A method and system for energy storage management for base stations

Abstract

The application discloses an energy storage method and system for base station management, and the method comprises the following steps: performing multi-modal data fusion processing according to historical traffic data, real-time user movement mode data and meteorological data of each virtual base station in a target area, to obtain a dynamic load prediction result of each virtual base station in a target period; based on the dynamic load prediction result, a distributed constraint optimization algorithm is used for dynamic allocation of energy storage resources, to generate an optimized allocation strategy containing a cross-base station energy scheduling path; according to the wireless network topology structure characteristics of the target area, an energy scheduling decision model based on graph theory is constructed, the energy storage scheduling time sequence is optimized, and a dynamic energy storage scheduling scheme with a fault-tolerant mechanism is generated, to realize energy storage for base station management. By using the embodiment of the application, the energy efficiency of the base station can be improved, energy waste can be reduced, and efficient configuration and real-time scheduling of energy storage resources can be realized.

Description

Technical Field

[0001] This invention belongs to the field of energy storage technology, and in particular to an energy storage method and system for base station management. Background Technology

[0002] In modern communication networks, base stations are a crucial component supporting mobile communications, and their operational efficiency directly impacts network service quality and user experience. With the rapid development of 5G and future communication technologies, base station traffic is increasing rapidly, especially during high-traffic periods and in densely populated user environments, placing immense pressure on energy demand. To address this challenge, energy storage technology is increasingly being incorporated into base station management to balance supply and demand, improve energy efficiency, and reduce operating costs.

[0003] Currently, traditional base station management methods largely rely on static energy scheduling strategies, which are insufficient to effectively address constantly changing traffic volumes, user mobility patterns, and weather conditions. These methods often depend on historical data for energy demand forecasting, lacking full utilization of real-time data, leading to energy shortages or waste during peak periods. Furthermore, the impact of factors such as the charging and discharging efficiency of energy storage units, state of charge, and energy transmission losses between adjacent base stations on energy storage scheduling is not effectively considered. This results in inefficient allocation and use of energy storage resources, affecting the overall efficiency of energy management. Summary of the Invention

[0004] The purpose of this invention is to provide an energy storage method and system for base station management, which addresses the shortcomings of existing technologies, improves the energy efficiency of base stations, reduces energy waste, and enables efficient allocation and real-time scheduling of energy storage resources.

[0005] One embodiment of this application provides an energy storage method for base station management, the method comprising:

[0006] Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, multimodal data fusion processing is performed to obtain the dynamic load prediction results of each virtual base station in the target time period.

[0007] Based on the dynamic load prediction results, combined with the current state of charge, charging and discharging efficiency curves of the energy storage units associated with each virtual base station and the energy transmission path loss parameters between adjacent base stations, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths.

[0008] Based on the characteristics of the wireless network topology in the target area, a graph theory-based energy scheduling decision model is constructed. Delay sensitivity analysis and energy efficiency balancing are performed on the optimization allocation strategy to optimize the energy storage scheduling sequence and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management.

[0009] Optionally, the step of performing multimodal data fusion processing based on historical traffic data, real-time user mobility pattern data, and meteorological data of each virtual base station within the target area to obtain the dynamic load prediction results of each virtual base station within the target time period includes:

[0010] Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, a multi-source data alignment algorithm based on dynamic time warping is adopted to eliminate the data timestamp inconsistency problem and fill in the missing values ​​by interpolation to obtain a time-synchronized multimodal dataset.

[0011] For time-synchronized multimodal datasets, a feature extraction network based on a multi-head attention mechanism is used to extract the temporal features of historical traffic volume, the spatial features of user movement patterns, and the environmental features of meteorological data. The features of different modalities are then weighted and fused through a cross-modal attention mechanism to generate a fused multimodal feature representation.

[0012] The fused multimodal feature representations are input into a pre-trained prediction model based on a spatiotemporal graph convolutional network. Combined with the base station topology of the target area, the spatiotemporal dependencies between base stations are captured, and the dynamic load change trend of each virtual base station in the target time period is predicted, thus obtaining the dynamic load prediction results of each virtual base station in the target time period.

[0013] Optionally, based on the dynamic load prediction results, combined with the current state of charge, charge / discharge efficiency curves, and energy transmission path loss parameters of the energy storage units associated with each virtual base station, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources, generating an optimized allocation strategy that includes cross-base station energy scheduling paths, including:

[0014] Based on the current state of charge, charge and discharge efficiency curves, and energy transmission path loss parameters between adjacent base stations of the energy storage units associated with each virtual base station, a multi-dimensional constraint model of the energy storage state is constructed. Through energy conservation constraints, charge and discharge rate constraints, and transmission loss constraints, the feasible solution space for energy storage resource allocation is defined.

[0015] By combining dynamic load prediction results with a multi-dimensional constraint model of energy storage status, a distributed optimization framework based on the Lagrange relaxation method is adopted to decompose the global optimization problem of energy storage resource allocation into multiple sub-problems. Each sub-problem corresponds to a local optimization objective of a virtual base station. The local optimization objectives include load balancing, energy transmission efficiency, and energy storage lifetime.

[0016] For the decomposed subproblems, a path optimization method based on genetic algorithm is adopted. Combined with the energy transmission path loss parameters between adjacent base stations, the optimal energy scheduling path is dynamically searched to generate a preliminary cross-base station energy scheduling scheme.

[0017] For the preliminary cross-base station energy scheduling scheme, a global optimization algorithm based on consensus mechanism is adopted. Through information exchange and iterative updates between distributed nodes, the solutions of each sub-problem are coordinated to generate an optimized allocation strategy that satisfies global constraints.

[0018] Optionally, based on the wireless network topology characteristics of the target area, a graph-based energy scheduling decision model is constructed to perform latency sensitivity analysis and energy efficiency balancing processing on the optimized allocation strategy, optimize the energy storage scheduling sequence, and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management, including:

[0019] Based on the characteristics of the wireless network topology and optimization allocation strategy of the target area, an energy scheduling decision model based on graph theory is constructed. In this model, nodes represent base stations or energy storage units, and edges represent energy scheduling paths. Weights are assigned to the edges using path loss parameters and transmission delay data to generate a weighted network topology graph.

[0020] For energy scheduling paths in the network topology graph, a latency sensitivity analysis method based on the shortest path algorithm is used to calculate the transmission latency of each path. By using a latency tolerance threshold, latency-sensitive critical paths are selected.

[0021] For latency-sensitive critical paths, a load-balanced energy efficiency optimization method is adopted. This method combines the dynamic load prediction results of base stations and the charging and discharging efficiency curves of energy storage units to adjust the allocation ratio of energy scheduling paths, ensuring that high-load base stations receive priority energy support and generating an energy-balanced scheduling scheme.

[0022] For energy efficiency balance scheduling schemes, a fault-tolerant mechanism based on redundant paths is introduced. By constructing backup energy scheduling paths and dynamic switching strategies, energy transmission can still be maintained when some paths fail, ultimately generating a dynamic energy storage scheduling scheme with fault tolerance.

[0023] Optionally, based on the wireless network topology characteristics and optimization allocation strategy of the target area, a graph-based energy scheduling decision model is constructed, wherein nodes represent base stations or energy storage units, edges represent energy scheduling paths, and weights are assigned to the edges using path loss parameters and transmission delay data to generate a weighted network topology graph, including:

[0024] Based on the characteristics of the wireless network topology in the target area, base stations and energy storage units are abstracted as nodes, and energy scheduling paths in the optimization allocation strategy are abstracted as edges. The set of nodes and edges is initialized through the physical connection relationship between base stations and energy transmission requirements to generate a preliminary network topology graph.

[0025] For each energy scheduling path, an environment-aware path loss calculation model is adopted. The path loss parameters are calculated by combining the distance between base stations, the characteristics of the transmission medium and environmental interference factors. The path loss parameters are normalized by optimizing the energy transmission efficiency target in the allocation strategy to generate path loss weights.

[0026] Based on the real-time scheduling requirements in the optimized allocation strategy, and combined with historical transmission delay data and the current network status, a delay prediction method based on time series analysis is adopted to dynamically update the transmission delay data of each energy scheduling path and use it as a delay weight.

[0027] The path loss weight and transmission delay weight are optimized and integrated through multi-objective optimization. A comprehensive weight assignment method based on weighted summation is adopted. Combined with the priority and fault tolerance requirements of energy scheduling paths, the edges in the preliminary network topology graph are assigned comprehensive weights to generate a weighted network topology graph.

[0028] Another embodiment of this application provides an energy storage system for base station management, the system comprising:

[0029] The fusion module is used to perform multimodal data fusion processing based on the historical traffic data, real-time user mobility pattern data and meteorological data of each virtual base station in the target area, and to obtain the dynamic load prediction results of each virtual base station in the target time period.

[0030] The allocation module is used to dynamically allocate energy storage resources based on dynamic load prediction results, combined with the current state of charge, charge and discharge efficiency curves of the energy storage units associated with each virtual base station, and energy transmission path loss parameters between adjacent base stations, using a distributed constraint optimization algorithm to generate an optimized allocation strategy that includes cross-base station energy scheduling paths.

[0031] The module is used to construct a graph theory-based energy scheduling decision model based on the wireless network topology characteristics of the target area, perform latency sensitivity analysis and energy efficiency balancing processing on the optimization allocation strategy, optimize the energy storage scheduling sequence, and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management.

[0032] Another embodiment of this application provides a storage medium storing a computer program, wherein the computer program is configured to execute the method described in any of the preceding claims when running.

[0033] Another embodiment of this application provides an electronic device including a memory and a processor, wherein the memory stores a computer program and the processor is configured to run the computer program to perform the method described in any of the preceding claims.

[0034] Compared with existing technologies, this invention provides an energy storage method for base station management. Based on historical traffic data, real-time user mobility pattern data, and meteorological data of each virtual base station within a target area, multimodal data fusion processing is performed to obtain dynamic load prediction results for each virtual base station within a target time period. Based on the dynamic load prediction results, a distributed constraint optimization algorithm is used for dynamic allocation of energy storage resources, generating an optimized allocation strategy that includes cross-base station energy scheduling paths. According to the wireless network topology characteristics of the target area, a graph theory-based energy scheduling decision model is constructed to optimize the energy storage scheduling sequence and generate a dynamic energy storage scheduling scheme with fault tolerance mechanisms. This enables energy storage for base station management, thereby improving base station energy efficiency, reducing energy waste, and achieving efficient configuration and real-time scheduling of energy storage resources. Attached Figure Description

[0035] Figure 1 A hardware structure block diagram of a computer terminal for an energy storage method for base station management, provided in an embodiment of the present invention;

[0036] Figure 2 This is a flowchart illustrating an energy storage method for base station management provided in an embodiment of the present invention.

[0037] Figure 3 This is a schematic diagram of an energy storage system for base station management provided in an embodiment of the present invention. Detailed Implementation

[0038] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0039] This invention first provides an energy storage method for base station management, which can be applied to electronic devices, such as computer terminals, specifically ordinary computers.

[0040] The following detailed explanation uses a computer terminal as an example. Figure 1 This is a hardware structure block diagram of a computer terminal for an energy storage method used in base station management, provided as an embodiment of the present invention. Figure 1 As shown, the computer device includes a processor, memory, and network interface connected via a system bus, wherein the memory may include non-volatile storage media and internal memory.

[0041] Non-volatile storage media can store operating systems and computer programs. These computer programs include program instructions that, when executed, cause the processor to perform any energy storage method for base station management.

[0042] The processor provides computing and control capabilities, supporting the operation of the entire computer device.

[0043] Internal memory provides an environment for the execution of computer programs in non-volatile storage media. When these computer programs are executed by a processor, the processor can execute any energy storage method used for base station management.

[0044] This network interface is used for network communication, such as sending assigned tasks. Those skilled in the art will understand that... Figure 1 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0045] It should be understood that the processor can be a Central Processing Unit (CPU), but it can also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among these, a general-purpose processor can be a microprocessor or any conventional processor.

[0046] See Figure 2 The present invention provides an energy storage method for base station management, which may include the following steps:

[0047] S201. Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, multimodal data fusion processing is performed to obtain the dynamic load prediction results of each virtual base station in the target time period.

[0048] This method utilizes multimodal data fusion processing of historical traffic data, real-time user mobility pattern data, and meteorological data for each virtual base station within a target area to obtain dynamic load prediction results for each virtual base station within a specific time period. Specifically, historical traffic data reflects the user request volume and service demand of the base station at different time periods; real-time user mobility pattern data provides real-time user behavior and movement trends in the network; and meteorological data may affect the base station's operating environment and user behavior, such as changes in user travel patterns caused by weather changes. By comprehensively analyzing these multi-dimensional data, more accurate load predictions can be obtained, helping base station managers to formulate energy allocation and scheduling strategies in advance.

[0049] This step is of profound significance. It not only improves the accuracy of predicting the load of each virtual base station but also effectively avoids energy shortages or overconsumption due to sudden load surges. This dynamic load prediction method makes base station energy management more intelligent and forward-looking, contributing to improved network stability and user experience. Especially during peak hours, accurate prediction ensures sufficient energy supply, preventing service interruptions or speed reductions, thereby greatly improving user satisfaction and network reliability.

[0050] Specifically, based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, a multi-source data alignment algorithm based on dynamic time warping can be used to eliminate the data timestamp inconsistency problem and fill in missing values ​​by interpolation to obtain a time-synchronized multimodal dataset.

[0051] In the process of multimodal data fusion, it is essential to ensure the consistency of timestamps across different data sources, which is crucial for cross-analysis. Dynamic Time Warping (DTW) algorithms can flexibly align different time series, even those with temporal variations or delays. Simultaneously, interpolation methods are used to appropriately fill in missing values ​​in the data sources, ensuring the generated multimodal dataset is complete and continuous.

[0052] This step ensures the accuracy and consistency of data during the fusion process, thereby enhancing the reliability of subsequent analysis and prediction. By eliminating timestamp inconsistencies and filling in missing values, a time-synchronized multimodal dataset is created, providing a solid foundation for dynamic load forecasting and accurately reflecting network status and user behavior within the target area.

[0053] In the data fusion process, the first step is to ensure the consistency of timestamps across various data sources. Data often originates from different sources and devices; for example, historical traffic data from base stations might be recorded hourly, while user mobility patterns and weather data might be recorded minute-by-minute. To address this issue, a method based on Dynamic Time Warping (DTW) is employed. This method allows for flexible comparison of different time series, even if they are subject to temporal variations or delays. DTW arranges time series in a non-linear manner, maximizing the similarity between two time series and thus achieving accurate time alignment.

[0054] While dealing with timestamp inconsistencies, we may also encounter situations where some data is missing. For example, a base station might fail to record user activity data within a certain time period. To address this issue, interpolation methods can be used to fill in missing values. Common interpolation methods include linear interpolation and spline interpolation. Through linear interpolation, we can extrapolate the missing value by linearly calculating the missing data based on the data points before and after it. This process not only helps maintain the integrity of the time series but also provides an accurate data foundation for subsequent analysis.

[0055] Finally, through the DTW algorithm and interpolation processing, a time-synchronized multimodal dataset was formed. This dataset will include information from multiple dimensions, such as historical traffic volume, real-time user mobility patterns, and meteorological data, so that it can be fully utilized in subsequent analysis. Effective data alignment and missing value handling can significantly improve the accuracy of subsequent load forecasting, providing a reliable basis for base station energy scheduling and optimization.

[0056] For time-synchronized multimodal datasets, a feature extraction network based on a multi-head attention mechanism is used to extract the temporal features of historical traffic volume, the spatial features of user movement patterns, and the environmental features of meteorological data. The features of different modalities are then weighted and fused through a cross-modal attention mechanism to generate a fused multimodal feature representation.

[0057] By leveraging a feature extraction network based on a multi-head attention mechanism, key features from various types of data can be effectively captured. For historical traffic volume data, extracting its temporal features helps understand traffic trends; spatial features of user mobility pattern data aid in analyzing user distribution within the network; and environmental features of meteorological data reveal the impact of the external environment on user behavior. The introduction of a cross-modal attention mechanism allows the model to dynamically adjust weights based on the importance of different modalities, thereby forming a comprehensive feature representation that lays the foundation for subsequent load prediction.

[0058] The ability to extract and fuse features in this process greatly enhances the processing of complex data, enabling the model to capture richer contextual information and thus improve the accuracy of dynamic load prediction. Weighted fusion of features from different modalities ensures the highlighting of important information while attenuating less important information, which is particularly important in the prediction process.

[0059] In the feature extraction process, a network structure based on a multi-head attention mechanism can effectively extract important features from time-synchronized multimodal datasets. For historical traffic data, the network needs to pay attention to the temporal changes of the data and extract the peaks and troughs of traffic. For example, at a specific time period, historical traffic data may show a peak phenomenon, clearly reflecting an increase in user access demand. Through the multi-head attention mechanism, the network can process multiple attention heads in parallel and learn different temporal features, enabling the model to adapt to complex traffic fluctuations.

[0060] Meanwhile, extracting spatial features is particularly important when processing user mobility pattern data. For example, the distribution of users in different geographical locations may affect the energy demand of base stations. In this process, based on the extraction of spatial features, it is possible to analyze the degree of user concentration in a certain area, and thus predict the load of base stations in that area. By extracting features from user mobility pattern data, high-density areas can be identified, providing an important basis for subsequent energy allocation.

[0061] Environmental feature extraction from meteorological data is equally important, as changes in the external environment can directly impact user behavior patterns. Meteorological factors such as temperature, precipitation, and wind speed can influence users' travel plans and network usage needs. Extracting environmental features allows for a better understanding of the impact of weather changes on traffic volume, enabling the consideration of meteorological factors in energy scheduling. Finally, these features are weighted and fused using a cross-modal attention mechanism to generate a multimodal feature representation containing comprehensive information. This feature representation provides a solid foundation for subsequent dynamic load forecasting models.

[0062] The fused multimodal feature representations are input into a pre-trained prediction model based on a spatiotemporal graph convolutional network. Combined with the base station topology of the target area, the spatiotemporal dependencies between base stations are captured, and the dynamic load change trend of each virtual base station in the target time period is predicted, thus obtaining the dynamic load prediction results of each virtual base station in the target time period.

[0063] In this step, the fused multimodal feature representations are fed into a pre-trained spatiotemporal graph convolutional network (ST-GCN). This network captures the spatial dependencies and temporal variations between base stations. By incorporating the base station topology, the model can simulate the interactions between base stations, thereby predicting load changes over specific time periods.

[0064] The core value of this step lies in its ability to leverage established spatiotemporal dependencies to provide more accurate dynamic load predictions. The interactions between base stations are complex, and modeling them using graph convolutional networks can better reflect this, ultimately making the predictions more consistent with reality and improving the system's responsiveness.

[0065] In this step, the fused multimodal feature representations are fed into a pre-trained spatiotemporal graph convolutional network (ST-GCN). In ST-GCN, the model can not only handle variations in time-series data but also consider the topological relationships between base stations. For example, the load of a base station may be affected by the load of surrounding base stations; user requests from neighboring base stations may be transferred to that base station to some extent. Through graph convolutional layers, the network can effectively capture these spatial correlations, thereby improving the accuracy of load prediction.

[0066] By incorporating the base station topology, the spatiotemporal graph convolutional network treats each base station as a node in a graph, with the energy transmission paths between them as edges. Using an adjacency matrix to represent the relationships between nodes, the model can simultaneously learn information about node characteristics and the graph structure. In this way, when a base station experiences a traffic peak at a certain time, neighboring base stations can also capture this change through graph convolution and influence each other, thus better predicting future load conditions.

[0067] Ultimately, the predictions output by the spatiotemporal graph convolutional network will reflect the dynamic load trends of each virtual base station within the target time period. For example, the model may predict a sharp increase in traffic to a specific base station during a large-scale event. Based on these predictions, base station managers can prepare in advance and allocate energy storage resources accordingly. This predictive mechanism not only improves the intelligence of resource scheduling but also ensures the stable operation of base stations amidst fluctuations in user demand.

[0068] S202, based on the dynamic load prediction results, combined with the current state of charge, charging and discharging efficiency curves of the energy storage units associated with each virtual base station and the energy transmission path loss parameters between adjacent base stations, adopts a distributed constraint optimization algorithm to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths;

[0069] Based on dynamic load forecasting results, this method combines the actual service demands of different virtual base stations with the current state of charge (SOC) of associated energy storage units, their charge / discharge efficiency curves, and energy transmission path loss parameters between adjacent base stations. A distributed constraint optimization algorithm is then used for dynamic allocation of energy storage resources. Specifically, the SOC of an energy storage unit determines the amount of available energy storage, the charge / discharge efficiency reflects the actual energy loss during charging and discharging, and the path loss parameter indicates the energy loss encountered when transmitting energy between different base stations. These factors are considered through the optimization algorithm, ensuring that the allocation of energy storage resources is based not only on current service demands but also on the economy and efficiency of energy transmission. This generates an optimized allocation strategy that includes cross-base station energy scheduling paths, ensuring that each base station receives adequate energy support during periods of high demand.

[0070] This optimized allocation strategy effectively improves the energy management efficiency of base stations, especially under conditions of significant service fluctuations. For example, when a base station anticipates a sharp increase in service volume during a specific period, energy storage units can provide the necessary energy support during peak demand, reducing reliance on the power grid and thus lowering operating costs. Simultaneously, considering energy transmission losses between adjacent base stations, the optimized allocation strategy ensures the selection of the optimal path during energy scheduling, achieving efficient energy distribution. This energy storage management approach, based on dynamic load forecasting and real-time optimization, not only enhances the responsiveness of base stations but also improves the overall network stability and service quality, providing users with a better experience.

[0071] Specifically, a multi-dimensional constraint model of the energy storage state can be constructed based on the current state of charge, charge and discharge efficiency curves, and energy transmission path loss parameters between adjacent base stations of the energy storage units associated with each virtual base station. The feasible solution space for energy storage resource allocation can be defined through energy conservation constraints, charge and discharge rate constraints, and transmission loss constraints.

[0072] In this step, a multi-dimensional constraint model needs to be established first to comprehensively understand the state of the energy storage units. The current state of charge of the energy storage unit associated with each virtual base station reflects the upper limit of its available energy, while the charge / discharge efficiency curve provides the efficiency changes during charging and discharging, which may be affected by factors such as temperature and aging. The energy transmission path loss parameters between adjacent base stations help the model calculate the potential losses when transferring energy between different base stations. Using this information, a comprehensive model is constructed that includes energy conservation constraints, charge / discharge rate constraints, and transmission loss constraints, defining the feasible solution space for energy storage resource allocation.

[0073] The construction of this model lays the foundation for subsequent optimization, ensuring the rationality and feasibility of resource allocation. By clearly defining the feasible solution space, it avoids system instability or energy shortages caused by resource allocation under extreme conditions, ensuring sufficient energy supply during peak periods. This constrained model not only makes the management of energy storage resources more systematic but also provides clear guidance for subsequent decision-making.

[0074] In this phase, the first step is to collect various data related to the virtual base stations, including the current state of charge (SOC) of each base station's energy storage unit, its charge / discharge efficiency curves, and energy transmission path loss parameters between adjacent base stations. Through comprehensive analysis of this data, a multi-dimensional energy storage state model can be constructed. Taking base stations A, B, and C as examples, assume that the current SOC of base station A is 60%, with a charge / discharge efficiency of 85%; the SOC of base station B is 75%, and that of base station C is 100%. Based on this, the energy transmission path loss between adjacent base stations is analyzed; for example, the loss from base station A to base station B is 10%, while the loss from base station B to base station C is 5%.

[0075] Next, by establishing energy conservation constraints, it is ensured that the energy released from the energy storage unit must meet the actual energy needs of each base station, and the energy transfer loss is calculated. Taking base station B as an example, if its peak energy requirement is predicted to be 100 kWh, then when transferring energy from base station A to base station B, a 10% loss needs to be considered. This means that base station A needs to release 111.11 kWh of energy in its state of charge. The charge / discharge rate constraint needs to be set according to the physical characteristics of each base station's battery. For example, base station A can only release energy at 80% of its maximum charge / discharge rate to prevent damage.

[0076] Through the above steps, a feasible solution space is finally defined, providing clear guidance for subsequent optimization. For example, suppose the constraints set in the model are: the maximum release capacity of each base station, the energy loss threshold, and the unit energy release time of each energy storage unit. This feasible solution space will lay the foundation for the allocation of energy storage resources among base stations, ensuring that the subsequent optimization process can be carried out within a reasonable range.

[0077] By combining dynamic load prediction results with a multi-dimensional constraint model of energy storage status, a distributed optimization framework based on the Lagrange relaxation method is adopted to decompose the global optimization problem of energy storage resource allocation into multiple sub-problems. Each sub-problem corresponds to a local optimization objective of a virtual base station. The local optimization objectives include load balancing, energy transmission efficiency, and energy storage lifetime.

[0078] In this step, the dynamic load forecast results are combined with a multi-dimensional constraint model to provide clear objectives and constraints for the optimal allocation of energy storage resources. The Lagrange relaxation method is applied to decompose the global optimization problem into multiple local sub-problems, each corresponding to the optimization objective of a virtual base station. These local optimization objectives include load balancing, meaning the load among base stations should be distributed as evenly as possible; energy transfer efficiency, i.e., minimizing energy loss while meeting energy demand; and energy storage lifetime, i.e., extending the lifespan of energy storage devices as much as possible during charging and discharging.

[0079] This decompositional optimization strategy effectively reduces computational complexity and improves algorithm execution efficiency. By breaking down the global problem into multiple local problems, each base station can perform optimization calculations independently, and then coordinate them to ensure the optimal solution to the overall problem. This method not only enhances the flexibility of resource allocation but also improves the base station's ability to respond quickly to dynamic changes, ensuring efficient energy scheduling even under extreme conditions.

[0080] At this stage, the previously constructed multi-dimensional constraint model of energy storage status is combined with the dynamic load forecast results to form an efficient optimization framework. Specifically, the dynamic load forecast results provide important information on the energy demand of each base station in the future. For example, during high-load periods, base station A is expected to require 150 kWh, base station B requires 120 kWh, while base station C actually requires 80 kWh. Based on these load forecasts, in order to achieve load balancing, the system needs to dynamically adjust the allocation of energy storage resources to ensure that the needs of all base stations are met.

[0081] Next, a distributed optimization strategy based on the Lagrange relaxation method decomposes the global optimization problem into multiple sub-problems. Each sub-problem corresponds to a virtual base station, allowing each to independently address its own optimization objective. For example, the local optimization objective of base station A might be to minimize the cost of released energy and maximize load balancing; base station B might focus on energy transmission efficiency, ensuring energy is received through the optimal path; while base station C's local optimization objective focuses on energy storage lifetime, maximizing battery lifespan. Through this method, each virtual base station can achieve efficient collaboration while satisfying its independent objectives.

[0082] Finally, the solutions to each sub-problem are fed back into the global optimization framework. Through iterative calculations, the results of local optimizations are aggregated to guide subsequent global optimization decisions. The decision-making process needs to consider the energy transmission efficiency, state of charge, and transmission losses between base stations to generate a reasonable energy allocation scheme, ensuring a balanced and reasonable energy supply among base stations during periods of high demand.

[0083] For the decomposed subproblems, a path optimization method based on genetic algorithm is adopted. Combined with the energy transmission path loss parameters between adjacent base stations, the optimal energy scheduling path is dynamically searched to generate a preliminary cross-base station energy scheduling scheme.

[0084] In this step, a genetic algorithm is applied to optimize the paths for the previously decomposed subproblems, taking into account the energy transmission path loss parameters between adjacent base stations. The genetic algorithm is an optimization algorithm that simulates natural selection and genetic mechanisms. Through steps such as selection, crossover, and mutation, it can effectively explore the optimal solution space. Combining the energy transmission path loss parameters, the genetic algorithm evaluates and selects energy scheduling paths in each iteration, minimizing the loss when transferring energy from one base station to another.

[0085] This step ensures the efficiency of the energy dispatch scheme, optimizes energy transmission between base stations, and minimizes losses. This efficient energy dispatch strategy not only guarantees energy supply to each base station during high-load periods but also improves the overall system operating efficiency, reduces energy costs, and provides a sustainable solution for future energy dispatch.

[0086] In this step, a path optimization method based on a genetic algorithm is employed for the decomposed subproblems to ensure the efficiency of energy scheduling. The genetic algorithm, by simulating the process of natural selection, including operations such as selection, crossover, and mutation, can find the optimal path in a large solution space. In the initial stage of this process, the system randomly generates multiple candidate energy scheduling paths, treating these paths as "individuals." For example, base station A can choose to transfer energy via base station B or base station C; these transfer paths can be diverse in the initial stage.

[0087] In each generation of the algorithm, the system evaluates the fitness of each candidate path based on the total energy loss of the transmission path. A higher fitness path indicates lower energy loss during transmission and is therefore preferred. For example, if the transmission loss from base station A to base station B is 10%, while the transmission loss from base station A to base station C is 5%, the latter will be considered more advantageous in the evaluation. The system then selects paths with higher fitness for crossover and mutation operations to generate new candidate scheduling schemes.

[0088] After multiple generations of iteration and optimization, the genetic algorithm will gradually converge to the optimal solution. The final preliminary cross-base station energy scheduling scheme will enable each base station to meet load requirements while optimizing energy transmission efficiency and minimizing energy loss. For example, if the path from base station A to base station B via base station C is considered optimal, the system will prioritize using this path for energy scheduling, ensuring rapid response and sufficient energy support during periods of high demand.

[0089] For the preliminary cross-base station energy scheduling scheme, a global optimization algorithm based on consensus mechanism is adopted. Through information exchange and iterative updates between distributed nodes, the solutions of each sub-problem are coordinated to generate an optimized allocation strategy that satisfies global constraints.

[0090] In this step, the initial cross-base station energy scheduling scheme needs further optimization using a global optimization algorithm based on a consensus mechanism. The consensus mechanism ensures consistency among different nodes in the distributed system during information exchange. This means that each virtual base station exchanges information by sharing its energy requirements, current load, and available energy storage data. During this process, each node can influence the others, continuously adjusting its energy allocation through iterative updates to achieve a more optimized overall allocation strategy.

[0091] By employing a consensus mechanism, the allocation strategies of different base stations and energy storage units can be coordinated, ensuring that the overall system achieves optimal performance while satisfying local optimization goals. This not only improves the efficiency of energy allocation but also effectively reduces losses during energy transmission, thereby enhancing the stability and reliability of the entire network. Furthermore, the real-time update capability provided by the consensus mechanism allows the system to dynamically adapt to load changes, achieving more efficient energy scheduling.

[0092] In the final step, the initial energy scheduling scheme is further coordinated and optimized using a global optimization algorithm based on a consensus mechanism. As a decentralized method, the consensus mechanism helps different virtual base stations achieve information sharing and consistency during resource allocation. Through information exchange between base stations, information such as energy demand, energy storage status, and load forecasts can be collected in real time. This process ensures that each base station is aware of each other's status, thus providing the necessary data support for global optimization.

[0093] During this process, each node iteratively updates its energy allocation strategy based on the previous energy scheduling plan. Suppose base station A initially allocated 80kWh to base station B, but after communication, it discovers that base station B's demand is 100kWh, while base station C has a lighter load, requiring only 50kWh. Through a consensus mechanism, each base station can negotiate a reallocation of resources to more rationally meet the load demand. For example, base station A can decide to increase its energy allocation to base station B to 100kWh, while correspondingly reducing the allocation to base station C, to enhance support for base station B.

[0094] Ultimately, through continuous communication and collaboration of this information, the system will generate an optimized allocation strategy that satisfies global constraints. This strategy must not only ensure that the energy needs of each base station are met, but also take into account the state of charge, charging and discharging efficiency, and path loss of the energy storage units, thereby achieving efficient energy scheduling across base stations. For example, the final scheduling scheme may designate base station A as the primary energy output point, while base stations B and C serve as load adjustment points, forming a highly efficient overall system.

[0095] This consensus-based global optimization strategy can significantly improve energy efficiency, reduce operating costs, and ensure network stability under extreme conditions. Through distributed decision-making, each base station can quickly respond to changing demands and implement flexible energy allocation, providing strong support for the deployment of future smart grids and renewable energy systems.

[0096] S203. Based on the characteristics of the wireless network topology in the target area, a graph theory-based energy scheduling decision model is constructed. The optimization allocation strategy is subjected to latency sensitivity analysis and energy efficiency balancing processing. The energy storage scheduling sequence is optimized, and a dynamic energy storage scheduling scheme with fault tolerance mechanism is generated to realize energy storage for base station management.

[0097] Based on the wireless network topology characteristics of the target area, a graph theory-based energy scheduling decision model is constructed to achieve efficient energy storage resource management. This model treats base stations and energy storage units in the target area as nodes in a graph, and the energy scheduling paths between them as edges. By analyzing path loss parameters and transmission delay data, weights are assigned to the edges to take into account the losses and delays that may be encountered during actual energy transmission. The generated weighted network topology graph will provide a foundation for subsequent scheduling optimization, ensuring that transmission delay and losses are minimized while outputting energy.

[0098] By constructing such a graph theory model, precise control of the energy scheduling process can be achieved, ensuring that high-load base stations receive priority access to necessary energy support. This not only improves the overall energy efficiency of the network but also enables intelligent and dynamic energy management, allowing for rapid adaptation to changes in user demand. Simultaneously, this strategy possesses a certain fault-tolerance mechanism, automatically adjusting when certain paths fail, achieving high system reliability and stability, and ensuring the smooth operation of base stations.

[0099] Specifically, based on the characteristics of the wireless network topology and optimization allocation strategy of the target area, an energy scheduling decision model based on graph theory can be constructed. In this model, nodes represent base stations or energy storage units, and edges represent energy scheduling paths. By using path loss parameters and transmission delay data, weights are assigned to the edges to generate a weighted network topology graph.

[0100] Constructing weighted network topology graphs provides a comprehensive foundation for subsequent energy scheduling decisions, helping to ensure rapid and efficient decision-making in actual energy transmission. By considering the weights of path loss and transmission delay, the decision-making system can comprehensively evaluate the economics and timeliness of different paths when selecting energy scheduling paths. This multi-dimensional weighting mechanism lays the foundation for optimizing energy scheduling schemes and provides important support for achieving efficient and flexible resource allocation.

[0101] When constructing an energy dispatch decision model based on graph theory, the first step is to identify all base stations and their associated energy storage units within the target area. Each base station and energy storage unit can be represented as a node in the graph, and the energy dispatch paths between them can be represented as edges. During this process, Geographic Information System (GIS) tools can be used to draw the physical layout of the wireless network. For example, assuming a city has base stations A, B, and C and energy storage unit D, A, B, and C can be considered as nodes, and D as another node, forming a preliminary network topology map.

[0102] Next, each generated edge needs to be weighted. This step is crucial because the weights directly affect subsequent scheduling decisions. Path loss is an important indicator for evaluating energy transmission efficiency. It is typically calculated by combining the physical distance between base stations, the characteristics of the transmission medium (e.g., radio waves, fiber optic cables, electrical cables), and environmental interference (e.g., building obstructions, weather conditions). Assuming the distance from base station A to base station B is 1 kilometer, and the path loss is 8 dB due to the urban environment and other interference, this means that energy will be lost proportionally during energy transmission.

[0103] Finally, in assigning weights to each edge, edge latency data must also be considered. This latency data can be obtained based on historical transmission records and analysis of the current network state. Combining path loss parameters and transmission latency, the resulting weighted network topology graph will provide a comprehensive foundation for subsequent energy scheduling decisions, ensuring fast and efficient decision-making in actual energy transmission.

[0104] Specifically, based on the characteristics of the wireless network topology in the target area, base stations and energy storage units can be abstracted as nodes, and energy scheduling paths in the optimization allocation strategy can be abstracted as edges. In this process, the set of nodes and edges is initialized through the physical connection relationship between base stations and energy transmission requirements to generate a preliminary network topology map.

[0105] In this stage, a comprehensive analysis of the wireless network structure within the target area is first required to identify the locations of all base stations and energy storage units. Defining each base station and energy storage unit as a node in the graph simplifies the complex network structure, making it suitable for further algorithmic processing. For example, if there are base stations A, B, and C, and energy storage unit D within the target area, A, B, C, and D can be abstracted as four nodes in the graph. Each node represents an entity, and the relationships between them need to be represented through physical connections.

[0106] Next, edges are constructed using the physical connections between base stations. These connections can take the form of fiber optic cables, wireless signals, etc. For example, suppose base station A is connected to base station B via fiber optic cable, and base station B is connected to base station C via wireless signal. Then, in the graph, the connection between A and B would be represented as edge 1, and the connection between B and C would be represented as edge 2. The system initializes the set of nodes and edges based on the energy transmission requirements and connection characteristics between base stations. At this point, the initial weight of each edge can be assigned based on the connection type and distance. For instance, edges connected by fiber optic cables might be assigned lower weights, while edges connected by wireless signals, due to greater environmental interference, might be assigned higher weights.

[0107] Ultimately, the preliminary network topology generated through the above analysis and processing provides a basic framework for subsequent energy scheduling decisions, ensuring that subsequent steps can be optimized and analyzed based on this graph. This graph theory model not only clearly presents the network structure but also provides the necessary data support for subsequent calculations.

[0108] Constructing such a network topology map is fundamental to the entire energy dispatch decision-making process, effectively reflecting the connectivity and energy transmission demands between base stations and energy storage units within the target area. This model will provide necessary data support and decision-making basis for subsequent energy allocation optimization. Through the abstract representation of nodes and edges, the interrelationships between various parts of the system can be quickly understood and analyzed, laying a solid foundation for real-time decision-making. This structure not only improves dispatch efficiency but also enables real-time monitoring and dynamic adjustment of complex network conditions.

[0109] For each energy scheduling path, an environment-aware path loss calculation model is adopted. The path loss parameters are calculated by combining the distance between base stations, the characteristics of the transmission medium and environmental interference factors. The path loss parameters are normalized by optimizing the energy transmission efficiency target in the allocation strategy to generate path loss weights.

[0110] In this step, the system uses environmentally sensed data to calculate the loss parameters for each energy dispatch path. First, it acquires the physical distance data between base stations, typically measured using GIS or existing network layout. For example, the distance between base station A and base station B is 1 kilometer, while the distance between A and C is 2 kilometers. Next, it needs to identify the transmission medium used for each path, such as fiber optic or radio waves, as different media have different signal transmission effects. Using this information, the system can assess the expected loss; for example, assuming that the loss of a wireless signal due to building obstruction in an urban environment is 8 dB, while the loss of a fiber optic connection is only 2 dB.

[0111] After calculating the path loss, the system normalizes these loss parameters. Normalization converts different loss values ​​into relative values, allowing them to be compared with other key indicators (such as transmission latency). For example, if the path loss values ​​are 5dB and 10dB, normalization adjusts them to values ​​between 0 and 1, facilitating subsequent processing. Ultimately, the generated path loss weights provide crucial information for subsequent energy scheduling decisions. This way, the loss performance of different paths can be compared using a unified standard, ensuring the rationality and effectiveness of the decision-making process. For instance, during the decision-making process, if the program finds that the path loss weight from base station A to base station B is significantly lower than that from A to C, the system will prioritize energy scheduling via B.

[0112] This calculation process not only allows for an accurate understanding of the performance of each energy dispatch path during actual transmission but also enables the optimization of the utilization efficiency of different paths. This environment-aware path loss calculation model provides a scientific basis for energy dispatch, enabling operators to formulate reasonable energy dispatch strategies based on specific environments and conditions. Normalizing the path loss weights helps improve the flexibility and adaptability of energy dispatch, ensuring efficient energy transmission even in complex environments.

[0113] Based on the real-time scheduling requirements in the optimized allocation strategy, and combined with historical transmission delay data and the current network status, a delay prediction method based on time series analysis is adopted to dynamically update the transmission delay data of each energy scheduling path and use it as a delay weight.

[0114] In this phase, the system will combine historical transmission delay data with the current network status to perform latency prediction. First, it needs to collect and store historical transmission delay data, including delay records for each path at different time periods. This data is crucial for discovering latency patterns and trends. For example, assuming that during peak periods, the transmission delay from base station A to base station B is typically between 300 and 400 milliseconds, while during off-peak periods it is between 80 and 150 milliseconds, the system can identify latency characteristics under high load conditions using this historical data.

[0115] Next, time series analysis methods (such as ARIMA or LSTM models) are used to predict the transmission delay data for each path. By inputting historical delay data into the model, the system can predict the delay changes of the path over a future period. For example, when the system recognizes that a weekend is approaching and historical data indicates that traffic is typically high during this period, the system will adjust the predicted transmission delay accordingly.

[0116] Finally, the transmission latency data for each energy scheduling path is dynamically updated, and the updated latency data is incorporated as a new latency weight into subsequent energy scheduling decisions. This ensures that the latest network status is fully considered during energy scheduling, thereby effectively addressing transmission latency issues under high load conditions and ensuring the timeliness and effectiveness of responding to demands.

[0117] This dynamically updated latency prediction mechanism significantly enhances the system's responsiveness to changes in network conditions. This approach ensures that the network maintains efficient and stable energy delivery even under conditions of drastic demand fluctuations, reducing potential service quality degradation due to latency. Furthermore, timely latency data updates provide decision-makers with real-time and accurate information, enabling more flexible and dynamic energy scheduling schemes to adapt to complex and changing network environments.

[0118] The path loss weight and transmission delay weight are optimized and integrated through multi-objective optimization. A comprehensive weight assignment method based on weighted summation is adopted. Combined with the priority and fault tolerance requirements of energy scheduling paths, the edges in the preliminary network topology graph are assigned comprehensive weights to generate a weighted network topology graph.

[0119] In the final step, the system performs multi-objective optimization and fusion of path loss weights and transmission delay weights. First, the relative importance of each component weight is determined. For example, in some business scenarios, ensuring low latency may be more important than reducing loss, so a higher weight can be assigned to the delay weight. Assuming the weight factor for loss is set to 0.4 and the factor for delay weight is 0.6, then the final weight of each path in the comprehensive calculation will be determined by the combination of these two factors.

[0120] Next, combining the priority and fault tolerance requirements of the energy scheduling path, the system assigns a comprehensive weight to each edge. This ensures that the optimal path is selected in a complex scheduling environment. For example, an edge that performs well in terms of transmission loss may be given a lower priority due to its weaker fault tolerance mechanism, while another edge that has slightly higher loss but stronger fault tolerance may be given a higher priority.

[0121] Ultimately, the generated weighted network topology will serve as the basis for determining energy scheduling paths. The comprehensive weighting method ensures that information from multiple dimensions is effectively integrated, making the decision-making process more scientific and rational. This optimized network topology can effectively support energy scheduling decisions, improve the overall efficiency and reliability of the system, and ensure timely and reliable energy support to meet the needs of base stations in complex situations.

[0122] By integrating path loss weights and transmission delay weights through multi-objective optimization, the system not only enhances the intelligence and flexibility of path selection and scheduling but also better meets energy delivery demands under high load conditions. This comprehensive weighting method allows for the full utilization of the characteristics of different paths, thereby improving the overall efficiency and reliability of the system. Furthermore, this process provides strong data support for real-time decision-making, ensuring the network can flexibly adjust to dynamic changes and guaranteeing the quality of service for base station users.

[0123] For energy scheduling paths in the network topology graph, a latency sensitivity analysis method based on the shortest path algorithm is used to calculate the transmission latency of each path. By using a latency tolerance threshold, latency-sensitive critical paths are selected.

[0124] In this stage, the shortest path algorithm is applied to perform latency sensitivity analysis on each energy dispatch path using the constructed weighted network topology. Specifically, starting from the source node, the shortest transmission latency to the target node is calculated based on the transmission latency weight of each path. For example, assuming base station A needs to send energy to base station B, the path could be AB or ACB. The calculation process needs to consider the sum of the latency of each path to determine which path is most optimal under the current conditions. After calculating the transmission latency of all paths, the system uses a set latency tolerance threshold to filter out those latency-sensitive critical paths. These critical paths are those that meet the energy transmission requirements while having high latency sensitivity. For example, if the transmission latency of a path exceeds the tolerance threshold, this path will be marked as unsuitable for use and should be replaced by a path with lower latency. Therefore, this analysis not only ensures response speed under high load conditions but also avoids the problem of low energy distribution efficiency due to latency.

[0125] By implementing latency sensitivity analysis, the accuracy of energy scheduling path selection can be improved, ensuring that the selected scheduling path has good timeliness during actual operation. This is of great significance for ensuring the energy needs of high-load base stations and helps the entire network respond quickly and allocate energy when facing sudden user growth. This method ensures a good balance between energy efficiency and stability, thereby better serving the needs of users within the target area.

[0126] In this stage, based on the constructed weighted network topology, shortest path algorithms are applied to perform latency sensitivity analysis. Shortest path algorithms, especially Dijkstra's algorithm, can effectively help us find the optimal path from the source node to the target node. For example, assuming base station A needs to transmit energy to base station B, the system will analyze all available paths from A to B, calculate the total latency of each path, and determine the fastest energy transmission method. This process involves not only the path length but also the weights of each edge in the selected path (i.e., transmission loss and latency).

[0127] After calculation, the system will obtain the transmission latency results for all possible paths. Next, it needs to filter these paths according to a set latency tolerance threshold. For example, if the threshold is set to 500 milliseconds, if the transmission latency of a path exceeds this value, the system will mark it as unacceptable. This filtering ensures that the selected paths meet energy transmission requirements while avoiding network performance degradation due to excessive latency. For instance, suppose the primary path latency from A to B is 450 milliseconds, while a backup path latency reaches 600 milliseconds. In this case, using the primary path would be more reasonable.

[0128] This step, which successfully identifies latency-sensitive critical paths, will provide important reference for subsequent energy scheduling optimization. Identifying these sensitive paths ensures that the system can respond more quickly when selecting energy transmission paths during peak load periods or sudden events, maximizing the fulfillment of base station load requirements.

[0129] For latency-sensitive critical paths, a load-balanced energy efficiency optimization method is adopted. This method combines the dynamic load prediction results of base stations and the charging and discharging efficiency curves of energy storage units to adjust the allocation ratio of energy scheduling paths, ensuring that high-load base stations receive priority energy support and generating an energy-balanced scheduling scheme.

[0130] In this process, for the identified latency-sensitive critical paths, the system will apply a load-balancing-based energy efficiency optimization method to adjust the energy dispatch scheme. The first step is to determine which base stations have energy demands during peak hours based on the dynamic load forecast results of the base stations. Assume that the predicted load of base station A is 150kWh, while the predicted load of base station B is 80kWh. To optimize energy allocation, the system will prioritize ensuring that base station A receives sufficient energy in a timely manner. Next, by combining the charge and discharge efficiency curves of the energy storage units, the available energy dispatch paths are further analyzed, and the allocation ratio of each path is adjusted accordingly. For example, if the energy storage unit of base station C has a higher charging efficiency and is closer to base station A, the system will tend to allocate more energy from base station C to base station A to reduce transmission losses. Through this allocation, the system ensures that the needs of high-load base stations are met to the maximum extent, optimizing overall energy efficiency. Finally, the generated energy efficiency balancing dispatch scheme will fully consider the load situation of each base station while ensuring the effective utilization of energy dispatch paths. This dynamic adjustment strategy guarantees flexibility and efficiency in responding to sudden demands, contributing to the stable operation of the overall network.

[0131] By employing a load-balancing-based energy efficiency optimization method, not only can high-load base stations receive the necessary energy support, but the overall energy utilization efficiency of the system can also be improved. This optimized scheduling scheme helps reduce energy waste in energy scheduling while ensuring that the system maintains dynamic balance in environments with frequent load fluctuations. Ultimately, it achieves a flexible and efficient energy scheduling mechanism that can better cope with complex network demands.

[0132] In this step, a load-balancing-based energy efficiency optimization method is adopted for the identified latency-sensitive critical paths. Before energy allocation, the system first needs to obtain the dynamic load forecast results of the base stations. Assume that at a certain time period, the expected load of base station A is 200 kWh, while that of base stations B and C is 100 kWh and 150 kWh, respectively. Since base station A has the greatest load demand, it must be prioritized to ensure that it receives the required energy in a timely manner during energy allocation.

[0133] Next, the system analyzes the current state of each energy storage unit by combining its charge and discharge efficiency curves to enable more effective energy dispatch. For example, if the charging efficiency of energy storage unit D is 90%, then during the energy dispatch process from D to A, B, and C, the system will consider a relatively high dispatch ratio from D to A, while ensuring that the needs of B and C are also met. In this way, the system will dynamically adjust the allocation ratio of each energy dispatch path to ensure that high-load base stations receive priority energy support and minimize energy loss.

[0134] Ultimately, considering all the above factors, the resulting energy efficiency balancing scheduling scheme will ensure the energy support of base stations under high load conditions, while also playing a positive role in optimizing the overall energy efficiency of the system. This flexible scheduling strategy not only improves the network's responsiveness but also provides an efficient solution for potential future load changes.

[0135] For energy efficiency balance scheduling schemes, a fault-tolerant mechanism based on redundant paths is introduced. By constructing backup energy scheduling paths and dynamic switching strategies, energy transmission can still be maintained when some paths fail, ultimately generating a dynamic energy storage scheduling scheme with fault tolerance.

[0136] In this step, a fault-tolerance mechanism for redundant paths is introduced for the previously generated energy efficiency balancing scheduling scheme. First, by analyzing the current network topology, feasible redundant paths are identified and designated as backup scheduling paths. For example, if the primary scheduling path from base station A to base station B fails, a backup path can be set up from base station A through base station C to finally reach base station B. The key to establishing these backup paths is the ability to quickly identify the status of the primary path and seamlessly switch to the backup path in case of failure. Next, a corresponding dynamic switching strategy needs to be formulated. This strategy should be able to monitor the status of the transmission path in real time and determine whether to continue using the primary path or switch to the backup path. For example, when the transmission delay of the primary path is detected to exceed a set threshold, the system will automatically switch to the backup path to ensure that energy can reach its destination in a timely manner. Through this dynamic switching mechanism, the scheduling scheme can be quickly adjusted when the network experiences abnormal conditions, ensuring the uninterrupted and stable overall energy transmission. Finally, a dynamic energy storage scheduling scheme with a complete fault-tolerance mechanism is constructed, which not only guarantees energy transmission in the event of a primary path failure but also improves the overall reliability of the system, providing a guarantee for the continuous operation of base stations. The introduction of this mechanism provides stronger resilience and security for energy allocation in complex networks.

[0137] Introducing a fault-tolerant mechanism with redundant paths significantly enhances the robustness and flexibility of the entire energy dispatching system. Even in the face of equipment failures or network anomalies, the system can still guarantee efficient energy delivery, reducing the risk of service interruptions due to single points of failure. Furthermore, the implementation of dynamic switching strategies ensures the real-time nature of energy policies, contributing to improved resource utilization and user experience across the entire network. This approach not only enhances system resilience but also provides a solid technical foundation for future large-scale network deployments.

[0138] In this step, a fault-tolerant mechanism based on redundant paths is introduced to increase the reliability of the energy efficiency balancing scheduling scheme. First, the existing network topology needs to be analyzed to identify available backup scheduling paths. This analysis aims to equip each primary scheduling path with one or more corresponding backup paths. For example, if the primary scheduling path from base station A to base station B fails, the system should be able to quickly switch to the backup path from A through base station C to base station B. This process ensures the continuity of energy transmission and reduces the risk of service interruption due to a single path failure.

[0139] Next, a dynamic switching strategy needs to be developed to monitor the status of the current transmission path in real time. If the system detects that the delay of the primary path exceeds a set threshold, or if a path fails, the backup path is immediately activated. For example, when signal packet loss occurs on the transmission path of base station A, the system will automatically switch to the backup path. This mechanism ensures the flexibility and adaptability of the network, helping to maintain efficient power transmission in complex network environments.

[0140] Finally, the fault-tolerant dynamic energy storage scheduling scheme generated in this stage can not only effectively cope with sudden failures on the main path, but also improve the overall system stability and reliability. In actual operation, it ensures that energy can be flexibly allocated between different paths, ensuring that base stations can operate smoothly even under high load conditions, thereby enhancing the service quality and user experience of the entire wireless network.

[0141] As can be seen, multimodal data fusion processing is performed on historical traffic data, real-time user mobility pattern data, and meteorological data of each virtual base station in the target area to obtain the dynamic load prediction results of each virtual base station in the target time period. Based on the dynamic load prediction results, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths. According to the wireless network topology characteristics of the target area, a graph theory-based energy scheduling decision model is constructed to optimize the energy storage scheduling sequence and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management, thereby improving the energy efficiency of base stations, reducing energy waste, and achieving efficient configuration and real-time scheduling of energy storage resources.

[0142] Another embodiment of the present invention provides an energy storage system for base station management, see [link to relevant documentation]. Figure 3 The system may include:

[0143] The fusion module 301 is used to perform multimodal data fusion processing based on the historical traffic data, real-time user mobility pattern data and meteorological data of each virtual base station in the target area, so as to obtain the dynamic load prediction results of each virtual base station in the target time period.

[0144] The allocation module 302 is used to dynamically allocate energy storage resources based on dynamic load prediction results, combined with the current state of charge, charging and discharging efficiency curves of the energy storage units associated with each virtual base station and the energy transmission path loss parameters between adjacent base stations, using a distributed constraint optimization algorithm to generate an optimized allocation strategy that includes cross-base station energy scheduling paths.

[0145] The construction module 303 is used to construct a graph theory-based energy scheduling decision model based on the wireless network topology characteristics of the target area, perform latency sensitivity analysis and energy efficiency balancing processing on the optimization allocation strategy, optimize the energy storage scheduling sequence, and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management.

[0146] As can be seen, multimodal data fusion processing is performed on historical traffic data, real-time user mobility pattern data, and meteorological data of each virtual base station in the target area to obtain the dynamic load prediction results of each virtual base station in the target time period. Based on the dynamic load prediction results, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths. According to the wireless network topology characteristics of the target area, a graph theory-based energy scheduling decision model is constructed to optimize the energy storage scheduling sequence and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management, thereby improving the energy efficiency of base stations, reducing energy waste, and achieving efficient configuration and real-time scheduling of energy storage resources.

[0147] This invention also provides a storage medium storing a computer program, wherein the computer program is configured to execute the steps in any of the above method embodiments when running.

[0148] Specifically, in this embodiment, the storage medium can be configured to store a computer program for performing the following steps:

[0149] S201. Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, multimodal data fusion processing is performed to obtain the dynamic load prediction results of each virtual base station in the target time period.

[0150] S202, based on the dynamic load prediction results, combined with the current state of charge, charging and discharging efficiency curves of the energy storage units associated with each virtual base station and the energy transmission path loss parameters between adjacent base stations, adopts a distributed constraint optimization algorithm to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths;

[0151] S203. Based on the characteristics of the wireless network topology in the target area, a graph theory-based energy scheduling decision model is constructed. The optimization allocation strategy is subjected to latency sensitivity analysis and energy efficiency balancing processing. The energy storage scheduling sequence is optimized, and a dynamic energy storage scheduling scheme with fault tolerance mechanism is generated to realize energy storage for base station management.

[0152] As can be seen, multimodal data fusion processing is performed on historical traffic data, real-time user mobility pattern data, and meteorological data of each virtual base station in the target area to obtain the dynamic load prediction results of each virtual base station in the target time period. Based on the dynamic load prediction results, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths. According to the wireless network topology characteristics of the target area, a graph theory-based energy scheduling decision model is constructed to optimize the energy storage scheduling sequence and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management, thereby improving the energy efficiency of base stations, reducing energy waste, and achieving efficient configuration and real-time scheduling of energy storage resources.

[0153] This invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the steps in any of the above method embodiments.

[0154] Specifically, the aforementioned electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the aforementioned processor, and the input / output device is connected to the aforementioned processor.

[0155] Specifically, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0156] S201. Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, multimodal data fusion processing is performed to obtain the dynamic load prediction results of each virtual base station in the target time period.

[0157] S202, based on the dynamic load prediction results, combined with the current state of charge, charging and discharging efficiency curves of the energy storage units associated with each virtual base station and the energy transmission path loss parameters between adjacent base stations, adopts a distributed constraint optimization algorithm to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths;

[0158] S203. Based on the characteristics of the wireless network topology in the target area, a graph theory-based energy scheduling decision model is constructed. The optimization allocation strategy is subjected to latency sensitivity analysis and energy efficiency balancing processing. The energy storage scheduling sequence is optimized, and a dynamic energy storage scheduling scheme with fault tolerance mechanism is generated to realize energy storage for base station management.

[0159] As can be seen, multimodal data fusion processing is performed on historical traffic data, real-time user mobility pattern data, and meteorological data of each virtual base station in the target area to obtain the dynamic load prediction results of each virtual base station in the target time period. Based on the dynamic load prediction results, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths. According to the wireless network topology characteristics of the target area, a graph theory-based energy scheduling decision model is constructed to optimize the energy storage scheduling sequence and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management, thereby improving the energy efficiency of base stations, reducing energy waste, and achieving efficient configuration and real-time scheduling of energy storage resources.

[0160] The above description, based on the embodiments shown in the figures, details the structure, features, and effects of the present invention. The above description is only a preferred embodiment of the present invention, but the present invention is not limited to the scope of implementation shown in the figures. Any changes made in accordance with the concept of the present invention, or equivalent embodiments modified to have equivalent changes, that do not exceed the spirit covered by the specification and figures, should be within the protection scope of the present invention.

Claims

1. An energy storage method for base station management, characterized in that, The method includes: Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, multimodal data fusion processing is performed to obtain the dynamic load prediction results of each virtual base station in the target time period. Based on the dynamic load prediction results, combined with the current state of charge, charging and discharging efficiency curves of the energy storage units associated with each virtual base station and the energy transmission path loss parameters between adjacent base stations, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources and generate an optimized allocation strategy that includes cross-base station energy scheduling paths. Based on the characteristics of the wireless network topology in the target area, a graph theory-based energy scheduling decision model is constructed. Delay sensitivity analysis and energy efficiency balancing are performed on the optimization allocation strategy to optimize the energy storage scheduling sequence and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management.

2. The method according to claim 1, characterized in that, The process involves multimodal data fusion processing based on historical traffic data, real-time user mobility pattern data, and meteorological data of each virtual base station within the target area to obtain dynamic load prediction results for each virtual base station during the target time period, including: Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, a multi-source data alignment algorithm based on dynamic time warping is adopted to eliminate the data timestamp inconsistency problem and fill in the missing values ​​by interpolation to obtain a time-synchronized multimodal dataset. For time-synchronized multimodal datasets, a feature extraction network based on a multi-head attention mechanism is used to extract the temporal features of historical traffic volume, the spatial features of user movement patterns, and the environmental features of meteorological data. The features of different modalities are then weighted and fused through a cross-modal attention mechanism to generate a fused multimodal feature representation. The fused multimodal feature representations are input into a pre-trained prediction model based on a spatiotemporal graph convolutional network. Combined with the base station topology of the target area, the spatiotemporal dependencies between base stations are captured, and the dynamic load change trend of each virtual base station in the target time period is predicted, thus obtaining the dynamic load prediction results of each virtual base station in the target time period.

3. The method according to claim 2, characterized in that, Based on the dynamic load prediction results, combined with the current state of charge, charge / discharge efficiency curves, and energy transmission path loss parameters of the energy storage units associated with each virtual base station, a distributed constraint optimization algorithm is used to dynamically allocate energy storage resources, generating an optimized allocation strategy that includes cross-base station energy scheduling paths, including: Based on the current state of charge, charge and discharge efficiency curves, and energy transmission path loss parameters between adjacent base stations of the energy storage units associated with each virtual base station, a multi-dimensional constraint model of the energy storage state is constructed. Through energy conservation constraints, charge and discharge rate constraints, and transmission loss constraints, the feasible solution space for energy storage resource allocation is defined. By combining dynamic load prediction results with a multi-dimensional constraint model of energy storage status, a distributed optimization framework based on the Lagrange relaxation method is adopted to decompose the global optimization problem of energy storage resource allocation into multiple sub-problems. Each sub-problem corresponds to a local optimization objective of a virtual base station. The local optimization objectives include load balancing, energy transmission efficiency, and energy storage lifetime. For the decomposed subproblems, a path optimization method based on genetic algorithm is adopted. Combined with the energy transmission path loss parameters between adjacent base stations, the optimal energy scheduling path is dynamically searched to generate a preliminary cross-base station energy scheduling scheme. For the preliminary cross-base station energy scheduling scheme, a global optimization algorithm based on consensus mechanism is adopted. Through information exchange and iterative updates between distributed nodes, the solutions of each sub-problem are coordinated to generate an optimized allocation strategy that satisfies global constraints.

4. The method according to claim 3, characterized in that, Based on the wireless network topology characteristics of the target area, a graph theory-based energy scheduling decision model is constructed. Delay sensitivity analysis and energy efficiency balancing are performed on the optimized allocation strategy to optimize energy storage scheduling timing and generate a dynamic energy storage scheduling scheme with fault tolerance, thereby realizing energy storage for base station management. This includes: Based on the characteristics of the wireless network topology and optimization allocation strategy of the target area, an energy scheduling decision model based on graph theory is constructed. In this model, nodes represent base stations or energy storage units, and edges represent energy scheduling paths. Weights are assigned to the edges using path loss parameters and transmission delay data to generate a weighted network topology graph. For energy scheduling paths in the network topology graph, a latency sensitivity analysis method based on the shortest path algorithm is used to calculate the transmission latency of each path. By using a latency tolerance threshold, latency-sensitive critical paths are selected. For latency-sensitive critical paths, a load-balanced energy efficiency optimization method is adopted. This method combines the dynamic load prediction results of base stations and the charging and discharging efficiency curves of energy storage units to adjust the allocation ratio of energy scheduling paths, ensuring that high-load base stations receive priority energy support and generating an energy-balanced scheduling scheme. For energy efficiency balance scheduling schemes, a fault-tolerant mechanism based on redundant paths is introduced. By constructing backup energy scheduling paths and dynamic switching strategies, energy transmission can still be maintained when some paths fail, ultimately generating a dynamic energy storage scheduling scheme with fault tolerance.

5. The method according to claim 4, characterized in that, The step involves constructing a graph-based energy scheduling decision model based on the wireless network topology characteristics and optimization allocation strategy of the target area. Nodes represent base stations or energy storage units, and edges represent energy scheduling paths. Weights are assigned to the edges using path loss parameters and transmission delay data to generate a weighted network topology graph, including: Based on the characteristics of the wireless network topology in the target area, base stations and energy storage units are abstracted as nodes, and energy scheduling paths in the optimization allocation strategy are abstracted as edges. The set of nodes and edges is initialized through the physical connection relationship between base stations and energy transmission requirements to generate a preliminary network topology graph. For each energy scheduling path, an environment-aware path loss calculation model is adopted. The path loss parameters are calculated by combining the distance between base stations, the characteristics of the transmission medium and environmental interference factors. The path loss parameters are normalized by optimizing the energy transmission efficiency target in the allocation strategy to generate path loss weights. Based on the real-time scheduling requirements in the optimized allocation strategy, and combined with historical transmission delay data and the current network status, a delay prediction method based on time series analysis is adopted to dynamically update the transmission delay data of each energy scheduling path and use it as a delay weight. The path loss weight and transmission delay weight are optimized and integrated through multi-objective optimization. A comprehensive weight assignment method based on weighted summation is adopted. Combined with the priority and fault tolerance requirements of energy scheduling paths, the edges in the preliminary network topology graph are assigned comprehensive weights to generate a weighted network topology graph.

6. An energy storage system for base station management, characterized in that, The system includes: The fusion module is used to perform multimodal data fusion processing based on the historical traffic data, real-time user mobility pattern data and meteorological data of each virtual base station in the target area, and to obtain the dynamic load prediction results of each virtual base station in the target time period. The allocation module is used to dynamically allocate energy storage resources based on dynamic load prediction results, combined with the current state of charge, charge and discharge efficiency curves of the energy storage units associated with each virtual base station, and energy transmission path loss parameters between adjacent base stations, using a distributed constraint optimization algorithm to generate an optimized allocation strategy that includes cross-base station energy scheduling paths. The module is used to construct a graph theory-based energy scheduling decision model based on the wireless network topology characteristics of the target area, perform latency sensitivity analysis and energy efficiency balancing processing on the optimization allocation strategy, optimize the energy storage scheduling sequence, and generate a dynamic energy storage scheduling scheme with fault tolerance mechanism to realize energy storage for base station management.

7. The system according to claim 6, characterized in that, The fusion module is specifically used for: Based on the historical traffic data, real-time user mobility data and meteorological data of each virtual base station in the target area, a multi-source data alignment algorithm based on dynamic time warping is adopted to eliminate the data timestamp inconsistency problem and fill in the missing values ​​by interpolation to obtain a time-synchronized multimodal dataset. For time-synchronized multimodal datasets, a feature extraction network based on a multi-head attention mechanism is used to extract the temporal features of historical traffic volume, the spatial features of user movement patterns, and the environmental features of meteorological data. The features of different modalities are then weighted and fused through a cross-modal attention mechanism to generate a fused multimodal feature representation. The fused multimodal feature representations are input into a pre-trained prediction model based on a spatiotemporal graph convolutional network. Combined with the base station topology of the target area, the spatiotemporal dependencies between base stations are captured, and the dynamic load change trend of each virtual base station in the target time period is predicted, thus obtaining the dynamic load prediction results of each virtual base station in the target time period.

8. The system according to claim 7, characterized in that, The allocation module is specifically used for: Based on the current state of charge, charge and discharge efficiency curves, and energy transmission path loss parameters between adjacent base stations of the energy storage units associated with each virtual base station, a multi-dimensional constraint model of the energy storage state is constructed. Through energy conservation constraints, charge and discharge rate constraints, and transmission loss constraints, the feasible solution space for energy storage resource allocation is defined. By combining dynamic load prediction results with a multi-dimensional constraint model of energy storage status, a distributed optimization framework based on the Lagrange relaxation method is adopted to decompose the global optimization problem of energy storage resource allocation into multiple sub-problems. Each sub-problem corresponds to a local optimization objective of a virtual base station. The local optimization objectives include load balancing, energy transmission efficiency, and energy storage lifetime. For the decomposed subproblems, a path optimization method based on genetic algorithm is adopted. Combined with the energy transmission path loss parameters between adjacent base stations, the optimal energy scheduling path is dynamically searched to generate a preliminary cross-base station energy scheduling scheme. For the preliminary cross-base station energy scheduling scheme, a global optimization algorithm based on consensus mechanism is adopted. Through information exchange and iterative updates between distributed nodes, the solutions of each sub-problem are coordinated to generate an optimized allocation strategy that satisfies global constraints.

9. A storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the method of any one of claims 1-5 when it is run.

10. An electronic device comprising a memory and a processor, characterized in that, The memory stores a computer program, and the processor is configured to run the computer program to perform the method of any one of claims 1-5.

Images