459] Although the preferred embodiments of the present application have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present application.

[0460] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application is also intended to include these modifications and variations.

Claims

1. 5G base station comprehensive energy efficiency evaluation method, characterized in that, Including: Based on the energy consumption, carbon emissions, 5G base station - power grid interaction, data flow and load characteristics of 5G base stations, set multi-level evaluation indicators, group the evaluation indicators according to the dominance relationship and link them into an evaluation hierarchy, and establish a comprehensive energy efficiency evaluation index system for 5G base stations based on the evaluation hierarchy; Calculate the entropy weight and subjective weight of each index in the comprehensive energy efficiency evaluation index system of 5G base stations based on the entropy weight method and the analytic hierarchy process respectively, and perform weighted fusion to obtain the comprehensive weight of each index; Based on the comprehensive energy efficiency evaluation index system of 5G base stations, the comprehensive weight of each index, and the typical operation scenario characteristics of 5G base stations, based on the fuzzy comprehensive evaluation method of hierarchical recursion, establish a fuzzy evaluation matrix for each index for different typical operation scenarios of 5G base stations respectively. Through the calculation of the fuzzy evaluation matrix, obtain the comprehensive evaluation score of each typical operation scenario of 5G base stations and the fuzzy evaluation results of each typical operation scenario of 5G base stations; According to the fuzzy evaluation results of each typical operation scenario of 5G base stations, adopt the dimensionality reduction and discrimination method of principal component analysis to obtain the key evaluation index criterion set of the comprehensive energy efficiency of 5G base stations; According to the key evaluation index criterion set of the comprehensive energy efficiency of 5G base stations, based on the label division clustering grouping including different regions, different load types and different equipment scales, construct a comprehensive energy efficiency baseline for 5G base stations for each typical operation scenario of 5G base stations, and set an efficient baseline and a tolerance threshold based on the comprehensive energy efficiency baseline of 5G base stations; Compare the comprehensive evaluation score of each typical operation scenario of 5G base stations with the set efficient baseline and tolerance threshold to conduct a comprehensive energy efficiency evaluation of 5G base stations.

2. The comprehensive energy efficiency evaluation method for a 5G base station according to claim 1, characterized in that: The multi-level evaluation indicators of the comprehensive energy efficiency evaluation index system of the 5G base stations include: primary evaluation indicators and secondary evaluation indicators; The primary evaluation indicators include: energy consumption evaluation indicators of 5G base stations, carbon emission evaluation indicators, 5G base station - power grid interaction evaluation indicators, data flow and load characteristic evaluation indicators; The secondary evaluation indicators are sub-indicators of each evaluation indicator in the primary evaluation indicators; The sub-indicators of the energy consumption evaluation indicators of the 5G base stations include: base station hardware energy consumption, energy storage system energy consumption / efficiency, standby / idle energy consumption ratio, energy consumption per unit traffic flow; The sub-indicators of the carbon emission evaluation indicators include: direct carbon emissions of the base station, indirect carbon emissions of the base station, renewable energy utilization rate, carbon emission intensity per unit energy consumption; The sub-indicators of the 5G base station - power grid interaction evaluation indicators include: regulation load capacity ratio, energy storage system peak shaving capacity, power grid interaction participation degree, auxiliary service revenue / emission reduction amount; The sub-indicators of the data flow and load characteristic evaluation indicators include: service traffic scale, service traffic bandwidth utilization rate, energy consumption per unit data flow, proportion of delay-sensitive services, load peak-valley ratio, load fluctuation index, load coupling degree, load adaptability.

3. The comprehensive energy efficiency evaluation method for a 5G base station according to claim 2, wherein: The calculation methods of the sub-indicators of the energy consumption evaluation indicators of the 5G base stations are as follows: Base station hardware energy consumption E HW Refers to the total energy consumption of the core hardware of a 5G base station during operation and the energy consumption of the corresponding cooling / auxiliary equipment. The formula is as follows: Where, P k represents the operating power of the k-th type of hardware device; t k represents the cumulative operating time; N represents the total number of types of devices counted; Energy consumption / efficiency η of energy storage system es Refers to the energy loss or efficiency of the base station energy storage system during the charging and discharging processes. The formula is as follows: where E in represents the energy stored in the energy storage system during charging; E out represents the actually available energy during discharging; Standby / idle energy consumption ratio α idle Refers to the ratio of the energy consumption of a 5G base station in a low-traffic or idle state to its full-load operation energy consumption. The formula is: Wherein, P idle represents the average power in the idle state; P full represents the average power during full-load operation; Unit business flow energy consumption E data Refers to the electric energy consumed by the base station for each unit of data transmission or processing. The formula is as follows: where, E total represents the total energy consumption of the base station during the statistical period, and V data represents the total amount of data processed or transmitted by the base station during the period; The calculation methods of the sub-indicators of the carbon emission evaluation indicators are as follows: Direct carbon emissions C of the base station direct It refers to the direct generation of greenhouse gas emissions by the base station using its own power supply in case of power shortage or emergency. The formula is as follows: Where, F k represents the usage amount of the k-th type of fuel; EF k represents the carbon emission factor corresponding to this type of fuel; N represents the total number of fuel types; Indirect carbon emissions C of the base station indirect It refers to the indirect carbon emissions generated by the base station when obtaining electricity from the power grid. The formula is as follows: C indirect = E grid × EF grid (6), where E grid represents the total power (kWh) obtained by the base station from the power grid; EF grid represents the average carbon emission factor of the power grid; Renewable energy utilization rate γ renew It refers to the proportion of the electricity of renewable energy in the total energy consumption for base station power supply. The formula is as follows: Where, E renew represents the electricity provided by renewable energy; E total represents the total energy consumption of the base station; Carbon emission intensity per unit energy consumption C eco Refers to the total carbon emissions corresponding to each 1 kWh of electrical energy consumed by the base station, including the sum of direct and indirect emissions. The formula is: The calculation methods of the sub-indicators of the 5G base station - power grid interaction evaluation indicators are as follows: Adjustable load capacity ratio θ adj It refers to the ratio of the load capacity that the base station participates in scheduling or can moderately turn off / reduce power, and the formula is: Where, P adj represents the adjustable load power, and P total represents the normal full load power of the base station; Peak shaving capacity θ of the energy storage system peak It refers to the contribution made to the power grid peak shaving when the energy storage of the base station cooperates with the load. The formula is as follows: where ΔE peak represents the electricity consumption reduced during peak hours by using energy storage, and E peak,total represents the total electricity consumption of the base station during the original peak hours; Grid interaction participation degree θ grid It refers to the ratio of the number of times, time, or load regulation amount of a base station participating in grid demand response or electricity price incentives within a cycle to its own adjustable load capacity. The formula is as follows: where P resp (t) represents the load power actually responded by the base station within the schedulable period t; P adj (t) represents the ideally adjustable load power; T resp represents a set of response time periods; T represents the set of schedulable time periods; Ancillary service revenue / Emission reduction ΔC AS It refers to the quantified reduction in carbon emissions or the economic benefits obtained by the base station and the power grid through active peak shaving, valley filling, and energy storage participation in ancillary services, which reduces the high-carbon units called by the power grid during peak periods. The formula is as follows: where ΔEF grid (t) represents the reduced unit power generation carbon emission factor after replacing high-carbon units during period t; Δt represents the length of the period; The calculation methods for the sub - indicators of the data flow and load characteristic evaluation indicators are as follows: Business traffic volume V data : Refers to the total amount of data transmitted by the base station within the statistical period. The formula is: In the formula, throughput(t) represents the downlink / uplink throughput at time t; Δt represents the time interval; Business traffic bandwidth utilization rate β bandwidth : It refers to the ratio of the bandwidth actually used by the base station to the allocable bandwidth. The formula is: Unit data stream energy consumption E data Refers to the energy consumed by the base station when processing a unit of data volume, and the formula is: Proportion δ of delay-sensitive services delay It refers to the proportion of the traffic of delay-sensitive services that can be counted among the services with different delay levels in the 5G network. The formula is as follows: In the formula, represents the total data volume of delay-sensitive services; Load peak-valley ratio ρ peak-valley It refers to the ratio of the highest load to the lowest load within the statistical period of the base station. The formula is: where P max and P min represent the maximum / minimum load power within the period, respectively; Load fluctuation index σ load It refers to quantitatively measuring the degree of fluctuation according to the change curve within the base station load statistical period, and the formula is: where P load (t) represents the base station load power at time t; represents the average load power; Load coupling degree φ data It refers to the correlation or the degree of temporal sequence overlap between the load curve and the service traffic curve. The formula is as follows: In the formula, Corr(·,·) represents the correlation coefficient; Load adaptability α load It refers to the ability of the base station to dynamically adjust its own power supply, energy storage, or start / stop strategy of the baseband unit according to the real-time service load. The formula is as follows: where, ΔP config (t) represents the load adjustment amount at the base station configuration end at time t; ΔP demand (t) represents the actual demand change.

4. The comprehensive energy efficiency evaluation method for a 5G base station according to claim 3, wherein: The calculation methods for the comprehensive evaluation scores of each typical operation scenario of the 5G base station and the fuzzy evaluation results of each typical operation scenario of the 5G base station are: Divide the typical operating scenarios of 5G base stations in different regions, with different load characteristics, different equipment scales, or business type dimensions, and let m s represent the number of typical operating scenarios, and the scenario set is denoted as Collect the actual data of each indicator in each scenario S j , denoted as X j =(x 1,j ,x 2,j ,…,x n,j ), where x i,j represents the actual observed value of the indicator I j in the scenario S i ; Set the fuzzy evaluation set \(V = \{v_1, v_2, \ldots, v\}\), where each evaluation level represents the corresponding performance level and has a fuzzy membership degree interval; adopt the triangular fuzzy function to define the index \(I\) m}, and the membership function of the index \(I\) on the evaluation level \(v\) i at the evaluation level \(v\) k is as follows: where μ i,k (x) represents that x belongs to the index I i is the membership degree at the evaluation level v k ; The membership degree results of all indicators for each evaluation level in the same scenario are formed into a fuzzy evaluation matrix for this scenario, which is expressed as follows: The comprehensive weight results of each indicator are weighted according to the comprehensive weight to obtain the fuzzy comprehensive evaluation vector or matrix from the secondary evaluation indicators to the primary evaluation indicators: In the formula, represents the fuzzy weight vector of each secondary index; B j represents the fuzzy evaluation matrix of each secondary index; represents the fuzzy operator, represents the scenario S j represents the comprehensive membership degree vector at each evaluation level; Perform secondary weighted fuzzy evaluation based on the comprehensive weight results of the first-level indicators to obtain the fuzzy comprehensive membership degree vector of the first-level indicators The expression for the membership degree vector of the 5G base station is as follows: In the formula, represents the comprehensive membership degree of this scenario at evaluation level v k ; According to the principle of maximum membership degree, the comprehensive energy efficiency level under the 5G base station scenario S j is judged as the level with the highest membership degree: Membership vector based on 5G base stations Calculate the comprehensive evaluation score S of each typical operation scenario of the 5G base station i , assign corresponding scoring values to the evaluation levels and perform weighted summation to obtain the fuzzy evaluation result Score(S j ), and the expression is as follows: where α k represents the score corresponding to the preset level v k .

5. The comprehensive energy efficiency evaluation method for a 5G base station according to claim 4, characterized in that: The method for obtaining the criterion set of key evaluation indicators for the comprehensive energy efficiency of the 5G base station is: For each index I i The obtained membership degree vectors a at each language evaluation V = {v1, v2,..., v m} level i = [μ i1 , μ i2 , …, μ im , satisfying 0 ≤ μ ik ≤ 1, Stack the membership degree vectors of all indexes by rows to form a matrix: In the formula, M represents the structured expression of the fuzzy membership degree of each indicator at each evaluation level; Normalize each column of M. Let the mean of the j-th column be and the standard deviation be σ j . Then the normalized matrix Z = [z ij is defined as: Calculate the covariance matrix S for the standardized matrix Z: In the formula, the dimension of S is m×m; Perform eigenvalue decomposition on the S matrix to obtain the eigenvalues λ1, λ2, …, λ sorted in order m and the corresponding eigenvectors e1, e2, …, e m : Se j = λ j e j , j = 1, 2, …, m. (43) Set the cumulative variance contribution rate threshold and select the first k principal components that satisfy ; Project the standardized matrix Z onto the selected principal components to obtain the score matrix: Y = ZE(44), where \(E = [e_1\ e_2\ \cdots\ e k \) represents the matrix of \(k\) selected eigenvectors; At the same time, the load of the indicator on each principal component is: In the formula, L represents the correlation between the indicator characteristics of each evaluation level and each principal component; For index I i The common factor variance is the sum of the squares of the loadings of the index on the retained principal components, i.e.: In the formula, represents the index I i is the proportion of variance that can be explained under the selected principal component; a threshold θ of the common factor variance is determined according to the cumulative contribution rate. Generally, if it indicates that the index has a high contribution to the principal component, it is retained; if considering that the redundancy of the index is high or the contribution is low, it can be eliminated; Finally, the retained indicator set is the criterion set of key evaluation indicators for the comprehensive energy efficiency of the 5G base station.

6. The key feature index criterion set for the comprehensive energy efficiency of 5G base stations according to claim 5, characterized in that: The method for setting the high - efficiency baseline and tolerance threshold is: Assume that there are already N 5G base station samples, and each sample contains the data of each index in the key evaluation index criterion set of the comprehensive energy efficiency of 5G base stations and the corresponding label information; for the key indexes of the i-th base station, they form a vector x i =(x i1 , x i2 , …, x ip ), where p represents the number of indexes in the criterion set; each sample contains the label information of the region R i , load type L i and equipment scale E i ; first, standardize each index, and let the standardized index vector be: z i = (z i1 , z i2 , …, z ip ) (47); Among them, for the j - th indicator, there is: wherein, represents the mean of the j-th index, and σ j represents the standard deviation; According to labels including different regions, different load types, and different equipment scales, initially group the samples, and define the grouping function: G:{1,2,…,N}→{1,2,…,M} (49), Divide the samples into M initial categories. Samples within each category have similar label attributes. For each group g (g = 1,,M), denote the samples in this group as: D g = {z i | G(i) = g} (50), Within each preliminary group D g the K-means clustering method is used to cluster the samples using the key evaluation index criterion set of the comprehensive energy efficiency of 5G base stations. Suppose there are K clusters within the group D g and the sample set D g is clustered. Its objective function is: where C k represents the k-th cluster; μ k represents the center vector of the cluster, Each cluster represents a typical operating scenario and obtains several clusters {C g , C g,1 , …, C g,2 , …, C g,k} within the grouping D; for the samples in a certain cluster C g,k , the set of comprehensive evaluation scores {S i |z i ∈ C g,k}, using the quantile method, set the efficient benchmark and the tolerance threshold, where the 75th percentile and the 25th percentile are taken respectively. For cluster C g,k , define the Efficient Baseline g,k and the Tolerable Threshold g,k The expressions are as follows: Among them, the median Q 50％ is defined as the overall reference score; Summarize the reference values of each clustering cluster C within all groups g = 1,..., M, establish a mapping relationship, and obtain the efficient reference and tolerance threshold sets for each typical operating scenario of the 5G base station: g,k ​ f:(region, load type, equipment scale)→ {Efficient Baseline,Tolerable Threshold,Median Score}(54). 7.5G base station integrated energy efficiency evaluation system, characterized in that: It includes an evaluation index system establishment module, a comprehensive weight calculation module, an evaluation calculation module, a key evaluation indicator criterion set acquisition module, a high - efficiency baseline and tolerance threshold setting module, and a comprehensive energy efficiency evaluation module; The evaluation index system establishment module is used to set multi - level evaluation indicators based on the energy consumption, carbon emissions, 5G base station - power grid interaction, data flow, and load characteristics of the 5G base station, group and link the evaluation indicators according to the dominance relationship to form an evaluation hierarchy structure, and establish a comprehensive energy efficiency evaluation index system for the 5G base station based on the evaluation hierarchy structure; The comprehensive weight calculation module is used to calculate the entropy weight and subjective weight of each indicator in the comprehensive energy efficiency evaluation index system of the 5G base station based on the entropy weight method and the analytic hierarchy process respectively, and perform weighted fusion to obtain the comprehensive weight of each indicator; The evaluation calculation module is used to establish a fuzzy evaluation matrix for each indicator for different typical operation scenarios of the 5G base station based on the comprehensive energy efficiency evaluation index system of the 5G base station, the comprehensive weight of each indicator, and the typical operation scenario characteristics of the 5G base station, and through the calculation of the fuzzy evaluation matrix, obtain the comprehensive evaluation scores of each typical operation scenario of the 5G base station and the fuzzy evaluation results of each typical operation scenario of the 5G base station; The key evaluation index criterion set acquisition module is used to obtain the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations by using the dimensionality reduction and discrimination method of principal component analysis based on the fuzzy evaluation results of various typical operation scenarios of 5G base stations; The high-efficiency benchmark and tolerance threshold setting module is used to construct the comprehensive energy efficiency baseline for 5G base stations for each typical operation scenario of 5G base stations based on the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations, and cluster and group according to label division including different regions, different load types, and different equipment scales, and set the high-efficiency benchmark and tolerance threshold by the quantile method based on the comprehensive energy efficiency baseline of 5G base stations; The comprehensive energy efficiency evaluation module is used to compare the comprehensive evaluation scores of various typical operation scenarios of 5G base stations with the set high-efficiency benchmark and tolerance threshold to evaluate the comprehensive energy efficiency of 5G base stations.

8. The comprehensive energy efficiency evaluation system of a 5G base station according to claim 7, characterized in that: In the evaluation index system establishment module, the multi-level evaluation indexes of the comprehensive energy efficiency evaluation index system of 5G base stations include: primary evaluation indexes and secondary evaluation indexes; The primary evaluation indexes include: energy consumption evaluation index of 5G base stations, carbon emission evaluation index, 5G base station-power grid interaction evaluation index, data flow and load characteristic evaluation index; the secondary evaluation indexes are sub-indexes of each evaluation index in the primary evaluation indexes; The sub-indexes of the energy consumption evaluation index of 5G base stations include: base station hardware energy consumption, energy storage system energy consumption / efficiency, standby / idle energy consumption ratio, energy consumption per unit traffic flow; The sub-indexes of the carbon emission evaluation index include: direct carbon emission of base stations, indirect carbon emission of base stations, renewable energy utilization rate, carbon emission intensity per unit energy consumption; The sub-indexes of the 5G base station-power grid interaction evaluation index include: regulation load capacity ratio, energy storage system peak shaving capacity, power grid interaction participation degree, auxiliary service revenue / emission reduction amount; The sub-indexes of the data flow and load characteristic evaluation index include: service traffic scale, service traffic bandwidth utilization rate, energy consumption per unit data flow, proportion of delay-sensitive services, load peak-valley ratio, load fluctuation index, load coupling degree, load adaptability.

9. The comprehensive energy efficiency index system of the 5G base station according to claim 8, wherein: The calculation methods of the sub-indexes of the energy consumption evaluation index of 5G base stations are as follows: Base station hardware energy consumption E HW Refers to the total energy consumption of the core hardware and the energy consumption of the corresponding refrigeration / auxiliary equipment during the operation of the 5G base station. The formula is as follows: Where, P k represents the operating power of the k-th type of hardware device; t k represents the cumulative operating time; N represents the total number of types of devices to be counted; Energy consumption / efficiency η of energy storage system es Refers to the energy loss or efficiency of the base station energy storage system during the charging and discharging processes. The formula is as follows: Where, E in represents the energy stored in the energy storage system during charging; E out represents the actually available energy during discharging; Standby / idle energy consumption ratio α idle It refers to the ratio of the energy consumption of a 5G base station in a state of low traffic volume or idle state to the energy consumption under full load operation. The formula is as follows: Wherein, P idle represents the average power in the idle state; P full represents the average power during full-load operation; Unit service flow energy consumption E data It refers to the electric energy consumed by the base station for each unit of data transmission or processing. The formula is: Where, E total represents the total energy consumption of the base station during the statistical period, and V data represents the total amount of data processed or transmitted by the base station during the period; The calculation methods of the sub-indexes of the carbon emission evaluation index are as follows: Direct carbon emissions C of the base station direct Refers to the direct generation of greenhouse gas emissions by the base station using its own power supply in case of power shortage or emergency. The formula is as follows: where F k represents the usage amount of the k-th type of fuel; EF k represents the carbon emission factor corresponding to this type of fuel; N represents the total number of fuel types; Indirect carbon emissions C of the base station indirect It refers to the indirect carbon emissions generated by the base station when obtaining electricity from the power grid. The formula is as follows: C indirect = E grid × EF grid (6), where E grid represents the total electricity (kWh) obtained by the base station from the power grid; EF grid represents the average carbon emission factor of the power grid; Renewable energy utilization rate γ renew It refers to the proportion of the electricity of renewable energy in the total energy consumption for the power supply of the base station. The formula is as follows: Where, E renew represents the electricity provided by renewable energy; E total represents the total energy consumption of the base station; Carbon emission intensity per unit energy consumption C eco Refers to the comprehensive carbon emissions corresponding to each 1 kWh of electrical energy consumed by the base station, including the sum of direct and indirect emissions. The formula is: The calculation methods of the sub-indexes of the 5G base station-power grid interaction evaluation index are as follows: Adjustable load capacity ratio θ adj It refers to the ratio of the load capacity that the base station participates in scheduling or can moderately turn off / reduce power. The formula is: where, P adj represents the adjustable load power, and P total represents the normal full load power of the base station; Peaking capacity θ of the energy storage system peak It refers to the contribution made to the peaking of the power grid when the energy storage of the base station is coordinated with the load. The formula is as follows: Where, ΔE peak represents the electricity consumption reduced during peak hours by using energy storage, and E peak,total represents the total electricity consumption of the base station during the original peak hours; Grid interaction participation degree θ grid It refers to the ratio of the number of times, time or load regulation amount of the base station participating in the grid demand response or electricity price incentive within a cycle to its own adjustable load capacity. The formula is as follows: Wherein, P resp (t) represents the load power actually responded by the base station within the schedulable period t; P adj (t) represents the ideally adjustable load power; T resp represents a set of response time periods; T represents the set of schedulable time periods; Ancillary service revenue / emission reduction ΔC AS It refers to the carbon emissions reduced or economic benefits obtained by the base station and the power grid through active peak shaving, valley filling, and energy storage participating in ancillary services to reduce the high-carbon units called by the power grid during peak periods. The formula is: where ΔEF grid (t) represents the reduced unit power generation carbon emission factor after replacing high-carbon units during period t; Δt represents the length of the period; The calculation methods of the sub-indexes of the data flow and load characteristic evaluation index are as follows: Service traffic volume V data : It refers to the total amount of data transmitted by the base station within the statistical period. The formula is: In the formula, throughput(t) represents the downlink / uplink throughput at time t; Δt represents the time interval; Business traffic bandwidth utilization rate β bandwidth : It refers to the ratio of the bandwidth actually used by the base station to the allocable bandwidth. The formula is: Unit data stream energy consumption E data Refers to the energy consumed by the base station when processing a unit of data volume, and the formula is: Proportion δ of delay-sensitive services delay It refers to the proportion of the traffic of delay-sensitive services that can be counted among the services with different delay levels in the 5G network. The formula is as follows: wherein, represents the total data volume of delay-sensitive services; Load peak-valley ratio ρ peak-valley It refers to the ratio of the highest load to the lowest load within the statistical period of the base station. The formula is as follows: Wherein, P max and P min respectively represent the maximum / minimum load power within the period; Load fluctuation index σ load It refers to quantitatively measuring the degree of fluctuation according to the change curve within the base station load statistical period, and the formula is: where P load (t) represents the base station load power at time t; represents the average load power; Load coupling degree φ data Refers to the correlation or temporal overlap degree between the load curve and the service traffic curve. The formula is as follows: In the formula, Corr(·,·) represents the correlation coefficient; Load adaptation degree α load It refers to the ability of the base station to dynamically adjust its own power supply, energy storage or start / stop strategy of the baseband unit according to the real-time service load. The formula is as follows: where, ΔP config (t) represents the load adjustment amount at the base station configuration end at time t; ΔP demand (t) represents the actual demand change.

10. The comprehensive energy efficiency evaluation system for a 5G base station according to claim 9, characterized in that: In the evaluation calculation module, the calculation methods of the comprehensive evaluation scores of various typical operation scenarios of 5G base stations and the fuzzy evaluation results of various typical operation scenarios of 5G base stations are as follows: Divide the typical operating scenarios of 5G base stations in different regions, with different load characteristics, different equipment scales, or different service types. Let m s represent the number of typical operating scenarios, and the set of scenarios is denoted as Collect the actual data of each metric in each scenario S j , denoted as X j =(x 1,j , x 2,j , …, x n,j ), where x i,j represents the actual observed value of metric I j in scenario S i ; Set the fuzzy evaluation set \(V = \{v_1, v_2, \ldots, v\}\), and each evaluation level represents the corresponding performance level, all of which have fuzzy membership intervals; adopt the triangular fuzzy function to define the index \(I\) m}. i The membership function of the index \(I\) on the evaluation level \(v\) k is as follows: where μ i,k (x) represents that x belongs to index I i is the membership degree at evaluation level v k ; The membership degree results of all indexes to each evaluation level in the same scenario are formed into a fuzzy evaluation matrix for this scenario as follows: The comprehensive weight results of each index are weighted according to the comprehensive weight to obtain the fuzzy comprehensive evaluation vector or matrix from the secondary evaluation index to the primary evaluation index: wherein, represents the fuzzy weight vector of each secondary index; B j represents the fuzzy evaluation matrix of each secondary index; represents the fuzzy operator, represents the scenario S j represents the comprehensive membership degree vector at each evaluation level; Perform secondary weighted fuzzy evaluation based on the comprehensive weight results of the first-level indicators to obtain the fuzzy comprehensive membership degree vector of the first-level indicators The expression of the membership degree vector of the 5G base station is as follows: In the formula, represents the comprehensive membership degree of this scenario at evaluation level v k ; According to the principle of maximum membership degree, the comprehensive energy efficiency level in the 5G base station scenario S j is determined as the level with the highest membership degree: Membership vector based on 5G base stations Calculate the comprehensive evaluation score S of each typical operation scenario of the 5G base station i , assign corresponding scoring values to the evaluation levels for weighted summation to obtain the fuzzy evaluation result Score(S j ), and the expression is as follows: where α k represents the score corresponding to the preset level v k .

11. The comprehensive energy efficiency evaluation system of the 5G base station according to claim 10, characterized in that: In the key evaluation index criterion set acquisition module, the acquisition method of the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations is: For each index I i The obtained membership degree vectors a at each language evaluation V = {v1, v2,..., v m} level i = [μ i1 , μ i2 , …, μ im , satisfying 0 ≤ μ ik ≤ 1, Stack the membership degree vectors of all indexes row by row to form a matrix: where M represents the structured expression of the fuzzy membership degrees of each index under each evaluation level; Normalize each column of M. Let the mean of the j-th column be and the standard deviation be σ j . Then the standardized matrix Z = [z ij is defined as: Calculate the covariance matrix S for the standardized matrix Z: where the dimension of S is m×m; Perform eigenvalue decomposition on the S matrix to obtain the eigenvalues λ1, λ2, …, λ sorted in ascending order m and the corresponding eigenvectors e1, e2, …, e m : Se j = λ j e j , j = 1, 2, …, m. (43) Set the cumulative variance contribution rate threshold and select the first k principal components that satisfy ; Project the standardized matrix Z onto the selected principal components to obtain a score matrix: Y = ZE(44), where \(E = [e_1 e_2 \ldots e k \) represents the matrix of \(k\) selected eigenvectors; Meanwhile, the load of the index on each principal component is: where L represents the correlation between the index characteristics of each evaluation level and each principal component; For index I i The common factor variance of i is the sum of the squares of the loadings of this index on the retained principal components, that is: In the formula, represents the index I i is the proportion of variance that can be explained under the selected principal component; a threshold θ of the common factor variance is determined according to the cumulative contribution rate. Generally, if it indicates that the index has a high contribution to the principal component, it is retained; if considering that the redundancy of the index is high or the contribution is low, it can be eliminated; Finally, the retained index set is the key evaluation index criterion set for the comprehensive energy efficiency of 5G base stations.

12. The key feature index criterion set for the comprehensive energy efficiency of a 5G base station according to claim 11, characterized in that: In the high-efficiency baseline and tolerance threshold setting module, the method for setting the high-efficiency baseline and tolerance threshold is: Suppose there are already N 5G base station samples, and each sample contains the data of each indicator in the key evaluation indicator criterion set of the comprehensive energy efficiency of 5G base stations and the corresponding label information; the key indicators of the i-th base station form a vector x i =(x i1 , x i2 , …, x ip ), where p represents the number of indicators in the criterion set; each sample contains the label information of the region R i , load type L i and equipment scale E i ; first, standardize each indicator. Let the standardized indicator vector be: z i = (z i1 , z i2 , …, z ip ) (47); Among them, for the jth index, there is: In the formula, represents the mean of the j-th index, and σ j represents the standard deviation; Preliminarily group the samples according to labels including different regions, different load types, and different equipment scales, and define a grouping function: G:{1,2,…,N}→{1,2,…,M} (49), Divide the samples into M preliminary categories. Samples within each category have similar label attributes. For each group g (g = 1,, M), denote the samples in this group as: D g = {z i | G(i) = g} (50), Within each preliminary group D g the K-means clustering method is used to cluster the samples using the criterion set of key evaluation indicators for the comprehensive energy efficiency of 5G base stations. Suppose there are K clusters within the group D g and the sample set D g is clustered. Its objective function is: where C k represents the k-th cluster; μ k represents the center vector of this cluster, Each cluster represents a typical operating scenario, and several clusters {C g , C g,1 , …, C g,2 , …, C g,k} are obtained within grouping D; for the samples in a certain cluster C g,k , the set of comprehensive evaluation scores {S i | z i ∈ C g,k}, using the quantile method, set the efficient benchmark and tolerance threshold, where the 75% quantile and the 25% quantile are taken respectively. For the cluster C g,k , define the Efficient Baseline g,k and the Tolerable Threshold g,k The expressions are as follows: Among them, the median Q 50％ is defined as the overall reference score; Summarize the reference values of each cluster C within all groups g = 1, …, M, establish a mapping relationship, and obtain the efficient reference and tolerance threshold sets for each typical operating scenario of the 5G base station: g,k ​ f:(region, load type, equipment scale)→ {Efficient Baseline,Tolerable Threshold,Median Score} (54).

13. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, it implements the comprehensive energy efficiency evaluation method for a 5G base station according to any one of claims 1-6.

Images