Vehicle-grid cooperative charging pile scheduling method and system based on retired battery energy storage

Abstract

The application discloses a vehicle-network cooperative charging pile scheduling method and system based on retired battery energy storage, and relates to the technical field of energy storage scheduling. The method comprises the following steps: S1, real-time acquisition of vehicle-network energy storage scheduling data for data preprocessing; S2, real-time judgment of energy storage unit scheduling priority, construction of energy storage charging and discharging strategy dataset; S3, construction of a price prediction model, reception of the energy storage charging and discharging strategy dataset, determination of the output of the energy storage charging and discharging power, and execution of the charging and discharging strategy; S4, evaluation of the charging and discharging power distribution of a single battery module during the execution of the charging and discharging strategy, differential charging and discharging scheduling of each battery module, and optimization of the charging and discharging strategy and the charging and discharging power distribution strategy. The method solves the problems of the existing charging pile system, such as lack of flexible energy storage and intelligent scheduling, difficulty in coping with price and load fluctuations, low utilization rate of retired power batteries, resource waste, battery safety, and insufficient power grid interaction capability.

Description

Technical Field

[0001] This invention relates to the field of energy storage scheduling technology, specifically to a vehicle-grid collaborative charging pile scheduling method and system based on retired battery energy storage. Background Technology

[0002] As the application of electric vehicles expands, charging piles are gradually evolving from single energy supply devices into integrated nodes with energy regulation and storage functions. Meanwhile, retired power batteries, due to their remaining usable energy storage characteristics, are being gradually introduced into charging facilities as distributed energy storage units to participate in vehicle-grid energy interaction. Against this backdrop, technologies related to the scheduling and control of charging demand, grid operation status, and the coordinated operation of energy storage units are continuously developing and attracting attention.

[0003] For example, the invention patent with announcement number CN120180250B discloses a method and system for deep sorting of retired batteries, relating to the field of battery energy storage technology. The method includes the following steps: generating multiple original feature curves of different types based on sampled data; performing two derivatives on the multiple original feature curves to obtain multiple velocity feature curves and multiple acceleration feature curves, obtaining statistical indices for each curve to obtain multiple baseline features, multiple velocity features, and multiple acceleration features; performing preliminary clustering on the multiple baseline features to obtain initial grouping results; obtaining the first Euclidean distance and the second Euclidean distance for each retired battery, and re-sorting the retired battery based on the first Euclidean distance and the second Euclidean distance to obtain the optimal grouping results. This invention significantly expands the feature space through multi-level feature extraction, thereby capturing more complex performance characteristics of the battery to accurately characterize the aging degree and future performance of the battery, achieving precise battery sorting.

[0004] For example, invention patent CN120087036A discloses a battery selection method, device, and energy storage plant. The method includes: determining the selection direction of the retired battery based on its state information and an application scenario selection model; constructing the application scenario selection model based on grid regulation scenarios; and determining the assembly strategy of the retired battery based on its selection direction and the energy storage parameters of the energy storage plant, provided the selection direction meets the application requirements. This method enables the application of retired batteries in the grid regulation scenarios to which they are applicable, and allows the construction of energy storage cores suitable for various grid regulation scenarios based on retired batteries.

[0005] However, in the process of implementing the inventive technical solution in the embodiments of this application, it was found that the above-mentioned technology has at least the following technical problems:

[0006] The existing charging pile system lacks flexible energy storage and intelligent scheduling capabilities, making it unable to cope with electricity price fluctuations and grid load changes. At the same time, retired power batteries are not used efficiently, resulting in serious resource waste and insufficient battery safety and grid interaction capabilities.

[0007] Therefore, in order to address the above issues, there is an urgent need for a vehicle-to-grid (V2G) collaborative charging pile scheduling method and system based on retired battery energy storage. Summary of the Invention

[0008] Technical problems to be solved

[0009] To address the shortcomings of existing technologies, this invention provides a vehicle-grid collaborative charging pile scheduling method and system based on retired battery energy storage. This solves the problems of resource waste, battery safety, and insufficient grid interaction capabilities caused by the lack of flexible energy storage and intelligent scheduling in existing charging pile systems, which make it difficult to cope with electricity price and load fluctuations. At the same time, the low utilization rate of retired power batteries leads to the problem of low grid interaction capabilities.

[0010] Technical solution

[0011] To achieve the above objectives, the present invention provides the following technical solution: a vehicle-grid collaborative charging pile scheduling method based on retired battery energy storage, comprising the following steps: S1, real-time collection of vehicle-grid energy storage scheduling data, data preprocessing and storage of the vehicle-grid energy storage scheduling data; S2, real-time determination of energy storage unit scheduling priority based on the preprocessed vehicle-grid energy storage scheduling data, and construction of energy storage charging and discharging strategy dataset based on the energy storage unit scheduling priority; S3, construction of electricity price prediction model based on historical vehicle-grid energy storage scheduling data, receiving energy storage charging and discharging strategy dataset, determining the output of energy storage charging and discharging power in combination with the electricity price prediction model, and executing the charging and discharging strategy; S4, during the execution of the charging and discharging strategy, evaluating the charging and discharging power allocation of a single battery module based on the preprocessed vehicle-grid energy storage scheduling data, performing differentiated charging and discharging scheduling for each battery module, and simultaneously optimizing the charging and discharging strategy and the charging and discharging power allocation strategy.

[0012] Furthermore, the specific process of real-time acquisition of vehicle-to-grid energy storage scheduling data, data preprocessing, and storage is as follows: Real-time acquisition of vehicle-to-grid energy storage scheduling data involves: acquiring charging request power through real-time communication between charging piles and the vehicle-to-grid platform; acquiring the maximum energy storage power of the energy storage unit through feedback from the bidirectional converter and battery management system; acquiring the grid load in real-time through smart meters and retrieving the maximum rated capacity parameters from the distribution system to acquire the maximum grid load; acquiring electricity prices in real-time through API integration with the power operation platform; obtaining the number of battery modules in the energy storage unit, and acquiring the overall SOC, overall SOH, overall temperature, and overall current of the energy storage unit, as well as the individual SOC of each battery module, through the battery management system. Data includes OC, SOH, temperature, and current of individual cells; meteorological information is collected via a meteorological API interface; the connection status signals between the charging gun and the vehicle are collected in real time; the integrity of the vehicle-to-grid energy storage scheduling data is checked, removing null values, duplicates, data exceeding physical limits, and data due to equipment failure, and smoothing data jumps using a sliding window method; the vehicle-to-grid energy storage scheduling data is standardized and dimensionless normalized; the data is aligned by timestamps, and all relevant data snapshots at the same time are grouped uniformly; a backup acquisition channel is switched in case of sudden data loss or communication interruption; a vehicle-to-grid energy storage time-series database is established, and the raw and pre-processed vehicle-to-grid energy storage scheduling data are stored in the vehicle-to-grid energy storage time-series database in real time.

[0013] Furthermore, the specific process for determining the scheduling priority of energy storage units in real time based on the preprocessed vehicle-to-grid energy storage scheduling data is as follows: Obtain the current charging request power and the maximum energy storage power, and calculate the ratio of the current charging request power to the maximum energy storage power to obtain the demand load value; obtain the current grid load and the maximum grid load, and calculate the ratio of the current grid load to the maximum grid load to obtain the grid pressure value; based on a sliding time window, obtain electricity price data, and calculate the mean and standard deviation to obtain the electricity price mean and standard deviation; calculate the absolute difference between the current electricity price and the electricity price mean, and divide it by the sum of the electricity price standard deviation and the minimum constant value to obtain the electricity price fluctuation value; subtract the electricity price fluctuation value from the sum of the demand load value and the grid pressure value to obtain the original priority value, and perform a hyperbolic tangent function operation on the original priority value to obtain the energy storage scheduling priority.

[0014] Furthermore, the specific process of constructing an energy storage charging and discharging strategy dataset based on the energy storage unit scheduling priority is as follows: The energy storage scheduling priority is written into the vehicle-to-grid energy storage time-series database and compared with the priority threshold. When no vehicle is connected to a charging pile, or the energy storage scheduling priority is less than or equal to the priority threshold, the energy storage unit enters a low-power standby state, does not perform charging or discharging operations, and supplies power to the vehicle and replenishes the energy storage battery through the grid, maintaining routine monitoring of the vehicle-to-grid energy storage scheduling data. When a vehicle is detected to be connected to a charging pile, and the energy storage scheduling priority is greater than the priority threshold, power is supplied through the energy storage unit, and the energy storage scheduling priority and the corresponding vehicle-to-grid energy storage scheduling data are combined to construct an energy storage charging and discharging strategy dataset, which is then output, triggering the next process to formulate the charging and discharging strategy. If the change in energy storage scheduling priority is detected to exceed the mutation threshold, a power limiting command for charging and discharging is generated to reduce the load in advance and activate the active cooling device. Simultaneously, the priority threshold is adjusted based on the historical energy storage scheduling priority distribution and the average energy storage scheduling priority using a sliding window self-learning algorithm.

[0015] Furthermore, the specific process of constructing an electricity price prediction model based on historical vehicle-grid energy storage scheduling data is as follows: acquire historical electricity price, grid load data and corresponding meteorological information, and statistically analyze the mean and standard deviation of historical electricity price and grid load data. Combine historical electricity price, grid load data, meteorological information and the mean and standard deviation of historical electricity price and grid load data to construct an electricity price time series feature set as input; use a long short-term memory network algorithm to train the electricity price time series feature set, and divide the training set and test set during training. Combine cross-validation and parameter tuning to construct an electricity price prediction model and output the predicted electricity price value for future times.

[0016] Further, the specific process of receiving the energy storage charging and discharging strategy dataset and determining the output of energy storage charging and discharging power in conjunction with the electricity price prediction model is as follows: Receive the energy storage charging and discharging strategy dataset, obtain the predicted electricity price and the current electricity price, divide the difference between the predicted and current electricity price by the sum of the absolute value of the predicted electricity price and the minimum constant value to obtain the time-series electricity price change term; based on the sliding time window, obtain the overall SOC data, and filter out the median to obtain the overall SOC median, while simultaneously calculating the overall SOC standard deviation; divide the difference between the current overall SOC and the overall SOC median by the sum of the overall SOC standard deviation and the minimum constant value to obtain the overall SOC deviation value; use the overall SOC deviation value as an exponent and perform natural exponential calculation to obtain the SOC deviation attenuation factor; add the SOC deviation attenuation factor to a constant and take the reciprocal to obtain the SOC adjustment term; obtain the maximum energy storage power and the overall SOH, and multiply the time-series electricity price change term, the maximum energy storage power, the SOC adjustment term, and the overall SOH to obtain the energy storage power decision value.

[0017] Furthermore, the specific process of implementing the charging and discharging strategy is as follows: The energy storage power decision value is written into the vehicle-to-grid energy storage time-series database; simultaneously, the upper and lower limits of the energy storage power decision value are checked, and a comprehensive safety check is performed with the overall SOH and overall SOC states; if the overall SOC, overall SOH, overall temperature, and overall current exceed the corresponding SOC safety threshold, SOH safety threshold, temperature safety threshold, and current safety threshold, the output of the energy storage charging and discharging power is limited and energy storage scheduling is suspended; the sign of the energy storage power decision value is determined, and the energy storage power decision value is... With charging threshold Discharge threshold and standby threshold The intelligent scheduling controller performs a comparison, generates instructions for executing the charging and discharging strategy, and executes them through the DC module controller; when > When the system is in charging mode, the charging process is initiated. Based on the energy storage power decision value, a charging command is issued. Coordinated by the charging pile and bidirectional converter, energy from the grid flows into the energy storage unit and the vehicle battery, and charging safety monitoring is activated. < When the system is in discharge mode, the discharge process is initiated. Based on the energy storage power decision value, a discharge command is issued, and the bidirectional converter delivers energy from the energy storage unit to the grid and external loads. Discharge safety monitoring is also activated. ≤ or ≥ When the system is in standby mode, the intelligent dispatch controller suspends all active charging and discharging operations of the energy storage unit, and the DC module controller shuts down the corresponding current channel, maintaining only status monitoring and self-testing. When the grid is detected to be unavailable, the system switches to independent charging and discharging and starts emergency power supply.

[0018] Furthermore, during the execution of the charging and discharging strategy, the specific process of evaluating the charging and discharging power allocation of a single battery module based on the preprocessed vehicle-to-grid energy storage scheduling data is as follows: During the execution of the charging and discharging strategy, the energy storage power decision value and the single-cell SOH and single-cell temperature data of each battery module are obtained. At the same time, based on the sliding time window, the single-cell temperature data is statistically analyzed and the median is selected as the target temperature value. The current single-cell temperature of battery module i is subtracted from the target temperature value of battery module i, and the absolute value is taken to obtain the absolute temperature deviation value. The current single-cell SOH of battery module i is subtracted from the constant and the absolute value is taken to obtain the absolute health deviation value. The absolute temperature deviation value and the absolute health deviation value are added together and the opposite number is taken as the exponent, and the natural exponentiation operation is performed to obtain the single-cell allocation weight of battery module i. Based on the number of battery modules, the single-cell allocation weights of all battery modules are summed to obtain the overall allocation weight. The ratio of the single-cell allocation weight to the overall allocation weight is multiplied by the current energy storage power decision value to obtain the single-cell differentiated power allocation value of battery module i.

[0019] Furthermore, the specific process of performing differentiated charging and discharging scheduling for each battery module, while optimizing the charging and discharging strategy and the charging and discharging power allocation strategy, is as follows: The differentiated power allocation value of each cell is written into the vehicle-to-grid energy storage time-series database. The intelligent scheduling controller generates power allocation instructions based on the differentiated power allocation value of each cell and sends them to the DC module controller in real time. The DC module controller adjusts the cell current of each battery module according to the power allocation instructions, controlling the charging and discharging of each battery module. Simultaneously, the charging and discharging status, cell temperature, and cell SOH of each battery module are transmitted back in real time during the power allocation process. If the continuous rise time of a cell temperature and the continuous fall time of a cell SOH of a battery module exceed the maximum allowable threshold, the differentiated power allocation value of the cell of the battery module is reduced, the connection with the main circuit is disconnected, and the active balancing circuit is activated to transfer, replenish, and eliminate energy differences in the battery modules, while simultaneously correcting the power allocation instructions. The energy storage power decision value and the differentiated power allocation value of each cell are monitored in real time throughout the entire process. Combined with historical vehicle-to-grid energy storage scheduling data, a genetic algorithm is used to optimize the charging and discharging strategy and the power allocation strategy.

[0020] The second aspect of this invention provides a vehicle-grid collaborative charging pile scheduling system based on retired battery energy storage, comprising: a multi-source data acquisition and preprocessing module for real-time acquisition of vehicle-grid energy storage scheduling data, preprocessing and storing the vehicle-grid energy storage scheduling data; an energy storage scheduling judgment and decision module for real-time judgment of energy storage unit scheduling priority based on the preprocessed vehicle-grid energy storage scheduling data, and constructing an energy storage charging and discharging strategy dataset based on the energy storage unit scheduling priority; a charging and discharging decision formulation and execution module for constructing an electricity price prediction model based on historical vehicle-grid energy storage scheduling data, receiving the energy storage charging and discharging strategy dataset, determining the output of energy storage charging and discharging power in conjunction with the electricity price prediction model, and executing the charging and discharging strategy; and a health temperature control and safety protection module for evaluating the charging and discharging power allocation of a single battery module based on the preprocessed vehicle-grid energy storage scheduling data during the execution of the charging and discharging strategy, performing differentiated charging and discharging scheduling for each battery module, and optimizing the charging and discharging strategy and the charging and discharging power allocation strategy.

[0021] Beneficial effects

[0022] The present invention has the following beneficial effects:

[0023] (1) This invention integrates multi-source heterogeneous data such as charging requests, grid load, electricity price, meteorology, SOC, SOH, temperature and current to build a unified data time-series archiving and preprocessing mechanism, providing accurate and comprehensive dynamic data support for vehicle-grid energy storage scheduling, and realizing a scheduling closed loop of intelligent perception and high-quality data-driven throughout the entire process.

[0024] (2) This invention, through priority criteria and electricity price prediction model, can dynamically determine when the energy storage unit enters charging, discharging and standby states, and flexibly respond to electricity price fluctuations, grid load changes and vehicle energy demand. Through sliding window and self-learning algorithm, the threshold and power strategy are continuously optimized to achieve a unified improvement in safety and energy utilization.

[0025] (3) Based on the health parameters of single cell SOH and temperature, this invention utilizes differentiated power allocation weights to achieve precise charge and discharge control for each battery module. For continuously abnormal single cells, derating, disconnection, energy balancing, and thermal control protection are implemented, significantly improving the health management level and operational safety of energy storage, and extending the secondary utilization life of retired batteries.

[0026] (4) This invention supports closed-loop data archiving and continuous monitoring of the entire process of energy storage scheduling strategy and power allocation. It can combine intelligent optimization algorithms to upgrade parameters and strategies through self-evolution, continuously improve the adaptive and optimization capabilities in complex operating scenarios, and promote the upgrading of energy storage systems towards intelligence, optimization and sustainable development.

[0027] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0028] Figure 1 The flowchart shows the vehicle-grid collaborative charging pile scheduling method based on retired battery energy storage.

[0029] Figure 2 This is a module diagram of a vehicle-grid collaborative charging pile scheduling system based on retired battery energy storage;

[0030] Figure 3 This is a schematic diagram illustrating the working principle of vehicle-grid collaborative charging pile scheduling based on retired battery energy storage.

[0031] Figure 4 This is a diagram showing the distribution of multiple influencing factors in energy storage power decision-making.

[0032] In the diagram, 1 is the energy storage unit; 2 is the power grid; 3 is the bidirectional converter; 4 is the charging pile; 5 is the vehicle battery; 6 is the battery management system; 7 is the intelligent dispatch controller; and 8 is the DC module controller. Detailed Implementation

[0033] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. As those skilled in the art will understand, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0034] Please see Figures 1-4 This invention provides a technical solution: a method and system for scheduling vehicle-to-grid (V2G) charging piles based on retired battery energy storage, such as... Figure 1 As shown, the process includes the following steps: S1, real-time acquisition of vehicle-to-grid energy storage scheduling data, data preprocessing and storage of the vehicle-to-grid energy storage scheduling data; S2, real-time determination of energy storage unit scheduling priority based on the preprocessed vehicle-to-grid energy storage scheduling data, and construction of energy storage charging and discharging strategy dataset based on the energy storage unit scheduling priority; S3, construction of electricity price prediction model based on historical vehicle-to-grid energy storage scheduling data, receiving energy storage charging and discharging strategy dataset, determining the output of energy storage charging and discharging power in combination with the electricity price prediction model, and executing the charging and discharging strategy; S4, during the execution of the charging and discharging strategy, evaluation of the charging and discharging power allocation of a single battery module based on the preprocessed vehicle-to-grid energy storage scheduling data, differentiated charging and discharging scheduling for each battery module, and optimization of the charging and discharging strategy and charging and discharging power allocation strategy.

[0035] Specifically, the process of real-time acquisition of vehicle-to-grid energy storage scheduling data, data preprocessing, and storage of this data is as follows: Real-time acquisition of vehicle-to-grid energy storage scheduling data involves: collecting charging request power through real-time communication between charging pile 4 and the vehicle-to-grid platform; collecting the maximum energy storage power of energy storage unit 1 through feedback from bidirectional converter 3 and battery management system 6; collecting grid load in real-time through smart meters and retrieving maximum rated capacity parameters from the distribution system to collect the maximum grid load; collecting electricity prices in real-time through API integration with the power operation platform; obtaining the number of battery modules in energy storage unit 1 and collecting the overall SO of energy storage unit 1 through battery management system 6. C. Overall SOH, overall temperature, and overall current, as well as individual cell SOC, individual cell SOH, individual cell temperature, and individual cell current data for each battery module; meteorological information, including temperature, humidity, air pressure, and wind speed, is collected via a meteorological API interface to assist in determining the energy storage operating conditions and safety boundaries; real-time acquisition of the charging gun's connection status signals to the vehicle ensures timely and complete responses to charging requests, dynamically monitoring vehicle connection and disconnection processes to guarantee accurate execution of charging and discharging commands; integrity checks are performed on vehicle-to-grid energy storage scheduling data, removing null values, duplicates, data exceeding physical limits, and data related to equipment failures. To address data disruptions, a multi-source redundancy verification mechanism is introduced, utilizing cross-validation and correction based on multi-sensor data acquisition to improve data accuracy and reliability. A sliding window method is employed to smooth data fluctuations; a set-length moving time window is used to perform mean, median, and variance statistical processing on continuously acquired data, eliminating abnormal fluctuations and short-term jitter, effectively suppressing erroneous responses from the acquisition system to sudden anomalies. Simultaneously, vehicle-to-grid energy storage scheduling data undergoes standardization and dimensionless normalization to facilitate subsequent multi-dimensional, multi-modal data fusion, feature extraction, and model input, ensuring that data with different dimensions and physical meanings can be processed and compared within a unified space. Furthermore, vehicle-to-grid energy storage... It can schedule data to be aligned by timestamp, and group all relevant data snapshots at the same time in a unified manner to achieve time synchronization of multi-channel asynchronous acquisition signals, eliminate time sequence errors caused by different sampling periods of different devices, and improve the accuracy of subsequent analysis and decision-making; it can switch to backup acquisition channels in case of sudden data loss and communication interruption failure, that is, when the main acquisition link is abnormal, it switches to the backup link to ensure uninterrupted and complete closed-loop data acquisition, effectively avoiding data loss caused by single point of failure; it establishes a vehicle-to-grid energy storage time series database, and stores the raw and pre-processed vehicle-to-grid energy storage scheduling data into the vehicle-to-grid energy storage time series database in real time, supporting high-frequency writing and efficient indexing of historical data.

[0036] like Figure 3The diagram illustrates the working principle of a vehicle-grid collaborative charging pile scheduling system based on retired battery energy storage. It showcases the overall architecture of the retired battery energy storage charging pile system for vehicle-grid collaboration, including key components such as energy storage unit 1, power grid 2, bidirectional converter 3, charging pile 4, on-board battery 5, battery management system 6, intelligent dispatch controller 7, and DC module controller 8. Energy storage unit 1 and on-board battery 5 can achieve bidirectional energy flow with power grid 2 through bidirectional converter 3, supporting multiple scenarios including grid-connected, off-grid, and emergency use. Battery management system 6 is responsible for monitoring the health parameters of each module, including SOC, SOH, temperature, and current. DC module controller 8 precisely executes power allocation and safety protection for each battery module. Intelligent dispatch controller 7 performs real-time status acquisition, scheduling decisions, and strategy distribution for the entire system, maintaining efficient data and signal linkage with all components to achieve collaborative optimization and intelligent scheduling of vehicle-grid, energy storage, and charging, improving flexibility, safety, and operational efficiency.

[0037] This implementation plan enables multi-source, high-frequency real-time acquisition and full-process data preprocessing of vehicle-to-grid (V2G) energy storage scheduling data. By uniformly integrating and verifying the integrity of the V2G energy storage scheduling data, removing anomalies, smoothing with sliding windows, standardizing and dimensionless normalizing the data, and aligning it with timestamps, the accuracy, timeliness, and consistency of the V2G energy storage scheduling data can be effectively improved. This provides a solid data foundation for subsequent core functional modules such as energy storage unit scheduling priority determination, charging and discharging strategy formulation, health assessment, and safety protection, helping to improve the system's intelligent perception capabilities and collaborative optimization level, achieving efficient, reliable, and intelligent operation.

[0038] Specifically, the process of determining the scheduling priority of energy storage units in real time based on preprocessed vehicle-to-grid energy storage scheduling data is as follows: First, obtain the current charging request power and the maximum energy storage power, calculate the ratio of the current charging request power to the maximum energy storage power to obtain the demand load value, dynamically reflecting the current charging demand intensity of vehicles for energy storage unit 1. Second, obtain the current grid load and the maximum grid load, calculate the ratio of the current grid load to the maximum grid load to obtain the grid pressure value, and characterize the load level and redundancy space of the grid operation in real time, which helps to flexibly allocate resources under multi-objective constraints and achieve load balance and supply-demand matching. Third, based on a sliding time window, obtain electricity price data, and calculate the mean and standard deviation to obtain the electricity price mean and standard deviation, reflecting the dynamic trend and fluctuation characteristics of electricity prices in real time. Calculate the absolute difference between the current electricity price and the electricity price mean, and divide it by the sum of the electricity price standard deviation and the minimum constant value to obtain the electricity price fluctuation value, measuring the position and fluctuation range of the current electricity price within the historical fluctuation range. Fourth, subtract the electricity price fluctuation value from the sum of the demand load value and the grid pressure value to obtain the original priority value, and perform a hyperbolic tangent function operation on the original priority value to obtain the energy storage scheduling priority. A hyperbolic tangent function (tanh) is used to normalize the original priority values ​​using a nonlinear mapping, thus limiting the final energy storage scheduling priority to the interval [-1, 1]. This avoids the impact of extreme outliers on subsequent scheduling and improves the robustness and sensitivity of scheduling decisions.

[0039] The specific formula for energy storage scheduling priority is as follows:

[0040] ;

[0041] In the formula, This indicates the priority of energy storage scheduling, used to determine whether energy storage scheduling should be initiated at the current moment, and utilizes the hyperbolic tangent function. All original priority values ​​are limited to the range of [-1, 1] to eliminate the impact of extreme outliers on decision-making and ensure robustness and stability. This indicates the current charging power request, which refers to the actual charging power required by the vehicle at this moment, reflecting the strength of the current charging demand. This indicates the maximum energy storage power, reflecting the maximum charge and discharge power that can be output, and is used to standardize current demand. This indicates the current grid load, reflects the overall pressure on the grid, and serves as a direct indicator of whether energy storage needs to help reduce the load. This represents the maximum load on the power grid and is used to standardize the current load into a relative pressure index. Indicates the current electricity price; This represents a very small constant value, with a value of 0.01. This represents the average electricity price, serving as a baseline for reference when electricity prices fluctuate. This represents the standard deviation of electricity prices. Standardizing the deviation of electricity prices makes it easier to add or subtract from other normalization factors. Adding a very small constant value to the denominator prevents division by zero errors when there are extreme fluctuations or no fluctuations.

[0042] This implementation scheme constructs a scientifically sound method for calculating demand load, grid pressure, and electricity price fluctuations by real-time collection and normalization of multi-dimensional parameters such as charging request power, maximum energy storage power, grid load, maximum grid load, and electricity price data. A sliding time window is used to extract the average and standard deviation of electricity prices, dynamically reflecting market fluctuation trends and effectively mitigating the impact of abnormal electricity prices on decision-making. The hyperbolic tangent function is used to process the original priority values, achieving adaptive and smooth output of energy storage unit scheduling priorities. This ensures that scheduling instructions are both timely and avoid execution risks caused by drastic fluctuations. Overall, this improves the intelligent sensing capabilities and adaptive optimization level of vehicle-grid energy storage under dynamic multi-source load conditions.

[0043] Specifically, the process of constructing an energy storage charging and discharging strategy dataset based on the energy storage unit scheduling priority is as follows: The energy storage scheduling priority is written into the vehicle-to-grid energy storage time-series database and compared with the priority threshold. When no vehicle is connected to charging pile 4, or the energy storage scheduling priority is less than or equal to the priority threshold, energy storage unit 1 enters a low-power standby state and does not perform charging or discharging operations. It supplies power to the vehicle and replenishes the energy storage battery through the grid 2, maintaining routine monitoring of the vehicle-to-grid energy storage scheduling data. When a vehicle is detected to be connected to charging pile 4 and the energy storage scheduling priority is greater than the priority threshold, power is supplied through energy storage unit 1. The energy storage scheduling priority and the corresponding vehicle-to-grid energy storage scheduling data are combined to construct an energy storage charging and discharging strategy dataset and output. This provides a data foundation for subsequent hierarchical regulation and refined power allocation, triggering the next process to formulate a charging and discharging strategy. If the change in energy storage scheduling priority is detected to exceed the sudden change threshold, a power limiting command is generated to reduce the load in advance and activate the active cooling device, including multiple temperature control methods such as air cooling and liquid cooling. To prevent the risk of thermal runaway and achieve dynamic protection of the operational safety of energy storage unit 1; at the same time, the priority threshold is adjusted according to the historical energy storage scheduling priority distribution and the average energy storage scheduling priority through a sliding window self-learning algorithm.

[0044] In this implementation scheme, by writing the energy storage scheduling priority into the vehicle-to-grid energy storage time-series database in real time and dynamically determining the working status of energy storage unit 1 based on the priority threshold, intelligent low-power standby management and precise charging and discharging strategy triggering of energy storage unit 1 are achieved. This effectively reduces energy loss caused by ineffective charging and discharging, and improves the flexibility and safety of response. Simultaneously, by monitoring changes in energy storage scheduling priority and generating power limiting commands, triggering load reduction and active cooling measures in a timely manner for sudden changes, fault prevention and extreme condition adaptability are significantly enhanced. A sliding window self-learning algorithm is adopted, combined with the historical energy storage scheduling priority distribution and its mean, to achieve intelligent dynamic optimization of the priority threshold, improving the adaptability and intelligence level of strategy switching.

[0045] Specifically, the process of constructing an electricity price prediction model based on historical vehicle-grid energy storage scheduling data is as follows: Historical electricity prices, grid load data, and corresponding meteorological information are acquired, and the mean and standard deviation of historical electricity prices and grid load data are statistically analyzed. The mean and standard deviation reflect the central tendency and dispersion of the data, which can help the model better capture time-series characteristics. A time-series feature set of electricity prices is constructed by combining historical electricity prices, grid load data, meteorological information, and the mean and standard deviation of historical electricity prices and grid load data as input. The time-series feature set of electricity prices is trained using a Long Short-Term Memory (LSTM) network algorithm. During training, training and testing sets are divided, and cross-validation and parameter tuning are combined. Cross-validation effectively prevents model overfitting and improves generalization ability. Parameter tuning includes adjusting the number of layers, hidden units, and learning rate hyperparameters of the LSTM to obtain the best prediction performance. Finally, an electricity price prediction model is constructed, outputting the predicted electricity price value for future times, providing a high-precision reference for electricity price trends for energy storage charging and discharging strategies and scheduling decisions.

[0046] In this implementation plan, regarding electricity price forecasting, a scientific time-series feature set of electricity prices was constructed by combining multi-source data such as historical electricity prices, grid load data, and meteorological information. This set was then trained and optimized using a Long Short-Term Memory (LSTM) network algorithm, achieving high-precision prediction of future electricity prices. This not only significantly improves the foresight of vehicle-grid coordinated energy storage scheduling but also provides reliable data support for subsequent energy storage charging and discharging decisions, contributing to a reduction in overall operational risk.

[0047] Specifically, the process of receiving the energy storage charging and discharging strategy dataset and determining the output of energy storage charging and discharging power in conjunction with the electricity price prediction model is as follows: The energy storage charging and discharging strategy dataset is received, the predicted electricity price and the current electricity price are obtained, and the difference between the predicted and current electricity prices is divided by the sum of the absolute value of the predicted electricity price and a minimum constant value to obtain the time-series electricity price change term. The minimum constant value is set to 0.01 to avoid the denominator being zero and to improve the numerical stability of the calculation process. The time-series electricity price change term reflects the dynamic relative change between the future and current electricity prices and is an important input for the economic dispatch of energy storage. Based on a sliding time window, overall SOC data is acquired, and the median is selected to obtain the overall SOC median. Simultaneously, the overall SOC standard deviation is calculated. The difference between the current overall SOC and the overall SOC median is divided by the sum of the overall SOC standard deviation and the minimum constant value to obtain the overall SOC deviation. This deviation is then used as an exponent for natural exponential calculation to obtain the SOC deviation attenuation factor. The SOC deviation attenuation factor is added to a constant, and the reciprocal is taken to obtain the SOC adjustment term. The overall SOC deviation is used to quantify the deviation between the current charge level of energy storage unit 1 and its historical median health status. Natural exponential calculation and normalization ensure that the SOC adjustment term always changes within a reasonable range, thereby automatically suppressing charging and discharging power under extreme SOC conditions, improving safety and health lifespan. The maximum energy storage power and the overall SOH are obtained. The maximum energy storage power reflects the instantaneous output capability of the current energy storage unit 1, and the overall SOH is used to reflect the life status and sustainable operation capability of the energy storage battery. The energy storage power decision value is obtained by multiplying the time-series electricity price change term, the maximum energy storage power, the SOC adjustment term and the overall SOH.

[0048] The specific formula for the energy storage power decision value is as follows:

[0049] ;

[0050] In the formula, This represents the energy storage power decision value, used to determine the optimal charging and discharging power command for energy storage at the current moment, enabling simultaneous response to multiple objectives such as electricity price trends, charge safety, and equipment health. This represents the electricity price forecast, reflecting the electricity price trend over a subsequent period, and is an important basis for judging the current charging and discharging trend; This indicates the current electricity price, reflecting the current electricity price level in real time, and is the basic operating condition input for the responsive dispatching of Power Grid 2. This represents a very small constant value, with a value of 0.01. This indicates the maximum energy storage power, ensuring that the actual output does not exceed the equipment's safety limits and providing physical boundary conditions for scheduling strategies. It represents the current overall SOC, which indicates the battery's current energy reserve and is a key safety parameter for adjusting charging and discharging capabilities and ensuring operation within limits. This represents the overall SOC median, serving as a reference baseline for safe charging and discharging to prevent extreme deviations in the overall SOC from causing abnormal scheduling. It represents the overall SOC standard deviation, describes the magnitude of overall SOC fluctuation, and is used to adaptively adjust the control sensitivity. The greater the fluctuation, the more moderate the adjustment factor, avoiding high-frequency switching and oscillation. It represents the overall SOH (State of Health), which dynamically reflects the battery's health status. The higher the health status, the greater the usable power. When the health status decreases, the rating is reduced to extend the lifespan and ensure safety. This represents the time-series electricity price change term, which measures the trend of electricity price changes relative to the current price over a future period. If the future electricity price is higher than the current price, the result is positive, indicating a tendency to charge now and discharge in the future to achieve peak shaving and valley filling; conversely, the result is negative, indicating a tendency to discharge now. This represents the SOC adjustment term, where, This indicates that the SOC deviates from the decay factor. This represents the overall SOC deviation value. When the overall SOC deviation value decreases, the SOC adjustment term becomes larger, enhancing the current power output. When the overall SOC deviation value increases, the SOC adjustment term becomes smaller, weakening the current power output. This achieves dynamic perception and global constraint of the overall state of charge of energy storage unit 1, preventing the SOC from deviating excessively from the target value and improving the balance and safety margin of operation.

[0051] In this embodiment, Table 1 is a data table of energy storage power decision values. The minimum constant value is set to 0.01, and the maximum energy storage power is set to 50. The table records in detail the predicted electricity price, current electricity price, current overall SOC, median overall SOC, standard deviation of overall SOC, overall SOH, and energy storage power decision value corresponding to five different sample numbers. Among them, the predicted electricity price for sample number 1 is 0.80, the current electricity price is 0.65, the current overall SOC is 70, the median overall SOC is 68, the standard deviation of overall SOC is 2.0, the overall SOH is 0.91, and the energy storage power decision value is 2.27; the predicted electricity price for sample number 2 is 0.82, the current electricity price is 0.85, the current overall SOC is 65, the median overall SOC is 70, the standard deviation of overall SOC is 3.0, the overall SOH is 0.87, and the energy storage power decision value is -. 1.32; Sample No. 3 has a predicted electricity price of 0.90, a current electricity price of 0.75, a current overall SOC of 73, a median overall SOC of 70, an overall SOC standard deviation of 2.5, an overall SOH of 0.89, and a decision value for energy storage power of 1.70; Sample No. 4 has a predicted electricity price of 0.72, a current electricity price of 0.69, a current overall SOC of 60, a median overall SOC of 68, an overall SOC standard deviation of 3.5, an overall SOH of 0.93, and a decision value for energy storage power of 1.73; Sample No. 5 has a predicted electricity price of 1.05, a current electricity price of 0.92, a current overall SOC of 76, a median overall SOC of 74, an overall SOC standard deviation of 2.2, an overall SOH of 0.88, and a decision value for energy storage power of 1.55.

[0052] Table 1 Energy Storage Power Decision Value Data Table

[0053]

[0054] like Figure 4 The figure shows the distribution of energy storage power decision and multiple influencing factors. A dual Y-axis joint display method is used, with bar charts representing the energy storage power decision values ​​corresponding to each sample number, and line charts in different formats displaying time-series electricity price changes. Overall SOC deviation And the overall SOH parameters, with the sample number on the x-axis. According to Table 1 and... Figure 4 It can be seen that the energy storage power decision values ​​of samples 1, 3, 4, and 5 are positive, indicating a tendency to charge and retain energy, while sample 2 is negative, indicating a demand for discharge. When the overall SOC deviation is large, i.e. far from the median, it will inhibit the energy storage power decision value and limit the output power. When the overall SOH is high, the energy storage power decision value is slightly higher under the same conditions, indicating that the health status can amplify the charging and discharging capabilities.

[0055] This implementation scheme achieves dynamic response and adjustment to market price fluctuations by integrating predicted and current electricity prices and calculating time-series electricity price changes based on the difference in predicted prices. It combines a sliding time window to obtain overall State of Charge (SOC) data, extracts the overall SOC median and standard deviation, dynamically calculates the overall SOC deviation, and constructs an SOC adjustment term through exponential decay and reciprocal normalization to achieve adaptive safety constraints on the energy storage's state of charge. The scheme incorporates the maximum energy storage power and actual overall SOH operating parameters, performing collaborative calculations with the aforementioned factors to output a multi-factor integrated control energy storage power decision value. This enables refined and intelligent power allocation of energy storage under complex markets and variable operating conditions, improving response flexibility and operational safety, and providing a solid data foundation and decision support for vehicle-to-grid (V2G) collaborative energy storage scheduling.

[0056] Specifically, the process of implementing the charging and discharging strategy is as follows: The energy storage power decision value is written into the vehicle-to-grid energy storage time-series database; simultaneously, the upper and lower limits of the energy storage power decision value are checked, and a comprehensive safety check is performed with the overall SOH and overall SOC states; the upper and lower limit checks include comparing the energy storage power decision value with the maximum and minimum allowable output power to ensure that it does not exceed the safe operating range of energy storage unit 1 and the power equipment. The overall SOH and overall SOC serve as comprehensive criteria for judging the battery pack's health and charge level, providing dynamic safety protection for energy storage unit 1. If the overall SOC, overall SOH, overall temperature, and overall current exceed the corresponding SOC safety threshold, SOH safety threshold, temperature safety threshold, and current safety threshold, the output of energy storage charging and discharging power is limited and energy storage scheduling is suspended to prevent overcharging, over-discharging, overheating, and overcurrent risks of energy storage unit 1; the sign of the energy storage power decision value is determined, and the energy storage power decision value is... With charging threshold Discharge threshold and standby threshold After comparison, the intelligent scheduling controller 7 generates and executes the charging and discharging strategy instructions, which are then executed by the DC module controller 8; when > When the charging mode is determined, the charging process is initiated. A charging command is issued based on the energy storage power decision value. Coordinated by the charging pile 4 and the bidirectional converter 3, energy from the grid 2 is allowed to flow into the energy storage unit 1 and the vehicle battery 5, and charging safety monitoring is activated. < When the system is in discharge mode, the discharge process is initiated. Based on the energy storage power decision value, a discharge command is issued, and the bidirectional converter 3 delivers energy from energy storage unit 1 to grid 2 and external loads. Discharge safety monitoring is also activated. Charging and discharging safety monitoring includes real-time monitoring of key parameters such as temperature, current, and voltage during charging and discharging. In case of abnormalities, an alarm is immediately issued, and control commands are automatically adjusted to prevent the abnormality from escalating. ≤ or ≥ When the system is in standby mode, the intelligent dispatch controller 7 suspends all active charging and discharging operations of energy storage unit 1, and the DC module controller 8 shuts down the corresponding current channel, maintaining only status monitoring and self-testing. In standby mode, energy storage unit 1 consumes very little energy, maintaining only signal monitoring, parameter acquisition, and fault self-testing, ensuring both smooth data link operation and extending the lifespan of the energy storage device. When the grid 2 is detected to be unavailable, it switches to independent charging and discharging and initiates emergency power supply. Independent charging and discharging mode, also known as islanded operation mode, can continuously supply power to important loads and emergency scenarios, improving fault resistance and operational resilience, and ensuring uninterrupted power supply in critical locations.

[0057] This implementation plan enables end-to-end safety verification of energy storage power decision values ​​and intelligent scheduling across multiple scenarios. By real-time monitoring of key parameters such as overall SOC, overall SOH, overall temperature, and overall current, it automatically identifies and addresses risks exceeding thresholds, effectively limiting and suspending charging and discharging to ensure system safety. The scheduling controller can intelligently switch between charging, discharging, and standby modes to ensure efficient and coordinated operation of energy storage unit 1. In the event of a grid outage, it can automatically enter independent charging / discharging and emergency power supply modes, significantly improving the safety, flexibility, and reliability of the vehicle-to-grid energy storage system.

[0058] Specifically, during the execution of the charging and discharging strategy, the process of evaluating the charging and discharging power allocation of a single battery module based on the preprocessed vehicle-to-grid energy storage scheduling data is as follows: During the execution of the charging and discharging strategy, the energy storage power decision value and the single-cell SOH and single-cell temperature data of each battery module are acquired. The single-cell SOH reflects the current health and lifespan of the battery module, while the single-cell temperature is used to dynamically monitor the thermal stability of the working environment of each module. Both are important parameters for achieving refined power management. At the same time, based on a sliding time window, the single-cell temperature data is statistically analyzed, and the median is selected as the target temperature value. The current single-cell temperature of battery module i is subtracted from the target temperature value of battery module i, and the absolute value is taken to obtain the absolute temperature deviation value. The current single-cell SOH of battery module i is subtracted from a constant, and the absolute value is taken to obtain the absolute health deviation value. The absolute temperature deviation value is used to measure the degree of deviation between the current temperature of each battery module and the target optimal temperature, while the absolute health deviation value reflects the distance between the health state of the battery cell and the optimal operating condition. Both serve as the core influencing factors for the allocation weight calculation. The absolute deviation values ​​of temperature and health are added together, and the negative of the sum is used as the exponent. This natural exponential operation yields the individual allocation weight of battery module i. The exponential operation ensures that the smaller the deviation, the greater the weight. That is, healthy modules with temperatures close to the target value are more likely to receive higher charge and discharge allocations, reflecting a dynamic priority for safety and health. Based on the number of battery modules, the individual allocation weights of all battery modules are summed to obtain the overall allocation weight. The ratio of the individual allocation weight to the overall allocation weight is multiplied by the current energy storage power decision value to obtain the individual differentiated power allocation value of battery module i.

[0059] The specific formula for the individual unit differentiated power allocation value is as follows:

[0060] ;

[0061] In the formula, This represents the individual differentiated power allocation value of battery module i. The total energy storage power decision value obtained from the system-level decision is allocated according to the current temperature and health of each battery module. This ensures that healthier battery modules with better temperatures receive more power, while abnormal and deviating battery modules are derated, thereby achieving differentiated protection and optimized resource utilization. This represents the energy storage power decision value, which is the total charging and discharging power allocated to all battery modules at the current scheduling moment, serving as the overall allocatable resource pool. This indicates the current temperature of a single cell in battery module i, reflecting the thermal state of the battery module. This represents the target temperature value of battery module i, which serves as an ideal reference baseline for battery module temperature control and is used to determine the degree of deviation of the current battery module temperature. This indicates the current state of energy (SOH) of a single cell in battery module i, reflecting the level of battery capacity retention and performance degradation. This represents the absolute temperature deviation value, reflecting the absolute value by which the current battery module temperature deviates from the target temperature; the smaller the deviation, the better. This represents the absolute deviation value of the battery health, which measures the deviation of the battery module's health from the ideal full health. The closer it is to 1, the healthier the battery is and the more suitable it is for power distribution. This represents the individual allocation weight of battery module i, achieving differentiated allocation so that higher-performing battery modules receive more power. The exponential term adds the two deviation values ​​as the disadvantage weight in power allocation, and the negative sign is used for subsequent exponential decay. Larger deviations result in smaller weights, and smaller deviations result in larger weights.

[0062] This implementation scheme can adaptively calculate the cell allocation weights of each battery module based on the energy storage power decision value and real-time data on the state of health (SOH) and temperature of each cell during the execution of the charging and discharging strategy. By constructing absolute temperature deviation values ​​and absolute health deviation values, the power allocation ratio of each cell is dynamically adjusted, achieving differentiated and precise energy management of battery modules. This effectively avoids the risks caused by cell overheating and abnormal health conditions, improves the balance, safety, and overall utilization efficiency of energy storage, and ensures high reliability and long lifespan operation of vehicle-grid collaborative energy storage under complex operating conditions.

[0063] Specifically, the process of performing differentiated charging and discharging scheduling for each battery module, while optimizing the charging and discharging strategy and the charging and discharging power allocation strategy, is as follows: The differentiated power allocation value of each cell is written into the vehicle-to-grid energy storage time-series database. The intelligent scheduling controller 7 generates a power allocation command based on the differentiated power allocation value of each cell and sends it to the DC module controller 8 in real time. The DC module controller 8 adjusts the cell current of each battery module according to the power allocation command to control the charging and discharging of each battery module. At the same time, the charging and discharging status, cell temperature, and cell SOH of each battery module are transmitted back in real time during the power allocation process. If it is detected that the cell temperature of a battery module continues to rise for a longer period of time and the cell SOH continues to fall for a longer period of time, the differentiated power allocation value of the cell of the battery module is reduced, the connection with the main circuit is disconnected, and the active balancing circuit is activated to transfer, replenish, and eliminate energy differences between the battery modules, while correcting the power allocation command. The active balancing circuit is an energy management device based on charge transfer. When it detects that the SOC or SOH deviation between battery modules is too large, it can transfer excess energy from cells in better condition to cells in worse condition, thereby achieving energy balance between modules. This not only improves overall capacity utilization but also significantly extends safe operating life. Real-time disconnection of abnormal modules and correction of power commands effectively prevents runaway risks such as individual unit overheating and capacity imbalance. Real-time monitoring of the entire process's energy storage power decision values ​​and individual unit differentiated power allocation values, combined with historical vehicle-to-grid energy storage scheduling data, utilizes a genetic algorithm to optimize charging / discharging strategies and power allocation strategies. The genetic algorithm is an intelligent optimization method based on natural selection and genetic mechanisms. Through continuous evolution and optimization using historical data, it dynamically adjusts charging / discharging thresholds, balancing strategies, and allocation coefficients, automatically adapting to changing operating conditions and achieving global optimization under multiple objectives. Combining real-time and historical data for bidirectional optimization ensures efficient, balanced, and safe operation even under high-frequency fluctuations and extreme events.

[0064] In this implementation scheme, differentiated energy management at the individual cell level is achieved by issuing power allocation commands based on the differentiated power allocation values ​​of each cell and dynamically adjusting the charging and discharging current of each battery module. Simultaneously, the charging and discharging status, cell temperature, and cell state of health (SOH) of each module can be monitored in real time. When abnormal cell temperature or health is detected, the power allocation is adjusted, the abnormal module is disconnected, and active balancing is initiated to eliminate energy differences and ensure overall safety and balance. Combining historical data, a genetic algorithm is used to continuously optimize the charging and discharging strategy and power allocation, ensuring that energy storage maintains high safety, high balance, and high utilization under varying operating conditions, significantly improving the intelligent management and healthy operation capabilities of vehicle-to-grid collaborative energy storage.

[0065] Reference Figure 2As shown, the second aspect of the present invention provides a vehicle-to-grid (V2G) collaborative charging pile scheduling system based on retired battery energy storage, applied to the aforementioned V2G collaborative charging pile scheduling method based on retired battery energy storage. The system includes: a multi-source data acquisition and preprocessing module for real-time acquisition of V2G energy storage scheduling data, preprocessing and storing the V2G energy storage scheduling data; an energy storage scheduling judgment and decision module for real-time judgment of energy storage unit scheduling priorities based on the preprocessed V2G energy storage scheduling data, and constructing an energy storage charging and discharging strategy dataset based on the energy storage unit scheduling priorities; a charging and discharging decision formulation and execution module for constructing an electricity price prediction model based on historical V2G energy storage scheduling data, receiving the energy storage charging and discharging strategy dataset, determining the energy storage charging and discharging power output based on the electricity price prediction model, and executing the charging and discharging strategy; and a health temperature control and safety protection module for evaluating the charging and discharging power allocation of a single battery module based on the preprocessed V2G energy storage scheduling data during the execution of the charging and discharging strategy, performing differentiated charging and discharging scheduling for each battery module, and simultaneously optimizing the charging and discharging strategy and the charging and discharging power allocation strategy.

[0066] This implementation plan integrates modules for multi-source data acquisition and preprocessing, energy storage scheduling judgment and decision-making, charge / discharge decision-making and execution, and health, temperature control, balancing, and safety protection. This enables real-time, efficient acquisition, precise preprocessing, and dynamic management of vehicle-to-grid (V2G) energy storage scheduling data. It can not only intelligently determine the scheduling priority of energy storage units based on the latest data and scientifically formulate charging / discharging strategies, but also combine electricity price prediction models for power decisions and strategy execution. Furthermore, it can perform real-time assessment and differentiated adjustment of the health status and temperature of each battery module during charging and discharging, achieving multi-dimensional energy balance and safety protection. This improves the data utilization efficiency, decision-making intelligence, and safety stability of the V2G energy storage system, providing strong support for achieving efficient, reliable, and intelligent V2G collaborative energy storage scheduling.

[0067] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0068] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. As those skilled in the art will understand, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for scheduling charging piles based on vehicle-to-grid coordination of retired battery energy storage, characterized in that, The method comprises the following steps: S1, real-time collection of vehicle-to-grid energy storage scheduling data, data preprocessing and storage of the vehicle-to-grid energy storage scheduling data; S2, real-time determination of energy storage unit scheduling priority based on the preprocessed vehicle-to-grid energy storage scheduling data, and construction of an energy storage charging and discharging strategy dataset according to the energy storage unit scheduling priority; the specific process of real-time determination of the energy storage unit scheduling priority based on the preprocessed vehicle-to-grid energy storage scheduling data is as follows: obtaining the current charging request power and the maximum energy storage power, calculating the ratio of the current charging request power to the maximum energy storage power to obtain a demand load value; obtaining the current grid load and the maximum grid load, calculating the ratio of the current grid load to the maximum grid load to obtain a grid pressure value; based on a sliding time window, obtaining price data and calculating the mean and standard deviation to obtain the price mean and price standard deviation, calculating the absolute difference between the current price and the price mean, and dividing by the sum of the price standard deviation and a minimum constant value to obtain a price fluctuation value; subtracting the price fluctuation value from the sum of the demand load value and the grid pressure value to obtain a priority original value, and performing a hyperbolic tangent function operation on the priority original value to obtain the energy storage scheduling priority; S3, construction of a price prediction model according to historical vehicle-to-grid energy storage scheduling data, reception of the energy storage charging and discharging strategy dataset, combination with the price prediction model to determine the output of the energy storage charging and discharging power, and execution of the charging and discharging strategy; S4, during execution of the charging and discharging strategy, evaluation of the charging and discharging power distribution of a single battery module according to the preprocessed vehicle-to-grid energy storage scheduling data, differential charging and discharging scheduling of each battery module, and optimization of the charging and discharging strategy and the charging and discharging power distribution strategy. 2.The retired battery storage based vehicle-grid collaborative charging pile scheduling method of claim 1, wherein, The specific process of real-time collection of vehicle-to-grid energy storage scheduling data, data preprocessing and storage of the vehicle-to-grid energy storage scheduling data is as follows: Real-time collection of vehicle-to-grid energy storage scheduling data, the specific process being as follows: real-time communication through a charging pile (4) and a vehicle networking platform to collect charging request power; feedback collection of the maximum energy storage power of an energy storage unit (1) through a bidirectional converter (3) and a battery management system (6); real-time collection of grid load through a smart meter, and collection of maximum grid load from a power distribution system by calling a maximum rated capacity parameter; real-time collection of electricity price through API docking with a power operation platform; obtaining the number of battery modules in the energy storage unit (1), and collecting overall SOC, overall SOH, overall temperature and overall current of the energy storage unit (1) and monomer SOC, monomer SOH, monomer temperature and monomer current data of each battery module through the battery management system (6); collection of weather information by calling a weather API interface; and real-time collection of a charging gun and vehicle connection state signal; The vehicle-to-grid energy storage scheduling data is subjected to integrity check, and data with null value, repetition, exceeding physical upper limit and device failure are removed, and the sliding window method is used to smooth data jump. Meanwhile, the vehicle-to-grid energy storage scheduling data is subjected to standardization and dimensionless normalization processing. The vehicle-to-grid energy storage scheduling data is aligned according to time stamp, and all related data snapshots at the same time are uniformly grouped. A backup collection channel is switched for sudden data loss and communication interruption fault state. A vehicle-to-grid energy storage time sequence database is established, and the original and preprocessed vehicle-to-grid energy storage scheduling data is stored in the vehicle-to-grid energy storage time sequence database in real time. 3.The retired battery storage based vehicle-grid collaborative charging pile scheduling method of claim 1, wherein, The specific process of constructing the energy storage charging and discharging strategy data set according to the energy storage unit scheduling priority is as follows: The energy storage scheduling priority is written into the vehicle-to-grid energy storage time sequence database, and compared with the priority threshold value. When no vehicle accesses the charging pile (4), or the energy storage scheduling priority is less than or equal to the priority threshold value, the energy storage unit (1) enters a low-power standby state, does not perform charging and discharging operation, supplies power to the vehicle through the power grid (2) and replenishes the energy storage battery, and maintains normal monitoring of the vehicle-to-grid energy storage scheduling data; When a vehicle accesses the charging pile (4) and the energy storage scheduling priority is greater than the priority threshold value, the energy storage unit (1) supplies power, combines the energy storage scheduling priority and the corresponding vehicle-to-grid energy storage scheduling data to construct an energy storage charging and discharging strategy data set, and outputs the data set to trigger the next process and develop a charging and discharging strategy; If the monitored energy storage scheduling priority change exceeds the mutation threshold value, a limit charging and discharging power instruction is generated to perform early load reduction and start the active cooling device. At the same time, the priority threshold value is adjusted according to the historical energy storage scheduling priority distribution and the energy storage scheduling priority mean value by using the sliding window self-learning algorithm.

4. The retired battery-based energy storage vehicle-to-grid cooperative charging pile scheduling method according to claim 1, characterized in that, The specific process of constructing the electricity price prediction model according to the historical vehicle-to-grid energy storage scheduling data is as follows: The historical electricity price, power grid load data and corresponding meteorological information are obtained, and the mean value and standard deviation of the historical electricity price and power grid load data are calculated. The historical electricity price, power grid load data, meteorological information and mean value and standard deviation of the historical electricity price and power grid load data are combined to construct an electricity price time sequence feature set as input. The long short-term memory network algorithm is used to train the electricity price time sequence feature set. The training set and test set are divided at the same time. The electricity price prediction model is constructed by combining cross-validation and parameter optimization, and the future electricity price prediction value is output.

5. The retired battery-based energy storage vehicle-to-grid cooperative charging pile scheduling method according to claim 4, characterized in that, The specific process of receiving the energy storage charging and discharging strategy data set and determining the output of the energy storage charging and discharging power in combination with the electricity price prediction model is as follows: The energy storage charging and discharging strategy data set is received, the electricity price prediction value and the current electricity price are obtained, and the time sequence electricity price variation term is obtained by dividing the difference between the electricity price prediction value and the current electricity price by the sum of the absolute value of the electricity price prediction value and a minimum constant value. Based on the sliding time window, the overall SOC data is obtained, and the median is screened out to obtain the overall SOC median, and the overall SOC standard deviation is calculated. The difference between the current overall SOC and the overall SOC median is divided by the sum of the overall SOC standard deviation and the minimum constant value to obtain the overall SOC deviation value. The overall SOC deviation value is used as an exponential power, and a natural exponential operation is performed to obtain an SOC deviation attenuation factor. The SOC deviation attenuation factor is added to a constant one, and the reciprocal is taken to obtain an SOC adjustment term; The maximum power of the energy storage and the overall SOH are obtained, and the time series price variation term, the maximum power of the energy storage, the SOC adjustment term, and the overall SOH are multiplied to obtain a storage power decision value.

6. The retired battery-based energy storage vehicle-to-grid cooperative charging pile scheduling method according to claim 5, characterized in that, The specific process of executing the charging and discharging strategy is: The storage power decision value is written into the vehicle network energy storage time series database; at the same time, the upper and lower limits of the storage power decision value are checked, and the overall SOH and the overall SOC state are comprehensively safety checked; if the overall SOC, the overall SOH, the overall temperature and the overall current exceed the corresponding SOC safety threshold, the SOH safety threshold, the temperature safety threshold and the current safety threshold, the output of the storage charging and discharging power is limited and the storage scheduling is suspended; determining the positive or negative of the energy storage power decision value, and comparing the energy storage power decision value with a charging threshold , a discharging threshold and a standby threshold , the intelligent scheduling controller (7) generates an execute charging and discharging strategy instruction and executes through the DC module controller (8); When > the charging mode is determined, the charging process is started, the charging instruction is issued according to the energy storage power decision value, the coordination of the charging pile (4) and the bidirectional converter (3) is allowed, the energy of the power grid (2) is allowed to flow into the energy storage unit (1) and the vehicle-mounted battery (5), and the charging safety monitoring is started. When discharge mode, start the discharge process, and issue a discharge command according to the energy storage power decision value to send the energy of the energy storage unit (1) to the power grid (2) and external load by the bidirectional converter (3), and start the discharge safety monitoring.​ When ≤ or ≥ , the standby mode is determined, the intelligent scheduling controller (7) suspends all active charging and discharging operations of the energy storage unit (1), the DC module controller (8) closes the corresponding current channel, and only state monitoring and self-checking are maintained; When it is monitored that the power grid (2) is unavailable, switch to independent charging and discharging and start emergency power supply.

7. The retired battery-based energy storage vehicle-to-grid cooperative charging pile scheduling method according to claim 6, characterized in that, The specific process of evaluating the charging and discharging power distribution of each battery module according to the pretreated vehicle network energy storage scheduling data when the charging and discharging strategy is executed is: During the execution of the charging and discharging strategy, the storage power decision value and the single SOH and single temperature data of each battery module are obtained, and based on the sliding time window, the single temperature data is counted and the median is screened out as the target temperature value; The absolute value of the difference between the current single temperature of the battery module i and the target temperature value of the battery module i is taken to obtain the temperature absolute deviation value; the absolute value of the difference between the constant one and the current single SOH of the battery module i is taken to obtain the health degree absolute deviation value; The temperature absolute deviation value and the health degree absolute deviation value are added and the reciprocal is taken as an exponential power, and a natural exponential operation is performed to obtain the single distribution weight of the battery module i; based on the number of battery modules, the single distribution weights of all battery modules are added to obtain the overall distribution weight; the ratio of the single distribution weight to the overall distribution weight is multiplied by the current storage power decision value to obtain the single differentiated power distribution value of the battery module i. 8.The retired battery storage based vehicle-to-grid cooperative charging pile scheduling method of claim 7, wherein, The specific process of performing differentiated charging and discharging scheduling for each battery module while optimizing the charging and discharging strategy and the charging and discharging power distribution strategy is: The single differentiated power distribution value is written into the vehicle network energy storage time series database, and the power distribution instruction is generated by the intelligent scheduling controller (7) according to the single differentiated power distribution value, and is real-time issued to the DC module controller (8). The DC module controller (8) adjusts the single current of each battery module according to the power distribution instruction to control the charging and discharging of each battery module; Meanwhile, the charging and discharging state, single temperature and single SOH of each battery module in the real-time backhaul power distribution process are detected. If the single temperature of a battery module continues to rise for a time exceeding a maximum allowable threshold and the single SOH continues to decrease for a time exceeding a maximum allowable threshold, the single differentiated power distribution value of the battery module is reduced, the connection with the main circuit is disconnected, the active balancing circuit is started, the energy transfer, supplement and elimination of the battery module are performed, and the power distribution instruction is corrected. The real-time monitoring of the energy storage power decision value and the single differentiated power distribution value in the whole process is performed, the historical vehicle network energy storage scheduling data are combined, the charging and discharging strategy and the power distribution strategy are optimized by using a genetic algorithm.

9. A vehicle-network collaborative charging pile scheduling system based on retired battery energy storage, applying the vehicle-network collaborative charging pile scheduling method based on retired battery energy storage according to any one of claims 1-8, characterized in that, The system comprises: a multi-source data acquisition and preprocessing module for acquiring vehicle network energy storage scheduling data in real time, preprocessing and storing the vehicle network energy storage scheduling data; an energy storage scheduling judgment and decision module for judging the energy storage unit scheduling priority in real time based on the preprocessed vehicle network energy storage scheduling data and constructing an energy storage charging and discharging strategy data set according to the energy storage unit scheduling priority; a charging and discharging decision making and execution module for constructing a price prediction model according to the historical vehicle network energy storage scheduling data, receiving the energy storage charging and discharging strategy data set, determining the output of the energy storage charging and discharging power in combination with the price prediction model, and executing the charging and discharging strategy; a health temperature control balancing and safety protection module for evaluating the charging and discharging power distribution of each battery module according to the preprocessed vehicle network energy storage scheduling data when the charging and discharging strategy is executed, performing the differentiated charging and discharging scheduling of each battery module, and optimizing the charging and discharging strategy and the charging and discharging power distribution strategy.

Images