A flexible regulation system for super-fast charging interface based on spatial distribution cooperation

Abstract

The application discloses a kind of based on spatial distribution cooperation's ultrafast charging interface flexible regulation system, including data acquisition unit, multi-source data fusion unit, load assessment and balance calculation unit, regulation strategy generation unit, distributed cooperative control unit, central coordination unit, execution control unit and state feedback and optimization unit.The device realizes multilevel operating state perception by multi-source data fusion, realizes power balance and cooperative optimization inside and outside the region using distributed autonomous node and central coordination mechanism, and combines closed-loop feedback to realize strategy adaptive correction.The system can maintain power supply stability and energy distribution balance under complex load fluctuation conditions, and is suitable for real-time flexible regulation of ultrafast charging facilities.

Description

Technical Field

[0001] This invention relates to the field of power system automation and intelligent control technology, specifically to an ultra-fast charging interface flexible control system based on spatially distributed collaborative multi-source data fusion and load balancing algorithm, which falls within the technical application scope of new energy charging infrastructure management, distribution network intelligent control, and distributed energy management system (EMS). Background Technology

[0002] The rapid increase in ultra-fast charging facilities in public power distribution networks brings about conditions characterized by high instantaneous power demand, rapid ramp-up, and strong simultaneity, which can easily lead to feeder voltage fluctuations, uneven branch currents, and localized heat load accumulation. The industry typically relies on existing distribution automation and station-level management systems for monitoring and control to improve access capacity and operational safety. However, due to limitations in measurement point distribution, measurement granularity, and communication bandwidth, the existing network has limited ability to identify and finely adjust rapid fluctuations in ultra-fast charging scenarios, making it difficult to reflect the true load status of lines and potential overload risks in a timely manner.

[0003] Existing engineering practices often employ a hierarchical, centralized control architecture: the upper-level platform performs load assessment and power allocation based on aggregated operating parameters, while lower-level equipment executes basic power limiting and queuing scheduling strategies according to instructions. Under this approach, parameter tuning largely relies on static thresholds or empirical coefficients, with control cycles primarily on the order of minutes. Faced with the suddenness and uncertainty of vehicle access, this often results in response lag and coarse scheduling. Furthermore, local optimization is frequently performed based on single stations or feeders, leading to insufficient cross-regional coordination capabilities and a tendency to create a phenomenon of "local relief, global imbalance."

[0004] On the data side, conventional systems primarily rely on measurement and operational data from each platform, often resulting in inconsistencies in data caliber, time base, and quality control. To ensure availability, engineering practices typically employ standardized sampling periods, simple filtering, and regularization to align the data. However, in highly dynamic scenarios, these methods can introduce latency and information attenuation, reducing the sensitivity of load assessment to short-term peak values ​​and phase differences. This misjudgment of the status can then propagate to subsequent power allocation and scheduling processes, impacting control effectiveness.

[0005] In terms of strategy generation, a common approach is to use heuristic or rule-based power allocation and charging priority control based on indicators such as feeder margin, transformer load rate, and user service level. While these methods are simple to implement and easy to deploy, they have limited adaptability to external factors such as equipment temperature rise, weather disturbances, and changes in vehicle state of charge. If external conditions change rapidly, existing strategies require manual intervention or conservative derating, leading to a passive reduction in available capacity and putting pressure on charging efficiency and user experience.

[0006] At the execution level, traditional solutions typically involve station-level controllers or platforms issuing unified commands, with terminals executing according to preset modes. This lacks fine-grained feedback and rapid correction of execution results. When communication links experience jitter or brief interruptions, terminals often enter a fixed safety mode, leading to rigid power allocation and decreased resource utilization. During high-concurrency periods, this vulnerability in execution-communication coupling is even more pronounced, potentially causing load imbalances to propagate upstream and increasing the risk of malfunctions and over-limit actions by protection devices.

[0007] In terms of closed-loop operation, the current system relies more on periodic reports and offline analysis to evaluate the control effect. Online optimization is often mainly based on threshold fine-tuning, lacking adaptive updates that couple historical execution with real-time status. As the scale of the station cluster expands and traffic flow fluctuations intensify, relying on manual experience and fixed parameters makes it difficult to balance safety margins and supply efficiency, easily leading to either overly conservative or insufficient control.

[0008] In summary, traditional technical approaches still have shortcomings in areas such as data consistency, timeliness of state identification, agility of strategy generation, robustness of execution, and self-optimization of the closed-loop operation. Furthermore, this approach suffers from issues such as response lag, insufficient coordination between local optima and global optimization, and limited fault tolerance to anomalies and communication disturbances. These problems may lead to feeder overruns, equipment overheating, service quality fluctuations, and decreased network stability under high concurrency and strong disturbance conditions in ultra-fast charging, thus hindering the safe, efficient access and widespread application of large-scale ultra-fast charging facilities. Summary of the Invention

[0009] To address the aforementioned problems in the prior art, this invention proposes a flexible control system for an ultra-fast charging interface based on spatial distribution coordination. This system includes: The data acquisition unit is used to collect operating parameter data from charging facilities, station-level management systems and power distribution control systems, and transmit the data to the control and management platform; The multi-source data fusion unit is used to receive and fuse data from different sources, and form a standardized multi-source measurement dataset based on time synchronization and data consistency processing. The load assessment and balance calculation unit is used to calculate the load balance status and overload risk level of the distribution line based on the multi-source measurement data set, and generate corresponding load control targets. The regulation strategy generation unit is used to generate a flexible power regulation strategy for constraining and guiding subsequent power allocation based on the load regulation target and in combination with system operation constraints and equipment status information. The distributed collaborative control unit includes multiple spatial autonomous nodes. Each autonomous node is equipped with a local load assessment module and a self-adjustment execution module. The local load assessment module and the self-adjustment execution module are used to negotiate and generate a regional power adjustment strategy within their respective regions based on a neighborhood information exchange mechanism. The central coordination unit is used to receive the regional regulation results from each autonomous node and to revise and coordinate the regulation strategies of each region based on the global load balance and overload risk status. The execution control unit is used to perform hierarchical control and dynamic allocation of the output power of each charging terminal according to the power adjustment strategy generated by the central coordination unit and each autonomous node; A status feedback and optimization unit is used to collect execution feedback data from each autonomous node and charging terminal, compare the load status before and after regulation, and automatically update the parameters of the regional regulation strategy based on the operational deviation of the load status between the central coordination unit and the distributed collaborative control unit. The data acquisition unit is located at the front-end data interface layer of the device and is used to establish data communication channels with charging facilities, station-level management systems, and power distribution control systems, and to collect operating parameter data. The data acquisition unit includes: A multi-protocol adapter module is used to ensure compatibility with different system communication protocols and to enable structured data input. The time reference alignment module is used to time-mark and synchronize the collected data based on a unified clock source; The data verification module is used to perform redundancy checks, logical range checks, and abnormal data processing on the collected data. The verified data is transmitted to the multi-source data fusion unit via the internal bus for subsequent fusion and calculation.

[0010] The multi-source data fusion unit is located between the data acquisition layer and the computing layer, and is used to perform unified processing on data from different systems to form a standardized measurement dataset. The multi-source data fusion unit includes: The format parsing submodule is used to perform format normalization and unit unification processing on data from different sources; The time alignment submodule is used to synchronize the collected data based on timestamp information to eliminate sampling frequency differences; The consistency processing submodule is used to perform cross-source consistency checks and logical integrity verifications on the synchronized data; The processed standardized measurement dataset is output to the load assessment and balancing calculation unit for subsequent calculations.

[0011] The load assessment and balancing calculation unit is located in the analysis and decision-making layer of the device. It is used to assess the operating load status, balance degree and overload risk of the power distribution line based on the standardized measurement dataset output by the multi-source data fusion unit, and generate load control targets. The load assessment and balancing calculation unit includes: The line condition assessment submodule is used to calculate the voltage deviation, branch load rate and heat load index of each node, and obtain the line balance index according to the weighting method. The target generation submodule is used to calculate the target power adjustment value of each controlled node based on the line status assessment results and preset operating constraints, so as to form a quantified load control target. The load control target is used to provide input parameters for the control strategy generation unit.

[0012] The regulation strategy generation unit is located in the decision control layer of the device and is used to generate a flexible power regulation strategy based on the load regulation target output by the load assessment and balance calculation unit, combined with system operation constraints and equipment status information. The regulation strategy generation unit includes: The target analysis module is used to analyze load regulation targets and generate node power setpoints; The constraint adaptation module is used to perform feasibility verification and correction of the power setting value based on the system operation constraints. The strategy orchestration module is used to generate a structured power regulation strategy set containing power target range, deviation tolerance and execution order based on the corrected power set value, and output the strategy set to the lower control module for execution; The control strategy generation unit can periodically adjust the strategy parameters based on feedback information.

[0013] The distributed collaborative control unit is located in the middle control layer of the system and is used to realize regional autonomous decision-making and cross-regional collaborative regulation. The distributed collaborative control unit includes multiple spatial autonomous nodes, each autonomous node comprising: The local load assessment module is used to calculate the power distribution, load rate and voltage deviation within the region based on the standardized measurement data output by the multi-source data fusion unit, and generate a region operation status vector. The self-adjusting execution module is used to execute local strategy solutions based on the upper-level target interval and boundary constraints, and output the regional power adjustment results. Among them, the multiple autonomous nodes share boundary state parameters and perform strategy negotiation through a neighborhood information exchange mechanism to make bidirectional adjustments to power allocation based on the detection results of boundary power difference or voltage difference, so as to achieve coordinated convergence between regions. After the negotiation is completed, a set of regional power adjustment strategies is formed and the results are reported to the central coordination unit for global optimization.

[0014] The central coordination unit is located in the global control layer of the system. It is used to summarize, analyze and macro-optimize the operation results of each region based on the regional control completed by the distributed autonomous nodes, so as to maintain the global load balance and system safety margin. The central coordination unit includes: The data aggregation module is used to receive policy summary information and operational status data uploaded from autonomous nodes in various regions, and to perform data integrity checks and time synchronization. The global assessment module is used to calculate the overall load balance and overload risk distribution of the system based on multi-regional operational data, and to identify key areas that affect the stability of the entire network. The strategy coordination module is used to make macro-level corrections and coordination of regional strategies based on the global assessment results, calculate cross-regional power transfer recommendation values, and generate an optimized global power allocation scheme under the condition of meeting line capacity and voltage fluctuation constraints.

[0015] The strategy coordination module calculates the cross-regional power transfer recommendation value using the following formula for generating the coordinated power transfer recommendation amount:

[0016] in: Indicates from the region To the region Coordination recommendations for power transfer amount; This is the upper bound of the transferable power of this region in the current cycle; This is the regional coordination gain coefficient, used to adjust the transfer amplitude; , This serves as an indicator of the balance between the two regions. These are measured values ​​of cross-regional tidal current. This is the global power flow limit for the system. , Regional load factor; It is a sensitive factor for differences in balance.

[0017] The execution control unit is located in the terminal control layer of the system and is used to perform hierarchical control and dynamic power allocation of the charging terminal according to the power adjustment strategy generated by the central coordination unit and the distributed autonomous nodes. The execution control unit includes: The policy receiving module is used to receive and verify the power adjustment command packets sent down from the upper layer; The instruction parsing module is used to convert upper-layer policies into control instructions that the terminal can recognize, and adjust the target power within a safe range according to the terminal's operating status. The power control module is used to execute power adjustment instructions in a hierarchical manner according to policy priority, and to perform power fine-tuning based on terminal feedback information to reduce deviation; The continuity protection module is used to execute sequence preservation and timeout protection mechanisms in the event of communication interruption or timeout, so as to maintain the terminal power continuously within a safe range; The execution control unit performs self-testing and statistical operations in each control cycle and feeds back the execution results to the upper-level module to achieve closed-loop optimization of the control strategy.

[0018] The state feedback and optimization unit is set in the closed-loop control layer of the system and is used to evaluate the execution results and optimize the strategy after each control cycle. The state feedback and optimization unit includes: The data acquisition module is used to receive operational feedback information from the distributed collaborative control unit and the execution control unit, and to align the multi-source feedback data through a timestamp synchronization mechanism; The deviation assessment module is used to compare the actual execution results with the target power distribution, calculate the operating deviation, and assess the global and regional load balance status of the system. The parameter update module is used to dynamically correct the key control parameters of the system based on the deviation evaluation results. It adopts an adaptive adjustment mechanism to adjust the correction magnitude or step size according to the deviation accumulation trend in order to achieve stable convergence and prevent over-response. The state feedback and optimization unit feeds back the updated parameter set and performance evaluation results to the multi-source data fusion unit and the load assessment and balancing calculation unit as input for the next control cycle, thereby realizing continuous optimization and self-learning of the system.

[0019] Beneficial effects: This system achieves data connectivity between charging facilities, station-level systems, and power distribution control systems through multi-source data fusion and time synchronization processing, significantly improving the accuracy of operational status identification. Distributed autonomous nodes and a central coordination unit work collaboratively to form a hierarchical, collaborative, and globally optimized control system, enabling rapid power balancing and limit-overflow risk suppression in multi-regional scenarios. The execution control and status feedback modules construct a closed-loop adaptive mechanism, giving the system real-time response and continuous optimization capabilities under dynamic load fluctuations, thereby improving the safety, stability, and energy efficiency of the power distribution network in ultra-fast charging scenarios. Attached Figure Description

[0020] The accompanying drawings, which are provided to further illustrate the invention and form part of this application, are not intended to unduly limit the invention. In the drawings: Figure 1 A schematic diagram of the overall structure of an ultra-fast charging flexible control system based on spatial distribution coordination is shown.

[0021] Figure 2The diagram shows the timing sequence of the control flow during the operation of the system. Detailed Implementation

[0022] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. The illustrative embodiments and descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0023] like Figure 1 As shown, this embodiment of a flexible control system for an ultra-fast charging interface based on multi-source data fusion and load balancing algorithms comprises a data acquisition unit 101, a multi-source data fusion unit 102, a load assessment and balancing calculation unit 103, a control strategy generation unit 104, a distributed collaborative control unit 105, a central coordination unit 106, an execution control unit 107, and a status feedback and optimization unit 108. These units interact with data and commands through a control management platform and form a closed-loop control system with the charging terminal through a power distribution control system. The device is suitable for power distribution network scenarios with multiple ultra-fast charging stations and can collaboratively perform load balancing and power allocation at the station and regional levels.

[0024] The data acquisition unit 101 establishes a data interface with the charging facilities, station-level management system, and power distribution control system to collect operational parameter data, including but not limited to terminal operating parameters, station-level power and load information, and power distribution line operating status. The data acquisition unit 101 sets up a time reference alignment mechanism to mark the measurement time and provides basic data verification capabilities to eliminate obviously abnormal data points. Verified data is transmitted to the multi-source data fusion unit 102 via the device's internal bus.

[0025] The multi-source data fusion unit 102 performs time synchronization and consistency processing on runtime data from different sources to form a standardized multi-source measurement dataset. This unit comprises three sub-modules: format parsing, time alignment, and consistency processing. The format parsing sub-module maps heterogeneous data to uniform fields and units; the time alignment sub-module aligns multi-source data based on timestamps; and the consistency processing sub-module performs rule-based integrity checks and cross-source consistency verification, ensuring that downstream algorithms run on a unified dataset. The processed standardized dataset is then available for use by the load assessment and balancing calculation unit 103.

[0026] The load assessment and balancing calculation unit 103 calculates the load balance status and overload risk level of the distribution line based on a standardized multi-source measurement dataset, and generates corresponding load control targets. This unit includes two sub-modules: line condition assessment and target generation. The line condition assessment sub-module calculates the load distribution of each feeder and branch, node voltage deviation, and heat load trends reflected by operational data, thereby outputting assessment results reflecting the line balance degree and overload risk. The target generation sub-module generates quantified load control targets based on the assessment results and preset operational constraints. These targets indicate the power distribution and load transfer direction to be achieved within a given control cycle.

[0027] The control strategy generation unit 104 generates a flexible power regulation strategy based on the load control target, combined with system operating constraints and equipment status information. This unit receives the target value from the load assessment and balancing calculation unit 103 and, in conjunction with the available capacity of station-level equipment, the current operating status of the terminals, and established service strategies, forms a control strategy that includes power setpoints, limit conditions, and priority rules. The structured control strategy includes elements such as target range, allowable deviation, execution sequence, and timeout protection conditions to adapt to downstream tiered execution.

[0028] The distributed collaborative control unit 105 includes multiple spatial autonomous nodes, which can be deployed at the station level or regional boundaries. Each autonomous node is equipped with a local load assessment module and a self-adjustment execution module, used to negotiate and generate regional power regulation strategies within its jurisdiction based on a neighborhood information exchange mechanism. The neighborhood information exchange mechanism is used to share operating status and strategy intentions related to the regional boundary among adjacent autonomous nodes, thereby reducing load unevenness and crosstalk effects at the boundary. When a control cycle is triggered, each autonomous node first completes the local strategy solution, and then completes the negotiation update within the neighborhood exchange window to form an executable strategy at the regional level.

[0029] The central coordination unit 106 is used to perform macro-level corrections and coordinated optimizations of the regional control results of each autonomous node. The central coordination unit 106 receives regional strategy summaries and operational evaluation results from the distributed collaborative control unit 105, and based on the global load balance and overload risk status, makes necessary global consistency corrections to the regional strategies, including cross-regional power transfer recommendations and dynamic constraint adjustments to global limits. The output of the central coordination unit is distributed in the form of lightweight parameters to reduce control link load and maintain the flexibility of regional autonomy.

[0030] The execution control unit 107 is used to perform hierarchical control and dynamic allocation of the output power of each charging terminal according to the power regulation strategy generated by the central coordination unit 106 and each autonomous node. Located in the station-side controller or terminal-side control device, the execution control unit 107 receives the upper-level strategy, decomposes it into control parameters recognizable by the terminal, and performs operations such as setting, power limiting, or restoration on the terminal within the control cycle. To ensure execution continuity in the event of communication jitter or short-term interruption, the execution control unit 107 has local sequence preservation and timeout protection mechanisms to ensure stable operation of the most recent effective strategy when upper-level control updates are temporarily unavailable.

[0031] The status feedback and optimization unit 108 collects execution feedback data from each autonomous node and charging terminal, compares the load status before and after regulation, and automatically updates regional strategy parameters based on global and local operational deviations to achieve closed-loop optimization control. At the end of the control cycle, this unit summarizes the execution results and line status changes, compares the target completion rate and deviation indicators of this cycle, and updates the weight coefficients and strategy parameters for the next cycle. The status feedback and optimization unit 108 transmits key evaluation quantities back to the multi-source data fusion unit 102 and the load assessment and balancing calculation unit 103 for calculation input in the next cycle, thereby forming a closed loop connecting data, assessment, strategy, and execution.

[0032] like Figure 2 As shown, the control flow of this embodiment includes a data acquisition stage, a fusion processing stage, an evaluation and target generation stage, a strategy solving and negotiation stage, a hierarchical execution stage, and a feedback optimization stage. In the data acquisition stage, the data acquisition unit 101 retrieves operating parameters according to a set sampling period; in the fusion processing stage, the multi-source data fusion unit 102 completes format parsing, time alignment, and consistency processing; in the evaluation and target generation stage, the load evaluation and balance calculation unit 103 outputs quantified control targets; in the strategy solving and negotiation stage, the control strategy generation unit 104 and the distributed collaborative control unit 105 jointly complete the process, where autonomous nodes first solve locally, then complete negotiation within a neighborhood exchange window to form a regional strategy, and the central coordination unit 106 performs macro-level corrections when necessary; in the hierarchical execution stage, the execution control unit 107 issues control parameters to the terminals and monitors the execution results; in the feedback optimization stage, the status feedback and optimization unit 108 completes index comparison and parameter updates, and uses the results for the next round of control.

[0033] During project implementation, the various units of the device can be deployed based on a modular hardware and software structure. The data acquisition unit 101 and the execution control unit 107 should be located close to the field equipment to shorten loopback latency; the multi-source data fusion unit 102, the load assessment and balancing calculation unit 103, and the control strategy generation unit 104 can be deployed on the computing resources of the management platform; the distributed collaborative control unit 105 is deployed regionally to improve autonomous response speed; the central coordination unit 106 operates on a control platform with a global perspective; and the status feedback and optimization unit 108 is interconnected with the above units to form data feedback. The device can ensure the integrity and availability of data and control commands during transmission through a secure communication mechanism.

[0034] Through the aforementioned structure and process, this embodiment, while ensuring data consistency, achieves periodic identification of the load balance status and overload risk of power distribution lines. It generates a flexible power regulation strategy by combining regional autonomy with global coordination, and then dynamically allocates and supervises the execution of charging terminals under a hierarchical control system. Execution deviations are fed back to the data and evaluation stages via a status feedback and optimization unit, achieving closed-loop optimization of the device during continuous operation and meeting the comprehensive requirements of timeliness, coordination, and stability in ultra-fast charging scenarios.

[0035] The specific implementation of each unit is as follows: The data acquisition unit 101 is located at the front-end data interface layer of the device, used to establish real-time data communication channels with charging facilities, station-level management systems, and power distribution control systems. To ensure the comprehensiveness and consistency of data acquisition, this unit achieves compatibility with different system communication protocols through a multi-protocol adaptation module, including commonly used industrial and information technology interface standards such as Modbus-TCP, IEC 61850, DL / T645, and RESTful API. Each protocol interface is abstracted and encapsulated through a unified data access management module, so that data from different sources can be input into the internal processing bus in a structured form.

[0036] In practical applications, the data acquisition unit 101 is configured with multiple acquisition ports, each corresponding to a specific data source category. For example, the data port from the charging facility is used to acquire real-time current, voltage, power, and charging mode status of the charging gun; the port from the station-level management system acquires total active power, reactive power, power factor, transformer temperature, and switch status information within the station; and the port from the power distribution control system acquires line power flow, voltage distribution, feeder status, and protection device operation records. To prevent data misalignment caused by different sampling frequencies, all acquisition ports are connected to a unified time reference alignment module.

[0037] The time reference alignment module uses the high-precision clock of the device's main control processor as its core, obtaining a global time reference from a higher-level time server via Network Time Protocol (NTP) or IEEE 1588 Precision Time Synchronization Protocol (PTP). Each set of acquired data is automatically marked with a precise timestamp before entering the buffer, with a time accuracy better than 1 millisecond. This module also has a clock drift detection mechanism, which automatically fine-tunes the clock by periodically comparing the deviation between the main clock and the external reference signal to ensure timing consistency during long-term operation.

[0038] To ensure data quality, the data acquisition unit 101 integrates a basic data verification module, which employs a multi-level verification strategy. First, CRC redundancy checks are performed on communication frames at the data receiving layer to identify transmission errors. Second, logical range checks are performed at the data parsing layer, marking values ​​exceeding reasonable physical ranges. Finally, statistical consistency judgments are performed at the data preprocessing layer, eliminating or marking invalid anomalous data points that show significant abrupt changes or distortions within a short period. For sampled data deemed anomalous, the system does not discard them directly but instead uses interpolation algorithms to perform linear or cubic spline compensation between adjacent valid data to ensure the continuity and integrity of the time-series data.

[0039] The verified and compensated data is stored in the internal bus structure in the form of standardized key-value pairs or JSON objects, and transmitted to the multi-source data fusion unit 102 through a message queue mechanism. The internal bus adopts a publish-subscribe model, with the data acquisition unit 101 acting as the publisher and periodically pushing the latest verified data packets to the bus. The multi-source data fusion unit 102 acts as the subscriber, automatically receiving the corresponding data streams according to the topic category (such as "pile-side operation", "station-level power", "line status"), thus achieving a highly reliable connection between the data acquisition layer and the fusion computing layer.

[0040] To improve system stability and security, the data acquisition unit 101 is equipped with a data buffer and retransmission mechanism. When communication is interrupted or the bus is congested, the receiving buffer can temporarily store data from multiple sampling periods, which will be retransmitted in chronological order after communication is restored. To prevent data duplication, the buffer uses a unique timestamp and sequence number index for identification, and an acknowledgment signal is returned by the receiving end after each transmission. All data transmission processes are executed through a TLS encrypted channel to ensure the confidentiality and integrity of operating parameters during acquisition and transmission.

[0041] Through the above-mentioned structure and technical means, the data acquisition unit 101 can stably and accurately collect key operating parameters of charging facilities, station-level systems and power distribution systems in multi-source heterogeneous, different sampling frequencies and complex communication environments, ensuring the consistency of data in time and quality, and providing a reliable data foundation for subsequent multi-source data fusion and load balance calculation.

[0042] The multi-source data fusion unit 102 is located between the data acquisition layer and the computing layer. It is used to unify the processing of running data from different sources, in different formats, and with different sampling frequencies to form a standardized measurement dataset that can be directly called by subsequent computing units. This unit consists of a format parsing submodule, a time alignment submodule, and a consistency processing submodule. The three submodules run sequentially according to the data flow to ensure that the fusion results are consistent in format, time, and logic.

[0043] The format parsing submodule is used to standardize the format of data from different systems. Due to differences in data structures and unit systems among charging facilities, station-level management systems, and power distribution control systems, the format parsing submodule pre-establishes a mapping rule table to define the correspondence between fields from different data sources and internal standard fields. For example, data from charging facilities primarily uses power P (kW) and current I (A), while similar data from power distribution control systems may be expressed in active power (kW) and apparent power (kVA). This module maps these to unified fields using the rule table and automatically performs unit conversion. The module also includes a format recognition engine capable of parsing various data structures such as JSON, XML, and CSV, and verifies field integrity and data type correctness through a data template validation mechanism. To ensure efficiency, the format parsing process employs a multi-threaded asynchronous parsing strategy, achieving millisecond-level parsing response during batch data input.

[0044] The time alignment submodule synchronizes data from different sources based on timestamp information provided by the data acquisition unit. Due to differences in sampling periods across different devices, the time alignment submodule uses a combination of interpolation and resampling to unify the data to the system-defined time step Δt. Specifically, the module establishes a time window within each synchronization cycle. When a time gap exists in a data source, an estimated value for the missing moment is calculated using linear interpolation or cubic spline interpolation. When the sampling frequency exceeds the set step size, a resampling operation is performed, using the mean or median value within the time window as representative data. The time alignment submodule also includes a delay detection mechanism to mark input data exceeding a preset delay threshold, preventing time drift during data fusion. After synchronization, the data is arranged using timestamps as indices, ensuring a one-to-one correspondence between measurements from multiple sources at the same time.

[0045] The consistency processing submodule performs cross-source consistency checks and logical integrity verification on time-aligned data. This module checks data based on predefined physical and logical constraints, such as ensuring that deviations in the relationship between current, voltage, and power (P≈U×I×cosφ) are within acceptable limits, and that voltage measurement differences from different sources at the same node are less than the specified tolerance ΔUmax. If significant deviations are found in cross-source data, the system automatically triggers a consistency correction mechanism, performing confidence-weighted averaging correction on data with large deviations, or marking it as invalid if correction is not possible to prevent misleading calculations. To ensure overall data integrity, the consistency processing submodule includes an integrity comparison program that statistically analyzes the number of data samples and field completeness for each time slice. When the missing proportion exceeds a set threshold, the module automatically records the error and outputs an anomaly report to the upper layer.

[0046] During operation, the multi-source data fusion unit 102 continuously processes the data stream using a sliding window approach through its built-in data buffer. The length of each window can be adaptively adjusted according to the system control cycle to ensure that the fusion result covers all valid data within the latest cycle. After format parsing, time alignment, and consistency processing, the data is repackaged into a standardized measurement dataset. This dataset is stored in the form of a structured table or object array, containing unified field definitions, standardized units, and a synchronization time index. The processed standardized dataset is transmitted to the load assessment and balancing calculation unit 103 via the device's internal bus for load assessment and algorithm calculation.

[0047] Through the above structure and process, the multi-source data fusion unit 102 can achieve unified processing of running data from different systems, different formats, and different sampling periods, ensuring that the fusion results are consistent in both the time and numerical dimensions, thereby providing high-precision and verifiable basic data support for subsequent load balancing calculations.

[0048] The load assessment and balancing calculation unit 103 is located in the analysis and decision-making layer of the device. Based on the standardized measurement dataset output by the multi-source data fusion unit, it assesses the operating load status, balance degree, and potential overload risk of the distribution lines in real time, and generates load control targets based on this, providing input parameters for the subsequent control strategy generation. This unit consists of a line status assessment submodule and a target generation submodule, which run sequentially and interact through an internal data channel.

[0049] After receiving the standardized measurement dataset, the line condition assessment submodule first performs statistical calculations on the real-time current, voltage, active power, and reactive power of each feeder and branch to obtain the power distribution of each node and the branch current load factor. This module pre-stores the topology data of the distribution network, including basic parameters such as node connections, feeder length, conductor impedance, and transformer capacity. Based on this, the module can calculate the voltage amplitude and phase angle distribution of each node in the network using power flow calculation methods. The power flow calculation employs an improved Newton-Raphson iterative algorithm to improve the calculation convergence speed and accuracy. Convergence tolerance parameters are set during the calculation process. , usually take Within this range, to ensure that the obtained voltage and power results are stable and reliable.

[0050] After obtaining the power flow data for each node, the line condition assessment submodule further calculates the node voltage offset. The module sets the voltage value of each node... With rated voltage Compare and calculate the voltage offset. Its definition is as follows:

[0051] The module will calculate the results Deviation from preset voltage limit Comparison. When At this point, the node is marked as a voltage over-limit point. The module simultaneously calculates the load distribution of each feeder and branch, using branch current... With rated current The ratio gives the load factor Its expression is:

[0052] when Approaching or exceeding a preset threshold At that time, it was determined that the branch circuit had an overload risk.

[0053] To reflect heat load trends, the line condition assessment submodule establishes a heat accumulation model using real-time power data and operating time series. The model calculates the heat load index by integrating the squared values ​​of branch currents over a time window. Its definition is:

[0054] This index is used to characterize the degree of thermal stress in a conductor or device. It can be correlated with the device's allowable heat capacity. Compare and determine the thermal safety status of the branch circuit or transformer. If... near If the temperature remains high, the module will generate a high-temperature warning signal and record it in the evaluation results.

[0055] After considering all indicators, the line status assessment submodule outputs a load rate. Node voltage offset Heat load index The module includes the state vector for exceeding limits. It uses a weighted scoring method to calculate the overall line balance index. Let the weighting coefficients be... , , Corresponding to voltage offset, load rate, and thermal load respectively, The calculation formula is:

[0056] in, This represents the average load rate of the system. This represents the average heat load. When... The closer the value is to 1, the more balanced the circuit is. The module classifies the operating status level based on this indicator, which is usually divided into four levels: normal, slightly unbalanced, moderately unbalanced, and severely unbalanced.

[0057] After receiving the balance and risk level results from the line condition assessment submodule, the target generation submodule generates quantified load regulation targets based on preset operating constraints and control cycle parameters. Operating constraints include the upper limit of feeder power. Node voltage limit range and equipment allowable load rate The target generation process first determines the set of controlled objects. This typically includes nodes and adjacent branches where the load rate or voltage deviation exceeds a threshold. The module calculates the target power adjustment value for each controlled node within each control cycle. To improve the overall network balance index It tends to maximize.

[0058] To ensure the target is achievable in engineering, the target generation submodule will adjust the power amount. Limiting the power adjustment to the range allowed by the equipment, i.e.:

[0059] in, The target power adjustment value is determined by the equipment type and operating status. The module also calculates the impact of power adjustments at each node on voltage and load distribution based on the power flow sensitivity matrix, ensuring the control target is globally feasible while meeting constraints. The final output load control target includes the target power adjustment value for each controlled node or region. Load transfer direction and priority identifiers are provided for use by the control strategy generation unit.

[0060] During operation, the load assessment and balancing calculation unit 103 executes the above process at fixed intervals (e.g., 1 second or 5 seconds) and can adaptively adjust the assessment frequency according to network load fluctuations. When the system detects line imbalance... When the risk level decreases significantly or increases due to overload, the module can trigger the target generation process in advance to shorten response latency. Both assessment and target results are stored in structured data format and include timestamps. And the running status identifier, which can be called by subsequent modules in the strategy generation and execution control phase.

[0061] Through the above structure and method, the load assessment and balancing calculation unit 103 can accurately identify uneven load distribution and overload trends in high-concurrency and dynamically changing operating environments, and generate quantitative and executable load control targets, providing a reliable decision-making basis for the system's flexible control and multi-level load balancing.

[0062] The control strategy generation unit 104 is located in the decision control layer of the device. It is used to generate flexible power regulation strategies based on the load control targets output by the load assessment and balancing calculation unit 103, combined with system operation constraints and equipment status information. This unit receives the target quantity from the load assessment and balancing calculation unit, and outputs a structured control strategy set by comprehensively analyzing the available capacity of station-level equipment, the current operating status of terminals, and the established service strategies, so as to achieve hierarchical execution and dynamic power adjustment.

[0063] The control strategy generation unit 104 includes a target parsing module, a constraint adaptation module, and a strategy orchestration module. The target parsing module analyzes and classifies the input load control targets. Assume the input targets are the target power adjustment values ​​for each controlled node. The target resolution module determines the node type, load level, and service priority based on these parameters. Grouping them into the corresponding execution units and generating target power setpoints. The calculation method is as follows:

[0064] in, Indicates the current power of the node. This indicates the required power increment or decrement to adjust to the target power adjustment value. To prevent power oscillation, the module... Applying smoothing factor ( ), calculate the dynamic adjustment step size This ensures a continuous and stable power regulation process.

[0065] The constraint adaptation module is used to verify the feasibility of the generated target power value. The module includes built-in system operating constraints, including the upper limit of feeder capacity. Node voltage limit Equipment load rate and temperature rise constraint During the calculation phase, this module performs constraint checks on the pre-adjusted power value of each node. When any constraint condition is detected as being triggered, the corresponding target value is automatically corrected. The constraint adaptation process employs a constraint correction algorithm based on a penalty function, the calculation expression of which is as follows:

[0066] in, , , To constrain the penalty coefficient, The current voltage of the node. This represents the current load rate of the node. Through this algorithm, the module dynamically corrects power allocation without disrupting the overall target direction, ensuring the output strategy is feasible for engineering implementation.

[0067] The strategy orchestration module is responsible for generating a structured set of flexible power regulation strategies based on the revised power setpoints. This strategy set includes elements such as the power target range, allowable deviations, execution sequence, and timeout protection conditions. The power target range is defined by... This indicates that the allowable deviation is... The execution unit makes dynamic adjustments within this range, satisfying the following formula:

[0068] in, Typically set to 2% to 5% of the target power to absorb short-term fluctuations. Execution order is based on load priority weights. Nodes with higher weights are prioritized to execute power adjustment commands. To prevent execution timeouts, the module is configured with timeout protection parameters. When a node fails to complete the adjustment within the specified period, the system automatically switches to the default safety policy, maintains the current power, and reports the anomaly.

[0069] After the control strategy generation unit 104 generates the strategy set, it packages it into a unified data structure, including strategy number, target power value, deviation tolerance, execution priority, and constraint status identifier. The strategy data is timestamped. As an index, it is stored in the policy buffer of the control and management platform and synchronized to the distributed collaborative control unit and execution control unit through the message bus.

[0070] During continuous operation, the control strategy generation unit 104 can periodically adjust the strategy parameters based on status feedback and the execution results returned by the optimization unit. If an execution deviation is detected... If the preset tolerance is exceeded, the module automatically recalculates the power target. Make it satisfy the following conditions:

[0071] in, This is the measured power value. This sets an upper limit for the allowable error. Through this feedback correction mechanism, the system achieves closed-loop self-adjustment of strategy generation and execution, ensuring the continuous effectiveness and response accuracy of the strategy throughout its operating cycle.

[0072] Through the above structure and method, the regulation strategy generation unit 104 can generate flexible power regulation strategies in real time under multiple constraints, realize refined control of power allocation and dynamic feasibility assurance, and provide accurate instruction basis for the hierarchical execution and flexible regulation of the system.

[0073] The distributed collaborative control unit 105 is located in the middle control layer of the system to realize autonomous decision-making at the regional level and cross-regional collaborative adjustment, thereby improving the response speed and stability of the overall control system. This unit consists of multiple spatial autonomous nodes, which can be deployed at the station level, regional power supply unit level, or geographical boundary. Each autonomous node has independent data processing, strategy solving, and local execution capabilities. Communication links are established between nodes through a neighborhood information exchange mechanism to achieve inter-regional strategy negotiation and boundary condition coordination, thereby generating regional power regulation strategies under a distributed architecture.

[0074] Each autonomous node contains a local load assessment module and a self-regulation execution module. The local load assessment module receives a subset of standardized measurement data relevant to the local area from the multi-source data fusion unit, including feeder current, voltage, transformer load rate, and substation power information for the area. Based on a preset local topology model, the module calculates the power distribution and line load rate of each node within the area and generates a regional load state vector. This state vector includes the total active power of the area. reactive power Maximum load rate With voltage offset These parameters are used to characterize the current operating status of the region.

[0075] After receiving the state vector output by the local load assessment module, the self-regulating execution module executes a local strategy solution process based on the upper-level target interval and regional boundary constraints issued by the central coordination unit. The module employs a constraint optimization-based regulation algorithm to adjust the power output of local charging terminals while maintaining regional power supply security and equipment constraints. To ensure solution speed, the local optimization uses a distributed Lagrange multiplier method, iteratively updating local power variables. and Lagrange multipliers Achieving convergence under the power balance condition. Convergence occurs when the region calculation reaches the convergence condition. When the time comes, output the local strategy solution for the current period.

[0076] Autonomous nodes share boundary state data and negotiate strategies through a neighborhood information exchange mechanism. This mechanism achieves information synchronization through point-to-point communication interfaces between nodes, with the communication cycle consistent with the control cycle, typically 1 to 5 seconds. Each node broadcasts its own set of boundary operating parameters, including boundary node voltage, to neighboring nodes within the neighborhood exchange window. Boundary inflow power With local strategic intentions After receiving parameters broadcast by each other, neighboring nodes determine whether there is an imbalance in the boundary conditions based on a matching algorithm. When a boundary power difference is detected... When the set threshold is exceeded, the two nodes automatically initiate a negotiation process, adjusting their respective boundary power allocation parameters bidirectionally until the constraints are met. .

[0077] To avoid oscillations and repeated adjustments during the negotiation process, the neighborhood information exchange mechanism introduces a synchronization time window and a weight decay factor. Synchronization Time Window Used to define the effective period for data exchange between adjacent nodes; weight decay factor This is used to gradually reduce the boundary adjustment magnitude during continuous negotiation, enabling the system to converge within a finite number of steps. After negotiation, each autonomous node updates its local power regulation parameters and forms a set of executable policies at the region level. This policy set includes the target power value, boundary balance correction coefficient, and execution sequence parameters for each controlled terminal, all with timestamps for identification by downstream execution control units.

[0078] When a control cycle is triggered, each autonomous node first completes the state assessment and policy solution for the current cycle within the local load assessment module, and then performs one or more negotiation updates within the neighborhood exchange window. The entire process is completed before the end of the control cycle, and the negotiation results are saved in the form of a regional policy matrix. The regional policy matrix is ​​represented by node power vectors. With boundary power vector This is the core data structure used to describe the results of power adjustment and boundary coordination within the region. At the start of the next cycle, the autonomous nodes synchronize the updated policy set to the central coordination unit for global consistency verification and macro-level optimization.

[0079] Through the aforementioned structure and operational process, the distributed collaborative control unit 105 achieves hierarchical decision-making and collaborative optimization among regional autonomous nodes. This unit can not only respond quickly to load fluctuations within the control cycle, but also reduce load unevenness and crosstalk effects at regional boundaries through a neighborhood information exchange mechanism, thereby realizing the collaborative control and overall load balance of ultra-fast charging facilities in multi-regional power distribution networks.

[0080] The central coordination unit 106 is located at the system's global control layer. Based on the regional control already completed by the distributed autonomous nodes, it summarizes, analyzes, and macroscopically optimizes the operational results of each region, thereby maintaining the overall load balance and safety margin of the power grid. This unit receives regional strategy summaries and operational evaluation results from the distributed cooperative control unit 105, comprehensively assesses the interaction status, load distribution, and potential risks among the autonomous nodes, and performs global consistency correction and coordinated optimization of regional strategies when necessary.

[0081] The central coordination unit 106 includes a data aggregation module, a global evaluation module, and a policy coordination module. The data aggregation module receives policy summary information and operational status data uploaded by autonomous nodes in each region. This module achieves parallel access through a multi-threaded asynchronous communication mechanism to ensure data reception efficiency in high-concurrency environments. Each autonomous node uploads a data packet containing information such as the average load rate within the region, node voltage statistics, boundary power exchange, and over-limit flags. After receiving complete data, the data aggregation module performs data integrity checks and time synchronization to ensure that all regional data corresponds to the same control cycle.

[0082] The global assessment module is used to calculate the overall load balance and overload risk distribution of the system based on aggregated multi-regional data. This module uses the load balance indicators uploaded by each region. With load factor Using the input as input, a weighted aggregation method is used to calculate the global balance index. The weighting coefficients are determined based on the regional capacity or power supply range. The module simultaneously assesses the global overload risk status, generating a risk distribution map by statistically analyzing the number and severity of overloaded branches in each region, and identifying the critical areas with the greatest impact on the entire network. The assessment process considers the impact of cross-regional power transmission on system power flow, calculating the cross-regional power exchange matrix to determine whether the current control strategy leads to concentrated loads on upstream lines or excessive voltage fluctuations.

[0083] After obtaining the global assessment results, the strategy coordination module performs macro-level modifications and coordination of regional strategies. This module employs an iterative optimization mechanism, first determining the set of regions requiring adjustment. This refers to regions where the load factor or voltage deviation exceeds a global threshold. Subsequently, based on the power flow sensitivity coefficient and the regional coupling matrix, a recommended value for cross-regional power transfer is calculated. , indicating from the region To the region The feasible power transfer amount is calculated. To ensure system safety, the module considers line capacity constraints and cross-regional voltage fluctuation limits when calculating the transfer amount, ensuring that the optimization results do not introduce new risks of exceeding limits. To improve coordination efficiency and convergence stability, the strategy coordination module adopts a coordination suggestion generation formula oriented towards global consistency correction, which is used to comprehensively consider the combined effects of balance degree differences, power flow margin, and load factor differences on the transfer suggestions within a single step:

[0084] in: Indicates from the region To the region Coordination recommendations for power transfer amount; This is the upper bound of the transferable power of this region in the current cycle; This is the regional coordination gain coefficient, used to adjust the transfer amplitude; , This serves as an indicator of the balance between the two regions. These are measured values ​​of cross-regional tidal current. This is the global power flow limit for the system. , Regional load factor; It is a sensitive factor for differences in balance; This means that only the non-negative part is taken, ensuring that the adjustment direction is consistent with the physical meaning.

[0085] This formula suppresses extreme regulation through hypercurvature functions, reflects power flow margin penalties in a quadratic form, and performs bounded normalization on load factor differences. It can provide feasible coordination suggestions without introducing additional complex constraint solutions, thereby accelerating the convergence of cross-regional strategies and reducing oscillation risks.

[0086] Balance driving term This reflects the "main driving force" of power transfer between regions.

[0087] When the target area balance Higher than the area At that time, the system tends to start from Towards Power transfer; The function ensures that the adjustment range is within Continuous internal variation to avoid excessive power surges; parameters Controlling and adjusting sensitivity: The smaller the value, the more sensitive the system is to differences in balance.

[0088] This achieves a continuous nonlinear coordinated response, which differs from the traditional linear power flow distribution formula.

[0089] Trend margin item This reflects the safety margin constraints of cross-regional lines. (When the power flow...) Approaching the upper limit When the power flow is significantly below the upper limit, this term approaches 0, automatically reducing power transfer; when the power flow is much below the upper limit, this term approaches 1, relaxing the power transfer capability; the quadratic decay form makes the constraint transition smoother and the calculation more stable. This design implements an adaptive power reduction mechanism with "embedded constraints" without explicitly calling external optimization constraints.

[0090] Load factor normalization term This item is used to provide bounded adjustment weights when there are significant differences in load rates between two regions. When When the result approaches +1, it indicates that it should be from... Towards Power transfer; when When the result approaches -1, the direction reverses; when the two are close, the result approaches 0, avoiding ineffective adjustments. This "saturated difference function" maintains directionality while preventing large fluctuations, and is a nonlinear equilibrium normalization mechanism.

[0091] This formula can be directly embedded into the strategy coordination module of the central coordination unit for rapid correction of cross-regional power allocation decisions. In simulation verification, when there is a sudden increase in load or significant differences in regional balance, the system can... It achieves rapid power redistribution, with an average convergence time that is 35% to 50% shorter than that of the linear model. At the same time, it reduces the power flow fluctuation amplitude of the boundary lines and significantly improves coordination efficiency and operational safety margin.

[0092] When a global assessment indicates that the total system load is approaching its limit or that some main lines show a potential overload trend, the central coordination unit 106 automatically triggers a global limit dynamic adjustment mechanism. This mechanism calculates new global operating constraints based on real-time operating margins, including the total power limit. Cross-regional trend limited edition and node voltage range The module updates the output to each autonomous node in a parameterized form. To reduce communication load, the module performs lightweight encapsulation of the output results, retaining only necessary parameters and adjustment coefficients, such as the regional power correction coefficient. Voltage constraint correction amount With priority adjustment vector .

[0093] During the control cycle, the central coordination unit 106 periodically performs a global coordination calculation and can immediately trigger temporary coordination upon detecting significant anomalies. The coordination output is broadcast to all autonomous nodes via the system bus. Each node updates its local policy based on the received correction parameters without re-executing the complete policy solution process, thus maintaining the independence and flexibility of regional autonomy. The module also maintains a coordination log, recording the input state, correction parameters, and output results for each round of optimization, for adaptive adjustment in subsequent cycles and long-term performance evaluation.

[0094] Through the aforementioned structure and working mechanism, the central coordination unit 106 can achieve global coordination and optimization without interfering with regional autonomous control. Its macro-correction mechanism, combined with the aforementioned coordination suggestion generation formula, effectively solves the problems of uneven power distribution and power flow coupling among multiple regions caused by differences in boundary conditions, ensuring that the system achieves load balance and risk control at the network-wide scale, and providing stable and reliable upper-level decision support for the wide-area collaborative operation of ultra-fast charging facilities.

[0095] The execution control unit 107, located at the system's terminal control layer, is the core unit for accurately implementing upper-level control strategies. Based on the power regulation strategies generated by the central coordination unit 106 and each autonomous node, this unit performs hierarchical control and dynamic power allocation for charging terminals, achieving flexible and sustainable power regulation at the regional and equipment levels. The execution control unit 107 is typically deployed in the station-side controller or terminal-side control device, maintaining real-time interaction with the upper-level control module via a high-speed communication interface to ensure that strategy commands are accurately issued and executed with millisecond-level latency.

[0096] The execution control unit 107 consists of a strategy receiving module, an instruction parsing module, a power control module, and a continuity protection module. The strategy receiving module receives power regulation instruction packets from the upper layer. These packets encapsulate key fields such as target power, allowable deviation, priority, and timeout protection time in a structured parameter format. Upon receiving the instruction, the module verifies its integrity, timestamp, and signature information, and uses a sequential numbering mechanism to ensure the orderly execution of the strategy. If duplicate or out-of-order instructions are detected, the system automatically ignores them to prevent erroneous execution or power fluctuations due to instruction conflicts.

[0097] The instruction parsing module is used to convert upper-level policies into control instructions that can be recognized by the terminals. The module first determines the charging terminals that need to be controlled based on the equipment topology within the station and obtains their current operating status, such as output power, voltage level, and communication status. Then, it calculates a new power setting target according to the policy requirements and automatically adjusts it within the safe range allowed by the equipment to prevent the risk of over-power or under-voltage operation. After parsing, the module generates a set of control parameters adapted to each terminal, including target power, execution mode, and control cycle, and then passes it to the lower-level control logic for execution.

[0098] The power control module is responsible for completing the specific power adjustment process within the control cycle. The module executes instructions hierarchically according to strategy priority, first adjusting critical load terminals and then controlling ordinary terminals sequentially to reduce instantaneous system fluctuations. Control commands are sent to each terminal device via the station-level communication bus. The terminals adjust their output power or current limits according to the instructions and transmit status information back in real time. The module continuously monitors terminal feedback. If it detects that the actual output deviates from the target setting by more than a reasonable range, it will automatically initiate a fine-tuning mechanism to gradually correct the output value towards the target value, thereby improving execution accuracy and system response stability.

[0099] To ensure the continuity and safety of the control process, the execution control unit 107 is equipped with local sequence preservation and timeout protection mechanisms. The local sequence preservation mechanism records the strategy number and execution status to enable resume control after a disconnection. When communication jitter or a short interruption occurs, the system can automatically recall the most recently successfully executed strategy to maintain operation and prevent sudden power fluctuations. The timeout protection mechanism automatically enters a conservative mode if no new strategy is received within a set time limit, maintaining the terminal power at a safe level or locking it at the current output state to avoid load instability caused by upper-layer disconnection.

[0100] The execution control unit 107 has self-testing and statistical functions during operation. After each control cycle, the module evaluates the command execution success rate, communication latency, and terminal response consistency, and generates an execution report which is sent to the central coordination unit 106 and the status feedback and optimization unit 108 for adjustment of upper-level strategies and system performance evaluation. The module also has a time synchronization mechanism to ensure that control actions between multiple terminals are synchronized, preventing power fluctuations caused by time drift.

[0101] Through the above design, the execution control unit 107 achieves high reliability in policy issuance and high precision in power regulation. Its hierarchical execution logic and local protection mechanism can effectively cope with complex operating scenarios such as communication anomalies, sudden load changes, and inconsistent device responses, thereby ensuring the safety, stability, and real-time controllability of the ultra-fast charging flexible control system in dynamic environments.

[0102] The status feedback and optimization unit 108 is located in the closed-loop control layer of the system and is used to comprehensively evaluate the execution results and optimize the strategy after each control cycle. This unit collects operational feedback data from each autonomous node and charging terminal, compares the changes in line load and power distribution before and after control, analyzes the execution effect of the current strategy, and automatically corrects key control parameters based on global and local operational deviations, thereby realizing adaptive closed-loop control of the system.

[0103] The status feedback and optimization unit 108 includes a data acquisition module, a deviation evaluation module, and a parameter update module. The data acquisition module receives operational feedback information from the distributed collaborative control unit 105 and the execution control unit 107, including data such as measured terminal power, node voltage, line load rate, regional balance, and execution status. To avoid data asynchrony caused by communication delays, the module uses a timestamp synchronization mechanism to align the multi-source feedback and aggregates all execution results within the current control cycle using a sliding window. If significant anomalies are detected in individual node data, the system automatically removes the anomalies to ensure the accuracy and stability of subsequent calculations.

[0104] After receiving feedback data, the deviation assessment module compares the actual performance of each node with the target power distribution generated by the load assessment and balancing calculation unit 103 in the previous stage to calculate deviations and evaluate performance. This module quantifies the system's performance at both global and regional levels, including changes in overall load balance, the completion rate of regulation in each region, and local stability indicators. If the assessment results show significant load shifts or a decrease in system balance in some regions, these regions are marked as key areas requiring optimization. The module also statistically analyzes deviation trends based on assessment results from multiple periods to determine whether the system is in a stable convergence state.

[0105] The parameter update module dynamically adjusts key parameters in the system strategy based on the deviation evaluation results. This module employs an adaptive adjustment mechanism: when deviations accumulate within consecutive periods, the system automatically increases the parameter adjustment magnitude to accelerate convergence; if the system is stable and the deviation approaches zero, the adjustment step size is appropriately reduced to prevent over-response. For autonomous nodes in different regions, the module can adjust their weight coefficients and power adjustment step sizes separately, making the optimization direction more targeted. Through this differentiated parameter update method, the system can balance local autonomous optimization with global coordination and consistency.

[0106] At the end of each control cycle, the state feedback and optimization unit 108 returns the updated parameter set, global balance improvement status, and performance evaluation results of each region in structured data form to the multi-source data fusion unit 102 and the load assessment and balancing calculation unit 103. This information serves as the calculation input and initial parameters for the next cycle, enabling continuous optimization of the system during periodic operation and forming a closed-loop control chain of data fusion, evaluation and analysis, strategy generation, and execution feedback.

[0107] During long-term operation, the state feedback and optimization unit 108 can automatically adjust the optimization coefficients based on historical data analysis of the system's dynamic response characteristics and parameter change patterns, ensuring the stability and adaptability of the control strategy under different load conditions. Through the continuous operation of this module, the ultra-fast charging flexible control system can achieve multi-cycle self-learning and dynamic optimization in complex power grid environments, thereby ensuring the balance of power distribution and the robustness of overall operation.

[0108] The above description is only a preferred embodiment of the present invention. Therefore, all equivalent changes or modifications made to the structure, features and principles described in the claims of this patent application are included in the scope of this patent application.

Claims

1. A flexible control system for an ultra-fast charging interface based on spatial distribution coordination, characterized in that: The system includes: The data acquisition unit is used to collect operating parameter data from charging facilities, station-level management systems and power distribution control systems, and transmit the data to the control and management platform; The multi-source data fusion unit is used to receive and fuse data from different sources, and form a standardized multi-source measurement dataset based on time synchronization and data consistency processing. The load assessment and balance calculation unit is used to calculate the load balance status and overload risk level of the distribution line based on the multi-source measurement data set, and generate corresponding load control targets. The regulation strategy generation unit is used to generate a flexible power regulation strategy for constraining and guiding subsequent power allocation based on the load regulation target and in combination with system operation constraints and equipment status information. The distributed collaborative control unit includes multiple spatial autonomous nodes. Each autonomous node is equipped with a local load assessment module and a self-adjustment execution module. The local load assessment module and the self-adjustment execution module are used to negotiate and generate a regional power adjustment strategy within their respective regions based on a neighborhood information exchange mechanism. The central coordination unit is used to receive the regional regulation results from each autonomous node and to revise and coordinate the regulation strategies of each region based on the global load balance and overload risk status. The execution control unit is used to perform hierarchical control and dynamic allocation of the output power of each charging terminal according to the power adjustment strategy generated by the central coordination unit and each autonomous node; The status feedback and optimization unit is used to collect execution feedback data from each autonomous node and charging terminal, compare the load status before and after regulation, and automatically update the parameters of the regional regulation strategy based on the operational deviation of the load status of the central coordination unit and the distributed collaborative control unit.

2. The flexible control system for ultra-fast charging interfaces based on spatial distribution coordination as described in claim 1, characterized in that: The data acquisition unit is located in the front-end data interface layer of the device and is used to establish a data communication channel with the charging facilities, station-level management system and power distribution control system and collect operating parameter data. The data acquisition unit includes: A multi-protocol adapter module is used to ensure compatibility with different system communication protocols and to enable structured data input. The time reference alignment module is used to time-mark and synchronize the collected data based on a unified clock source; The data verification module is used to perform redundancy checks, logical range checks, and abnormal data processing on the collected data. The verified data is transmitted to the multi-source data fusion unit via the internal bus for subsequent fusion and calculation.

3. The flexible control system for ultra-fast charging interfaces based on spatial distribution coordination as described in claim 1, characterized in that: The multi-source data fusion unit is located between the data acquisition layer and the computing layer, and is used to perform unified processing on data from different systems to form a standardized measurement dataset. The multi-source data fusion unit includes: The format parsing submodule is used to perform format normalization and unit unification processing on data from different sources; The time alignment submodule is used to synchronize the collected data based on timestamp information to eliminate sampling frequency differences; The consistency processing submodule is used to perform cross-source consistency checks and logical integrity verifications on the synchronized data; The processed standardized measurement dataset is output to the load assessment and balancing calculation unit for subsequent calculations.

4. The flexible control system for ultra-fast charging interfaces based on spatial distribution coordination as described in claim 1, characterized in that: The load assessment and balancing calculation unit is located in the analysis and decision-making layer of the device. It is used to assess the operating load status, balance degree and overload risk of the power distribution line based on the standardized measurement dataset output by the multi-source data fusion unit, and generate load control targets. The load assessment and balancing calculation unit includes: The line condition assessment submodule is used to calculate the voltage deviation, branch load rate and heat load index of each node, and obtain the line balance index according to the weighting method. The target generation submodule is used to calculate the target power adjustment value of each controlled node based on the line status assessment results and preset operating constraints, so as to form a quantified load control target. The load control target is used to provide input parameters for the control strategy generation unit.

5. The flexible control system for ultra-fast charging interfaces based on spatial distribution coordination as described in claim 1, characterized in that: The regulation strategy generation unit is located in the decision control layer of the device and is used to generate a flexible power regulation strategy based on the load regulation target output by the load assessment and balance calculation unit, combined with system operation constraints and equipment status information. The regulation strategy generation unit includes: The target analysis module is used to analyze load regulation targets and generate node power setpoints; The constraint adaptation module is used to perform feasibility verification and correction of the power setting value based on the system operation constraints. The strategy orchestration module is used to generate a structured power regulation strategy set containing power target range, deviation tolerance and execution order based on the corrected power set value, and output the strategy set to the lower control module for execution; The control strategy generation unit can periodically adjust the strategy parameters based on feedback information.

6. The flexible control system for ultra-fast charging interfaces based on spatial distribution coordination as described in claim 1, characterized in that: The distributed collaborative control unit is located in the middle control layer of the system and is used to realize regional autonomous decision-making and cross-regional collaborative regulation. The distributed collaborative control unit includes multiple spatial autonomous nodes, each autonomous node comprising: The local load assessment module is used to calculate the power distribution, load rate and voltage deviation within the region based on the standardized measurement data output by the multi-source data fusion unit, and generate a region operation status vector. The self-adjusting execution module is used to execute local strategy solutions based on the upper-level target interval and boundary constraints, and output the regional power adjustment results. Among them, the multiple autonomous nodes share boundary state parameters and perform strategy negotiation through a neighborhood information exchange mechanism to make bidirectional adjustments to power allocation based on the detection results of boundary power difference or voltage difference, so as to achieve coordinated convergence between regions. After the negotiation is completed, a set of regional power adjustment strategies is formed and the results are reported to the central coordination unit for global optimization.

7. The flexible control system for ultra-fast charging interfaces based on spatial distribution coordination as described in claim 1, characterized in that: The central coordination unit is located in the global control layer of the system. It is used to summarize, analyze and macro-optimize the operation results of each region based on the regional control completed by the distributed autonomous nodes, so as to maintain the global load balance and system safety margin. The central coordination unit includes: The data aggregation module is used to receive policy summary information and operational status data uploaded from autonomous nodes in various regions, and to perform data integrity checks and time synchronization. The global assessment module is used to calculate the overall load balance and overload risk distribution of the system based on multi-regional operational data, and to identify key areas that affect the stability of the entire network. The strategy coordination module is used to make macro-level corrections and coordination of regional strategies based on the global assessment results, calculate cross-regional power transfer recommendation values, and generate an optimized global power allocation scheme under the condition of meeting line capacity and voltage fluctuation constraints.

8. The flexible control system for ultra-fast charging interfaces based on spatial distribution coordination as described in claim 7, characterized in that: The strategy coordination module calculates the cross-regional power transfer recommendation value using the following formula for generating the coordinated power transfer recommendation amount: in: Indicates from the region To the region The coordination recommendation is the amount of power transfer; This is the upper bound of the transferable power of this region in the current cycle; This is the regional coordination gain coefficient, used to adjust the transfer amplitude; , This serves as an indicator of the balance between the two regions. These are measured values ​​of cross-regional tidal current. This is the global power flow limit for the system. , Regional load factor; It is a sensitive factor for differences in balance.

9. The flexible control system for an ultra-fast charging interface based on spatial distribution coordination as described in claim 1, characterized in that: The execution control unit is located in the terminal control layer of the system and is used to perform hierarchical control and dynamic power allocation of the charging terminal according to the power adjustment strategy generated by the central coordination unit and the distributed autonomous nodes. The execution control unit includes: The policy receiving module is used to receive and verify the power adjustment command packets sent down from the upper layer; The instruction parsing module is used to convert upper-layer policies into control instructions that the terminal can recognize, and adjust the target power within a safe range according to the terminal's operating status. The power control module is used to execute power adjustment instructions in a hierarchical manner according to policy priority, and to perform power fine-tuning based on terminal feedback information to reduce deviation; The continuity protection module is used to execute sequence preservation and timeout protection mechanisms in the event of communication interruption or timeout, so as to maintain the terminal power continuously within a safe range; The execution control unit performs self-testing and statistical operations in each control cycle and feeds back the execution results to the upper-level module to achieve closed-loop optimization of the control strategy.

10. The flexible control system for an ultra-fast charging interface based on spatial distribution coordination as described in claim 1, characterized in that: The state feedback and optimization unit is set in the closed-loop control layer of the system and is used to evaluate the execution results and optimize the strategy after each control cycle. The state feedback and optimization unit includes: The data acquisition module is used to receive operational feedback information from the distributed collaborative control unit and the execution control unit, and to align the multi-source feedback data through a timestamp synchronization mechanism; The deviation assessment module is used to compare the actual execution results with the target power distribution, calculate the operating deviation, and assess the global and regional load balance status of the system. The parameter update module is used to dynamically correct the key control parameters of the system based on the deviation evaluation results. It adopts an adaptive adjustment mechanism to adjust the correction magnitude or step size according to the deviation accumulation trend in order to achieve stable convergence and prevent over-response. The state feedback and optimization unit feeds back the updated parameter set and performance evaluation results to the multi-source data fusion unit and the load assessment and balancing calculation unit as input for the next control cycle, thereby realizing continuous optimization and self-learning of the system.

Images