An energy management method, system, device and storage medium of a distributed storage and charging system

Abstract

The application belongs to the technical field of power distribution and discloses an energy management method, system and device of a distributed storage and charging system and a storage medium, comprising: constructing a machine networking architecture containing a plurality of storage and charging units; when there is external power input or power assistance demand in the machine networking architecture, calculating a corresponding competition value based on the residual power data and the predicted load of each storage and charging unit; and calculating an input distribution scheme of external power and a power assistance scheme between the storage and charging units based on the corresponding competition value of each storage and charging unit. The technical scheme disclosed by the application can effectively avoid the risk of system collapse under centralized control, while reducing the cost of the system and improving the efficiency and stability of the system.

Description

Technical Field

[0001] This invention belongs to the field of power distribution technology, and in particular relates to an energy management method, system, device and storage medium for a distributed energy storage and charging system. Background Technology

[0002] As energy storage systems become increasingly important in power systems, especially against the backdrop of the rapid development of the electric vehicle industry, traditional large-scale centralized energy storage systems are gradually revealing some limitations. To address these challenges, decentralized storage / charge systems (DSCS) have emerged as a new option for users due to their flexibility, rapid response, and convenience. By combining distributed storage / charge systems with DC fast charging and supercharging stations, and leveraging the shared DC bus characteristic of the storage / charge system, not only is the operating efficiency of charging stations improved, but the capacity requirements of dedicated transformer equipment for charging stations are also reduced.

[0003] However, most current distributed energy storage and charging system clusters adopt a centralized control approach. While this approach can achieve overall coordination, it also has some drawbacks in practical applications, such as high requirements for central control equipment and transmission channels, increased risk of system crashes as the number of devices increases, and rising hardware and software costs.

[0004] Current decentralized distributed energy storage and charging system cluster coordination and control schemes, while proposing the concept of decentralization, still rely on a mechanism to designate a host to control the remaining devices, failing to achieve complete decentralization. Host failures and the intervals between the election process can still cause the cluster to malfunction. Furthermore, the cluster only handles control command exchanges and does not coordinate energy exchanges between the various energy storage and charging system units via the DC bus. Therefore, it does not resolve the issues of reduced energy utilization caused by transmission losses in the process of units directly sharing the DC bus.

[0005] While existing technologies, when combined with DC fast charging and supercharging stations, improve the operating efficiency of charging stations and reduce the demand for power equipment capacity by sharing a DC bus, the lack of intelligent planning often results in suboptimal energy scheduling among the various energy storage and charging units when sharing the DC bus. This limitation leads to uneven energy distribution, increased transmission losses, and difficulty in dynamically adjusting the charging load according to real-time demand, posing a risk of system collapse and ultimately affecting the overall system's energy efficiency. Summary of the Invention

[0006] The purpose of this invention is to provide an energy management method, system, device, and storage medium for a distributed energy storage and charging system, so as to solve the problems existing in the prior art.

[0007] To achieve the above objectives, the present invention provides an energy management method for a distributed energy storage and charging system, comprising:

[0008] Step 1: Construct a machine network architecture containing several storage and charging units;

[0009] Step 2: When there is an external power input or power assistance requirement in the machine network architecture, calculate the corresponding competition value based on the remaining energy storage power data and predicted load of each of the energy storage and charging units;

[0010] Step 3: Calculate the external power input allocation scheme and the power assistance scheme between the power storage and charging units based on the competition value corresponding to each of the aforementioned power storage and charging units.

[0011] Optionally, step two specifically includes:

[0012] Acquire training data, which includes historical charging data, relative geographical location data, and load data for each of the storage and charging units;

[0013] A load prediction model is constructed based on a spatiotemporal graph convolutional network;

[0014] Based on the training data, the load prediction model is trained and its parameters are optimized using the Grey Wolf optimization algorithm to obtain the optimized load prediction model.

[0015] The load forecasting of each of the storage and charging units is performed based on the optimized load forecasting model to obtain the predicted load data of each of the storage and charging units.

[0016] Obtain the remaining energy storage capacity and electricity price data of each of the energy storage and charging units;

[0017] The remaining energy storage capacity, electricity price level, and predicted load data of each energy storage and charging unit are weighted and summed to obtain the corresponding competition value.

[0018] Optionally, the load prediction model is trained and its parameters optimized based on the training data and the Grey Wolf optimization algorithm, specifically including:

[0019] Extract the hyperparameter combinations included in the load forecasting model;

[0020] The load prediction model corresponding to each of the hyperparameter combinations is trained based on the training data, and the training results are evaluated. The evaluation results are used as the evaluation scores corresponding to each of the hyperparameter combinations.

[0021] The gray wolf population is generated using the aforementioned hyperparameter combinations;

[0022] Based on the evaluation scores corresponding to each hyperparameter combination, the optimal hyperparameter combination is obtained by combining the gray wolf optimization algorithm to optimize the parameters in the gray wolf population.

[0023] An optimized load prediction model is constructed based on the optimal combination of hyperparameters.

[0024] Optionally, the load prediction model corresponding to each of the hyperparameter combinations is trained based on the training data, specifically including:

[0025] The training data is divided into a training set, a test set, and a validation set according to a preset ratio;

[0026] The load prediction model corresponding to each hyperparameter combination is trained based on the training set, and the performance of the load prediction model corresponding to each hyperparameter combination is verified and evaluated through the validation set. The evaluation result is used as the evaluation score corresponding to each hyperparameter combination.

[0027] Optionally, based on the evaluation scores corresponding to each of the hyperparameter combinations, the gray wolf optimization algorithm is used to optimize the parameters within the gray wolf population, specifically including:

[0028] Set the range of values, population size and number of iterations for the hyperparameter combinations, and use each of the hyperparameter combinations as gray wolves to generate the gray wolf population;

[0029] The gray wolf population is iteratively updated multiple times based on the gray wolf optimization algorithm for each hyperparameter combination. In each iteration, the evaluation score corresponding to each hyperparameter combination is calculated, and the gray wolf population is updated based on the evaluation score corresponding to each hyperparameter combination until the number of iterations reaches a preset value, at which point the optimal hyperparameter combination is output.

[0030] Optionally, load forecasting for each of the storage and charging units is performed based on the optimized load forecasting model, specifically including:

[0031] Extract the temporal and relative geographical location features of the storage and charging unit.

[0032] The temporal features are subjected to mode decomposition, and the decomposed mode functions are combined with relative geographical location features to construct a feature dataset with spatiotemporal features;

[0033] The load prediction model includes a first temporal convolutional layer, a graph convolutional layer, a second temporal convolutional layer, and a fully connected prediction layer connected in sequence. The feature dataset is input into the load prediction model, and features are extracted through each temporal convolutional layer and the graph convolutional layer.

[0034] The extracted features are input into the fully connected prediction layer for classification and prediction to obtain the predicted load data.

[0035] Optionally, an input power allocation scheme for external power and a power assistance scheme between power storage and charging units are calculated based on the competition value corresponding to each of the aforementioned power storage and charging units, specifically including:

[0036] External power input allocation scheme: sort the competition values ​​corresponding to each of the aforementioned power storage units, and select several power storage units as candidate power storage units based on a preset competition threshold in the sorted list;

[0037] Based on the competition value corresponding to each candidate energy storage unit, the input power of external electrical energy is allocated proportionally to complete the external electrical energy input allocation for the current cycle.

[0038] Energy assistance scheme between energy storage and charging units: The energy storage and charging unit to be assisted initiates an assistance request, and the remaining energy storage and charging units other than the one to be assisted are sorted according to the competition value. The energy storage and charging unit to be assisted is selected based on the sorting result.

[0039] An energy management system for a distributed energy storage and charging system includes:

[0040] The machine-to-machine network (MTN) construction module is used to build an MTN architecture that includes several storage and charging units;

[0041] The competition mechanism module is used to calculate the corresponding competition value based on the remaining energy storage capacity data and predicted load of each of the energy storage and charging units when there is an external power input or power assistance demand in the machine network architecture.

[0042] The assistance module is used to calculate the input distribution scheme of external power and the power assistance scheme between the storage and charging units based on the competition value corresponding to each of the storage and charging units.

[0043] An electronic device includes a memory and a processor, the memory storing a computer program, and the processor running the computer program to enable the electronic device to perform an energy management method for a distributed energy storage and charging system.

[0044] A computer-readable storage medium storing a computer program that, when executed by a processor, implements an energy management method for a distributed energy storage and charging system.

[0045] The technical effects of this invention are as follows:

[0046] This invention achieves energy sharing and information exchange between units through a hardware-level scheduling and response mechanism in a decentralized "machine-to-machine network" structure for energy storage and charging system units. This structure is compatible with flexible networking methods for multiple units, reducing the deployment difficulty of energy storage and charging system clusters and expanding their applicability. This invention effectively avoids the risk of system collapse under centralized control by using a hostless decentralized control algorithm, while also reducing system costs. Utilizing a competitive algorithm, the system can dynamically decide the subject issuing commands based on the total grid power, the input demand of each energy storage and charging system unit, and load forecasts, thereby improving system reliability and flexibility. This invention further incorporates deep learning algorithms to enhance the scientific nature of "machine-to-machine network" energy scheduling. By dynamically predicting the load of each unit, the system can provide more accurate competitive weights during energy scheduling, thereby optimizing the overall energy management strategy and improving system efficiency and stability. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0049] Figure 1 This is a functional allocation diagram of the PS and PL terminals of the storage and charging unit in this embodiment of the invention.

[0050] Figure 2 This is a diagram showing the structure of the special cables connecting each storage and charging unit in the machine network according to an embodiment of the present invention;

[0051] Figure 3 This is a schematic diagram of the machine network cluster structure in an embodiment of the present invention;

[0052] Figure 4 This is a schematic diagram of the competition mechanism process in an embodiment of the present invention;

[0053] Figure 5 This is a flowchart illustrating the specific calculation process of the competition value in this embodiment of the invention.

[0054] Figure 6 This is a flowchart illustrating the specific steps involved in constructing a deep computing model using the VMD-GWO-ST-GCN algorithm in an embodiment of the present invention.

[0055] Figure 7This is a graph showing the full service rate of supercharging for each strategy in a daily scenario under different time periods in an embodiment of the present invention.

[0056] Figure 8 This is a graph showing the daily average power consumption to electricity cost ratio for each strategy in the embodiments of the present invention. Detailed Implementation

[0057] Various exemplary embodiments of the present invention will now be described in detail. This detailed description should not be considered as a limitation of the present invention, but rather as a more detailed description of certain aspects, features, and embodiments of the present invention.

[0058] It should be understood that the terminology used in this invention is merely for describing particular embodiments and is not intended to limit the invention. Furthermore, with respect to numerical ranges in this invention, it should be understood that each intermediate value between the upper and lower limits of the range is also specifically disclosed. Every smaller range between any stated value or intermediate value within a stated range, and any other stated value or intermediate value within said range, is also included in this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0059] Unless otherwise stated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. While only preferred methods have been described herein, any methods similar or equivalent to those described herein may be used in the implementation or testing of this invention. All references to this specification are incorporated by way of citation to disclose and describe the methods associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0060] Various modifications and variations can be made to the specific embodiments described in this specification without departing from the scope or spirit of the invention, as will be apparent to those skilled in the art. Other embodiments derived from this specification will also be obvious to those skilled in the art. This application specification and embodiments are merely exemplary.

[0061] The terms “include,” “including,” “have,” “contain,” etc., used in this article are all open-ended terms, meaning that they include but are not limited to.

[0062] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0063] Example 1

[0064] like Figure 1 - Figure 8 As shown, this embodiment provides an energy management method for a distributed energy storage and charging system, including:

[0065] Step 1: Construct a machine network architecture containing several storage and charging units;

[0066] Step 2: When there is an external power input or power assistance requirement in the machine network architecture, calculate the corresponding competition value based on the remaining energy storage power data and predicted load of each of the energy storage and charging units;

[0067] Step 3: Calculate the external power input allocation scheme and the power assistance scheme between the power storage and charging units based on the competition value corresponding to each of the aforementioned power storage and charging units.

[0068] The problem to be solved in this embodiment is: how to achieve efficient coordination and exchange of energy and information among multiple energy storage and charging system units, and to plan energy storage and charging in an orderly manner according to the total power of the grid and equipment requirements, while avoiding the risk of system collapse and cost increase caused by centralized control.

[0069] To address the aforementioned issues, this embodiment proposes a decentralized "machine-to-machine network" structure for energy storage and charging system units, compatible with networking modes of 1-n units. Utilizing a competitive algorithm, it dynamically determines the entity issuing commands between nodes based on factors such as the total power input from the power grid, the input demands and geographical locations of each energy storage and charging system unit, and load predictions calculated by deep learning algorithms. This method leverages deep learning algorithms to improve the scientific nature of energy scheduling in the "machine-to-machine network," dynamically predicting the load of each individual unit and providing a basis for competitive weighting during energy scheduling.

[0070] This embodiment achieves energy sharing and information exchange between units through a hardware-level scheduling and response mechanism in a decentralized "machine-to-machine network" structure for energy storage and charging system units. This structure is compatible with flexible networking methods of 1-n units, reducing the deployment difficulty of energy storage and charging system clusters and expanding its applicability. The hostless decentralized control algorithm effectively avoids the risk of system collapse under centralized control, while also reducing system costs. Utilizing a competition algorithm, the system can dynamically decide the subject issuing commands based on the total grid power, the input demand of each energy storage and charging system unit, geographical location, peak and valley electricity prices, and load forecasts, thereby improving system reliability and flexibility. Deep learning algorithms enhance the scientific nature of the "machine-to-machine network" energy scheduling. By dynamically predicting the load of each unit, the system can provide more accurate competitive weights during energy scheduling, thereby optimizing the overall energy management strategy and improving system efficiency and stability. This embodiment proposes a decentralized "machine-to-machine network" structure for energy storage and charging system units, compatible with flexible networking methods of 1-n energy storage and charging units, and achieving hostless decentralized control. Through a competitive algorithm, each energy storage and charging unit dynamically decides which entity to issue commands based on factors such as the total power input from the grid, its own input demand, geographical location, peak-valley electricity prices, and load forecasts calculated by deep learning algorithms, thereby optimizing the operation of the entire cluster. Furthermore, this embodiment proposes an energy scheduling optimization method based on deep learning algorithms, which can dynamically predict the load of each unit and provide a basis for competitive weights during the "machine-networked" energy scheduling process, further improving the system's energy efficiency and scheduling scientificity. Through autonomous scheduling mechanisms and intelligent algorithms, each energy storage and charging unit can dynamically optimize the energy allocation on the DC bus based on real-time energy demand and load forecasts, avoiding the limitations of centralized control methods and improving the system's flexibility and stability.

[0071] like Figure 1 As shown, an EMS (Energy Management System) for the energy storage and charging unit is constructed using ZYNQ (FPGA + ARM combined with main control chip). A hardware-level control response system design is built on the PL (Programmable Logic) side, and an energy management strategy and cloud server interconnection function are built on the PS (Processing System) side. The ARM processing unit provides rich interfaces and powerful computing capabilities, which are suitable for running complex energy management algorithms and exchanging data with the cloud.

[0072] like Figure 2 As shown, each storage and charging unit is connected using a specially designed cable containing two signal lines (S+, S-), two power lines (V+, V-), one ground line (GND), and two heat transfer fluid pipes.

[0073] like Figure 3 As shown, each energy storage unit is connected in parallel within the network using specialized cables, and the internal circuitry of each unit is accessed through the network's interconnect controller. Communication between nodes in the network utilizes the Flexible Rate Controller Area Network (CAN-FD) protocol. This establishes the hardware connection and communication control response mechanism for the network cluster structure.

[0074] This system architecture combines the powerful capabilities of the ZYNQ chip, the flexible connectivity of specialized cables, and the reliable communication of the CAN-FD protocol to construct a highly efficient, stable, and scalable energy management system (EMS). This system not only achieves real-time performance and high efficiency in hardware-level control response but also enables intelligent energy management strategies and remote monitoring through the interconnection between the ARM processing unit and the cloud server. This system architecture has broad application prospects in the fields of energy storage and charging.

[0075] like Figure 4 As shown, after each storage and charging unit is connected to the network, when a storage and charging unit receives external power input or needs power assistance, it will actively initiate a competition mechanism in the network. After receiving the competition initiation command, other nodes will calculate their own competition weight factors and send their own weight information to the network within a specified time window. The competition mechanism initiating unit collects the weight data of each unit and determines the recipient of external power input and the recipient of assistance.

[0076] The specific calculation method of the competition value is as follows: Figure 5 As shown, each node calculates its own competition value based on the following two factors:

[0077] 1. Remaining battery state of charge (SoC);

[0078] 2. Predicted Load from the Model (PL);

[0079] 3. Price Factor (PF);

[0080] Each factor is weighted and summed to obtain the node's total competition value. The formula is as follows:

[0081] CV i =w1·SoC i +w2·PL i +w3·PF i

[0082] Among them, CV i The contention value for node i; SoC i PL represents the remaining energy storage capacity of node i;i For the predicted load of node i; PF i The relative factor of the electricity price level at which node i calculates the competition value is 1 during peak hours, -1 during off-peak hours, and 0 during normal hours; w1, w2, and w3 are the weights of the corresponding factors, which are determined by regression analysis of historical data to determine the degree of influence of each factor on the system, and are dynamically adjusted through a feedback mechanism during actual operation.

[0083] Among them, PL, as the predicted load of the node, is constructed by extracting historical charging data and relative geographical location features of charging piles using the Variational Mode Decomposition (VMD) algorithm to build a dataset with spatiotemporal characteristics. A charging load prediction model for charging piles is built using the Spatiotemporal Graph Convolutional Network (ST-GCN) algorithm, and the Gray Wolf Optimization (GWO) algorithm is used to optimize the model parameters to improve model performance. Specific implementation steps are as follows: Figure 6 As shown:

[0084] The explanation for each step is as follows:

[0085] 1. Data Preprocessing: Data preprocessing consists of two stages: data collection and feature extraction.

[0086] Data collection: Collect historical charging data for each storage and charging unit, including charging amount, charging time, and relative geographical location for each time period.

[0087] Feature Extraction: Temporal and Relative Geographical Position (RGP) features of the charging data from charging piles are extracted. RGP features represent the relative geographical location of a charging unit within a network of charging stations. Assuming a network of charging units located in a gas station parking lot, the charging unit closest to the parking lot entrance should have the highest RGP feature value. Units farther from the entrance will have lower RGP feature values. This is because distance from the entrance directly affects the charging unit's load; units closer to the entrance have higher utilization rates.

[0088] 2. Variational Mode Decomposition (VMD):

[0089] Mode Decomposition: The collected historical charging data from charging stations is processed using VMD to decompose the original time series into several mode functions. Each mode function represents different frequency components and trends.

[0090] Spatiotemporal feature dataset construction: The decomposed modal functions are combined with relative geographical location features to form a dataset with spatiotemporal features.

[0091] 3. Construct a Spatiotemporal Graph Convolutional Network (ST-GCN) model:

[0092] Model architecture design:

[0093] Input layer: The input is a dataset with spatiotemporal features. The dimension of the input data is (T×N×F), where (T) is the time step, (N) is the number of nodes (i.e. the number of charging piles), and (F) is the feature dimension.

[0094] Temporal convolutional layer: One-dimensional convolution (1D-CNN) is used to extract temporal features, and the kernel size is (k t The output dimension is (T'×N×F'), where (T') is the length of the time step after convolution, and (F') is the output feature dimension.

[0095] Graph Convolutional Layer: Graph Convolution (GCN) is used to extract spatial features. A graph structure is constructed based on the relative geographical location of each storage unit. The output dimension of the graph convolutional layer is (T'×N×F”), where (F”) is the feature dimension after graph convolution.

[0096] Temporal convolutional layer: One-dimensional convolution is applied again to further extract temporal features, and the output dimension is (T”×N×F”’), where (T”) is the length of the time step after convolution, and (F”’) is the final feature dimension.

[0097] Fully connected layer: Flatten the features output by the convolutional layer and output the predicted value through the fully connected layer.

[0098] 4. Optimization using the Gray Wolf Algorithm (GWO):

[0099] Objective function definition: The objective function is defined as the mean square error (MSE) between the model prediction result and the actual charging load.

[0100] Parameter optimization: The parameters of the ST-GCN model are optimized using GWO. Initially, several gray wolf populations are set up, with each individual representing a set of model parameters. By iteratively updating the population positions, the optimal parameter combination is found to minimize the objective function value.

[0101] 5. Model Training and Validation:

[0102] Data segmentation: The constructed spatiotemporal feature dataset is divided into a training set and a validation set.

[0103] Model training: Train the ST-GCN model, optimize parameters using GWO, and continuously improve model performance through multiple iterations.

[0104] Model validation: Evaluate the model performance on the validation set to ensure that the model has good generalization ability.

[0105] 6. Model Deployment and Application:

[0106] Model Deployment: The trained ST-GCN model is deployed to the charging pile management system to achieve real-time charging load prediction.

[0107] Strategy optimization: Based on the prediction results, optimize the scheduling and management strategies of charging piles to improve charging efficiency and user experience.

[0108] 7. Performance Evaluation and Improvement:

[0109] Model evaluation: Regularly evaluate the model's performance in practical applications and collect feedback data for analysis.

[0110] Continuous optimization: Based on the evaluation results, the model is continuously optimized and improved to ensure that it maintains good predictive performance under different environments and conditions.

[0111] Through the above steps, this embodiment can efficiently predict the charging load of charging piles, provide a scientific basis for the management and scheduling of charging piles, and improve the utilization rate of charging piles and user satisfaction.

[0112] The specific application methods of the competition value in the two use cases are as follows:

[0113] When a storage and charging unit that has received external power input initiates an input power allocation request, power allocation is performed based on the contention value of the nodes. The specific steps are as follows:

[0114] 1. All nodes calculate and report the contention value (CV). i ;

[0115] 2. Storage and charging nodes with competition values ​​below a certain threshold are included in the input allocation candidate list. This is because during peak electricity price periods, the electricity price factor is positive when calculating the competition value, which increases the competition value of all storage and charging units. Conversely, during off-peak electricity price periods, the electricity price factor is negative when calculating the competition value, which decreases the competition value of all storage and charging units. By setting a reasonable threshold for the competition value of receiving input allocation, it is possible to maximize charging during off-peak periods and minimize charging during peak periods, thereby achieving the goal of peak shaving and valley filling.

[0116] 3. Based on the contention value, the input power P is allocated proportionally, with storage cells having higher priority in input allocation due to lower contention values. In one possible implementation, the number of cells (Np) receiving input is calculated based on the maximum acceptable input power (Pd) for a single node, i.e., Np = P / Pd. The result is rounded to obtain the number of cells receiving input, and these are sorted according to contention values, with the storage cells having the lowest required number receiving input allocation.

[0117] 4. After a certain time window (e.g., 10 minutes), the energy storage units that have received external power input will initiate a competition mechanism again to re-determine which energy storage unit will receive the input allocation. Since the energy storage unit with the lowest weight previously received the power input, its competition value will increase, causing a change in ranking. The energy input will then be obtained by the energy storage unit with the lowest new competition value.

[0118] When a node needs power assistance, the node that provides the assistance is determined based on the contention score. The specific steps are as follows:

[0119] The unit in need of assistance initiates a request, and all nodes calculate and report the contention value CV. i .

[0120] The unit that needs assistance sorts other units according to the size of the competition value, and selects the node with the highest competition value to provide assistance.

[0121] If the power of one node is insufficient to meet the demand, multiple nodes can be selected to provide assistance, and the assistance power can be allocated according to the competition value ratio.

[0122] The above methods effectively implement a power competition mechanism, ensuring that each node competes for power resources rationally based on its own circumstances and provides effective power assistance when needed. This not only improves system efficiency but also enhances system flexibility and reliability.

[0123] The proposed device-to-grid network (DGG) application in this embodiment was applied to a storage and charging unit cluster for performance testing. Daily usage data (excluding weekends and holidays) was collected over a three-month trial period. The full supercharging service rate and the average ratio of electricity provided to electricity cost for each strategy were summarized and calculated to obtain the following results: Figure 7 and Figure 8 The data shown illustrates that constructing a grid network using the implementation method described in this application can significantly improve the supercharging full service rate. However, compared to not using a grid network, the increased service rate also significantly increases the amount of electricity drawn from the grid during peak daytime hours, raising the average charging amount to electricity cost ratio and thus increasing service costs. However, by employing a neural network optimization strategy, through flexible control of grid power draw, a supercharging full service rate similar to that of a fixed strategy is achieved while significantly reducing the charging amount to electricity cost ratio and thus significantly lowering service costs.

[0124] An energy management system for a distributed energy storage and charging system includes:

[0125] The machine-to-machine network (MTN) construction module is used to build an MTN architecture that includes several storage and charging units;

[0126] The competition mechanism module is used to calculate the corresponding competition value based on the remaining energy storage capacity data and predicted load of each of the energy storage and charging units when there is an external power input or power assistance demand in the machine network architecture.

[0127] The assistance module is used to calculate the input distribution scheme of external power and the power assistance scheme between the storage and charging units based on the competition value corresponding to each of the storage and charging units.

[0128] An electronic device includes a memory and a processor, the memory storing a computer program, and the processor running the computer program to enable the electronic device to perform an energy management method for a distributed energy storage and charging system.

[0129] A computer-readable storage medium storing a computer program that, when executed by a processor, implements an energy management method for a distributed energy storage and charging system.

[0130] The above description is merely a preferred embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An energy management method for a distributed energy storage and charging system, characterized in that, include: Step 1: Construct a decentralized machine network architecture containing several storage and charging units; Step 2: When there is an external power input or power assistance requirement in the machine network architecture, calculate the corresponding competition value based on the remaining energy storage power data and predicted load of each of the energy storage and charging units; Step 3: Calculate the external power input allocation scheme and the power assistance scheme between the storage and charging units based on the competition value corresponding to each of the aforementioned storage and charging units, specifically including: External power input allocation scheme: When a storage and charging unit that has obtained external power input initiates an input power allocation request, the input power of external power is allocated proportionally based on the competition value corresponding to each candidate storage and charging unit. The storage and charging unit with the smaller competition value has priority in input allocation. After a certain time window, the storage and charging unit that has obtained external power input will initiate the competition mechanism again to re-determine the storage and charging unit that will receive the input allocation. Power assistance scheme between energy storage and charging units: When the energy storage and charging unit to be assisted initiates an assistance request, all nodes calculate and report the competition value, and sort other units according to the size of the competition value, and select the node with the highest competition value to provide assistance.

2. The energy management method according to claim 1, characterized in that, The calculation of the corresponding competitive value based on the remaining energy storage power data and predicted load of each energy storage and charging unit specifically includes: weighted summation of the remaining energy storage power data, electricity price level data and predicted load data of each energy storage and charging unit to obtain the corresponding competitive value; wherein the weight of each factor is determined by regression analysis of historical data and dynamically adjusted through a feedback mechanism during actual operation.

3. The energy management method according to claim 2, characterized in that, The predicted load is obtained in the following manner: Acquire training data, which includes historical charging data, relative geographical location data, and load data for each of the storage and charging units; A load prediction model is constructed based on a spatiotemporal graph convolutional network; Based on the training data, the load prediction model is trained and its parameters are optimized using the Grey Wolf optimization algorithm to obtain the optimized load prediction model. The load forecasting of each of the storage and charging units is performed based on the optimized load forecasting model, specifically including: extracting the time-series features and relative geographical location features of the storage and charging units; The temporal features are subjected to mode decomposition, and the decomposed mode functions are combined with relative geographical location features to construct a feature dataset with spatiotemporal characteristics. The load prediction model includes a first temporal convolutional layer, a graph convolutional layer, a second temporal convolutional layer, and a fully connected prediction layer connected in sequence. The feature dataset is input into the load prediction model, and features are extracted through each temporal convolutional layer and the graph convolutional layer. The extracted features are input into the fully connected prediction layer for classification and prediction to obtain the predicted load data.

4. The energy management method according to claim 1, characterized in that, When external electrical energy is input, the number of units Np to receive the input is determined based on the total external input power P and the maximum input power Pd that a single storage and charging unit can accept, i.e., Np = P / Pd, and the result is rounded up to get the number of units to receive the input.

5. An energy management system for a distributed energy storage and charging system, using the method as described in any one of claims 1-4, characterized in that, include: The machine network construction module is used to build a decentralized machine network architecture that includes several storage and charging units; The competition mechanism module is used to calculate the corresponding competition value based on the remaining energy storage capacity data and predicted load of each of the energy storage and charging units when there is an external power input or power assistance demand in the machine network architecture. The assistance module is used to calculate the input distribution scheme of external power and the power assistance scheme between the storage and charging units based on the competition value corresponding to each of the storage and charging units.

6. An electronic device, characterized in that, The device includes a memory and a processor, the memory being used to store a computer program, and the processor running the computer program to cause the electronic device to perform the energy management method of the distributed energy storage and charging system according to any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, It stores a computer program, which, when executed by a processor, implements the energy management method of the distributed energy storage and charging system as described in any one of claims 1-4.

Images