Methods and systems for load balancing and optimization scheduling of battery swapping networks

Abstract

This invention relates to the field of battery swapping scheduling technology, and discloses a method and system for load balancing and optimization scheduling in a battery swapping network. The method involves: each battery swapping station's hardware module collecting real-time data on battery cell voltage, temperature, state of charge, and health status via a BMS acquisition line, obtaining battery cell status data; a cloud-based intelligent control platform performing multi-site fusion analysis on the battery cell status data to obtain a load balancing scheduling strategy; the cloud-based intelligent control platform generating scheduling instructions for each battery swapping station based on the load balancing scheduling strategy; and each battery swapping station's hardware module receiving the scheduling instructions and performing multi-level load balancing processing to obtain the load redistribution result. This invention achieves coordinated load balancing and load redistribution among multiple stations in a battery swapping network, thereby realizing intelligent operation and management of the battery swapping network.

Description

Technical Field

[0001] This invention relates to the field of battery swapping scheduling technology, and in particular to a method and system for balanced and optimized scheduling of battery swapping network load. Background Technology

[0002] Significant differences exist in battery status and load distribution across battery swapping networks. Due to factors such as geographical location, usage frequency, and battery aging, different swapping stations exhibit uneven load distribution characteristics, leading to overload at some stations and underutilization at others, severely impacting the overall operational efficiency of the network. While existing passive balancing technologies, which dissipate load through resistance, are low-cost, they suffer from significant energy loss and heat generation. Active balancing technologies, although capable of charge transfer, are complex, costly, and limited to local optimization within a single device, lacking a global understanding and intelligent scheduling capability for the entire battery swapping network's load distribution. Summary of the Invention

[0003] The main objective of this invention is to provide a method and system for load balancing and optimization scheduling in a battery swapping network. This invention achieves coordinated balancing and load redistribution among multiple sites in the battery swapping network, thereby realizing intelligent operation and management of the battery swapping network.

[0004] To achieve the above objectives, the present invention provides a method for load balancing and optimization scheduling of a battery swapping network, comprising the following steps:

[0005] Each battery swapping station's hardware modules collect real-time data on cell voltage, temperature, state of charge, and health status via the BMS acquisition line, thus obtaining cell status data.

[0006] The cloud-based intelligent control platform performs multi-site fusion analysis on the cell status data to obtain a balanced scheduling strategy;

[0007] The cloud-based intelligent control platform generates scheduling instructions for each battery swapping station based on the aforementioned balanced scheduling strategy.

[0008] After receiving the scheduling instructions, the hardware modules of each battery swapping station perform multi-level load balancing processing to obtain the load redistribution results.

[0009] Optionally, in a first implementation of the first aspect of the present invention, the hardware modules of each battery swapping station collect real-time data on cell voltage, temperature, state of charge, and health status via a BMS acquisition line to obtain cell status data, including:

[0010] Each battery swapping station's hardware modules synchronously collect data on cell voltage, temperature, state of charge, health status, and internal resistance via the BMS acquisition line to obtain raw cell data.

[0011] The original cell data is filtered to obtain a set of cell parameters;

[0012] The site load index is calculated based on the set of battery cell parameters, and a unique site identifier and geographic coordinate information are embedded to obtain the identified battery cell data.

[0013] The identified cell data is encapsulated into a data packet format that conforms to the battery swapping network communication protocol to obtain cell status data.

[0014] Optionally, in a second implementation of the first aspect of the present invention, the cloud-based intelligent control platform performs multi-site fusion analysis on the cell status data to obtain a balanced scheduling strategy, including:

[0015] The cloud-based intelligent control platform receives the battery cell status data transmitted from each battery swapping station and constructs a real-time dataset containing battery cell parameters from all stations based on the battery cell status data.

[0016] By combining the historical operation data and geographical coordinate information of each battery swapping station, a spatiotemporal correlation analysis is performed on the real-time dataset to obtain the distance weighting coefficient and energy transmission efficiency coefficient between stations.

[0017] Based on the distance weighting coefficient and the energy transmission efficiency coefficient, a network-wide battery consistency assessment is conducted to obtain the cell health weight and maximum power capacity parameters of each battery swapping station.

[0018] A network-wide load allocation function is established based on the cell health weight and the maximum power capacity parameter.

[0019] Based on the network-wide load distribution function, load oscillation analysis is performed to obtain a balanced scheduling strategy.

[0020] Optionally, in a third implementation of the first aspect of the present invention, the step of combining historical operating data and geographical coordinate information of each battery swapping station to perform spatiotemporal correlation analysis on the real-time dataset to obtain distance weighting coefficients and energy transmission efficiency coefficients between stations includes:

[0021] Historical operation data is obtained by extracting charging and discharging cycle data, temperature change data, and time decay data of each battery swapping station from the cloud database.

[0022] The geographical coordinates of each battery swapping station are spatially matched with the historical operation data to obtain the physical distance matrix and operation correlation matrix between the stations.

[0023] Spatiotemporal correlation analysis is performed based on the physical distance matrix and the operational correlation matrix to obtain the distance weight coefficients between stations;

[0024] The energy transmission efficiency coefficient is calculated based on the cell health status of each battery swapping station in the real-time dataset and the distance weighting coefficient.

[0025] Optionally, in a fourth implementation of the first aspect of the present invention, the step of performing load oscillation analysis based on the network-wide load allocation function to obtain a balanced scheduling strategy includes:

[0026] The real-time load value of each battery swapping station is calculated using the network-wide load distribution function, and the deviation between the real-time load value and the load reference value is analyzed to obtain the load droop coefficient.

[0027] The health droop coefficient is obtained by calculating the health status and health reference value of each battery swapping station. At the same time, the capacity droop coefficient is obtained by calculating the deviation between the capacity and capacity reference value of each battery swapping station.

[0028] The load droop coefficient, the healthy droop coefficient, and the capacity droop coefficient are combined and input into the three-dimensional droop control algorithm, and a predictive damping factor is added to calculate the load oscillation suppression, thereby obtaining the oscillation control parameters.

[0029] Small-signal stability analysis is performed on the oscillation control parameters to obtain stability analysis results. Based on the stability analysis results, load areas are divided for each battery swapping station to obtain a balanced scheduling strategy.

[0030] Optionally, in a fifth implementation of the first aspect of the present invention, the step of combining the load droop coefficient, the healthy droop coefficient, and the capacity droop coefficient and inputting them into a three-dimensional droop control algorithm, and adding a predictive damping factor to perform load oscillation suppression calculations to obtain oscillation control parameters, includes:

[0031] The load droop coefficient is used as the load deviation adjustment coefficient, the health droop coefficient is used as the cell health adjustment coefficient, and the capacity droop coefficient is used as the capacity adjustment coefficient.

[0032] The load deviation adjustment coefficient, the cell health adjustment coefficient, and the capacity adjustment coefficient are combined to obtain the adjustment coefficient combination;

[0033] Load prediction patterns are extracted from historical load change data of each battery swapping station, and predictive damping factors are calculated.

[0034] The predictive damping factor and the adjustment coefficient are combined and input into the three-dimensional droop control algorithm to perform three-dimensional droop control calculations, thereby obtaining the power adjustment amount of each battery swapping station. The oscillation suppression effect of the battery swapping network is verified and the stability conditions are confirmed through the power adjustment amount, and the oscillation control parameters are obtained.

[0035] Optionally, in a sixth implementation of the first aspect of the present invention, the cloud-based intelligent control platform generates scheduling instructions for each battery swapping station according to the balanced scheduling strategy, including:

[0036] The cloud-based intelligent control platform identifies the load status characteristics of each battery swapping station based on the aforementioned balanced scheduling strategy;

[0037] Based on the load status characteristics, the corresponding operation mode is selected and the results are allocated for each battery swapping station from the following modes: battery health priority mode, load balancing priority mode, user distance priority mode, grid friendliness priority mode, cost optimization mode, or hybrid intelligent mode.

[0038] The corresponding equalization threshold angle is calculated based on the inconsistency data of the first cell of each battery swapping station in the operation mode allocation result, and the health degradation coefficient is calculated in combination with the average health status of each battery swapping station.

[0039] Based on the equilibrium threshold angle and the health decay coefficient, specific control parameters are generated for each battery swapping station.

[0040] Based on the dedicated control parameters, the level of coordination and scheduling of each battery swapping station is determined and the coordination relationship between stations is established to obtain the scheduling level allocation scheme.

[0041] The operation mode allocation results, dedicated control parameters, and coordination relationships in the scheduling hierarchy allocation scheme are encapsulated into a standard scheduling protocol format to obtain the scheduling instructions for each battery swapping station.

[0042] Optionally, in the seventh implementation of the first aspect of the present invention, after receiving the scheduling instruction, the hardware modules of each battery swapping station perform multi-level load balancing processing to obtain the load redistribution result, including:

[0043] Each battery swapping station's hardware module receives the scheduling instructions issued by the cloud-based intelligent control platform and parses the station's dedicated load balancing configuration parameters;

[0044] Based on the equalization execution configuration parameters, detect the inconsistency data of the second cell and compare it with the equalization threshold angle to determine the equalization processing status;

[0045] Based on the balanced processing status, establish communication links with adjacent battery swapping stations and perform inter-station load coordination according to the hierarchical relationship of two-station coordination, three-station coordination, four-station coordination or five-level linkage to obtain the load coordination processing result.

[0046] The load coordination processing results are used to complete the load transfer and energy redistribution operations between various sites in the battery swapping network, resulting in a load redistribution result.

[0047] Optionally, in an eighth implementation of the first aspect of the present invention, the step of detecting the inconsistency data of the second cell based on the equalization execution configuration parameters and comparing it with the equalization threshold angle to determine the equalization processing state includes:

[0048] The equalization threshold angle is parsed from the equalization execution configuration parameters and the maximum voltage difference and state of charge difference between each cell are collected synchronously to obtain the second cell inconsistency data.

[0049] The second cell inconsistency data is compared with the equalization threshold angle. When the second cell inconsistency data exceeds the trigger value range, an equalization start trigger signal is generated.

[0050] The intelligent switch matrix in the equalization execution unit is awakened according to the equalization start trigger signal, and the PWM duty cycle control parameters are extracted from the health decay coefficient to obtain the current intensity adjustment command that matches the health status of the battery cell.

[0051] The current intensity adjustment command is executed to control the direction and magnitude of the equalization current between each cell, while simultaneously monitoring temperature fluctuations and voltage change trends during the equalization process to obtain the equalization processing status.

[0052] This invention also provides a load balancing and optimization scheduling system for battery swapping networks, comprising:

[0053] The real-time acquisition subsystem is used by the hardware modules of each battery swapping station to collect real-time data on battery cell voltage, temperature, state of charge, and health status through the BMS acquisition line, thereby obtaining battery cell status data.

[0054] The multi-site fusion analysis subsystem is used by the cloud-based intelligent control platform to perform multi-site fusion analysis on the cell status data to obtain a balanced scheduling strategy.

[0055] A generation subsystem is used by the cloud-based intelligent control platform to generate scheduling instructions for each battery swapping station according to the balanced scheduling strategy.

[0056] The multi-level load balancing subsystem is used by each battery swapping station hardware module to perform multi-level load balancing after receiving the scheduling instructions, and obtain the load redistribution result.

[0057] In summary, the technical solution provided by this invention integrates and analyzes the cell status data of multiple battery swapping stations through a cloud-based intelligent control platform, constructing a network-wide load distribution function. This overcomes the limitations of traditional technologies that can only handle the balancing within a single device, achieving coordinated balancing and load redistribution among multiple stations in the battery swapping network. The use of a three-dimensional droop control algorithm combined with predictive damping factors effectively suppresses load oscillations in the battery swapping network, ensuring stable system operation. A multi-level scheduling architecture, from two-station collaboration to five-level linkage, is established, providing six dynamically switchable operating modes according to different application scenarios, enhancing system adaptability. Deep learning algorithms analyze historical data to achieve forward-looking scheduling optimization, and a closed-loop feedback parameter adaptive system is constructed by combining a dynamic adjustment mechanism for the balancing threshold angle and health decay coefficient. The cloud-edge collaborative distributed processing architecture moves complex calculations to the cloud, simplifying local hardware functions, reducing system deployment costs and maintenance difficulty, and realizing intelligent operation and management of the battery swapping network. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the steps of a load balancing and optimization scheduling method for a battery swapping network in one embodiment of the present invention;

[0059] Figure 2 This is a block diagram of a load balancing and optimization scheduling system for a battery swapping network according to an embodiment of the present invention.

[0060] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0062] Reference Figure 1 This embodiment provides a method for load balancing and optimization scheduling of a battery swapping network, including the following steps:

[0063] S1, the hardware modules of each battery swapping station collect the battery cell voltage, temperature, state of charge, and health status in real time through the BMS acquisition line to obtain battery cell status data;

[0064] In this system, each battery swapping station's hardware module synchronously acquires key operating parameters of each individual battery cell, such as voltage, temperature, state of charge, health status, and internal resistance, through a data acquisition line directly connected to the battery management system, forming a raw cell data set. The data processing unit within the hardware module filters this raw cell data, removing outliers caused by sampling noise, instantaneous fluctuations, or external electromagnetic interference, resulting in a set of cell parameters. Based on the state of charge, health status, voltage margin, and internal resistance indicators in the cell parameter set, a load index for the station is calculated according to a preset calculation model, quantifying the station's current operating pressure and resource carrying capacity within the battery swapping network. The hardware module embeds each station's unique identifier and precise geographic coordinates into the load index, forming identifiable cell data with station-specific characteristics. Following the dedicated communication protocol of the battery swapping network, the identifiable cell data is packaged and encapsulated to generate a cell status data packet with complete header information, verification fields, and a data body, which is then transmitted to the cloud-based intelligent control platform via a dual-redundant communication link.

[0065] S2, the cloud-based intelligent control platform performs multi-site fusion analysis on cell status data to obtain a balanced scheduling strategy;

[0066] Specifically, the cloud-based intelligent control platform receives cell status data uploaded by hardware modules through a secure communication link with each battery swapping station. It integrates this data, including multi-dimensional parameters such as voltage, temperature, state of charge, health status, and internal resistance, with station identification information in timestamp order to construct a real-time dataset covering all battery swapping stations. This dataset dynamically reflects the operational status and resource distribution of all stations across the network. The platform then merges this real-time dataset with historical operational data and precise geographic coordinates of each station. Through spatiotemporal correlation analysis, it identifies spatial proximity and historical load interaction patterns between different stations, calculates distance weighting coefficients and energy transmission efficiency coefficients between stations, and quantifies the spatial cost and transmission loss during energy allocation. Based on the distance weighting coefficients and energy transmission efficiency coefficients, it performs a network-wide battery consistency assessment, transforming the cell health status of each station into a health weighting index. This index, combined with the maximum power capacity parameter of each station, forms an input reflecting the station's load-bearing capacity in load allocation. A network-wide load allocation function is established based on the cell health weights and maximum power capacity parameters. This ensures that load allocation considers not only capacity and health status but also constraints such as distance and efficiency, achieving a globally optimal resource scheduling model. The cloud platform uses the network-wide load distribution function to perform load oscillation analysis, predicts the load change trend and fluctuation amplitude of each site under different scheduling strategies, and generates a balanced scheduling strategy while suppressing low-frequency oscillations and avoiding local overload.

[0067] S3, the cloud-based intelligent control platform generates scheduling instructions for each battery swapping station based on the balanced scheduling strategy;

[0068] It should be noted that the cloud-based intelligent control platform analyzes and identifies the operating status of all battery swapping stations across the network based on the balanced scheduling strategy, extracting core status characteristics reflecting the load level, including the current load index, cell health weight, battery capacity utilization, and service demand density of the station. Based on the load status characteristics, and considering the actual operating conditions of each battery swapping station, the platform selects the operating mode that best matches its status characteristics from various operating strategies, such as battery health priority mode, load balancing priority mode, user distance priority mode, grid friendliness priority mode, cost optimization mode, or hybrid intelligent mode, forming the network-wide operating mode allocation result. Based on the inconsistency data of the first cell of each battery swapping station in the operating mode allocation result, the platform calculates the corresponding balancing threshold angle using a preset calculation model to determine the trigger boundary conditions of the balancing function; simultaneously, the platform calculates the health decay coefficient based on the average health status of the cells of each battery swapping station to control the intensity and frequency of the balancing process. Using the balancing threshold angle and the health decay coefficient, the cloud platform generates a set of dedicated control parameters for each battery swapping station. Based on this, the platform determines the participation level of each battery swapping station in coordinated scheduling according to the differences in dedicated control parameters and the geographical distribution of site loads. It then establishes a cross-site coordination relationship network, forming a scheduling hierarchy allocation scheme covering the entire network. The operation mode allocation results, corresponding dedicated control parameters, and site coordination relationships in the scheduling hierarchy allocation scheme are encapsulated according to the format of the standard scheduling protocol for battery swapping networks. This generates scheduling instructions with structured fields, version identifiers, and security verification information, which are then sent to each battery swapping station via encrypted communication links.

[0069] S4: After receiving the scheduling instructions, the hardware modules of each battery swapping station perform multi-level load balancing processing to obtain the load redistribution results.

[0070] Specifically, each battery swapping station's hardware module receives dispatch instructions through a secure communication channel established with the cloud-based intelligent control platform. The instructions are then structured and parsed to extract specific balancing execution configuration parameters for the station, including the balancing threshold angle, health attenuation coefficient, balancing current target value, and coordination level information. Based on these parameters, the hardware module detects the cell status of the station, calculates inconsistency data for the second cell, reflecting voltage, state of charge, and internal resistance differences between cells within the current operating cycle, and compares the calculation results with the balancing threshold angle to determine if balancing processing is required. When balancing processing is deemed necessary, the hardware module establishes communication links with adjacent battery swapping stations according to the coordination level defined in the dispatch instructions. The communication process employs encrypted handshakes and low-latency transmission protocols to ensure data security and real-time performance. Based on these established communication links, each battery swapping station coordinates load across different levels, such as two-station coordination, three-station coordination, four-station coordination, or five-level linkage. Higher-level modes can cover a wider range of stations, achieving broader load optimization. During load coordination, each site dynamically calculates battery swapping demand and energy allocation based on real-time load indices, available battery quantity, and energy transmission efficiency, resulting in a load coordination processing result. The hardware module then performs load transfer and energy redistribution operations at the physical level based on this result, redirecting some battery swapping requests and battery resources from high-load sites to low-load sites, achieving network-wide operational pressure balancing, and outputting the completed load redistribution result.

[0071] In one example, the hardware modules of each battery swapping station collect real-time data on cell voltage, temperature, state of charge, and health status via the BMS acquisition line, obtaining cell status data, including:

[0072] Each battery swapping station's hardware modules synchronously collect data on cell voltage, temperature, state of charge, health status, and internal resistance via the BMS acquisition line to obtain raw cell data.

[0073] The raw cell data is filtered to obtain a set of cell parameters.

[0074] The site load index is calculated based on the set of battery cell parameters, and a unique site identifier and geographic coordinate information are embedded to obtain the identified battery cell data.

[0075] The cell identification data is encapsulated into a data packet format that conforms to the battery swapping network communication protocol to obtain cell status data.

[0076] In this example, each battery swapping station's hardware module establishes a real-time acquisition path for cell status parameters via a BMS acquisition line directly connected to the battery management system. Key operating parameters of each individual cell are synchronously acquired under a unified time reference. Synchronous acquisition covers core indicators reflecting cell performance and degradation trends, such as cell voltage, temperature, state of charge, health status, and internal resistance. By standardizing the acquisition trigger signals for different parameters, time consistency among multi-dimensional parameters is ensured, avoiding delay deviations and data misalignment issues introduced by asynchronous acquisition methods. During acquisition, the hardware module captures transient changes at millisecond-level time resolution using a high-frequency sampling strategy, recording the dynamic behavior of the cells during different operating stages such as charging, discharging, and resting, generating raw cell data, and storing it in a local cache module using a sequential buffer. The raw cell data is input into a built-in data preprocessing unit, where multi-level filtering algorithms denoise and remove anomalies. Filtering techniques include low-pass filtering, Kalman filtering, or adaptive smoothing to suppress data fluctuations caused by sampling circuit noise, transient spike interference, or electromagnetic interference. Simultaneously, outliers that clearly violate physical characteristics are detected and eliminated, such as voltage jumps or temperature abrupt changes outside the normal operating range. The filtered data is organized into a multi-dimensional set of cell parameters, reflecting the actual operating status of the cells. Based on this set of cell parameters, the hardware module calculates the site load index according to a preset mathematical model. The site load index integrates multiple factors, including the health status of each cell, real-time power output capability, capacity utilization, and internal resistance change trends, to quantify the current resource carrying capacity and load pressure level of the site within the battery swapping network. A unique site identifier and high-precision geographic coordinates for each battery swapping station are embedded into the load index to obtain identified cell data. The site identifier uses a unique coding rule across the entire network; the geographic coordinates are obtained through real-time interaction with GPS or BeiDou positioning modules, marking the spatial location of the site. The hardware module encapsulates the identified cell data, including the site load index, cell parameter set, and location identifier, according to the dedicated communication protocol for the battery swapping network. The encapsulation process includes constructing the data packet header, inserting the protocol version number, writing the data length field, and adding a checksum to ensure the integrity and correctness of the data during transmission. The actual battery cell identification data is then formatted and arranged as part of the data body. The data packet design conforms to the communication standards of battery swapping networks across multiple sites and devices, supporting transmission over wired Ethernet and dual-redundant 5G / 4G wireless links, ensuring reliable data transmission with low latency and low packet loss even under high-load network environments. The resulting battery cell status data includes battery cell operating status information and carries the site's unique identity and geographic location.

[0077] In one example, the cloud-based intelligent control platform performs multi-site fusion analysis on cell status data to obtain a balanced scheduling strategy, including:

[0078] The cloud-based intelligent control platform receives cell status data transmitted from each battery swapping station and builds a real-time dataset containing cell parameters from all stations based on the cell status data.

[0079] By combining historical operation data and geographic coordinate information of each battery swapping station, spatiotemporal correlation analysis is performed on the real-time dataset to obtain the distance weighting coefficient and energy transmission efficiency coefficient between stations.

[0080] Based on the distance weighting coefficient and energy transmission efficiency coefficient, the consistency assessment of batteries across the entire network is carried out to obtain the cell health weight and maximum power capacity parameters of each battery swapping station.

[0081] A network-wide load distribution function is established based on cell health weights and maximum power capacity parameters.

[0082] Load oscillation analysis is performed based on the network-wide load distribution function to obtain a balanced scheduling strategy.

[0083] In this example, the cloud-based intelligent control platform continuously receives cell status data transmitted from various battery swapping stations via encrypted, dual-redundant communication links. This data includes core operating parameters for each individual cell within the station, such as voltage, temperature, state of charge, health status, and internal resistance, along with a unique station identifier and geographic coordinates. Upon receiving cell status data from multiple swapping stations, the platform sorts and synchronizes the data according to timestamps to eliminate the impact of upload delays between different stations on data timeliness. It then structurally integrates the cell parameters from all stations to form a real-time dataset covering the entire network. Logically, the real-time dataset is a multi-dimensional matrix, with rows corresponding to different swapping stations and their internal cell numbers, and columns corresponding to various operating parameters, thus reflecting the current operating status and spatial distribution characteristics of the entire network's cells. The real-time dataset is then fused with historical operating data from each swapping station stored in the cloud database, along with corresponding geographic coordinates. A spatiotemporal correlation analysis algorithm is used to identify the coupling relationships between different stations in terms of geographic space and historical operating patterns. The platform derives distance weighting coefficients between stations by calculating the spatial Euclidean distance, road travel time, and historical energy dispatch records. Simultaneously, it estimates energy transmission efficiency coefficients under different load conditions by combining transmission loss and efficiency curves from historical energy transmission records. The distance weighting coefficient reflects the spatial cost of load transfer between stations, while the energy transmission efficiency coefficient reflects the feasibility and economy of cross-site energy allocation. Based on the distance weighting coefficient and energy transmission efficiency coefficient, a network-wide battery consistency assessment is conducted, comparing the cell operating status of each swapping station with the network average to identify the degree of difference in cell health status. The platform calculates a cell health weight for each swapping station based on factors such as cell health status, historical degradation rate, and temperature stability; a higher weight indicates that the overall cell performance within the station is closer to its optimal state. Furthermore, the platform calculates the maximum power capacity parameter of each station by combining the number of cells, rated capacity, and historical peak output power, to characterize its potential load-bearing capacity in load distribution. A network-wide load allocation function is established based on cell health weights and maximum power capacity parameters. These parameters are used as core input variables, while distance weighting and energy transmission efficiency coefficients are introduced as constraint factors. This ensures the load allocation function considers multiple dimensions, including site performance, capacity resources, spatial distance, and transmission losses, when allocating load. The load allocation function outputs the optimal load allocation ratio for each site through weighted summation and normalization. Load oscillation analysis is performed based on the network-wide load allocation function to predict the temporal fluctuation trend of load levels at each battery swapping station and the overall network load stability under different scheduling strategies. The oscillation analysis employs a combination of frequency and time domain methods to identify low-frequency oscillation modes and assess their impact on network operational stability.When the analysis results show that certain scheduling schemes may cause periodic load fluctuations or local overload risks, the platform will adjust the parameter weights of the load distribution function and iteratively generate a balanced scheduling strategy.

[0084] In one example, by combining historical operational data and geographic coordinate information of each battery swapping station, a spatiotemporal correlation analysis is performed on the real-time dataset to obtain the distance weighting coefficient and energy transmission efficiency coefficient between stations, including:

[0085] Historical operation data is obtained by extracting charging and discharging cycle data, temperature change data, and time decay data of each battery swapping station from the cloud database.

[0086] Spatial matching of the geographical coordinates of each battery swapping station with historical operational data yields the physical distance matrix and operational correlation matrix between the stations.

[0087] Spatiotemporal correlation analysis was performed based on the physical distance matrix and the operational correlation matrix to obtain the distance weight coefficients between stations;

[0088] The energy transmission efficiency coefficient is calculated based on the cell health status and distance weighting coefficient of each battery swapping station in the real-time dataset.

[0089] In this example, the cloud-based intelligent control platform accesses and retrieves historical operating data from each battery swapping station stored in the cloud database. This historical operating data includes charge / discharge cycle data, temperature change data, and time decay data. Charge / discharge cycle data records the number of charge / discharge cycles, depth of charge, and discharge rate for each cell at different time periods, used to analyze the cell's capacity decay pattern and energy output stability. Temperature change data reflects the dynamic temperature changes of the cell due to variations in ambient temperature, workload, and heat dissipation conditions during operation. Time decay data characterizes the performance degradation trend of the cell over long-term use, including the annual decrease in capacity retention and the rate of increase in internal resistance over time. The geographical coordinates of each battery swapping station are spatially matched with the corresponding historical operating data. Geographical coordinates are obtained through positioning systems (such as GPS or BeiDou) to determine the physical location of the battery swapping station. The spatial matching process establishes a correspondence between each set of historical operating data and its corresponding geographical location, thereby creating a comprehensive data structure for the entire network of battery swapping stations that simultaneously possesses operational characteristics and spatial location attributes. Based on a comprehensive data structure, the physical distances between each battery swapping station are calculated, forming a physical distance matrix. Each element of the matrix represents the straight-line geographical distance or road travel distance between two battery swapping stations. Simultaneously, the platform calculates an operational correlation matrix based on the historical operational data of each station. The element values ​​of this matrix reflect the similarity of two stations in their operational characteristics, such as the matching degree of load curves, the overlap rate of peak charging and discharging periods, and the consistency of cell aging patterns. Spatiotemporal correlation analysis is performed based on the physical distance matrix and the operational correlation matrix, fusing the spatial dimension of physical distance with the temporal dimension of operational correlation. Through weighted combination and normalization, distance weight coefficients between stations are obtained. For example, stations with highly similar operational patterns do not require significant adjustments to their control strategies when sharing load, therefore their distance weight coefficients are relatively lower, increasing their priority in scheduling optimization. The distance weight coefficients are combined with the cell health status of each battery swapping station in the real-time dataset to calculate the energy transmission efficiency coefficient. Cell health status is comprehensively evaluated by multiple indicators such as voltage consistency, capacity retention rate, internal resistance level, and temperature stability, reflecting the station's ability to handle high-power transmission tasks in energy allocation. In calculating the energy transmission efficiency coefficient, the platform comprehensively considers the interaction between the distance weight coefficient and the health status: for sites with good health status and low distance weight coefficient, their energy transmission efficiency coefficient is assigned a higher value, indicating that lower loss and higher available power can be achieved in cross-site dispatch; for sites with poor health status or high distance weight coefficient, their efficiency coefficient is reduced to reflect that they should undertake fewer cross-site transmission tasks in scheduling optimization.

[0090] The study involves a spatiotemporal correlation analysis based on the physical distance matrix and the operational correlation matrix to obtain distance weighting coefficients between stations. This includes: normalizing the straight-line distance data between each battery swapping station in the physical distance matrix and calculating the actual reachable distance based on the urban road network and traffic conditions to obtain a corrected spatial distance relationship matrix; analyzing the synchronicity and complementarity of load changes of each battery swapping station during historical operation based on the operational correlation matrix, and quantifying the operational correlation degree by calculating the Pearson correlation coefficient and mutual information of load changes between stations to obtain quantitative indicators of operational correlation between stations; and weighting and fusing the spatial distance relationship matrix and the quantitative indicators of operational correlation. The weight ratios of spatial and temporal factors are determined using the analytic hierarchy process (AHP) and a comprehensive evaluation model of spatiotemporal correlation is established to obtain the spatiotemporal correlation assessment results between sites. Based on the spatiotemporal correlation assessment results, a fuzzy clustering algorithm is used to spatially partition and temporally group the battery swapping stations, identifying strongly correlated and weakly correlated site groups to obtain the spatiotemporal correlation partitioning results of the battery swapping network. Based on the spatiotemporal correlation partitioning results, the ease of load transfer within and between each site group is calculated, and the numerical allocation rules of the distance weight coefficient are determined in conjunction with the energy transmission loss model to obtain the distance weight coefficient reflecting the spatial layout and operational characteristics of the battery swapping network.

[0091] In one example, load oscillation analysis is performed based on the network-wide load distribution function to obtain a balanced scheduling strategy, including:

[0092] The real-time load value of each battery swapping station is calculated using the network-wide load distribution function, and the deviation between the real-time load value and the load reference value is analyzed to obtain the load droop coefficient.

[0093] The health droop coefficient is obtained by calculating the health status and health reference value of each battery swapping station. At the same time, the capacity droop coefficient is obtained by calculating the deviation between the capacity and capacity reference value of each battery swapping station.

[0094] The load droop coefficient, healthy droop coefficient, and capacity droop coefficient are combined and input into the three-dimensional droop control algorithm. A predictive damping factor is added to calculate the load oscillation suppression and obtain the oscillation control parameters.

[0095] Small-signal stability analysis was performed on the oscillation control parameters to obtain the stability analysis results. Based on the stability analysis results, the load areas of each battery swapping station were divided to obtain a balanced scheduling strategy.

[0096] In this example, the cloud-based intelligent control platform evaluates the real-time dataset using a network-wide load distribution function to obtain the real-time load value for each battery swapping station. It then performs continuous deviation analysis between the real-time load value and a pre-set load reference value for the operational period. The magnitude and duration of the deviation jointly characterize the current load deviation, thereby calculating the load droop coefficient to describe the load-side droop response sensitivity. The platform extracts the cell health status of each battery swapping station from the real-time dataset and compares it with the health reference value. A health status deviation metric is generated based on factors such as the rate of health status decline, temperature stability range, and consistency distribution. This health status deviation metric is mapped to a health droop coefficient. Simultaneously, the cloud-based intelligent control platform uses the available capacity, maximum allowable output capacity, and historical capacity recovery curves of each battery swapping station as constraints to calculate the deviation between the capacity and the capacity reference value, forming a capacity droop coefficient. These three types of coefficients together characterize the instantaneous adjustable space across the three dimensions of load, health, and capacity. The load droop coefficient, health droop coefficient, and capacity droop coefficient are vectorized and combined, and then input into the three-dimensional droop control algorithm. At the same time, a predictive damping factor learned from historical operating data is superimposed to offset overshoot and low-frequency fluctuations that may occur in the next time window. The resulting oscillation control parameters are used to correct the allocation weights and station power output limits in real time. The three channels of the three-dimensional droop control correspond to the load, capacity, and health dimensions, respectively. The predictive damping factor is used to introduce a feedforward vibration suppression term into the control law, enabling the scheduling to exhibit stronger adaptive and forward-looking characteristics across the entire network. The cloud-based intelligent control platform performs small-signal stability analysis on the oscillation control parameters. It assesses the stability margin of the control loop through state equations and eigenvalue locations. Combining the minimum requirement for the predictive damping factor and the network's low-frequency mode suppression capability in the 0.1–1 Hz range, it determines whether the current scheduling strategy might induce periodic load swings or localized coupled oscillations. If the analysis indicates insufficient stability margin, the cloud-based intelligent control platform rewrites the weighting coefficient and droop coefficient ranges in the load distribution function, forming a closed-loop parameter calibration process to ensure that the control loop poles are always located in the stable half-plane and that the vibration suppression effect meets the standards. After obtaining the stability analysis results, the cloud-based intelligent control platform divides the battery swapping stations into three operating areas—high-load area, balanced buffer zone, and low-load area—based on load deviation distribution, health status stratification, and capacity redundancy. The area division results, along with the inter-site distance weighting coefficient and energy transmission efficiency coefficient, are input into the network-wide load distribution function for an iterative update, outputting a balanced scheduling strategy that simultaneously satisfies stability constraints and efficiency targets.

[0097] This process involves extracting load prediction patterns and calculating predictive damping factors from historical load change data of each battery swapping station. This includes: extracting hourly load change data, daily load peak-to-valley difference data, and weekly load cycle pattern data for the past 30 days from a cloud database, and performing time-series standardization to obtain a standardized historical load dataset; using Fourier transform to analyze the frequency domain characteristics of load changes on the historical load dataset, identifying the main periodic components and harmonic characteristics of load changes, and analyzing the time domain regularity of load changes through autocorrelation functions to obtain the spectral characteristics and time-domain periodic patterns of load changes; and based on the spectral characteristics and time-domain periodic patterns of load changes... A mathematical model for load forecasting is established, and the ARIMA time series algorithm is used to fit the load change trend and extract the seasonal and trend factors of load change to obtain load forecasting regularity parameters. The load forecasting regularity parameters are input into a machine learning algorithm to identify load abrupt change patterns, analyze the triggering conditions and magnitude of load jump changes, and calculate the confidence interval and prediction error distribution of load forecasting to obtain the load forecasting accuracy evaluation index. Based on the load forecasting accuracy evaluation index and the oscillation suppression requirements of the battery swapping network, the optimal value of the predictive damping factor is calculated, and it is verified whether the value meets the damping factor threshold condition for system stability to obtain the predictive damping factor.

[0098] In one example, the load droop coefficient, healthy droop coefficient, and capacity droop coefficient are combined and input into a three-dimensional droop control algorithm, and a predictive damping factor is added to calculate load oscillation suppression, resulting in oscillation control parameters, including:

[0099] The load droop coefficient is used as the load deviation adjustment coefficient, the health droop coefficient is used as the cell health adjustment coefficient, and the capacity droop coefficient is used as the capacity adjustment coefficient.

[0100] By combining the load deviation adjustment coefficient, the cell health adjustment coefficient, and the capacity adjustment coefficient, a combination of adjustment coefficients is obtained.

[0101] Load prediction patterns are extracted from historical load change data of each battery swapping station, and predictive damping factors are calculated.

[0102] The predictive damping factor and the adjustment coefficient are combined and input into the three-dimensional droop control algorithm to perform three-dimensional droop control calculations, obtain the power adjustment amount of each battery swapping station, verify the oscillation suppression effect of the battery swapping network and confirm the stability conditions through the power adjustment amount, and obtain the oscillation control parameters.

[0103] In this example, the load droop coefficient, health droop coefficient, and capacity droop coefficient are each assigned a specific control role. The load droop coefficient, defined as the load deviation adjustment coefficient, directly reflects the deviation between the current load and the reference load of each battery swapping station in the scheduling calculation and determines the magnitude of output correction. The health droop coefficient, defined as the cell health adjustment coefficient, dynamically adjusts the priority and intensity of cell participation in load sharing based on the overall health status of each station's cells, thereby achieving a balance between extending battery life and maintaining balanced service capabilities. The capacity droop coefficient, defined as the capacity adjustment coefficient, reflects the proportional relationship between the available capacity of each battery swapping station and the rated capacity reference value at different time periods, thereby rationally allocating peak power and reserve capacity in network scheduling. The load deviation adjustment coefficient, cell health adjustment coefficient, and capacity adjustment coefficient are vectorized and combined to form a multi-dimensional adjustment coefficient combination. Simultaneously, the platform accesses historical load change data from each battery swapping station stored in a cloud database. This data covers load time-series changes under different seasons, weather conditions, and user travel patterns. Through time series analysis, spectral decomposition, and regression modeling, it extracts load prediction patterns, reflecting short- and medium-term load trends. A predictive damping factor is then generated through feature parameterization. This predictive damping factor pre-corrects the dynamic response of the control loop to suppress low-frequency oscillations and overshoot risks caused by load change inertia and network coupling effects. The predictive damping factor is combined with adjustment coefficients and input into a three-dimensional droop control algorithm for three-dimensional droop control calculations. The algorithm uses three control channels corresponding to three adjustment targets: load, health, and capacity. A corresponding adjustment coefficient is introduced into each channel for proportional control. Simultaneously, the predictive damping factor is superimposed as a feedforward vibration suppression term in the global part of the control law, pre-suppressing potential oscillations while performing power allocation calculations. The algorithm solves the coupled control equations and outputs the power adjustment amount for each battery swapping station. The platform verifies the oscillation suppression effect of the battery swapping network using power adjustment. The verification process involves reproducing the current network operating state in a simulation environment and evaluating the system's dynamic response curves and frequency domain characteristics under different load disturbance scenarios with the power adjustment applied, focusing on the attenuation of low-frequency modes in the 0.1Hz to 1Hz range. Stability is confirmed when the analysis results show that the power adjustment effectively suppresses load oscillations and maintains system response stability, and eigenvalue analysis shows that the poles of the control loop are all within the stable half-plane. The platform converts the verified control output into oscillation control parameters, including the target power adjustment value for each battery swapping station, the adjustment coefficient correction range, and the dynamic update frequency of the damping factor.

[0104] The process involves inputting a combination of predictive damping factors and adjustment coefficients into a three-dimensional droop control algorithm to perform three-dimensional droop control calculations. This yields the power adjustment amount for each battery swapping station. The power adjustment amount is used to verify the oscillation suppression effect of the battery swapping network and confirm the stability conditions, resulting in oscillation control parameters. These parameters include: constructing a three-dimensional droop control mathematical model by combining predictive damping factors and adjustment coefficients; establishing a power adjustment calculation equation set based on load deviation, health deviation, and capacity deviation terms to obtain the power balance calculation model for the battery swapping network; inputting the real-time load value, health status value, and capacity value of each battery swapping station into the power balance calculation model for solution calculation; and determining the power adjustment direction and adjustment magnitude based on the degree of deviation between each station and the reference value. The power adjustment values ​​of each battery swapping station are obtained. Based on the power adjustment values, a battery swapping network oscillation simulation model is established, and a low-frequency oscillation test signal of 0.1Hz to 1Hz is set. The network response characteristics and oscillation attenuation after power adjustment are analyzed through simulation calculation to obtain oscillation suppression effect evaluation data. The oscillation suppression effect evaluation data are compared and analyzed with the preset oscillation suppression standard, and the system characteristic value is calculated to verify whether the small signal stability condition meets the stability criterion requirements, and the system stability verification result is obtained. According to the system stability verification result, the range of the predictive damping factor is adjusted and the configuration of the three-dimensional droop control parameters is optimized to obtain oscillation control parameters that meet the oscillation suppression requirements and stability conditions.

[0105] In one example, the cloud-based intelligent control platform generates scheduling instructions for each battery swapping station based on a balanced scheduling strategy, including:

[0106] The cloud-based intelligent control platform identifies the load status characteristics of each battery swapping station based on the balanced scheduling strategy.

[0107] Based on load status characteristics, the corresponding operation mode is selected and the results are allocated for each battery swapping station from the following modes: battery health priority mode, load balancing priority mode, user distance priority mode, grid friendliness priority mode, cost optimization mode, or hybrid intelligent mode.

[0108] The equalization threshold angle is calculated based on the inconsistency data of the first cell of each battery swapping station in the operation mode allocation results, and the health degradation coefficient is calculated in combination with the average health status of each battery swapping station.

[0109] Dedicated control parameters for each battery swapping station are generated based on the balanced threshold angle and health decay coefficient.

[0110] Based on the dedicated control parameters, the level of coordination and scheduling of each battery swapping station is determined and the coordination relationship between the stations is established to obtain the scheduling level allocation scheme.

[0111] The operation mode allocation results, dedicated control parameters, and coordination relationships in the scheduling hierarchy allocation scheme are encapsulated into a standard scheduling protocol format to obtain the scheduling instructions for each battery swapping station.

[0112] In this example, based on a balanced scheduling strategy, the operational data of all battery swapping stations across the network is analyzed and filtered to extract key indicators reflecting the current load characteristics, forming a load state characteristic set for each station. This set includes the deviation between the real-time load value and the load reference value, as well as elements such as the state of charge distribution of the battery pack, discharge rate, historical load fluctuation curves, and user battery swapping demand density. The cloud platform, through its built-in mode selection logic, matches the load state characteristics of each station with the adaptation conditions of various operating modes. Among battery health priority mode, load balancing priority mode, user distance priority mode, grid-friendly priority mode, cost-optimal mode, and hybrid intelligent mode, it selects the mode that best matches the current operational needs of the station, generating the network-wide operating mode allocation result. The inconsistency data of the first battery cell in each station is extracted from the operating mode allocation result to quantify the performance differences between cells within the station, including multi-dimensional indicators such as voltage difference, capacity difference, and internal resistance difference. Based on a preset mathematical calculation model, the cloud platform converts the inconsistency data into a balancing threshold angle to determine when to trigger in-station balancing processing. Simultaneously, the platform calculates a health degradation coefficient based on the average health status of all battery cells within the station. This coefficient reflects the degree of protection for battery cell health during balancing operations and is used to control the intensity and duration of the balancing current. Based on the balancing threshold angle and the health degradation coefficient, and combined with the specific requirements of the operating mode, a dedicated set of control parameters is generated for each battery swapping station. This set includes control information such as the station's balancing triggering conditions, balancing execution strategy, power adjustment range, and collaborative scheduling priority. This enables each battery swapping station to efficiently perform balancing processing according to its own operating characteristics and the overall network scheduling strategy. Based on the dedicated control parameters of each battery swapping station, the cloud platform comprehensively considers the station's load-bearing capacity, geographical location, and energy transmission efficiency to determine the station's hierarchical level in the overall network coordination and scheduling system, and establishes a coordination relationship network between stations. The hierarchical system ranges from local two-station coordination to small-scale three-station coordination, then to larger-area four-station coordination, and finally to five-level linkage across the entire network. This allows for flexible adjustment of the scheduling scope and resource allocation strategy based on network conditions. The coordination network, by defining direct and backup collaborative links between stations, ensures that in the event of overload or failure at a particular station, other stations at the same or higher levels can quickly take over the load, achieving robust operation of the entire network. The operation mode allocation results in the scheduling hierarchy allocation scheme, the dedicated control parameters for each battery swapping station, and the coordination relationships between stations are encapsulated according to the standard scheduling protocol for battery swapping networks. During encapsulation, the platform constructs a standard data packet format containing the protocol version number, station identifier code, parameter field definitions, coordination relationship mapping table, and data integrity check code to obtain the scheduling instructions for each battery swapping station.

[0113] The process involves determining the coordination and scheduling levels of each battery swapping station based on dedicated control parameters and establishing coordination relationships between stations to obtain a scheduling level allocation scheme. This includes: analyzing the load carrying capacity and cell health level of each battery swapping station based on the balancing threshold angle and health decay coefficient in the dedicated control parameters, and calculating the station coordination complexity index in conjunction with the geographical distribution density of the battery swapping stations to obtain a battery swapping station coordination capability evaluation matrix; inputting the battery swapping station coordination capability evaluation matrix into a mode selection decision algorithm and performing mode matching analysis based on the current network load distribution status to determine the optimal operating mode for each battery swapping station from battery health priority mode, load balancing priority mode, user distance priority mode, grid friendliness priority mode, cost-optimal mode, and hybrid intelligent mode, thus obtaining a dynamic mode allocation result. Based on the operating mode type and coordination capability assessment data of each battery swapping station in the dynamic mode allocation results, the scheduling hierarchy is calculated. By analyzing the physical distance, load correlation, and communication delay between stations, the hierarchical affiliation of two-station coordination, three-station coordination, four-station coordination, and five-level linkage is determined, resulting in a hierarchical coordinated scheduling configuration. Based on the hierarchical coordinated scheduling configuration, different decision weights and communication protocol types are assigned to each scheduling level, and hierarchical command relationships and peer coordination mechanisms are established between levels, resulting in a hierarchical scheduling architecture that includes weight allocation and coordination relationships. The mode allocation, hierarchical affiliation, and coordination relationships in the hierarchical scheduling architecture are verified in real time, and the mode and hierarchical configuration of each battery swapping station are dynamically adjusted according to the load changes of the battery swapping network, resulting in a scheduling hierarchy allocation scheme.

[0114] In one example, after receiving scheduling instructions, the hardware modules of each battery swapping station perform multi-level load balancing processing to obtain the load redistribution results, including:

[0115] Each battery swapping station's hardware module receives scheduling instructions from the cloud-based intelligent control platform and parses the station's dedicated load balancing configuration parameters;

[0116] Based on the equalization execution configuration parameters, detect the inconsistency data of the second battery cell and compare it with the equalization threshold angle to determine the equalization processing status;

[0117] Based on the balanced processing status, establish communication links with adjacent battery swapping stations and perform inter-station load coordination according to the hierarchical relationship of two-station coordination, three-station coordination, four-station coordination or five-level linkage to obtain the load coordination processing result.

[0118] The load coordination process is used to complete the load transfer and energy redistribution operations between sites in the battery swapping network, and the load redistribution results are obtained.

[0119] In this example, each battery swapping station's hardware module receives dispatch instructions through an encrypted communication channel established with the cloud-based intelligent control platform. The instructions are then structured and parsed to extract customized equalization execution configuration parameters tailored to the station's operational status. These parameters include the equalization threshold angle, health degradation coefficient, target equalization current, power adjustment range, and the hierarchical identifier for inter-station collaboration. The parsing process verifies the integrity and version of the parameter fields to ensure compatibility with the current station's hardware and software versions, thus preventing control anomalies caused by parameter mismatches. Based on the equalization execution configuration parameters, the hardware module, through its local monitoring unit, performs second-cell inconsistency data detection on the station's battery pack according to the detection cycle and accuracy requirements specified in the instructions. This second-cell inconsistency data reflects voltage, capacity, and internal resistance differences between different cells within the current operating cycle. The detection results undergo data filtering and anomaly removal. The second-cell inconsistency data is compared to the equalization threshold angle. If the data exceeds the threshold angle, equalization processing is initiated; otherwise, monitoring remains in place without starting the equalization process. After establishing a balanced processing state, the hardware module initiates the communication link establishment process with adjacent battery swapping stations based on the coordination level information in the scheduling instructions. The communication link employs a low-latency, high-reliability transmission protocol and performs bidirectional handshakes and encryption authentication during the link establishment phase to ensure the security and consistency of cross-station data exchange. Depending on the coordination mode, the communication topology between stations differs: in the two-station coordination mode, the two stations directly establish a point-to-point data link to achieve bidirectional load and energy exchange; in the three-station coordination mode, a relay station is introduced, and multi-directional load allocation is achieved through a triangle structure; in the four-station coordination mode, a star topology is used, with the station with the lowest load acting as the coordination center for energy allocation; and in the five-level linkage mode, a hierarchical, full-network collaborative network is constructed, achieving wide-area load balancing through multi-level allocation by central nodes and regional nodes. After establishing inter-site communication links and entering the corresponding collaborative mode, each participating site performs inter-site load coordination calculations based on real-time collected load information, available battery capacity, and energy transmission efficiency parameters. This results in a load coordination process, including the number of battery swapping requests to be transferred, energy transmission paths, transmission power, and execution timing. Transmission losses and delays are also estimated to optimize scheduling. The load coordination results are updated synchronously among the participating sites. Based on the load coordination results, the hardware module performs load transfer and energy redistribution operations at the physical level, including dynamic adjustment of battery pack charging and discharging power, intelligent diversion of battery swapping requests, and real-time control of energy transmission across sites. During execution, changes in cell temperature, voltage, and balancing current are continuously monitored, and protection mechanisms are triggered promptly to ensure operational safety in case of anomalies.Once all transfer and allocation operations are completed, the new load distribution and site operating status are recorded as the load redistribution results and transmitted back to the cloud-based intelligent control platform via the communication link.

[0120] In one example, based on the equalization execution configuration parameters, inconsistency data of the second cell is detected and compared with the equalization threshold angle to determine the equalization processing status, including:

[0121] The equalization threshold angle is parsed from the equalization execution configuration parameters and the maximum voltage difference and state of charge difference between each cell are collected synchronously to obtain the second cell inconsistency data.

[0122] The inconsistency data of the second cell is compared with the equalization threshold angle. When the inconsistency data of the second cell exceeds the trigger value range, an equalization start trigger signal is generated.

[0123] The intelligent switch matrix in the equalization execution unit is awakened by the equalization start-up trigger signal, and the PWM duty cycle control parameters are extracted from the health decay coefficient to obtain the current intensity adjustment command that matches the health status of the battery cell.

[0124] The system executes current intensity adjustment commands to control the direction and magnitude of the equalization current between each cell, while simultaneously monitoring temperature fluctuations and voltage change trends during the equalization process to obtain the equalization status.

[0125] In this example, the local instruction parsing unit parses the equalization threshold angle from the equalization execution configuration parameters to determine whether the current cell consistency status requires entering the equalization process. Simultaneously, the hardware module activates the synchronous sampling channel connected to the BMS acquisition line to perform high-precision synchronous acquisition of real-time operating parameters for all cells within the station. It extracts the maximum voltage difference and state-of-charge difference between cells within the current sampling period and generates second-cell inconsistency data through difference calculation and timestamp alignment. This second-cell inconsistency data is compared with the equalization threshold angle. If the comparison shows that the second-cell inconsistency data exceeds the trigger value range set by the equalization threshold angle, equalization processing is determined to be required, and the control logic generates an equalization start trigger signal. This signal wakes up the intelligent switch matrix within the equalization execution unit. The intelligent switch matrix establishes a controllable energy transfer path between different cells, and its switching state is dynamically adjusted by the controller to allocate equalization current based on the voltage levels and state-of-charge differences of different cells. During this process, PWM duty cycle control parameters matching the current cell health status are extracted from the health decay coefficient field in the equalization execution configuration parameters. The PWM duty cycle control parameter automatically adjusts the intensity of the balancing current based on the cell's State of Health (SOH). For example, for cells with weaker health, the duty cycle is reduced to decrease the balancing current intensity, thereby reducing heat load and loss risk; while for cells with better health, the duty cycle is appropriately increased to accelerate the balancing process. After obtaining the PWM duty cycle control parameter, the hardware module generates corresponding current intensity adjustment commands, which are directly used to control the drive signal of the intelligent switch matrix and the output waveform of the PWM control module, thereby controlling the direction and magnitude of the balancing current between cells. Simultaneously, a full-process monitoring mode is activated to collect and analyze the cell temperature fluctuations and voltage change trends in real time during the balancing process. Temperature monitoring ensures that the temperature rise of any cell or connecting component does not exceed a safe threshold during the balancing process, while voltage change monitoring is used to dynamically assess the balancing progress and determine whether the balancing target is nearing completion. When an excessively rapid temperature rise or a voltage difference narrowing to a preset range is detected, the balancing current intensity is adjusted promptly according to a safety strategy. The temperature change curves, voltage change curves, actual balancing current waveforms, and balancing duration recorded during the balancing process are compiled to form balancing processing status data.

[0126] Reference Figure 2 This embodiment provides a load balancing and optimization scheduling system for a battery swapping network, including:

[0127] The real-time acquisition subsystem 201 is used by the hardware modules of each battery swapping station to collect real-time data on battery cell voltage, temperature, state of charge, and health status through the BMS acquisition line, thereby obtaining battery cell status data.

[0128] The multi-site fusion analysis subsystem 202 is used by the cloud-based intelligent control platform to perform multi-site fusion analysis on the cell status data to obtain a balanced scheduling strategy.

[0129] The generation subsystem 203 is used by the cloud-based intelligent control platform to generate scheduling instructions for each battery swapping station according to the balanced scheduling strategy.

[0130] The multi-level load balancing subsystem 204 is used by each battery swapping station hardware module to perform multi-level load balancing after receiving scheduling instructions, and to obtain the load redistribution result.

[0131] Figure 2 This is a schematic diagram of the overall architecture of the battery swapping network load balancing and optimization scheduling system of the present invention. Figure 2 As shown, the load balancing and optimization scheduling system for the battery swapping network includes: a cloud-based intelligent control platform: located at the upper layer of the system, responsible for network-wide data processing and scheduling decisions, including the following modules: a multi-site fusion analysis subsystem 202: used to receive cell status data transmitted from each battery swapping station, construct a real-time dataset containing cell parameters from all stations, perform spatiotemporal correlation analysis based on historical operating data and geographic coordinate information, and establish a network-wide load allocation function. a generation subsystem 203: used to generate scheduling instructions for each battery swapping station according to the balanced scheduling strategy, including operating mode allocation, dedicated control parameter configuration, and scheduling level allocation. a load oscillation analysis module: used to perform load oscillation analysis based on the network-wide load allocation function, employ a three-dimensional droop control algorithm for oscillation suppression calculation, and generate a balanced scheduling strategy. a cloud database: storing historical operating data such as charge / discharge cycle data, temperature change data, and time decay data from each battery swapping station. an AI algorithm engine: providing deep learning algorithms and intelligent analysis functions for network-wide battery consistency assessment and predictive scheduling optimization. a 5G / 4G wireless communication network: located at the middle layer of the system, providing a data transmission channel between the cloud-based intelligent control platform and the hardware modules of each battery swapping station. Battery swapping stations A, B, C, and N represent hardware module groups of battery swapping stations distributed in different geographical locations. Each battery swapping station includes: Real-time acquisition subsystem 201: containing BMS acquisition lines, used to synchronously acquire cell voltage, temperature, state of charge, health status, and internal resistance, generating cell status data. Multi-level balancing processing subsystem 204: used to receive cloud scheduling instructions and perform multi-level balancing processing, including balancing execution units, intelligent switch matrices, and PWM control technology. Communication interface module: responsible for data transmission, uploading cell status data to the cloud platform, and receiving and executing scheduling instructions issued by the cloud. Connection relationship description: Solid arrows indicate the main data flow for data upload and scheduling instruction issuance; dashed lines indicate coordination connections between battery swapping stations, used to achieve multi-level scheduling of two-station collaboration, three-station collaboration, four-station collaboration, or five-level linkage; arrows inside the cloud platform indicate the data processing flow between each subsystem.

[0132] In this embodiment, the specific implementation of each unit in the above system embodiment is described in the above method embodiment, and will not be repeated here.

[0133] In this embodiment of the invention, a cloud-based intelligent control platform is used to fuse and analyze the cell status data of multiple battery swapping stations, constructing a network-wide load distribution function. This overcomes the limitations of traditional technologies that can only handle the balancing within a single device, achieving coordinated balancing and load redistribution among multiple stations in the battery swapping network. A three-dimensional droop control algorithm, combined with a predictive damping factor, effectively suppresses load oscillations in the battery swapping network. Small-signal stability analysis ensures stable system operation under dynamic load changes. A multi-level scheduling architecture, from two-station collaboration to five-level linkage, is established. The scheduling level is dynamically configured based on the geographical distribution and load characteristics of the battery swapping stations, achieving hierarchical load management from local optimization to network-wide coordination. Deep learning algorithms on the cloud platform analyze historical operating data and real-time status, constructing a battery aging prediction model and a load prediction function. Based on different application scenarios and operational needs, it provides battery health priority, load balancing priority, user distance priority, grid friendliness priority, cost optimization, and hybrid intelligent modes, enhancing the system's adaptability and flexibility. Through dynamic adjustment of the balancing threshold angle and health decay coefficient, combined with real-time status feedback and cloud-based optimization adjustment, adaptive optimization of the scheduling strategy and intelligent adjustment of system parameters are achieved. By working together with the cloud-based intelligent control platform and the battery swapping station hardware modules, complex algorithm calculations are moved to the cloud, simplifying the functions of local hardware modules and reducing system deployment costs and maintenance difficulties.

[0134] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, system, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, system, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, system, article, or method that includes that element.

[0135] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Any equivalent structural or procedural transformations made based on the content of the present invention specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of the present invention.

Claims

1. A method for load balancing and optimized scheduling of a battery swapping network, characterized in that, include: Each battery swapping station's hardware modules collect real-time data on cell voltage, temperature, state of charge, and health status via the BMS acquisition line, thus obtaining cell status data. The cloud-based intelligent control platform performs multi-site fusion analysis on the battery cell status data to obtain a balanced scheduling strategy. Specifically, this includes: the cloud-based intelligent control platform receiving the battery cell status data transmitted from each battery swapping station and constructing a real-time dataset containing battery cell parameters from all stations based on the battery cell status data; combining historical operating data and geographic coordinate information from each battery swapping station, performing spatiotemporal correlation analysis on the real-time dataset to obtain distance weighting coefficients and energy transmission efficiency coefficients between stations; conducting a network-wide battery consistency assessment based on the distance weighting coefficients and energy transmission efficiency coefficients to obtain the battery cell health weights and maximum power capacity parameters for each battery swapping station; establishing a network-wide load allocation function based on the battery cell health weights and maximum power capacity parameters; and performing load oscillation analysis based on the network-wide load allocation function to obtain a balanced scheduling strategy. The cloud-based intelligent control platform generates scheduling instructions for each battery swapping station based on the aforementioned balanced scheduling strategy. After receiving the scheduling instructions, the hardware modules of each battery swapping station perform multi-level load balancing processing to obtain the load redistribution results.

2. The load balancing and optimization scheduling method for a battery swapping network according to claim 1, characterized in that, The hardware modules of each battery swapping station collect real-time data on cell voltage, temperature, state of charge, and health status via the BMS acquisition line, obtaining cell status data, including: Each battery swapping station's hardware modules synchronously collect data on cell voltage, temperature, state of charge, health status, and internal resistance via the BMS acquisition line to obtain raw cell data. The original cell data is filtered to obtain a set of cell parameters; The site load index is calculated based on the set of battery cell parameters, and a unique site identifier and geographic coordinate information are embedded to obtain the identified battery cell data. The identified cell data is encapsulated into a data packet format that conforms to the battery swapping network communication protocol to obtain cell status data.

3. The load balancing and optimization scheduling method for battery swapping networks according to claim 1, characterized in that, The method combines historical operational data and geographic coordinate information from each battery swapping station to perform spatiotemporal correlation analysis on the real-time dataset, obtaining distance weighting coefficients and energy transmission efficiency coefficients between stations, including: Historical operation data is obtained by extracting charging and discharging cycle data, temperature change data, and time decay data of each battery swapping station from the cloud database. The geographical coordinates of each battery swapping station are spatially matched with the historical operation data to obtain the physical distance matrix and operation correlation matrix between the stations. Spatiotemporal correlation analysis is performed based on the physical distance matrix and the operational correlation matrix to obtain the distance weight coefficients between stations; The energy transmission efficiency coefficient is calculated based on the cell health status of each battery swapping station in the real-time dataset and the distance weighting coefficient.

4. The load balancing and optimization scheduling method for battery swapping networks according to claim 3, characterized in that, The load oscillation analysis based on the network-wide load distribution function to obtain the balanced scheduling strategy includes: The real-time load value of each battery swapping station is calculated using the network-wide load distribution function, and the deviation between the real-time load value and the load reference value is analyzed to obtain the load droop coefficient. The health droop coefficient is obtained by calculating the health status and health reference value of each battery swapping station. At the same time, the capacity droop coefficient is obtained by calculating the deviation between the capacity and capacity reference value of each battery swapping station. The load droop coefficient, the healthy droop coefficient, and the capacity droop coefficient are combined and input into the three-dimensional droop control algorithm, and a predictive damping factor is added to calculate the load oscillation suppression, thereby obtaining the oscillation control parameters. Small-signal stability analysis is performed on the oscillation control parameters to obtain stability analysis results. Based on the stability analysis results, load areas are divided for each battery swapping station to obtain a balanced scheduling strategy.

5. The load balancing and optimization scheduling method for a battery swapping network according to claim 4, characterized in that, The process involves combining the load droop coefficient, the healthy droop coefficient, and the capacity droop coefficient, inputting them into a three-dimensional droop control algorithm, and adding a predictive damping factor to perform load oscillation suppression calculations to obtain oscillation control parameters, including: The load droop coefficient is used as the load deviation adjustment coefficient, the health droop coefficient is used as the cell health adjustment coefficient, and the capacity droop coefficient is used as the capacity adjustment coefficient. The load deviation adjustment coefficient, the cell health adjustment coefficient, and the capacity adjustment coefficient are combined to obtain the adjustment coefficient combination; Load prediction patterns are extracted from historical load change data of each battery swapping station, and predictive damping factors are calculated. The predictive damping factor and the adjustment coefficient are combined and input into the three-dimensional droop control algorithm to perform three-dimensional droop control calculations, thereby obtaining the power adjustment amount of each battery swapping station. The oscillation suppression effect of the battery swapping network is verified and the stability conditions are confirmed through the power adjustment amount, and the oscillation control parameters are obtained.

6. The load balancing and optimization scheduling method for a battery swapping network according to claim 1, characterized in that, The cloud-based intelligent control platform generates scheduling instructions for each battery swapping station based on the balanced scheduling strategy, including: The cloud-based intelligent control platform identifies the load status characteristics of each battery swapping station based on the aforementioned balanced scheduling strategy; Based on the load status characteristics, the corresponding operation mode is selected and the results are allocated for each battery swapping station from the following modes: battery health priority mode, load balancing priority mode, user distance priority mode, grid friendliness priority mode, cost optimization mode, or hybrid intelligent mode. The corresponding equalization threshold angle is calculated based on the inconsistency data of the first cell of each battery swapping station in the operation mode allocation result, and the health degradation coefficient is calculated in combination with the average health status of each battery swapping station. Based on the equilibrium threshold angle and the health decay coefficient, specific control parameters are generated for each battery swapping station. Based on the dedicated control parameters, the level of coordination and scheduling of each battery swapping station is determined and the coordination relationship between stations is established to obtain the scheduling level allocation scheme. The operation mode allocation results, dedicated control parameters, and coordination relationships in the scheduling hierarchy allocation scheme are encapsulated into a standard scheduling protocol format to obtain the scheduling instructions for each battery swapping station.

7. The load balancing and optimization scheduling method for a battery swapping network according to claim 1, characterized in that, After receiving the scheduling command, the hardware modules of each battery swapping station perform multi-level load balancing processing to obtain the load redistribution results, including: Each battery swapping station's hardware module receives the scheduling instructions issued by the cloud-based intelligent control platform and parses the station's dedicated load balancing configuration parameters; Based on the equalization execution configuration parameters, detect the inconsistency data of the second cell and compare it with the equalization threshold angle to determine the equalization processing status; Based on the balanced processing status, establish communication links with adjacent battery swapping stations and perform inter-station load coordination according to the hierarchical relationship of two-station coordination, three-station coordination, four-station coordination or five-level linkage to obtain the load coordination processing result. The load coordination processing results are used to complete the load transfer and energy redistribution operations between stations in the battery swapping network, resulting in a load redistribution result.

8. The load balancing and optimization scheduling method for a battery swapping network according to claim 7, characterized in that, The step of detecting inconsistency data of the second cell based on the equalization execution configuration parameters and comparing it with the equalization threshold angle to determine the equalization processing status includes: The equalization threshold angle is parsed from the equalization execution configuration parameters and the maximum voltage difference and state of charge difference between each cell are collected synchronously to obtain the second cell inconsistency data. The second cell inconsistency data is compared with the equalization threshold angle. When the second cell inconsistency data exceeds the trigger value range, an equalization start trigger signal is generated. The intelligent switch matrix in the equalization execution unit is awakened according to the equalization start trigger signal, and the PWM duty cycle control parameters are extracted from the health decay coefficient to obtain the current intensity adjustment command that matches the health status of the battery cell. The current intensity adjustment command is executed to control the direction and magnitude of the equalization current between each cell, while simultaneously monitoring temperature fluctuations and voltage change trends during the equalization process to obtain the equalization processing status.

9. A load balancing and optimization scheduling system for a battery swapping network, characterized in that, The steps for implementing the load balancing and optimization scheduling method for the battery swapping network according to any one of claims 1 to 8 include: The real-time acquisition subsystem is used by the hardware modules of each battery swapping station to collect real-time data on battery cell voltage, temperature, state of charge, and health status through the BMS acquisition line, thereby obtaining battery cell status data. The multi-site fusion analysis subsystem is used by the cloud-based intelligent control platform to perform multi-site fusion analysis on the cell status data to obtain a balanced scheduling strategy. A generation subsystem is used by the cloud-based intelligent control platform to generate scheduling instructions for each battery swapping station according to the balanced scheduling strategy. The multi-level load balancing subsystem is used by each battery swapping station hardware module to perform multi-level load balancing after receiving the scheduling instructions, and obtain the load redistribution result.

Images