City-level 5G base station cluster participation power grid interactive scheduling method and system based on population thermodynamic diagram

Abstract

The invention discloses a population thermodynamic diagram-based urban 5G base station cluster participation power grid interactive scheduling method and system, and relates to the field of 5G base station cluster participation power grid interactive scheduling. According to the method, the high-fidelity communication flow distribution is estimated by using the public population thermodynamic diagram and the base station coverage characteristics, and the flow distribution simulation problem caused by insufficient base station communication data is avoided. The invention provides a city-level communication system energy consumption optimization model based on distance-flow collaboration, variables and constraints corresponding to communication users are converted into simple geometric calculation results, and quantitative analysis of flow space transfer and diffusion effects among regional 5G base station clusters is realized. According to the invention, an interactive excitation scheduling mechanism between a power distribution system operator and a communication system operator is provided, a 5G base station cluster is actively attracted to participate in demand response by using a DLMP signal, and a targeted operation strategy is formulated to meet diverse demands of a regional power grid.

Description

Technical Field

[0001] This invention relates to the field of 5G base station clusters participating in power grid interactive scheduling, specifically a method and system for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps. Background Technology

[0002] With the increasing demand for highly reliable and low-latency communication from users, 5G base stations, as information transmission intermediaries, are forming 5G communication networks in cities in a more widespread and denser manner. 5G base station clusters are characterized by flexible control methods and a large overall scale, enabling them to respond orderly to grid regulation needs through aggregated management and control in the power market environment.

[0003] Although research on the participation of 5G base station clusters in grid interaction has received widespread attention, the following challenges remain for city-level 5G base station clusters in a power market environment: City-level 5G base station clusters typically contain tens of thousands of base stations, and important parameters such as traffic data from each base station are difficult to obtain directly due to commercial privacy concerns. Furthermore, the operational characteristics of 5G base stations are closely related to factors such as noise and distance, resulting in highly non-convex and nonlinear characteristics. For base station clusters of this scale, not only are there obstacles to direct, refined modeling, but also difficulties in solving the model. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for city-level 5G base station clusters to participate in power grid interactive scheduling based on population heat maps, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for city-level 5G base station clusters to participate in power grid interactive scheduling based on population heat maps includes the following steps: (1) Based on the population heat map, complete the city-level 5G base station cluster communication demand estimation and energy consumption optimization model. The model includes the 5G base station and user connection relationship model, the adjustable capacity model of 5G base station backup energy storage, the 5G base station operation energy consumption model, and the 5G base station cluster traffic distribution model based on the population heat map. Output base station communication demand data, energy consumption calculation rules and traffic distribution characteristics. (2) Based on the communication demand data and traffic distribution characteristics output in step (1), and combined with the load transfer chain reaction caused by base station hibernation, a dynamic hibernation model of 5G base station cluster considering hibernation diffusion effect is constructed. The model includes quantification and model construction of 5G base station cluster hibernation diffusion effect, construction of 5G base station dynamic hibernation model, and clarification of the constraints and load transfer rules of base station hibernation. (3) Integrate the energy consumption calculation rules and flow distribution characteristics of step (1) with the dormancy constraints and load transfer rules of step (2) to build a city-level 5G communication network participating in the power distribution system interactive regulation optimization model based on the marginal electricity price DLMP of distribution nodes. The model includes the city-level 5G base station cluster and power distribution network interaction framework, the construction of the city-level 5G base station cluster operation optimization model, and the construction of the power distribution network economic dispatch model. (4) For the city-level 5G communication network participating in the power distribution system interactive control optimization model in step (3), an iterative solution algorithm for the interactive optimization model between the city-level 5G base station cluster and the power distribution system operator DSO side is proposed.

[0006] As a further aspect of the present invention, step (1) is specifically as follows: Modeling the connection between 5G base stations and users: The transmission power is calculated based on the base station transmission power, channel fading coefficient, user data traffic and distance, and the upper limit constraints of bandwidth and transmission power are met. Communication quality is ensured by the signal-to-interference-plus-noise ratio, and mobile users access the nearest base station to maximize the access volume. Base station transmit power refers to the energy intensity of radio frequency signals radiated by a 5G base station within its coverage area. 5G base station Transmission power The result is obtained by calculation using equation (1): (1) in, It is connected to the base station A collection of users, User The corresponding transmission power, , , ; , These represent the sets of base stations and users, respectively. The calculation formula is as shown in equation (2): (2) in, This represents the noise power, specifically the average power of unwanted random interference at the receiver. and The channel fading coefficient is used to quantify the power attenuation and distortion of the signal during propagation; both are constants. User Data traffic, User With base station The distance between them and These are base stations Total bandwidth and its allocation to users The bandwidth of the two is related to the bandwidth of the two as shown in equation (3), where it is assumed that the transmission power of all users connected to the same base station is 1. and bandwidth All are consistent: (3) Equations (4)-(5) indicate the base station Both the transmit power and bandwidth have upper limits, namely... and : (4) (5) Meanwhile, transmission power Affecting base stations With users The signal-to-interference-plus-noise ratio and unit data transmission rate Their expressions are shown in equations (6) and (7) respectively: (6) (7) This refers to the ratio of the desired signal power to the total power of all interference and noise. In equation (6), the signal, interference, and noise are respectively expressed as... , and , Indicates 5G base station To users The maximum data rate that can be achieved when transmitting data. Except for base stations External users The set of other base stations that can be connected. Representing base stations With users Channel gain between Calculated using equation (8), when Less than the reference distance hour, Take fixed path loss value ;on the contrary, Follow The increase decreases exponentially. This is the path loss coefficient. The constraints satisfying equation (9) are: (8) (9) Combined formulas (1)-(9), mobile users under the same conditions With base station distance The smaller the value, the lower the corresponding transmission power of the base station. The smaller the number, the more mobile users and Communication quality can be guaranteed when connected to any base station, but when mobile users... When connected to a distant base station, mobile users are assigned to access the nearest 5G base station. Adjustable capability model for backup energy storage of 5G base stations: Based on the uninterrupted power supply requirements, the charge state limit is set, and the charging and discharging meet the upper limit of power, efficiency, and upper and lower limits of charge state. Simultaneous charging and discharging are not allowed in the same period, and the charge state is consistent at the beginning and end of the optimization cycle. The energy storage charge state limit considering the reliability of base station power supply is shown in equation (10), where the uninterrupted power supply duration of the 5G base station is... Not less than 3 hours; (10) in, For base stations Backup energy storage The minimum charge state is required to maintain continuous operation of the base station. For base stations Backup energy storage capacity, For base stations exist The standby continuous operating time at any given moment, i.e., the amount of backup energy storage must meet the needs of the base station. In the future Internal operational requirements For base stations Backup energy storage The lower limit refers to the minimum state of charge that an energy storage device should maintain to operate normally for a long period of time. The charging and discharging constraints of the base station backup energy storage are shown in equations (11)-(14): (11) (12) (13) (14) in, and Base station Backup energy storage The charging and discharging power at any given time and Base station The upper limit of charging and discharging power of backup energy storage. and This is a 0-1 variable used to determine whether the energy storage is being charged or discharged, and it is stipulated that the energy storage cannot be charged and discharged simultaneously within a certain time period. For base stations Backup energy storage State of charge at time t, and For base stations The charging and discharging efficiency of backup energy storage. For base stations The upper limit of the state of charge of backup energy storage is expressed by equation (14). After one optimization cycle, it needs to be compared with the state before optimization. Keep it the same at all times; 5G base station operating energy consumption model: The operating power includes constant power and variable power that varies with the load. During sleep, some devices are shut down to reduce the power to a fixed sleep state. The sleep state is represented by 0-1 variables and linearized by the Big M method. 5G base station operating power Divided into AAU power BBU power Power of other transmission equipment Power of heat dissipation equipment With auxiliary equipment power As shown in equation (15): (15) The power of heat dissipation and auxiliary equipment, as well as the power of some communication equipment that does not change with the communication load, is considered as the rated power, while the power of the communication equipment itself is... It changes with the communication load, therefore, equation (15) is replaced by equation (16), where Base station exist Operating power at any given time This refers to the constant portion of the power consumed by a base station under normal operating conditions. For base stations Fixed transmission power, For the size of the communication load, It is a base station unit Increased due to improvement : (16) When the base station is in sleep mode, some communication and auxiliary equipment are shut down, and the operating power is reduced accordingly but remains unchanged. The operating power after the base station goes into sleep mode is shown in equation (17), where This refers to the operating power of a base station in sleep mode. To characterize the base station exist A 01 variable indicating whether a time is sleepy. When the base station is in sleep mode, When the base station is working normally, equation (17) is linearized using the Big M method to obtain equation (18), where For a maximum value: (17) (18); Traffic distribution model: Population heat map data from Baidu Maps is used to simulate traffic distribution; Define base station Two types of areas within the coverage area and and will / The total flow in is defined as / : When the communication user is in the area And base station During normal operation, users will connect to the base station. ; When the communication user is in the area At that time, users can only access the base station. ; From the above definition, we know that Indicates all possible access base stations The set of user locations, This means that access can only be made to base stations outside the coverage area of ​​all other base stations. The set of user locations, satisfying Correspondingly, and Reflects and The sum of the internal flows satisfies ; For calculation and The value needs to be determined. and Specific scope; assuming base station The position coordinates are Service coverage area For base station Position coordinates are the center of the circle A circle with radius and the base station Other base stations with overlapping coverage are numbered as follows: ,Depend on As can be seen from the definition, Satisfying equation (19), where For base stations Complement of the covered area, Users at base stations Within the coverage area, and when it is simultaneously located at the base station When within the coverage area, it is related to the base station The distance should be greater than that to the base station. The distance must be sufficient; otherwise, it will connect to the base station. ;therefore, Satisfy equation (20), where Compared to base stations Distance from base station The set of closer coordinates is expressed as in equation (21); Traffic heatmaps divide the space into several regions, which are related to base stations. The overlapping areas of coverage are denoted as The corresponding traffic flow set is ,therefore, and The result is obtained by calculation using equation (22), where It is an area function; In addition, the coverage area of ​​all base stations is set as a circle with an equal radius: (twenty two).

[0007] As a further aspect of the present invention, in step (2): The dormancy diffusion effect refers to the phenomenon where the load of a single base station shifts to neighboring base stations after it goes into dormancy, triggering a chain reaction that affects the dormancy states of those neighboring base stations. This effect is achieved by calculating the load distribution of neighboring base stations before and after they go into dormancy. and The numerical changes can be used to quantitatively analyze the impact of 5G base station dormancy on adjacent base stations; Considering the combined sleep mode of adjacent base stations and It is calculated from equation (23), where and To disregard base station sleep mode and , Including base stations A certain combination of all external base stations, and for After all base stations go into hibernation and The numerical change, and Then it means to let and Non-zero The set of: (twenty three); There is an upper limit to the communication load that a base station can handle, therefore there is a corresponding limit. upper limit Meanwhile, considering service quality, if a base station needs to go into hibernation, the communication load within its coverage area that results in the loss of 5G communication service due to the inability to access other base stations must be less than a certain value. Also smaller than , and The specific values ​​can be determined by the communication system operator (MNO) based on historical data and equations (1)-(9), with the following constraints: (twenty four); and Calculated using formula (25), where This represents a type of user. express The set of user types at that time express The set of user types at that time Indicates when type The distance between the user and the base station The nearest base station number, Representation type The total number of users in the formula (25) is established. , The correlation between the number of the two types of users is established by introducing a coefficient that correlates base station traffic with the number of users. Formula (25) achieves the following: and Calculation: (25); The dynamic sleep model introduces 0-1 variables to represent the sleep combination state of the base station. After linearization, the variable product term is eliminated. The model includes the influence of all base station sleep combinations and reflects the sleep diffusion effect brought about by the spatiotemporal coupling characteristics of the load. Considering the impact of adjacent base station hibernation, the dynamic hibernation model of the 5G base station cluster is constructed using equations (23)-(26). As a binary variable, it is calculated using equation (27) that when all base stations in the set are in a dormant state, The value is 1 if it is not 0 otherwise. Compared with equation (24), equation (26) takes into account the dormant state of all base stations. Different base stations can still pass through and influence each other even if their coverage areas do not overlap. and ; (26) (27) Since there are more than two 0-1 variables in equation (27), it can be linearized using equations (28)-(29), where The value is equal to and Therefore, equation (26) is equivalent to the linear constraint equation (30), and the linear dynamic dormancy model of 5G base stations is equation (28)-(30): (28) (29) (30).

[0008] As a further aspect of the present invention, the interaction framework between the city-level 5G base station cluster and the power distribution network in step (3) is a three-layer architecture: The upper-level flow distribution layer estimates flow demand using population heat maps and historical flow data; The middle layer of the mobile network is where the communication system operator (MNO) formulates base station operation strategies, including dynamic hibernation and energy storage management. At the lower distribution network layer, the distribution system operator (DSO) releases dynamic marginal electricity prices for distribution network nodes based on predicted node demand. In this scenario, the communication system operator (MNO) first estimates the traffic distribution of each 5G base station based on the overall signal coverage and the location and density of people obtained from population heat maps. Subsequently, the communication system operator (MNO) constructs an optimized energy management model for all 5G base stations.

[0009] As a further aspect of the present invention, in step (3): City-level 5G base station cluster operation optimization model: The objective function is to minimize the power purchase cost and state switching energy consumption, and the constraints include backup energy storage charging and discharging constraints, dynamic hibernation constraints, and energy consumption constraints. objective function (31) (32) (33) In formula (31) For base stations The power node is located at DLMP value at time, The power consumption required to switch the operating status of a base station once. To characterize the base station exist The 01 variable indicating whether the running state changes at any given time can be represented by equation (32) and linearized by equation (33). To characterize and Size of a 01 variable, when hour ,when hour ; Constraints (1) 5G base station backup energy storage constraints (34) (35) (36) (37) in and For charging and discharging power, and This is the upper limit of charging and discharging power. and The 01 variables are used to determine whether the energy storage is being charged or discharged; For backup energy storage , and For charging and discharging efficiency, For backup energy storage Upper limit; (2) Dynamic sleep constraints of 5G base stations (38) (39) (40) The constraints for each part are the same as those in formulas (28)-(30); (3) Energy consumption constraints of 5G base stations (41) in For base station operating power, The constant power when the base station is activated. For base station sleep power, For the size of the communication load, It is a base station unit Increased power due to upgrades For base station sleep variables, and It is a maximum value; Distribution network economic dispatch model: The objective function is to minimize the electricity purchase cost of the upper-level grid, distributed wind power and photovoltaics. The constraints include node power balance, line current and node voltage constraints. By solving the dual problem, the marginal electricity price (DLMP) of the distribution node, which includes energy, congestion and loss costs, is obtained. The marginal price of distribution nodes (DLMP) refers to the marginal cost to the entire distribution system when an additional unit of electricity consumption is added or reduced at a specific power node in the distribution network. DLMP represents the dynamic and refined electricity price in the distribution network in both time and space. It not only reflects the energy cost of electricity consumption, but also includes the congestion cost and loss cost of the distribution network transporting electricity to a specific power node. Equation (42) is the economic dispatch model for the distribution network, and its objective function is to minimize the power purchase cost of the distribution network. Number the nodes. A set of distribution network nodes. , and These represent the unit cost of purchasing electricity from the upstream power grid, distributed wind power, and photovoltaic power, respectively. , , and , , This represents the corresponding purchased active and reactive power. Equations (42b) and (42c) are the node power balance constraints, where... and For the line The active and reactive power flowing through, and For node load power, The square of the line current. This represents the Lagrange multiplier corresponding to constraint (42b). For line impedance, and Then it is the square of the node voltage; The marginal electricity price (DLMP) at the distribution node can be obtained by solving the dual problem of equation (42), i.e. Equations (42f)-(42j) represent the upper and lower bound constraints for some variables, where... The upper and lower limits are and , The upper limit is , The upper and lower limits are and , The upper and lower limits are and , The upper and lower limits are and ; (42a) (42b) (42c) (42d) (42e) (42f) (42g) (42h) (42i) (42j).

[0010] As a further aspect of the present invention, the iterative solution algorithm in step (4) includes: Distribution system operator (DSO) side: (1) Predict the total load of the distribution network based on historical data With 5G base station load And distributed new energy output, making , ,in For the number of iterations, This refers to the 5G base station load during the first iteration. (2) Update the 5G base station load based on the base station load data reported by the communication system operator MNO: , ,in The difference between the updated 5G base station load and the original forecast. (3) Construct an economic dispatch model for the distribution network by combining the above data with relevant parameters of the distribution network; (4) Solve the dual problem of the model in step (3) to obtain , For the first During the next iteration ; (5) If , or If so, the process will terminate; otherwise, it will... Provided to the MNO (Mobile Network Operator) and proceeding with step (3) of the MNO's process; wherein , It is used to determine the time interval between iterations. The threshold of minute change This is the upper limit of the number of iterations; On the MNO side of the communication system operator: (1) Conduct traffic distribution simulation based on the pedestrian flow heat map and historical traffic data of the base station; (2) Calculate the impact of all base station sleep combinations on the traffic of other base stations based on the traffic distribution; (3) Based on the data in steps (1)-(2) and the data provided by the DSO Construct an optimization model for 5G base station cluster operation; (4) Solve the model in step (3) to obtain And report it to the DSO, so that ; In the first iteration, the distribution system operator (DSO) directly executes steps (1)-(4), while the communication system operator (MNO) executes steps (1)-(2) in parallel. Subsequently, the distribution system operator (DSO) will... The information is passed to the communication system operator MNO, enabling it to continue with steps (3)-(4). Then, in the second iteration, the communication system operator MNO provides the distribution system operator DSO with... Afterwards, the Distribution System Operator (DSO) executes steps (2)-(5); if the termination condition is met, the process ends; otherwise, the Distribution System Operator (DSO) will... The information is passed to the communication system operator MNO, which then performs steps (3)-(4) again.

[0011] As a further aspect of the present invention, the linearization process of the dynamic dormancy model eliminates the product terms of more than two 0-1 variables by introducing intermediate variables, ensuring that the model can be solved efficiently.

[0012] As a further aspect of the present invention, the power purchase cost of the city-level 5G base station cluster operation optimization model is positively correlated with DLMP and base station load, and the energy consumption for state switching is determined by the number of switching times and the power consumption per switching.

[0013] Furthermore, the present invention also provides a city-level 5G base station cluster participating in power grid interactive scheduling system based on population heat maps, comprising: The communication demand and energy consumption modeling module is used to perform the aforementioned communication demand estimation and energy consumption optimization modeling. The dynamic dormancy modeling module is used for dynamic dormancy modeling that considers the dormancy diffusion effect. The interactive control framework modeling module is used to perform the interactive control optimization framework modeling based on the marginal electricity price of distribution nodes (DLMP). The iterative solution module is used to execute the iterative solution algorithm for the interactive optimization model.

[0014] Compared with the prior art, the beneficial effects of the present invention are: Due to the sheer number of 5G base station clusters in cities and the difficulty in obtaining complete real traffic data for each base station, existing research often employs simplified modeling methods using uniform sampling when estimating communication traffic distribution. This approach fails to accurately reflect the true distribution characteristics of wide-area communication traffic within cities. This invention utilizes public population heat maps and base station coverage characteristics to estimate high-fidelity communication traffic distribution, avoiding the traffic distribution simulation challenges caused by insufficient base station communication data.

[0015] Due to the highly spatiotemporally coupled characteristics of urban 5G base station clusters, existing base station operation modeling methods struggle to construct operational optimization models that balance modeling accuracy and computational feasibility. This invention proposes an energy consumption optimization model for urban communication systems based on distance-traffic coordination. This model transforms the variables and constraints corresponding to communication users into simplified geometric calculation results, enabling quantitative analysis of the spatial transfer and diffusion effects of traffic between regional 5G base station clusters.

[0016] Existing methods for the interactive regulation of communication networks and power grids typically focus on the macro-level aggregation and regulation capabilities of 5G base station clusters, generally neglecting their widespread distribution within urban geographical spaces and their specific impact on differentiated power nodes in the distribution network. This invention proposes an interactive incentive scheduling mechanism between distribution system operators and communication system operators. It utilizes DLMP signals to proactively attract 5G base station clusters to participate in demand response and formulates targeted operation strategies to meet the diverse needs of the regional power grid. Attached Figure Description

[0017] Figure 1 This is a 5G base station traffic allocation map based on a population heatmap in an embodiment of the present invention.

[0018] Figure 2 As described in the embodiments of the present invention and Schematic diagram.

[0019] Figure 3 This is a schematic diagram of the base station dormancy diffusion effect in an embodiment of the present invention.

[0020] Figure 4 This is a diagram of a three-layer communication-power interaction framework in an embodiment of the present invention.

[0021] Figure 5 This is a flowchart illustrating the solution process of the coordination and scheduling method in this embodiment of the invention.

[0022] Figure 6 This is an improved IEEE 123 node distribution network diagram in an embodiment of the present invention.

[0023] Figure 7(a) is a geographical distribution map of 5G base stations in an embodiment of the present invention.

[0024] Figure 7(b) is a typical load curve of various 5G base stations in the embodiments of the present invention.

[0025] Figure 8 This is a coordinate diagram of nodes 1, 2, and 3 and their corresponding 5G base stations in an embodiment of the present invention.

[0026] Figure 9 This is a population heat map of Shanghai in an embodiment of the present invention.

[0027] Figure 10 This is a diagram showing the base station communication load distribution obtained using two different methods from Case 3 and Case 4 in this embodiment of the invention.

[0028] Figure 11 This is a graph showing the total power consumption of a 5G base station cluster in an embodiment of the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Please see Figure 1 In this embodiment of the invention, a method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps includes the following steps: (1) Based on the population heat map, complete the city-level 5G base station cluster communication demand estimation and energy consumption optimization model. The model includes the 5G base station and user connection relationship model, the adjustable capacity model of 5G base station backup energy storage, the 5G base station operation energy consumption model, and the 5G base station cluster traffic distribution model based on the population heat map. Output base station communication demand data, energy consumption calculation rules and traffic distribution characteristics. (2) Based on the communication demand data and traffic distribution characteristics output in step (1), and combined with the load transfer chain reaction caused by base station hibernation, a dynamic hibernation model of 5G base station cluster considering hibernation diffusion effect is constructed. The model includes quantification and model construction of 5G base station cluster hibernation diffusion effect, construction of 5G base station dynamic hibernation model, and clarification of the constraints and load transfer rules of base station hibernation. (3) Integrate the energy consumption calculation rules and flow distribution characteristics of step (1) with the dormancy constraints and load transfer rules of step (2) to build a city-level 5G communication network participating in the power distribution system interactive regulation optimization model based on the marginal electricity price DLMP of distribution nodes. The model includes the city-level 5G base station cluster and power distribution network interaction framework, the construction of the city-level 5G base station cluster operation optimization model, and the construction of the power distribution network economic dispatch model. (4) For the city-level 5G communication network participating in the power distribution system interactive control optimization model in step (3), an iterative solution algorithm for the interactive optimization model between the city-level 5G base station cluster and the power distribution system operator DSO side is proposed.

[0031] The specific steps of step (1) above are as follows: Step 1.1 Modeling the connection relationship between 5G base stations and users The connection between users and 5G base stations is primarily determined by the base station's power consumption and communication quality, both of which are closely related to the base station's transmit power and communication rate. Base station transmit power refers to the energy intensity of the radio frequency signals radiated by the 5G base station within its coverage area. 5G base station Transmission power It can be calculated using equation (1).

[0032] (1) in, It is connected to the base station A collection of users, User The corresponding transmission power. , , . , These represent the sets of base stations and users, respectively. The calculation formula is as shown in equation (2): (2) in, This represents the noise power, i.e., the average power of unwanted random interference at the receiver. and The channel fading coefficient is used to quantify the power attenuation and distortion of the signal during propagation; both are constants. User Data traffic, User With base station The distance between them. and These are base stations Total bandwidth and its allocation to users The bandwidth of the two is shown in equation (3). In this patent, it is assumed that the transmission power of all users connected to the same base station is... and bandwidth All are consistent.

[0033] (3) Equations (4)-(5) indicate the base station Both the transmit power and bandwidth have upper limits, namely... and .

[0034] (4) (5) Meanwhile, transmission power It also affects base stations With users Signal-to-Interference plus Noise Ratio (SINR) and unit data transmission rate The expressions are shown in equations (6) and (7), respectively.

[0035] (6) (7) This refers to the ratio of the desired signal power to the total power of all interference and noise. In equation (6), the signal, interference, and noise are respectively expressed as... , and . Indicates 5G base station To users The maximum data rate that can be achieved when transmitting data. Except for base stations External users A collection of other base stations that can be connected. Representing base stations With users Channel gain between. It can be calculated using equation (8), when Less than the reference distance hour, Take fixed path loss value ;on the contrary, Follow The increase shows an exponential decrease. This is the path loss coefficient. To ensure user communication quality, The constraint of equation (9) must be satisfied.

[0036] (8) (9) From the combined equations (1)-(9), it can be seen that, under the same conditions, mobile users With base station distance The smaller the value, the lower the corresponding transmission power of the base station. The smaller, but and The relationship between them is quite complex. It can be considered that mobile users... and Communication quality can be guaranteed when connected to any base station in the system, but when connected to a base station that is farther away, The increase in the number of 5G base stations not only increases the power costs of the base stations but also limits the number of other users that can connect to them. Therefore, mobile users will be assigned to connect to the nearest 5G base station, thereby providing service to as many users as possible while ensuring communication quality.

[0037] Step 1.2 Adjustable Capability Model of Backup Energy Storage for 5G Base Stations The adjustable capability model of backup energy storage for 5G base stations is directly related to the power demand of the backup power supply supporting uninterrupted power supply. The state-of-charge limit of energy storage considering the reliability of base station power supply is shown in equation (10). Wherein, the uninterrupted power supply duration of the 5G base station... It is usually no less than 3 hours.

[0038] (10) in, For base stations Backup energy storage The minimum charge state is required to maintain continuous operation of the base station. For base stations The capacity of backup energy storage. For base stations exist The standby continuous operating time at any given moment, i.e., the amount of backup energy storage must meet the needs of the base station. In the future Internal operational requirements. For base stations Backup energy storage The lower limit refers to the minimum state of charge that an energy storage device should maintain to operate normally for an extended period of time.

[0039] The charging and discharging constraints of the base station backup energy storage are shown in equations (11)-(14).

[0040] (11) (12) (13) (14) in, and Base station Backup energy storage The charging and discharging power at any given time and Base station The upper limit of charging and discharging power of backup energy storage. and This is a 0-1 variable used to determine whether the energy storage is being charged or discharged, and it is stipulated that the energy storage cannot be charged or discharged simultaneously within a certain time period. For base stations Backup energy storage State of charge at time t, and For base stations The charging and discharging efficiency of backup energy storage. For base stations The upper limit of the state of charge of backup energy storage. Equation (14) then represents the upper limit of the state of charge of backup energy storage. After one optimization cycle, it needs to be compared with the state before optimization. Always keep the same.

[0041] Step 1.3 5G Base Station Operation Energy Consumption Model 5G base station operating power It can be divided into AAU power BBU power Power of other transmission equipment Power of heat dissipation equipment With auxiliary equipment power As shown in equation (15).

[0042] (15) Among these, the power of heat dissipation and auxiliary equipment, as well as the power of some communication equipment that does not change with the communication load, can be considered as the rated power, while the power of the communication equipment itself... This will change with the communication load. Therefore, equation (16) can be used instead of equation (15), where Base station exist Operating power at any given time This refers to the constant portion of the power consumed by a base station under normal operating conditions. For base stations Fixed transmission power, For the size of the communication load, It is a base station unit Increased due to improvement .

[0043] (16) When the base station is in sleep mode, its operating power will decrease accordingly and remain constant because some communication and auxiliary equipment will be shut down. The operating power of the base station after sleep mode is shown in equation (17), where This refers to the operating power of a base station in sleep mode. To characterize the base station exist A 01 variable indicating whether a time is sleepy. When the base station is in sleep mode, The base station is working normally. Using the Big M method, equation (17) can be linearized to obtain equation (18), where... It is a maximum value.

[0044] (17) (18) Step 1.4 5G Base Station Cluster Traffic Distribution Model Based on Population Heat Map The distribution of communication traffic has a significant impact on the operational strategies of 5G base station clusters. However, since the specific locations of all users connected to 5G base stations in urban mobile networks cannot be directly obtained, existing studies typically assume that users are evenly distributed within the signal coverage area. But due to the large coverage area of ​​5G base stations, there is a significant difference in service distribution between densely populated areas and sparsely populated surrounding areas, making it difficult to accurately estimate actual traffic demand using the traditional assumption of uniform traffic distribution.

[0045] Therefore, this patent introduces a population heatmap from Baidu Maps to simulate traffic distribution. Given the widespread use of Baidu Maps in my country and the significant 5G communication demands generated when using this software, this population heatmap data can effectively reflect the traffic distribution of 5G base stations.

[0046] In the appendix Figure 1In the image, each rectangular area with dimensions of 100-200 meters represents a heatmap value in Baidu Maps. The darker the rectangle's color, the higher the corresponding heatmap value. Users are evenly distributed within each small rectangular area, and user density is positively correlated with heatmap values. Therefore, for 5G base stations with large coverage areas, traffic distribution within their coverage area can be simulated more accurately within a smaller area. The circles in the image represent the communication coverage of different 5G base stations, with different colored dots representing users connected to different base stations; a denser distribution of dots indicates higher communication traffic in that area. Solid black triangles and circles centered on them represent 5G base stations and their coverage areas, while pink dashed lines separate users connected to different base stations.

[0047] While the aforementioned communication traffic estimation method based on population heatmaps can simulate the geographical distribution of users and their communication traffic demands, the highly overlapping coverage areas between base stations make it difficult to accurately determine the traffic load borne by each base station. To investigate the traffic distribution and dynamic changes of each base station under a base station dormancy strategy, we now define a base station... Two types of areas within the coverage area and and will / The total flow in is defined as / : 1. When the communication user is in the area And base station During normal operation, users will connect to the base station. ; 2. When the communication user is in the area At that time, users can only access the base station. .

[0048] From the above definition, it can be seen that, Indicates all possible access base stations The set of user locations, This means that access can only be made to base stations outside the coverage area of ​​all other base stations. The set of user locations, therefore satisfying Correspondingly, and Reflects and The sum of the internal flows satisfies .

[0049] For calculation and The value needs to be determined. and The specific scope. Assuming a base station. The position coordinates are Service coverage area For base station Position coordinates are the center of the circle A circle with radius and the base station Other base stations with overlapping coverage are numbered as follows: .Depend on As can be seen from the definition, Satisfying equation (19), where For base stations The complement of the covered area. Users at base stations Within the coverage area, and when it is simultaneously located at the base station When within the coverage area, it is related to the base station The distance should be greater than that to the base station. The distance must be sufficient; otherwise, it will connect to the base station. .therefore, Satisfy equation (20), where Compared to base stations Distance from base station The set of closer coordinates is expressed as in equation (21). (See attached...) Figure 2 Taking the three 5G base stations in the image as an example, the red area represents... The blue area is The sum of the red and blue areas is .

[0050] (19) (20) (twenty one) Traffic heatmaps divide the space into several regions, which are related to base stations. The overlapping areas of coverage are denoted as The corresponding traffic flow set is .therefore, and It can be calculated using equation (22), where It is an area function. It is worth noting that, as shown in equations (19)-(22), and The calculation only involves union and intersection operations on different sets of regions, while and The value of depends on the proportional relationship between the areas of different regions. Because... , , and The calculations do not involve the specific shape of the region, so this method is universally applicable to coverage areas of any shape.

[0051] In real-world scenarios, base station coverage areas often exhibit irregular shapes due to differences in equipment parameters and obstruction from surrounding buildings. However, since this patent does not focus on signal obstruction and related issues, and follows the main approach of existing research, all base station coverage areas in this patent are set as circles of equal radius.

[0052] (twenty two) The specific steps of step (2) above are as follows: Step 2.1 5G Base Station Cluster Dormant Diffusion Effect and its Model Construction The 5G base station hibernation diffusion effect refers to the phenomenon where, when a single 5G base station enters hibernation, the communication load within its original service area shifts to neighboring base stations, potentially triggering a chain reaction. This can cause some neighboring base stations to postpone or exit hibernation due to increased load, and even further affect the operation of outermost base stations. This phenomenon reflects the dynamic coupling characteristics of the spatial distribution of communication load within a base station cluster.

[0053] The following is attached Figure 3 For example, base station 2 goes into hibernation and transfers its communication load to adjacent base stations 1 and 3, preventing base stations 1 and 3 from going into hibernation. Base station 3 then continues to influence the hibernation plan of its adjacent base station 4, thus spreading all the way to base station n, allowing base station n to go into hibernation. Conversely, if base station 2 is working normally, the communication load of base stations 1 and 3 can be transferred to base station 2, allowing base stations 1 and 3 to go into hibernation, which in turn spreads to base station n, preventing base station n from going into hibernation.

[0054] By calculating the time before and after the adjacent base station goes into sleep mode and The numerical changes can be used to quantitatively analyze the impact of 5G base station hibernation on adjacent base stations. It is worth noting that this section only considers the impact of different combinations of adjacent base station hibernation on a single base station, without considering the base station hibernation diffusion effect.

[0055] Considering the combined sleep mode of adjacent base stations and It can be calculated from equation (23), where and To disregard base station sleep mode and . Including base stations A certain combination of all external base stations, and for After all base stations go into hibernation and The numerical change. and Then it means to let and Non-zero A set of.

[0056] (twenty three) To ensure the communication quality for accessing users, this patent proposes the constraint of equation (24). Since there is an upper limit to the communication load that a base station can handle, there is a corresponding constraint. upper limit Meanwhile, considering service quality, if a base station needs to go into hibernation, the communication load within its coverage area that results in the loss of 5G communication service due to the inability to access other base stations should be less than a certain value. It should also be smaller than . and The specific values ​​can be determined by MNO based on historical data and formulas (1)-(9).

[0057] (twenty four) Based on the above analysis, and It can be calculated using formula (25), where This represents a type of user. express The set of user types at that time express The set of user types at that time Indicates when type The distance between the user and the base station The nearest base station number, Representation type The total number of users in the system. Formula (25) establishes... , The correlation between the number of users and the number of users in both categories is established by introducing a coefficient that correlates base station traffic with the number of users. This formula achieves the following: and The calculation.

[0058] (25) Step 2.2 Construction of 5G Base Station Dynamic Sleep Model Considering the impact of adjacent base station hibernation, the dynamic hibernation model of 5G base station cluster can be constructed using equations (23)-(26). As a binary variable, it can be calculated using equation (27). When all base stations in the set are in a dormant state, The value is 1 if it is not 0 otherwise. Compared to equation (24), equation (26) additionally considers the dormant state of all base stations. Different base stations, even if their coverage areas do not overlap, can still be affected by [various factors]. and Since equation (26) includes all base station sleep combinations, this means that the model can reflect the sleep diffusion effect described above.

[0059] (26) (27) Since there are more than two 0-1 variables in equation (27), it can be linearized using equations (28)-(29), where The value is equal to and The product of . Therefore, equation (26) can be equivalent to linear constraint equation (30), and the linear dynamic dormancy model of 5G base station is equation (28)-(30).

[0060] (28) (29) (30) The specific steps of step (3) above are as follows: Step 3.1 City-level 5G base station cluster and power distribution network interaction framework To explore the regulatory potential of urban communication systems in providing flexibility to the operation of power distribution systems, this patent constructs an interactive framework between urban-level 5G base station clusters and power distribution networks.

[0061] As attached Figure 4 As shown, the framework comprises three layers. The upper layer, the traffic distribution layer, reflects the traffic demand of communication devices and can be estimated using population heatmaps and historical traffic data from 5G base stations. The middle layer, the mobile network layer, contains all 5G base stations within the regional communication system. The Mobile Network Operator (MNO) will formulate operational strategies based on the location and communication coverage parameters of the 5G base stations to support the communication needs of corresponding users. The lower layer, the power distribution network, corresponds to the power distribution system in the same area. The Distribution Network Operator (DSO) must ensure the safe and stable operation of the power system.

[0062] Specifically, in the city-level communication-power interaction process, the Distribution Locational Marginal Price (DLMP) will be dynamically released based on predicted node demand to attract price-sensitive users to adjust their electricity consumption strategies. In this context, the Management Network Operator (MNO) will first estimate the traffic distribution of each 5G base station based on overall signal coverage and population location and density data obtained from population heat maps. Subsequently, the MNO will construct an optimized energy management model for all 5G base stations, aiming to reduce overall electricity purchase costs while meeting communication demands, while ensuring service quality. This model will consider the communication traffic transfer and diffusion effects generated by 5G base station sleep modes when formulating dynamic sleep scheduling and backup energy storage management strategies for city-level 5G base station clusters, thereby developing an optimal operating strategy that balances the spatiotemporal coupling characteristics of communication traffic.

[0063] Step 3.2 Construction of City-level 5G Base Station Cluster Operation Optimization Model Mobile network operators (MNOs) can reduce overall electricity costs by comprehensively considering battery energy storage system operation strategies and base station sleep strategies. The optimization model for city-level 5G base station cluster operation is as follows: 1. Objective function (31) (32) (33) In formula (31) For base stations The power node is located at The specific meaning and calculation method of the DLMP value at time will be explained in step 3.3. The power consumption required to switch the operating status of a base station once. To characterize the base station exist The 01 variable indicating whether the running state changes at any given time can be represented by equation (32) and linearized using equation (33). To characterize and Size of a 01 variable, when hour ,when hour .

[0064] 2. Constraints (1) 5G base station backup energy storage constraints (34) (35) (36) (37) in and For charging and discharging power, and This is the upper limit of charging and discharging power. and The 01 variables are used to determine whether the energy storage is being charged or discharged. For backup energy storage , and For charging and discharging efficiency, For backup energy storage Upper limit.

[0065] (2) Dynamic sleep constraints of 5G base stations (38) (39) (40) The constraints for each part are the same as those in formulas (28)-(30).

[0066] (3) Energy consumption constraints of 5G base stations (41) in For base station operating power, The constant power when the base station is activated. This refers to the base station's sleep power. For the size of the communication load, It is a base station unit Increased power due to enhancement. For base station sleep variables, and It is a maximum value.

[0067] Step 3.3 Construction of Economic Dispatch Model for Distribution Network DLMP refers to the marginal cost incurred by the entire distribution system when an additional unit of electricity consumption is added or removed at a specific power node in the distribution network. Therefore, DLMP characterizes the dynamic, fine-grained electricity price in the distribution network across both time and space. It not only reflects the energy cost of electricity consumption but also includes the congestion and loss costs of transporting electricity from the distribution network to specific power nodes.

[0068] Equation (42) is the economic dispatch model for the distribution network, and its objective function is to minimize the power purchase cost of the distribution network. Number the nodes. It is a set of distribution network nodes. , and These represent the unit cost of purchasing electricity from the upstream power grid, distributed wind power, and photovoltaic power, respectively. , , and , , This represents the corresponding purchased active and reactive power. Equations (42b) and (42c) are the node power balance constraints, where... and For the line The active and reactive power flowing through, and For node load power, The square of the line current. Then it represents the Lagrange multiplier corresponding to constraint (42b). For line impedance, and This is the square of the node voltage.

[0069] The DLMP can be obtained by solving the dual problem of equation (42), i.e. Equations (42f)-(42j) represent the upper and lower bound constraints for some variables, where... The upper and lower limits are and , The upper limit is , The upper and lower limits are and , The upper and lower limits are and , The upper and lower limits are and .

[0070] (42a) (42b) (42c) (42d) (42e) (42f) (42g) (42h) (42i) (42j) The specific steps of step (4) above are as follows: Appendix Figure 5 The specific solution process for the city-level 5G base station cluster participating in the power distribution network scheduling model is shown, which can be divided into two parts: DSO and MNO. 1. DSO (1) Predict the total load of the distribution network based on historical data With 5G base station load And distributed renewable energy output. (This is followed by an unrelated sentence fragment: "Order...") , ,in For the number of iterations, This refers to the 5G base station load during the first iteration.

[0071] (2) Update the 5G base station load based on the base station load data reported by the MNO: , ,in This represents the difference between the updated 5G base station load and the original forecast.

[0072] (3) Combine the above data with relevant parameters of the distribution network to construct an economic dispatch model for the distribution network.

[0073] (4) Solve the dual problem of the model in step (3) to obtain . For the first During the next iteration .

[0074] (5) If ( )or If so, the process will terminate. Otherwise, it will... Provided to the MNO, and the third step of the MNO process continues. . It is used to determine the time interval between iterations. The threshold of minute change. This represents the upper limit of the number of iterations. To control the overall solution time, in this section... Set to 5.

[0075] 2. MNO (1) Simulate the flow distribution based on the heat map of pedestrian flow and the historical flow data of the base station.

[0076] (2) Calculate the impact of all base station dormant combinations on the traffic of other base stations based on the traffic distribution.

[0077] (3) Based on the data in steps (1)-(2) and the data provided by the DSO Construct an optimization model for the operation of 5G base station clusters.

[0078] (4) Solve the model in step (3) to obtain And report it to the DSO, so that .

[0079] In the first iteration, the DSO directly executes steps 1-4, while the MNO executes steps 1-2 in parallel. Subsequently, the DSO will... This is passed to the MNO, enabling it to continue with steps 3-4. Then, in the second iteration, the MNO provides the DSO with... Then, the DSO executes steps 2-5. If the termination condition is met, the process ends; otherwise, the DSO will... Pass it to MNO so that it can execute steps 3-4 again.

[0080] To verify the effectiveness of the proposed method, an improved IEEE 123-node system was used for testing. The system topology is shown in the attached figure. Figure 6 As shown, this includes 8 nodes connected to distributed photovoltaic power and 7 nodes connected to distributed wind power. The 5G base station data comes from the three major mobile communication operators in China: China Mobile, China Unicom, and China Telecom. 25,166 5G base stations located in Shanghai were selected, and their geographical distribution is shown in Figure 7(a). Based on the specific location of the base stations and the changes in communication load, they are divided into four categories: low load, office, residential, and commercial, represented by black, red, green, and blue, respectively. Figure 7(b) shows the typical load curves for the four types of base stations. Low-load base stations are often located in sparsely populated areas, therefore the demand for 5G communication is lower.

[0081] Given the dense distribution of 5G base stations in Shanghai, it can be assumed that a certain number of base stations are connected to each power node in the improved IEEE 123 node system. (See attached image) Figure 8 As shown in the figure, the red dots and the numbers to their right represent the locations and corresponding numbers of the power nodes to which the base stations are connected. Dots of the same color surrounding the red dots represent 5G base stations connected to that node. The pedestrian flow heatmap data used in this embodiment is sourced from Baidu Maps. Figure 9 This is a heat map of pedestrian flow in Shanghai on a typical day in 2024. The colors in the map, from white to dark red, represent pedestrian density from low to high.

[0082] Table 1 provides some parameters of the 5G base station and its backup energy storage in the example, including the capacity of the backup energy storage. The power configurations are based on the maximum power of each base station, at 10 / 15 / 20 kWh respectively. In the examples presented in this paper, the total optimization time for the optimization model is 24 hours, with a 1-hour time interval. Appropriate optimization software is used for problem modeling and solving.

[0083] Table 1 Parameters of 5G Base Stations and Backup Energy Storage To verify the effectiveness of the method presented in this paper, the following different cases are set up for comparative analysis: Case 1: The method used in this patent is adopted.

[0084] Case 2: Determine whether a base station can go into sleep mode based on its communication load at a given moment. Time base station It can hibernate. Traffic distribution and the impact of base station hibernation on adjacent base stations are not considered.

[0085] Case 3: Assuming base station users are evenly distributed within the signal coverage area and calculating... , Time base station It can hibernate. Traffic distribution is considered, but the impact of base station hibernation on adjacent base stations is not considered.

[0086] Case 4: Generating a traffic flow heatmap based on a pedestrian flow heatmap and calculating... , Time base station It can hibernate. Traffic distribution is considered, but the impact of base station hibernation on adjacent base stations is not considered.

[0087] To verify the effectiveness of the proposed methods for traffic simulation and quantification of the impact of base station sleep mode, Cases 2-4 are compared with Case 1. To compare the impact of different traffic distributions on dynamic base station sleep mode, Case 2 is configured not to perform traffic distribution simulation, while Cases 3 and 4 employ two different methods for traffic distribution simulation, with Case 4 using the same method as Case 1. As a control variable, Cases 2-4 are configured not to consider the impact of base station sleep mode on other base stations, thus verifying the necessity of dynamic sleep mode by comparing Case 1 and Case 4.

[0088] Appendix Figure 10 The base station communication load distribution is obtained using two different methods from Case 3 and Case 4. Since the method used in Case 3 simply assumes that the communication load density is the same within the base station coverage area, [further details are needed]. Figure 10 The above diagram shows the base station communication load distribution and attachments. Figure 10 The image below is more uniform, and the maximum value is relatively smaller. (See attached image for comparison.) Figure 9The actual distribution of pedestrian traffic and the characteristics of 5G base station communication load distribution in reality are shown in the attached figure. Figure 10 The communication load distribution in the image below is clearly more in line with reality.

[0089] Appendix Figure 11 The diagram illustrates the total power curve of a 5G base station, where the original load curve represents the total power load when all base stations are active. The yellow and gray areas in the diagram represent the approximate feasible and approximate infeasible regions of the base station power curve, respectively. By observing the regions where the power curve lies, a preliminary assessment of the feasibility of the applied strategy can be made. When the power curve lies entirely within the yellow area, a corresponding base station operation strategy exists; conversely, when the curve lies entirely within the gray area, no base station operation strategy satisfies the constraints. If the power curve is partially within the yellow area and partially within the gray area, it is difficult to directly determine whether a feasible solution exists; therefore, both areas are considered approximate regions.

[0090] With appendix Figure 11 Compared to the original load, the curves for all other calculation cases that considered base station sleep mode are below it, showing the specific percentage decrease in total power consumption. The percentage decrease in total electricity costs Please refer to the data in Table 3.3. According to the appendix... Figure 11 As shown in Table 2, Case 2's sleep strategy is the most conservative because it does not consider the distribution of communication load and thus ignores the possibility that some users can access neighboring base stations after the base station goes into sleep mode. However, since Case 2 also does not consider the impact of base station sleep mode on neighboring base stations, this conservative sleep strategy may still cause some base stations to not satisfy equation (28), as shown in Table 2. This indicates that the conditions are not met in each time period. The number of base stations activated, and This indicates that the conditions are not met in any of the time periods. The number of dormant base stations. Table 2 shows... and refer to and The proportion of the base station's communication load to the total communication load.

[0091] Both Case 3 and Case 4 consider the impact of traffic distribution on base station sleep mode. However, because they do not consider the impact of base station sleep mode on adjacent base stations, the resulting sleep mode strategies are very aggressive. While they can reduce electricity costs significantly, they also have a major impact on the communication quality of a large number of users. Compared to Case 4, Case 3's simulation of traffic distribution is further removed from the real situation, leading to greater uncertainty in the effectiveness of its sleep mode strategy.

[0092] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0093] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. A method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps, characterized in that, Includes the following steps: (1) Based on the population heat map, complete the city-level 5G base station cluster communication demand estimation and energy consumption optimization model. The model includes the 5G base station and user connection relationship model, the adjustable capacity model of 5G base station backup energy storage, the 5G base station operation energy consumption model, and the 5G base station cluster traffic distribution model based on the population heat map. Output base station communication demand data, energy consumption calculation rules and traffic distribution characteristics. (2) Based on the communication demand data and traffic distribution characteristics output in step (1), and combined with the load transfer chain reaction caused by base station hibernation, a dynamic hibernation model of 5G base station cluster considering hibernation diffusion effect is constructed. The model includes quantification and model construction of 5G base station cluster hibernation diffusion effect, construction of 5G base station dynamic hibernation model, and clarification of the constraints and load transfer rules of base station hibernation. (3) Integrate the energy consumption calculation rules and flow distribution characteristics of step (1) with the dormancy constraints and load transfer rules of step (2) to build a city-level 5G communication network participating in the power distribution system interactive regulation optimization model based on the marginal electricity price DLMP of distribution nodes. The model includes the city-level 5G base station cluster and power distribution network interaction framework, the construction of the city-level 5G base station cluster operation optimization model, and the construction of the power distribution network economic dispatch model. (4) For the city-level 5G communication network participating in the power distribution system interactive control optimization model in step (3), an iterative solution algorithm for the interactive optimization model between the city-level 5G base station cluster and the power distribution system operator DSO side is proposed.

2. The method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps according to claim 1, characterized in that, The specific steps (1) are as follows: Modeling the connection between 5G base stations and users: The transmission power is calculated based on the base station transmission power, channel fading coefficient, user data traffic and distance, and the upper limit constraints of bandwidth and transmission power are met. The communication quality is guaranteed by the signal-to-interference-plus-noise ratio, and mobile users access the nearest base station to maximize the access volume. Base station transmit power refers to the energy intensity of radio frequency signals radiated by a 5G base station within its coverage area. 5G base station Transmission power The result is obtained by calculation using equation (1): (1) in, It is connected to the base station A collection of users, User The corresponding transmission power, , , ; , These represent the sets of base stations and users, respectively. The calculation formula is as shown in equation (2): (2) in, This represents the noise power, specifically the average power of unwanted random interference at the receiver. and The channel fading coefficient is used to quantify the power attenuation and distortion of the signal during propagation; both are constants. User Data traffic, User With base station The distance between them and These are base stations Total bandwidth and its allocation to users The bandwidth of the two is related to the bandwidth of the two as shown in equation (3), where it is assumed that the transmission power of all users connected to the same base station is 1. and bandwidth All are consistent: (3) Equations (4)-(5) indicate the base station Both the transmit power and bandwidth have upper limits, namely... and : (4) (5) Meanwhile, transmission power Affecting base stations With users The signal-to-interference-plus-noise ratio and unit data transmission rate Their expressions are shown in equations (6) and (7) respectively: (6) (7) This refers to the ratio of the desired signal power to the total power of all interference and noise. In equation (6), the signal, interference, and noise are respectively expressed as... , and , Indicates 5G base station To users The maximum data rate that can be achieved when transmitting data. Except for base stations External users The set of other base stations that can be connected. Representing base stations With users Channel gain between Calculated using equation (8), when Less than the reference distance hour, Take fixed path loss value ;on the contrary, Follow The increase decreases exponentially. This is the path loss coefficient. The constraints satisfying equation (9) are: (8) (9) Combined formulas (1)-(9), mobile users under the same conditions With base station distance The smaller the value, the higher the corresponding transmission power of the base station. The smaller the number, the more mobile users and Communication quality can be guaranteed when connected to any base station, but when mobile users... When connected to a distant base station, mobile users are assigned to access the 5G base station closest to them; Adjustable capability model for backup energy storage of 5G base stations: Based on the uninterrupted power supply requirements, the charge state limit is set, and the charging and discharging meet the upper limit of power, efficiency, and upper and lower limits of charge state. Simultaneous charging and discharging are not allowed in the same period, and the charge state is consistent at the beginning and end of the optimization cycle. The energy storage charge state limit considering the reliability of base station power supply is shown in equation (10), where the uninterrupted power supply duration of the 5G base station is... Not less than 3 hours; (10) in, For base stations Backup energy storage The minimum charge state is required to maintain continuous operation of the base station. For base stations Backup energy storage capacity, For base stations exist The standby continuous operating time at any given moment, i.e., the amount of backup energy storage must meet the needs of the base station. In the future Internal operational requirements For base stations Backup energy storage The lower limit refers to the minimum state of charge that an energy storage device should maintain to operate normally for a long period of time. The charging and discharging constraints of the base station backup energy storage are shown in equations (11)-(14): (11) (12) (13) (14) in, and Base station Backup energy storage The charging and discharging power at any given time and Base station The upper limit of charging and discharging power of backup energy storage. and This is a 0-1 variable used to determine whether the energy storage is being charged or discharged, and it is stipulated that the energy storage cannot be charged and discharged simultaneously within a certain time period. For base stations Backup energy storage State of charge at time t, and For base stations The charging and discharging efficiency of backup energy storage. For base stations The upper limit of the state of charge of backup energy storage is expressed by equation (14). After one optimization cycle, it needs to be compared with the state before optimization. Keep it the same at all times; 5G base station operating energy consumption model: The operating power includes constant power and variable power that varies with the load. During sleep, some devices are shut down to reduce the power to a fixed sleep state. The sleep state is represented by 0-1 variables and linearized by the Big M method. 5G base station operating power Divided into AAU power BBU power Power of other transmission equipment Power of heat dissipation equipment With auxiliary equipment power As shown in equation (15): (15) The power of heat dissipation and auxiliary equipment, as well as the power of some communication equipment that does not change with the communication load, is considered as the rated power, while the power of the communication equipment itself is... It changes with the communication load, therefore, equation (15) is replaced by equation (16), where Base station exist Operating power at any given time This refers to the constant portion of the power consumed by a base station under normal operating conditions. For base stations Fixed transmission power, For the size of the communication load, It is a base station unit Increased due to improvement : (16) When the base station is in sleep mode, some communication and auxiliary equipment are shut down, and the operating power is reduced accordingly but remains unchanged. The operating power after the base station goes into sleep mode is shown in equation (17), where This refers to the operating power of a base station in sleep mode. To characterize the base station exist A 01 variable indicating whether a time is sleepy. When the base station is in sleep mode, When the base station is working normally, equation (17) is linearized using the Big M method to obtain equation (18), where For a maximum value: (17) (18)； Traffic distribution model: Population heat map data from Baidu Maps is used to simulate traffic distribution; Define base station Two types of areas within the coverage area and and will / The total flow in is defined as / : When the communication user is in the area And base station During normal operation, users will connect to the base station. ; When the communication user is in the area At that time, users can only access the base station. ; From the above definition, we know that Indicates all possible access base stations The set of user locations, This means that access can only be made to base stations outside the coverage area of ​​all other base stations. The set of user locations, satisfying Correspondingly, and Reflects and The sum of the internal flows satisfies ; For calculation and The value needs to be determined. and Specific scope; assuming base station The position coordinates are Service coverage area For base station Position coordinates are the center of the circle A circle with radius and the base station Other base stations with overlapping coverage are numbered as follows: ,Depend on As can be seen from the definition, Satisfying equation (19), where For base stations The complement of the covered area, Users at base stations Within the coverage area, and when it is simultaneously located at the base station When within the coverage area, it is related to the base station The distance should be greater than that to the base station. The distance must be sufficient; otherwise, it will connect to the base station. ;therefore, Satisfy equation (20), where Compared to base stations Distance from base station The set of closer coordinates is expressed as in equation (21); Traffic heatmaps divide the space into several regions, which are related to base stations. The overlapping areas of coverage are denoted as The corresponding traffic flow set is ,therefore, and The result is obtained by calculation using equation (22), where It is an area function; In addition, the coverage area of ​​all base stations is set as a circle with an equal radius: （22）。 3. The method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps according to claim 2, characterized in that, In step (2): The dormancy diffusion effect refers to the phenomenon where the load of a single base station shifts to neighboring base stations after it goes into dormancy, triggering a chain reaction that affects the dormancy states of those neighboring base stations. This effect is achieved by calculating the load distribution of neighboring base stations before and after they go into dormancy. and The numerical changes can be used to quantitatively analyze the impact of 5G base station dormancy on adjacent base stations; Considering the combined sleep mode of adjacent base stations and It is calculated from equation (23), where and To disregard base station sleep mode and , Including base stations A certain combination of all external base stations, and for After all the base stations go into hibernation and The numerical change, and Then it means let and Non-zero The set of: (23)； There is an upper limit to the communication load that a base station can handle, therefore there is a corresponding limit. upper limit Meanwhile, considering service quality, if a base station needs to go into hibernation, the communication load within its coverage area that results in the loss of 5G communication service due to the inability to access other base stations must be less than a certain value. Also smaller than , and The specific values ​​can be determined by the communication system operator (MNO) based on historical data and equations (1)-(9), with the following constraints: (24)； and Calculated using formula (25), where This represents a type of user. express The set of user types at that time express The set of user types at that time Indicates when type The distance between the user and the base station The nearest base station number, Representation type The total number of users in the formula (25) is established. , The correlation between the number of the two types of users is established by introducing a coefficient that correlates base station traffic with the number of users. Formula (25) achieves the and Calculation: (25)； The dynamic sleep model introduces 0-1 variables to represent the sleep combination state of the base station. After linearization, the variable product term is eliminated. The model includes the influence of all base station sleep combinations and reflects the sleep diffusion effect brought about by the spatiotemporal coupling characteristics of the load. Considering the impact of adjacent base station hibernation, the dynamic hibernation model of the 5G base station cluster is constructed using equations (23)-(26). As a binary variable, it is calculated using equation (27) that when all base stations in the set are in a dormant state, The value is 1 if it is not 0 otherwise. Compared with equation (24), equation (26) takes into account the dormant state of all base stations. Different base stations can still pass through and influence each other even if their coverage areas do not overlap. and ; (26) (27) Since there are more than two 0-1 variables in equation (27), it can be linearized using equations (28)-(29), where The value is equal to and Therefore, equation (26) is equivalent to the linear constraint equation (30), and the linear dynamic dormancy model of 5G base stations is equation (28)-(30): (28) (29) (30)。 4. The method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps according to claim 3, characterized in that, The interaction framework between the city-level 5G base station cluster and the power distribution network in step (3) is a three-layer architecture: The upper-level flow distribution layer estimates flow demand using population heat maps and historical flow data; The middle layer of the mobile network is where the communication system operator (MNO) formulates base station operation strategies, including dynamic hibernation and energy storage management. At the lower distribution network layer, the distribution system operator (DSO) releases dynamic marginal electricity prices for distribution network nodes based on predicted node demand. In this scenario, the communication system operator (MNO) first estimates the traffic distribution of each 5G base station based on the overall signal coverage and the location and density of people obtained from the population heat map. Subsequently, the communication system operator (MNO) constructs an optimized energy management model for all 5G base stations.

5. A method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps according to claim 4, characterized in that, In step (3): City-level 5G base station cluster operation optimization model: The objective function is to minimize the power purchase cost and state switching energy consumption, and the constraints include backup energy storage charging and discharging constraints, dynamic hibernation constraints, and energy consumption constraints. objective function (31) (32) (33) In formula (31) For base stations The power node is located at DLMP value at time, The power consumption required to switch the operating status of a base station once. To characterize the base station exist The 01 variable indicating whether the running state changes at any given time can be represented by equation (32) and linearized by equation (33). To characterize and Size of a 01 variable, when hour ,when hour ; Constraints (1) 5G base station backup energy storage constraints (34) (35) (36) (37) in and For charging and discharging power, and This is the upper limit of charging and discharging power. and The 01 variables are used to determine whether the energy storage is being charged or discharged; For backup energy storage , and For charging and discharging efficiency, For backup energy storage Upper limit; (2) Dynamic sleep constraints of 5G base stations (38) (39) (40) The constraints for each part are the same as those in formulas (28)-(30); (3) Energy consumption constraints of 5G base stations (41) in For base station operating power, The constant power when the base station is activated. For base station sleep power, For the size of the communication load, It is a base station unit The increased power due to the upgrade For base station sleep variables, and It is a maximum value; Distribution network economic dispatch model: The objective function is to minimize the electricity purchase cost of the upper-level grid, distributed wind power and photovoltaics. The constraints include node power balance, line current and node voltage constraints. By solving the dual problem, the marginal electricity price (DLMP) of the distribution node, which includes energy, congestion and loss costs, is obtained. The marginal price of distribution nodes (DLMP) refers to the marginal cost to the entire distribution system when an additional unit of electricity consumption is added or reduced at a specific power node in the distribution network. DLMP represents the dynamic and refined electricity price in the distribution network in both time and space. It not only reflects the energy cost of electricity consumption, but also includes the congestion cost and loss cost of the distribution network transporting electricity to a specific power node. Equation (42) is the economic dispatch model for the distribution network, and its objective function is to minimize the power purchase cost of the distribution network. Number the nodes. A set of distribution network nodes. , and These represent the unit cost of purchasing electricity from the upstream power grid, distributed wind power, and photovoltaic power, respectively. , , and , , This represents the corresponding purchased active and reactive power. Equations (42b) and (42c) are the node power balance constraints, where... and For the line The active and reactive power flowing through, and For node load power, The square of the line current. This represents the Lagrange multiplier corresponding to constraint (42b). For line impedance, and Then it is the square of the node voltage; The marginal electricity price (DLMP) at the distribution node can be obtained by solving the dual problem of equation (42), i.e. Equations (42f)-(42j) represent the upper and lower bound constraints for some variables, where... The upper and lower limits are and , The upper limit is , The upper and lower limits are and , The upper and lower limits are and , The upper and lower limits are and ; (42a) (42b) (42c) (42d) (42e) (42f) (42g) (42h) (42i) (42j)。 6. A method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps according to claim 5, characterized in that, The iterative solution algorithm in step (4) includes: Distribution system operator (DSO) side: (1) Predict the total load of the distribution network based on historical data With 5G base station load And distributed new energy output, making , ,in For the number of iterations, This refers to the 5G base station load during the first iteration. (2) Update the 5G base station load based on the base station load data reported by the communication system operator MNO: , ,in The difference between the updated 5G base station load and the original forecast. (3) Construct an economic dispatch model for the distribution network by combining the above data with relevant parameters of the distribution network; (4) Solve the dual problem of the model in step (3) to obtain , For the first During the next iteration ; (5) If , or If so, the process will terminate; otherwise, it will... Provided to the MNO (Mobile Network Operator) and proceeding with step (3) of the MNO's process; wherein , It is used to determine the time interval between iterations. The threshold of minute change This is the upper limit of the number of iterations; On the MNO side of the communication system operator: (1) Conduct traffic distribution simulation based on the pedestrian flow heat map and historical traffic data of the base station; (2) Calculate the impact of all base station sleep combinations on the traffic of other base stations based on the traffic distribution; (3) Based on the data in steps (1)-(2) and the data provided by the DSO Construct an optimization model for 5G base station cluster operation; (4) Solve the model in step (3) to obtain And report it to the DSO, so that ; In the first iteration, the distribution system operator (DSO) directly executes steps (1)-(4), while the communication system operator (MNO) executes steps (1)-(2) in parallel. Subsequently, the distribution system operator (DSO) will... The information is passed to the communication system operator MNO, enabling it to continue with steps (3)-(4). Then, in the second iteration, the communication system operator MNO provides the distribution system operator DSO with... Afterwards, the Distribution System Operator (DSO) executes steps (2)-(5); if the termination condition is met, the process ends; otherwise, the Distribution System Operator (DSO) will... The information is passed to the communication system operator MNO, which then performs steps (3)-(4) again.

7. A method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps according to claim 6, characterized in that, The linearization process of the dynamic dormancy model eliminates the product terms of more than two 0-1 variables by introducing intermediate variables, ensuring that the model can be solved efficiently.

8. A method for city-level 5G base station clusters participating in power grid interactive scheduling based on population heat maps according to claim 7, characterized in that, The electricity purchase cost of the city-level 5G base station cluster operation optimization model is positively correlated with DLMP and base station load, and the energy consumption of state switching is determined by the number of switching and the electricity consumption of a single switching.

9. A city-level 5G base station cluster participating in power grid interactive dispatching system based on population heatmaps, applying the method described in any one of claims 1-8, characterized in that, include: The communication demand and energy consumption modeling module is used to perform the aforementioned communication demand estimation and energy consumption optimization modeling. The dynamic dormancy modeling module is used for dynamic dormancy modeling that considers the dormancy diffusion effect. The interactive control framework modeling module is used to perform the interactive control optimization framework modeling based on the marginal electricity price of distribution nodes (DLMP). The iterative solution module is used to execute the iterative solution algorithm for the interactive optimization model.

Images