A load balancing method and system for vehicle charging demand

Abstract

The application provides a load balancing method and system for automobile charging demand, which is applied to the field of electric data processing; through identification and dynamic management and control of charging load distribution information under a multi-layer architecture of a power distribution network, the application realizes closed-loop control from "load identification-model construction-risk prediction-constraint update", can realize real-time sensing of charging pile operation state and power distribution network operation constraint under the condition of concentrated charging of electric vehicles, can predict power demand and identify potential risk areas in advance, can prevent local feeder overload, voltage out-of-limit and deterioration of power quality, and can also actively reduce power of a risk node and balance three-phase load through introduction of risk upper limit judgment and protection measures, so as to guarantee power supply stability.

Description

Technical Field

[0001] This invention relates to the field of electrical data processing, and in particular to a load balancing method and system for automobile charging needs. Background Technology

[0002] With the rapid growth in the number of electric vehicles (EVs), a large number of concentrated or simultaneous charging loads are becoming a new type of "peak load source" for the power distribution network. In many early charging load balancing studies, the goal was to smooth the load curve globally, such as distributing the charging power of a city or region evenly over time and avoiding system peaks.

[0003] However, the power system is not a "uniform energy pool" but a topological network composed of lines, transformers and nodes. Therefore, even if the total charging power of the city remains unchanged, local nodes may still experience voltage drops, overloads or power quality degradation due to uneven distribution of charging piles and limited line capacity. Summary of the Invention

[0004] This invention aims to solve the physical operational risks caused by centralized charging of electric vehicles, such as local overload of the power distribution network, voltage exceeding limits, and power quality degradation, and provides a load balancing method and system for electric vehicle charging demand.

[0005] The present invention employs the following technical means to solve the technical problem:

[0006] This invention provides a load balancing method for vehicle charging demand, comprising:

[0007] Based on the pre-defined architecture layer of the power distribution network, the charging load allocation information of the charging station is identified. Specifically, the architecture layer includes a data perception layer, a prediction and optimization layer, and a control execution layer.

[0008] Determine whether the charging load allocation information meets the preset safety constraints of the power distribution network;

[0009] If so, then based on the charging load allocation information, a power operation model of the charging station is constructed, the operating status of the distribution network is collected, and the predicted power demand of the charging station within a preset time period is generated through the power operation model. Based on the predicted power demand, the risk areas of the distribution network are identified. The power operation model specifically includes topology, line impedance, transformer capacity, node voltage level and feeder connection relationship, and the operating status specifically includes charging pile power, node voltage and current and transformer load.

[0010] Determine whether the number of risk areas reaches the preset risk limit of the charging station;

[0011] If the risk is reached, the risk factors of the risk area are obtained. Based on the risk factors, the preset protection measures of the distribution network are executed, the charging indicators of the charging station are collected, and the constraint parameters of the distribution network are dynamically updated according to the charging indicators. Specifically, the risk factors include low power factor, harmonic over-limit, and three-phase imbalance. The protection measures specifically include configuring reactive power compensation, limiting harmonic source power in dispatching, and adjusting single-phase pile power to balance the load. The charging indicators specifically include voltage over-limit number, equipment overload rate, power quality indicators, and user charging completion rate. The constraint parameters specifically include increasing distributed energy storage, upgrading transformers, and optimizing line structure.

[0012] Furthermore, after the step of constructing the power operation model of the charging station based on the charging load allocation information, the method further includes:

[0013] Based on the preset physical configuration of the charging station, the operating parameters of the power distribution network are obtained, wherein the physical configuration specifically includes the overall configuration of the station, power supply equipment, power distribution lines and charging pile configuration;

[0014] Determine whether the operating parameters match the boundary conditions of the power operation model;

[0015] If so, the branch connection relationship between each node is identified according to the pre-established electrical topology of the charging station and the distribution network. Based on the branch connection relationship, the equipment operation constraints of the charging station are dynamically adjusted. The branch connection relationship specifically includes the node level, branch direction and single-phase load distribution. The equipment operation constraints specifically include the overload allowable coefficient, allowable fluctuation range and power qualified range.

[0016] Furthermore, the step of identifying risk areas of the distribution network based on the predicted power demand also includes:

[0017] Based on the predicted power demand, the load concentration areas of the distribution network are divided, and the corresponding load growth trends are obtained from the load concentration areas.

[0018] Determine whether the load growth trend is caused by concentrated charging demand;

[0019] If so, the node power distribution of the load concentration area is obtained, and based on the node power distribution, abnormal operating indicators of the load concentration area are identified. Based on the abnormal operating indicators, potential risks of the load concentration area are obtained. The abnormal operating indicators specifically include load rate indicators, node voltage level indicators, and power factor indicators. The potential risks specifically include feeders approaching current limits, transformer load rates being too high, node voltages showing a downward trend, and three-phase imbalance in the line.

[0020] Furthermore, the step of collecting the charging indicators of the charging station and dynamically updating the constraint parameters of the distribution network based on the charging indicators also includes:

[0021] Based on the total charging power pre-collected by the charging station, the feeder capacity of the distribution network is obtained;

[0022] Determine whether the feeder capacity has reached the preset safe load;

[0023] If so, the feeder topology of the distribution network is identified, and the charging access request of the charging station is delayed according to the feeder topology. Based on the operating status of the risk area, the operating charging equipment of the charging station is dynamically adjusted. The feeder topology specifically includes lower-level nodes, branches, and charging range.

[0024] Furthermore, the step of determining whether the charging load allocation information meets the preset safety constraints of the distribution network also includes:

[0025] Based on the pre-classified equipment types of the charging station, the power data of the charging load allocation information is detected, wherein the equipment types specifically include linear equipment and nonlinear equipment, and the power data specifically includes harmonic content and neutral line current;

[0026] Determine whether the electrical energy data exceeds a preset constraint range;

[0027] If so, then the harmonic index of the power data is obtained, and the corresponding harmonic over-limit node is marked from the power data according to the harmonic index. Based on the harmonic over-limit node, the corresponding harmonic source type is identified. The harmonic index specifically includes harmonic distortion rate, harmonic component amplitude and waveform distortion coefficient, and the harmonic source type specifically includes rectifier harmonic source, inverter harmonic source and unbalanced harmonic source.

[0028] Furthermore, the step of determining whether the number of risk areas reaches the preset risk limit of the charging station also includes:

[0029] Based on the spatial location between risk areas, the feeder concentration of the risk areas is collected;

[0030] Determine whether the feeder concentration exceeds a preset upper limit for distribution;

[0031] If so, the clustering risk caused by the concentration of the feeder is identified. Based on the clustering risk, the voltage change direction and amplitude of each node in the risk area are monitored. Based on the voltage change direction and amplitude, the voltage linkage characteristics corresponding to the clustering risk are obtained. Specifically, the clustering risk includes local congestion and voltage linkage.

[0032] Furthermore, the step of identifying the charging load allocation information of the charging station based on the preset architecture layer of the distribution network also includes:

[0033] Based on the structural relationship of the architecture layer, the access point information of the charging station is obtained, wherein the structural relationship specifically includes the trunk, branches and the end, and the access point information specifically includes the access transformer, feeder number and voltage level;

[0034] Determine whether the access point information matches the distribution network node, wherein the distribution network node specifically includes the node number, the feeder to which it belongs, and the branch line path;

[0035] If so, then based on the mapping relationship of the distribution network nodes, the load characteristic information of each node is extracted, and based on the load characteristic information, the power distribution ratio of the charging load is generated. Specifically, the load characteristic information includes the node charging power, the connected phase, and the node load ratio.

[0036] The present invention also provides a load balancing system for vehicle charging demand, comprising:

[0037] The identification module is used to identify the charging load allocation information of the charging station based on the preset architecture layer of the power distribution network. The architecture layer specifically includes a data perception layer, a prediction and optimization layer, and a control execution layer.

[0038] The judgment module is used to determine whether the charging load allocation information meets the preset safety constraints of the power distribution network;

[0039] An execution module is configured to, if so, construct a power operation model for the charging station based on the charging load allocation information, collect the operating status of the distribution network, generate a predicted power demand for the charging station within a preset time period using the power operation model, and identify risk areas of the distribution network based on the predicted power demand. Specifically, the power operation model includes topology, line impedance, transformer capacity, node voltage level, and feeder connection relationship, and the operating status specifically includes charging pile power, node voltage and current, and transformer load.

[0040] The second judgment module is used to determine whether the number of risk areas reaches the preset risk limit of the charging station;

[0041] The second execution module is used to, if the risk is reached, acquire the risk factors of the risk area, execute the preset protection measures of the distribution network based on the risk factors, collect the charging indicators of the charging station, and dynamically update the constraint parameters of the distribution network according to the charging indicators. Specifically, the risk factors include low power factor, harmonic over-limit, and three-phase imbalance. The protection measures specifically include configuring reactive power compensation, limiting harmonic source power in scheduling, and adjusting single-phase pile power to balance the load. The charging indicators specifically include voltage over-limit number of times, equipment overload rate, power quality indicators, and user charging completion rate. The constraint parameters specifically include increasing distributed energy storage, upgrading transformers, and optimizing line structure.

[0042] Furthermore, it also includes:

[0043] The acquisition module is used to acquire the operating parameters of the power distribution network based on the preset physical structure of the charging station, wherein the physical structure specifically includes the overall configuration of the station, power supply equipment, power distribution lines and charging pile configuration;

[0044] The third judgment module is used to determine whether the operating parameters match the boundary conditions of the power operation model;

[0045] The third execution module is used to identify the branch connection relationship between each node according to the pre-established electrical topology of the charging station and the distribution network if the condition is met, and to dynamically adjust the equipment operation constraints of the charging station according to the branch connection relationship. The branch connection relationship specifically includes the node level, branch direction and single-phase load distribution, and the equipment operation constraints specifically include the overload allowable coefficient, allowable fluctuation range and power qualified range.

[0046] Furthermore, the execution module also includes:

[0047] The division unit is used to divide the load concentration area of ​​the distribution network based on the predicted power demand, and to obtain the corresponding load growth trend from the load concentration area;

[0048] The judgment unit is used to determine whether the load growth trend is caused by concentrated charging demand;

[0049] The execution unit is configured to, if so, acquire the node power distribution of the load concentration area, identify abnormal operating indicators of the load concentration area based on the node power distribution, and acquire potential risks of the load concentration area based on the abnormal operating indicators. The abnormal operating indicators specifically include load rate indicators, node voltage level indicators, and power factor indicators. The potential risks specifically include feeders approaching current limits, transformer load rates being too high, node voltages showing a downward trend, and three-phase imbalance in the line.

[0050] This invention provides a load balancing method and system for vehicle charging demand, which has the following beneficial effects:

[0051] This invention achieves closed-loop control from "load identification - model building - risk prediction - constraint update" by identifying and dynamically managing charging load distribution information in a multi-layered distribution network architecture. In addition to being able to perceive the operating status of charging piles and distribution network constraints in real time under centralized electric vehicle charging conditions, predict power demand in advance and identify potential risk areas to prevent local feeder overload, voltage exceeding limits and power quality deterioration, it can also proactively reduce the power of risk nodes and balance three-phase loads by introducing risk upper limit judgment and protection measures to ensure stable power supply. At the same time, it dynamically updates constraint parameters based on charging indicators, enabling the distribution network structure to adaptively optimize. Overall, it significantly improves the safety margin and operating efficiency of the distribution network under centralized charging scenarios, realizing intelligent and sustainable operation control under large-scale electric vehicle access conditions. Attached Figure Description

[0052] Figure 1 A schematic flowchart of one embodiment of the load balancing method for vehicle charging demand of the present invention;

[0053] Figure 2 This is a structural block diagram of one embodiment of the load balancing system for vehicle charging demand according to the present invention. Detailed Implementation

[0054] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. The realization of the purpose, functional features, and advantages of the invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings.

[0055] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0056] Reference Appendix Figure 1 A load balancing method for vehicle charging demand according to an embodiment of the present invention includes:

[0057] S1: Based on the pre-set architecture layer of the power distribution network, identify the charging load allocation information of the charging station, wherein the architecture layer specifically includes a data perception layer, a prediction optimization layer and a control execution layer;

[0058] S2: Determine whether the charging load allocation information meets the preset safety constraints of the power distribution network;

[0059] S3: If so, then based on the charging load allocation information, construct the power operation model of the charging station, collect the operating status of the distribution network, generate the predicted power demand of the charging station within a preset time period through the power operation model, and identify the risk areas of the distribution network based on the predicted power demand. The power operation model specifically includes topology, line impedance, transformer capacity, node voltage level and feeder connection relationship, and the operating status specifically includes charging pile power, node voltage and current and transformer load.

[0060] S4: Determine whether the number of risk areas reaches the preset risk limit of the charging station;

[0061] S5: If the risk is reached, the risk factors of the risk area are obtained. Based on the risk factors, the preset protection measures of the distribution network are executed, the charging indicators of the charging station are collected, and the constraint parameters of the distribution network are dynamically updated according to the charging indicators. The risk factors specifically include low power factor, harmonic over-limit, and three-phase imbalance. The protection measures specifically include configuring reactive power compensation, limiting harmonic source power in scheduling, and adjusting single-phase pile power to balance the load. The charging indicators specifically include voltage over-limit number, equipment overload rate, power quality indicators, and user charging completion rate. The constraint parameters specifically include increasing distributed energy storage, upgrading transformers, and optimizing line structure.

[0062] In this embodiment, the system identifies the charging load allocation information of the charging station based on a pre-established architecture layer of the distribution network. This architecture layer specifically includes a data sensing layer, a prediction and optimization layer, and a control execution layer. The system then determines whether this charging load allocation information meets the pre-set safety constraints of the distribution network and executes the corresponding steps accordingly. For example, if the system determines that the charging load allocation information of the charging station cannot meet the pre-set safety constraints of the distribution network, the system considers that the spatial distribution and power characteristics of the current charging load have violated the safety constraints of the distribution network operation. If no adjustment is made, it may cause local overload, voltage drop, or abnormal power quality, thereby affecting the stability and power supply reliability of the entire distribution system. Based on load allocation information and real-time monitoring data, the system will locate areas causing constraint violations and simultaneously suspend or delay the scheduling of new charging tasks in those areas to prevent further increases in local current. It will also reduce the power of already connected charging loads or implement time-sharing and zone-based current limiting to reduce instantaneous load peaks. For example, when the system determines that the charging load allocation information of the charging station meets the pre-set safety constraints of the distribution network, it will consider the spatial distribution and power characteristics of the current charging load to conform to the safety constraints of the distribution network operation. Based on the charging load allocation information, the system will construct a power operation model of the charging station, which specifically includes the topology, line impedance, and transformer... The system collects data on charging station capacity, node voltage levels, and feeder connections to determine the operational status of the distribution network. This operational status includes charging pile power, node voltage and current, and transformer load. Using this power operation model, the system generates predicted charging power demand for charging stations within a pre-set time period. Based on these predicted power demands, it identifies risk areas in the distribution network. By ensuring that the charging load allocation information meets the distribution network's safety constraints before proceeding with subsequent modeling and prediction, the system ensures that the spatial distribution and power characteristics of electric vehicle charging loads remain within a safe operating range, avoiding operational deviations caused by blind modeling or scheduling. Furthermore, the system constructs a power operation model based on load allocation information under safety constraints. This model includes the distribution network topology, line impedance, transformer capacity, node voltage levels, and feeder connections. Combined with real-time collected operating status parameters such as charging pile power, node voltage and current, and transformer load, it can dynamically calculate the predicted charging power demand within a preset time period. Based on generating the predicted charging power demand and identifying risk areas, it enables adaptive adjustments to the distribution network. For example, by optimizing the charging sequence in advance, adjusting power distribution, or balancing the load structure, it can effectively improve line utilization and power distribution efficiency without increasing hardware investment. Then, the system determines whether the number of these risk areas reaches the pre-set risk limit of the charging station and executes the corresponding steps.For example, if the system determines that a risk area in the distribution network has not reached the pre-set risk limit of the charging station, the system will consider the distribution network as still within a safe operating range. The system will continue to perform prediction, optimization, and monitoring based on this. It will maintain the current charging task allocation and power output strategy, without actively reducing charging power to ensure user charging experience and plan completion rate. Simultaneously, it will perform real-time monitoring of the few identified risk areas, tracking changes in load, voltage, and power factor to determine if there are signs of risk spread or worsening. Furthermore, based on the latest operating data, it will recalculate the predicted power demand for future periods using the power operation model to ensure the prediction results are accurate and reliable. To reflect the latest load dynamics and maintain the timeliness and accuracy of the model, for example, when the system determines that a risk area in the distribution network has reached the pre-set risk limit of the charging station, the system will consider the distribution network to be in an abnormal operating range. The system will collect risk factors for these risk areas, specifically including low power factor, harmonic over-limit, and three-phase imbalance. Based on different risk factors, the system will execute the pre-set protection measures of the distribution network, specifically including configuring reactive power compensation, limiting harmonic source power in dispatching, and adjusting single-phase pile power to balance the load. The system will also collect charging indicators of the charging station, specifically including the number of voltage over-limit occurrences, equipment overload rate, power quality indicators, and user charging completion. Based on these charging indicators, the system dynamically updates the constraint parameters of the distribution network. These constraints include increasing distributed energy storage, upgrading transformers, and optimizing line structure. When the system determines that the number of risk areas in the distribution network reaches the preset risk limit for charging stations, it indicates that the local operating state of the distribution network has entered an abnormal range. By collecting risk factors such as low power factor, excessive harmonics, and three-phase imbalance, the system can accurately identify the specific causes and impact range of the anomalies. Based on the differences in risk factors, the system executes corresponding protection measures, such as configuring reactive power compensation, limiting harmonic source power, or adjusting the power distribution of single-phase charging piles, thereby proactively suppressing the fault source before the problem spreads and quickly restoring the system. Operational stability is significantly improved, enhancing the safety and self-healing capabilities of the distribution network. Furthermore, by implementing various protection measures, the system can improve the operational quality of the distribution network in multiple dimensions. Reactive power compensation effectively improves power factor and voltage stability; limiting harmonic source power reduces harmonic distortion rate, ensuring the safe operation of critical equipment; adjusting single-phase pile power balances three-phase loads and reduces neutral current. After implementing protection measures, the system dynamically updates the distribution network's constraint parameters based on real-time collected charging indicators, including voltage overruns, equipment overload rate, power quality indicators, and user charging completion rate. This includes measures such as increasing distributed energy storage capacity, upgrading transformers, or optimizing line structure.

[0063] It should be noted that, based on the charging load allocation information, a power operation model for the charging station is constructed, the operating status of the distribution network is collected, and the predicted power demand for the charging station within a preset time period is generated through the power operation model. Based on the predicted power demand, risk areas of the distribution network are identified, specifically:

[0064] First, a power operation model for the charging station is constructed based on the charging load allocation information. This model is based on the physical connection relationship of the charging station in the distribution network, and comprehensively considers power parameters such as its topology, line impedance, transformer capacity, node voltage level and feeder connection relationship. It is used to accurately describe the electrical characteristics of the charging station and its energy interaction with the distribution network. The input of the model includes the load power, connected phase, location number and feeder of each charging node, and the output is the node voltage, current distribution and power flow direction.

[0065] After the model is established, the system collects the operating status of the distribution network, including real-time monitored data such as charging pile power, node voltage and current, and transformer load rate. This information reflects the current operating conditions and load level of the distribution network. By combining the operating status with the model parameters, the system can reflect the dynamic coupling relationship between the charging load and the distribution network.

[0066] Subsequently, based on the power operation model, the system generates the predicted power demand of charging stations within a preset time period. Specifically, the system uses historical charging behavior data, user charging reservation information, and time characteristics (such as morning and evening peak hours, holidays, temperature, etc.) to predict the load change trend in each time period in the future. The model calculates the voltage, current, and active and reactive power flow distribution of different nodes accordingly, thereby obtaining the power demand curve within the preset time period.

[0067] Finally, the system identifies risk areas in the distribution network based on predicted power demand. The core of risk identification is to compare the prediction results with the distribution network safety constraint thresholds: if the current of a feeder exceeds its rated capacity, it is identified as an "overload risk area"; if the predicted voltage value of a node is lower than the allowable lower limit (e.g., 0.9 pu), it is identified as a "voltage drop risk area"; if the load rate of a transformer exceeds the safety limit, it is identified as a "transformer high load risk area"; if the three-phase current is unbalanced or the power factor is low, it is marked as a "power quality risk area".

[0068] Specific examples are as follows:

[0069] Assume that there is a 10 kV power distribution feeder connecting three electric vehicle charging stations within the park; each charging station includes 20 DC fast charging piles (60 kW per pile) and several slow charging piles; the system operates as follows:

[0070] Constructing a power operation model: The system establishes an electrical connection model between the three charging stations and the distribution feeder, inputting line impedance, transformer capacity (e.g., 800 kVA), node voltage level (400 V / 10 kV), and load mapping relationship;

[0071] Real-time data collection of charging pile power (e.g., current output of fast charging pile is 45 kW), voltage of each node (e.g., 0.96 pu), and current load rate of transformer (e.g., 85%).

[0072] Generate predicted power demand: Based on historical data and user charging reservations, it is predicted that the charging power will increase by 20% in the next two hours, and the predicted voltage of the feeder end node is calculated to be 0.88 pu.

[0073] Risk area identification: The system determines that the voltage of the node is lower than the set lower limit of 0.9 pu and marks it as a "voltage drop risk area"; at the same time, it detects that the predicted value of the transformer load rate will reach 98% and marks it as an "equipment overload risk area"; therefore, the system outputs a risk warning and includes this area in the subsequent load balancing and scheduling optimization scope;

[0074] In summary, through the above steps, the system can establish an accurate power operation model based on charging load allocation information, and combine real-time operating status to predict future power and identify risk areas. This process realizes a closed-loop analysis from data perception to risk pre-judgment, which not only improves prediction accuracy and operational visualization, but also enables the distribution network to identify potential bottlenecks in the scenario of centralized charging of electric vehicles in advance, ensuring power supply security and stable network operation.

[0075] In this embodiment, after step S3 of constructing the power operation model of the charging station based on the charging load allocation information, the method further includes:

[0076] S301: Based on the preset physical configuration of the charging station, obtain the operating parameters of the power distribution network, wherein the physical configuration specifically includes the overall configuration of the station, power supply equipment, power distribution lines and charging pile configuration;

[0077] S302: Determine whether the operating parameters match the boundary conditions of the power operation model;

[0078] S303: If so, then based on the pre-established electrical topology between the charging station and the distribution network, identify the branch connection relationship between each node, and dynamically adjust the equipment operation constraints of the charging station according to the branch connection relationship. The branch connection relationship specifically includes node level, branch direction and single-phase load distribution, and the equipment operation constraints specifically include overload allowable coefficient, allowable fluctuation range and power qualified range.

[0079] In this embodiment, the system obtains the operating parameters of the power distribution network based on the pre-defined physical configuration of the charging station, which specifically includes the overall site configuration, power supply equipment, power distribution lines, and charging pile configuration. The system then determines whether these operating parameters match the boundary conditions of the power operation model to execute corresponding steps. For example, if the system determines that the operating parameters of the power distribution network cannot match the boundary conditions of the power operation model, the system will consider that the power operation model can no longer accurately reflect the current physical state or operating boundary of the power distribution network. Continuing to rely on the original model for prediction and scheduling may lead to errors in risk identification, delayed control response, or even trigger [unspecified events]. For operational risks such as localized overload or voltage exceeding limits, the system compares the differences between operating parameters and model input boundaries to identify the source of mismatch, such as equipment parameter drift, topology changes, or data acquisition errors. This clarifies the source of the deviation, determines subsequent correction strategies, and simultaneously suspends the use of the current power operation model for risk prediction or load optimization to prevent erroneous operations caused by decisions based on incorrect boundaries. Furthermore, if the mismatch originates from structural changes (such as topology alterations or equipment replacement), the system automatically reconstructs the power operation model, regenerating node connections, power distribution, and constraint boundaries. For example, when the system determines that the operating parameters of the distribution network match the power operation model... The system assumes that the power operation model can still accurately reflect the current physical state or operating boundary of the distribution network under the boundary conditions. Based on the pre-established electrical topology between the charging station and the distribution network, the system identifies the branch connections between nodes. These branch connections specifically include node level, branch direction, and single-phase load distribution. Based on different branch connections, the system dynamically adjusts the equipment operation constraints of the charging station. These constraints specifically include overload allowable coefficients, allowable fluctuation ranges, and power compliance intervals. By maintaining a high degree of consistency between the model and actual operating conditions, the system avoids risk misjudgments and control errors caused by model deviations. To ensure the reliability and real-time performance of subsequent charging scheduling, load balancing, and energy distribution calculations, the system significantly improves prediction accuracy and operational stability. After confirming the model's effectiveness, the system identifies the branch connections between nodes based on the pre-established electrical topology of the charging station and distribution network, including node hierarchy, branch direction, and single-phase and three-phase load distribution. Utilizing the topology identification results and equipment constraint parameters, the system achieves adaptive adjustment of charging power distribution and equipment load. When a local node is detected to be approaching its operating limit, the system can automatically reduce the overload allowable coefficient of that node or narrow the fluctuation range to prevent local overload or voltage exceeding limits.

[0080] It should be noted that the operating parameters mainly include: the voltage amplitude and phase angle range of the upstream bus, the maximum allowable current or power of the feeder connected to the charging station, the possible voltage fluctuation range and short-circuit capacity, and the real-time load information provided by the distribution automation system (such as other loads and distributed photovoltaic access status).

[0081] It should be added that, based on the pre-established electrical topology between the charging station and the power distribution network, the branch connection relationships between each node are identified, and the equipment operation constraints of the charging station are dynamically adjusted according to these branch connection relationships. Specifically:

[0082] The system first identifies the physical access location of the charging station in the entire power distribution network and its relationship with its superior and subordinate nodes based on the pre-established electrical topology of the charging station and the power distribution network. This electrical topology is based on the components of the power distribution network, such as lines, transformers, switching equipment and charging piles, and describes the multi-level connection method from the main feeder to the end charging pile, which is used to reflect the energy flow path and node dependency relationship.

[0083] Based on this, the system identifies the branch connections between nodes; the specific branch connections include:

[0084] Node hierarchy: This represents the hierarchical structure between charging stations, distribution transformers, branch lines, and end loads, and is used to determine the direction of power transmission.

[0085] Branch direction relationship: reflects the power flow from the upper-level node to the lower-level node, and is used to determine the current flow direction and load distribution path;

[0086] Single-phase and three-phase load distribution relationship: used to distinguish the phase connection between nodes and identify possible phase current imbalances;

[0087] After identification, the system will dynamically adjust the equipment operation constraints of the charging station based on different branch connection relationships. The equipment operation constraints are a set of parameters for boundary control of the operating state of the charging equipment, specifically including:

[0088] Overload tolerance factor: used to limit the maximum current or power that the equipment can withstand for a short period of time;

[0089] Allowable fluctuation range: Defines the safe range within which node voltage and current can fluctuate within a certain range;

[0090] Power acceptable range: It stipulates that charging equipment should operate within a range where the power factor and output power meet the power quality standards;

[0091] During real-time operation, the system dynamically updates these constraint parameters based on the topological location and load transmission characteristics of each node. For example, if the load rate of the upstream feeder is close to the limit, the system will reduce the overload allowable factor of the downstream node; if the load distribution of the branch node is unbalanced, the operating power of the single-phase pile will be adjusted; if the voltage fluctuation at the end of a branch is large, the voltage allowable range of that node will be tightened. Through this topologically related constraint adjustment mechanism, the system achieves refined control over the operation of charging equipment, ensuring the safety and stability of the entire distribution network.

[0092] Specific examples are as follows:

[0093] Let's take a 10 kV distribution feeder in a certain city as an example:

[0094] The feeder connects to two branch nodes, A and B. Node A connects to a large public charging station (containing 20 fast charging piles), and node B connects several residential areas with small slow charging piles. The system has the following pre-established electrical topology in the distribution network:

[0095] Main feeder line → Node A (charging station) → Fast charging piles A1~A20

[0096] Main feeder → Node B (residential load)

[0097] (1) Identifying branch connection relationships, the system identifies that nodes A and B are both at the downstream level of the same feeder, but node A has concentrated power and strong fluctuations, while node B is a continuous load; the topology analysis results show that:

[0098] The impedance between node A and the main feeder is relatively large, and power changes can easily affect the main line voltage.

[0099] The three-phase load imbalance in branch A at node A is 7%, indicating a slight phase imbalance.

[0100] (2) Dynamically adjust operational constraints. Based on the above branching relationships, the system performs the following dynamic constraint adjustments:

[0101] When the main feeder load rate increases to 90%, the overload allowable factor of node A branch is automatically reduced from 1.2 to 1.05 to suppress the simultaneous start-up behavior of fast charging piles; when the A phase current is detected to be too high, the system reduces the output power of the corresponding charging pile by 10% and appropriately increases the output of the B and C phase piles to balance the three-phase current; if the voltage at the end of node A is lower than 0.93pu, the system adjusts the allowable voltage fluctuation range of the node from ±7% to ±5% to prevent the voltage drop from further expanding.

[0102] (3) Operation results: Through this dynamic adjustment method, the main feeder voltage is stabilized within the allowable range (0.95–1.05 pu), the instantaneous load fluctuation of the charging pile group is reduced by about 15%, and the phase current imbalance is reduced to 4%; the system realizes precise control based on topology, so that the local load adjustment is consistent with the safety constraints of the whole network, and avoids the problem of local overload or voltage over-limit caused by centralized charging.

[0103] In summary, in the examples above, the system identifies node branch relationships based on the electrical topology and dynamically adjusts equipment operating constraints accordingly. The system can accurately reflect network structure characteristics and power flow direction, achieving node-level monitoring and control. At the same time, it flexibly adjusts constraint parameters according to the operating status of different branches, enhancing the load adaptability of the distribution network. Furthermore, it maintains overall network stability and qualified power quality, improving equipment utilization while ensuring safety.

[0104] In this embodiment, step S3, which identifies risk areas of the distribution network based on the predicted power demand, further includes:

[0105] S31: Based on the predicted power demand, divide the load concentration areas of the distribution network and obtain the corresponding load growth trends from the load concentration areas;

[0106] S32: Determine whether the load growth trend is caused by concentrated charging demand;

[0107] S33: If so, obtain the node power distribution of the load concentration area, identify the abnormal operation indicators of the load concentration area based on the node power distribution, and obtain the potential risks of the load concentration area based on the abnormal operation indicators. The abnormal operation indicators specifically include load rate indicators, node voltage level indicators and power factor indicators. The potential risks specifically include feeders approaching current limits, transformer load rates being too high, node voltages showing a downward trend and lines having three-phase imbalance.

[0108] In this embodiment, the system divides the power grid into load concentration areas based on predicted power demand, obtains the corresponding load growth trends from these load concentration areas, and then determines whether the load growth trend is caused by concentrated charging demand to execute corresponding steps. For example, when the system determines that the load growth trend in a load concentration area is not caused by concentrated charging demand, the system considers the load change in that area to be a non-charging power increase, that is, the load increase is mainly caused by other types of electricity consumption behavior or external factors, rather than the concentrated operation of electric vehicle charging stations. The system will then remove that area from the "Charging Concentration Monitoring Area" label and mark it as... The "Regular Load Growth Zone" is used to distinguish it from controllable load growth zones caused by charging. It does not adjust the operating status, power allocation, or scheduling strategies of charging piles, maintaining the existing power distribution operation mode to prevent interference with normal load power supply due to misjudgment. The system collects voltage, current, and power changes in this zone during subsequent monitoring periods. If the growth trend continues and remains unrelated to charging demand, it is confirmed as stable load growth; otherwise, a re-judgment process is triggered to dynamically track potential charging load intervention. For example, when the system determines that the load growth trend in a concentrated load area is caused by concentrated charging demand, the system will consider the load change in that area as... For normal charging power increases, the system acquires the node power distribution in these load-concentrated areas. Based on the different node power distributions, it identifies abnormal operating indicators in these load-concentrated areas. These abnormal operating indicators include load rate indicators, node voltage level indicators, and power factor indicators. Based on these different abnormal operating indicators, the system identifies potential risks in these load-concentrated areas, including feeders approaching current limits, transformer overload, node voltage showing a downward trend, and three-phase imbalance in the line. By extracting and analyzing the node power distribution in these areas, the system can clearly reflect the load differences and power flow direction between different nodes. This provides a reliable foundation for subsequent risk assessment and scheduling optimization. After acquiring the node power distribution in the load-concentrated area, the system can identify abnormal operating indicators based on real-time monitoring data, including load rate indicators, node voltage level indicators, and power factor indicators. After identifying different abnormal operating indicators, the system further analyzes their corresponding potential risk types, such as feeder approaching current limits, transformer overload, node voltage drop trends, or three-phase imbalance of the line. The system can implement hierarchical management of the load-concentrated area according to the degree of risk: optimize operating parameters for mild risks, trigger early warnings for moderate risks, and perform load diversion or power limiting operations for severe risks.

[0109] It should be noted that obtaining the node power distribution of the load concentration area, identifying abnormal operating indicators of the load concentration area based on the node power distribution, and obtaining potential risks of the load concentration area based on the abnormal operating indicators are specifically as follows:

[0110] After determining that the power increase in the concentrated load area is caused by the charging demand of electric vehicles, the system first obtains the node power distribution in the concentrated load area. The node power distribution reflects the active power, reactive power and their rate of change of different nodes in the same time period, and can characterize the spatial distribution differences of the load and the direction of power flow. By collecting the current, voltage and phase information of each node in real time, the system can construct a node power distribution map to assess whether the power flow direction within the area is balanced and whether there is a sudden increase in load in local branches.

[0111] Next, the system identifies abnormal operating indicators in the concentrated load area based on the node power distribution. These abnormal operating indicators reflect the degree to which the operating status of a node or branch deviates from the normal range, and mainly include the following three categories:

[0112] Load factor: measures the ratio of the actual power of a node or device to its rated capacity. If it is continuously higher than the safety threshold (such as 90%), it is considered an overload risk signal.

[0113] Node voltage level index: reflects the extent to which the node voltage deviates from the rated voltage. If it is below 0.9 pu or above 1.1 pu, it indicates a tendency to exceed the voltage limit.

[0114] Power factor: This indicates the ratio of active to reactive power. If the power factor is below 0.9, it means that the reactive power consumption in the area is too high and the line loss is increased.

[0115] The system analyzes the above indicators to determine whether there are operational deviations or uneven energy distribution in the region. Then, based on different abnormal operational indicators, the system further identifies potential risks in the load-concentrated area. These potential risks are specific grid operation hazards caused by abnormal indicators, mainly including:

[0116] Feeder current close to limit: When the load rate is abnormally high, it indicates that the feeder current is approaching the upper limit of the allowable limit, which may cause thermal damage or protection action.

[0117] High transformer load rate: This indicates that the transformer has been under high load for a long time, which may cause overheating, shortened lifespan, or tripping.

[0118] There is a downward trend in node voltage: When the power of downstream nodes increases and the voltage drop of upstream lines increases, it may lead to the voltage exceeding the limit on the user side.

[0119] The line has a three-phase imbalance: if the three-phase load is not distributed evenly, it can easily cause problems such as excessive neutral current and deterioration of power quality.

[0120] Through this process, the system can automatically identify "power data" and "operational risks," providing a basis for subsequent optimization scheduling and protection strategies.

[0121] In this embodiment, step S5, which involves collecting the charging indicators of the charging station and dynamically updating the constraint parameters of the distribution network based on the charging indicators, further includes:

[0122] S51: Based on the total charging power pre-collected by the charging station, obtain the feeder capacity of the distribution network;

[0123] S52: Determine whether the feeder capacity has reached the preset safe load;

[0124] S53: If so, identify the feeder topology of the distribution network, delay the charging access request of the charging station according to the feeder topology, and dynamically adjust the operating charging equipment of the charging station according to the operating status of the risk area. The feeder topology specifically includes lower-level nodes, branches and charging range.

[0125] In this embodiment, the system obtains the feeder capacity of the distribution network based on the total charging power pre-collected by the charging station. The system then determines whether these feeder capacities have reached a pre-set safe load and executes corresponding steps accordingly. For example, if the system determines that the feeder capacity of the distribution network has not reached the pre-set safe load, the system considers the current operating state of the distribution network to be within a safe margin range, meaning the actual transmission power of the feeder is lower than its rated carrying capacity. The system will maintain the existing charging pile power allocation scheme without forcibly reducing the charging load or delaying the charging plan. If it detects that some nodes are under light load, the system can appropriately guide more charging loads to that area to improve overall power distribution efficiency. Simultaneously, it continuously monitors the feeder current change trend based on time-series data and records the charging... If the load growth rate trend indicates that future power may approach the safe limit, the system can issue a mild early warning signal. Under safe margin conditions, the system can further improve grid energy efficiency through optimization (such as power balancing or electricity price guidance). For example, it can encourage fast charging piles to operate during low load periods to distribute peak load pressure, reduce line losses, and improve power supply economy. For example, when the system determines that the feeder capacity of the distribution network has reached the preset safe load, the system will consider the current operating state of the distribution network to be in an abnormal range. The system will identify the feeder topology of the distribution network, which specifically includes lower-level nodes, branches, and charging range. Based on these feeder topologies, the system will delay the charging access request of the charging station and dynamically adjust the operating charging equipment of the charging station according to the operating status of the risk area.By identifying the feeder topology of the distribution network (including downstream nodes, branches, and charging range), the system can accurately locate load concentration areas and affected downstream nodes. This allows it to determine which lines, electrical nodes, or charging groups are in a high-risk operating state, thus suppressing overload at its source. By delaying some charging access requests, the system effectively prevents feeder current from continuing to rise, avoiding problems such as line overheating, protection device malfunctions, or accelerated equipment aging, ensuring the electrical safety and operational stability of the distribution network. Furthermore, after identifying the feeder topology, the system can dynamically schedule and adjust the power of charging equipment based on the operating status of risk areas (such as current, node voltage, and load rate). Specifically, the system can automatically reduce the output power of some high-power charging piles or prioritize suspending charging at points located downstream of bottleneck branches. The system receives connection requests, thereby redistributing power flow among different branches to achieve load balancing. This process is adaptive, capable of multi-level load adjustment based on topology and real-time state changes. This ensures the entire charging network maintains orderly operation within capacity limits. By dynamically adjusting charging equipment and delaying load access, the system not only prevents local line overload risks but also maximizes the utilization of remaining feeder capacity and available power resources. This distributed, topology-aware control mechanism gives the distribution network greater operational resilience, maintaining service continuity under load fluctuations and sudden increases in charging demand. During the adjustment process, the system continuously updates node operating parameters, providing a more accurate data foundation for subsequent power prediction and scheduling optimization, thereby improving the overall energy efficiency and adaptive management capabilities of the power grid.

[0126] It should be noted that identifying the feeder topology of the distribution network, delaying the charging access request of the charging station based on the feeder topology, and dynamically adjusting the operating charging equipment of the charging station according to the operating status of the risk area, specifically involves:

[0127] The system first identifies the feeder topology of the distribution network; the feeder topology is used to describe the nodes, branches, and their electrical connections within the distribution network.

[0128] Specifically, it includes:

[0129] Lower-level node information: reflects the hierarchical connection between transformers, branch lines and end-point charging piles;

[0130] Branch structure: describes the power transmission path, power distribution direction, and the mutual influence between branches;

[0131] Charging range mapping: Defines the charging area covered by each feeder and the corresponding load concentration area;

[0132] Through topology identification, the system can determine the spatial location of the load peak impact and its hierarchical relationship, and distinguish the risk levels of the main feeder and the terminal branch; then, the system delays the charging access request of the charging station according to the identified feeder topology.

[0133] Specifically, when the current of a feeder or branch is detected to be close to the safety threshold, the system will temporarily suspend new access requests for some charging piles under that branch, or adjust the queuing strategy to postpone their start-up time. This delay control mechanism can reduce the instantaneous charging load in a short time and avoid power spikes causing line overload or a sharp drop in node voltage.

[0134] Finally, the system dynamically adjusts the operating charging equipment of the charging station based on the operational status of the risk area. The operational status includes real-time load rate, voltage level, power factor, and equipment temperature rise within the risk area. The system uses this status data to perform differentiated control on the charging pile group, for example:

[0135] Reduce the power of some fast charging stations or switch to slow charging mode;

[0136] Suspend the operation of equipment with low power factor or high harmonic distortion;

[0137] Activate backup charging stations in low-risk areas to distribute the load.

[0138] Through this process, the system achieves topology-aware adjustment and dynamic risk mitigation for high-load feeders, thereby keeping the entire distribution network operating within safe and stable boundaries.

[0139] Specific examples are as follows:

[0140] Taking a city's integrated charging station as an example, this station is connected to the regional power distribution network through two 10kV feeders (F1 and F2):

[0141] Feeder F1's power supply range covers the main charging area, with a total of 60 fast charging piles and a rated total power of 3 MW;

[0142] Feeder F2 supplies power to the auxiliary parking area, which has 40 slow charging piles with a total rated power of 1 MW.

[0143] During the evening peak hours, the system monitored that the real-time load of feeder F1 had reached 98% of its rated capacity, and the current was close to the protection action threshold.

[0144] Based on this, the system enters the topology identification stage and reads the network structure of the F1 feeder:

[0145] The upstream node is the area transformer T1 (capacity 4 MVA).

[0146] The lower-level node is divided into three branches: B1, B2, and B3;

[0147] Among them, the B2 branch line connects to 20 fast charging piles, and has the highest load concentration;

[0148] By analyzing the topology and power flow distribution, the system identified branch B2 as the main risk area.

[0149] At this point, the system performs the following steps:

[0150] Delayed charging access request:

[0151] The activation requests for the 8 newly added fast charging piles under the B2 branch line will be suspended for approximately 10 minutes to prevent peak charging spikes from accumulating.

[0152] Dynamically adjust charging equipment:

[0153] Increase the power of the slow charging piles in the lower part of B1 branch road by 20% to guide some vehicles to switch to the standby charging area;

[0154] At the same time, the power of the existing fast charging piles on the B2 branch will be reduced to 80% of the rated value;

[0155] Continuously monitor the risk status:

[0156] The system continuously monitors the current and node voltage of branch B2. Once the current drops to 90% of the rated value, normal access is gradually restored.

[0157] The final results show that:

[0158] The peak current of feeder F1 decreased by approximately 9%, and the node voltage recovered from 0.91 pu to 0.95 pu;

[0159] The charging service delay time was kept within an acceptable range, and the overall charging completion rate remained above 97%.

[0160] The power distribution network is operating stably, with no overloads or protection tripping occurring.

[0161] In summary, the above examples demonstrate that the system achieves accurate risk location based on topology. The system can quickly identify load-concentrated branches and high-risk nodes, while using delay control to achieve flexible load shaving, prevent instantaneous overload, ensure the safe operation of the distribution network, and dynamically schedule charging equipment based on risk status, maintaining overall service efficiency while controlling risks.

[0162] In this embodiment, step S2, which determines whether the charging load allocation information meets the preset safety constraints of the distribution network, further includes:

[0163] S21: Based on the pre-divided equipment types of the charging station, detect the power data of the charging load allocation information, wherein the equipment types specifically include linear equipment and nonlinear equipment, and the power data specifically includes harmonic content and neutral line current;

[0164] S22: Determine whether the electrical energy data exceeds the preset constraint range;

[0165] S23: If so, obtain the harmonic index of the power data, mark the corresponding harmonic over-limit node from the power data according to the harmonic index, and identify the corresponding harmonic source type according to the harmonic over-limit node. The harmonic index specifically includes harmonic distortion rate, harmonic component amplitude and waveform distortion coefficient, and the harmonic source type specifically includes rectifier type harmonic source, inverter type harmonic source and unbalanced harmonic source.

[0166] In this embodiment, the system detects the electrical data of the charging load allocation information based on the pre-defined equipment types of the charging station, specifically linear and nonlinear equipment. This electrical data includes harmonic content and neutral current. The system then determines whether this electrical data exceeds a pre-set constraint range and executes corresponding steps accordingly. For example, if the system determines that the electrical data of the charging load allocation information does not exceed the pre-set constraint range, the system assumes that the harmonics, current imbalances, and neutral current generated by all equipment in the current charging station are within acceptable limits. The system maintains the existing power allocation and scheduling plan, without adjusting the output or access sequence of charging piles, ensuring continuous charging for users. Simultaneously, it continues to periodically collect harmonic content and neutral current data, performs waveform analysis on key nodes (such as the main feeder and transformer neutral point), and establishes a power quality time series model for subsequent risk prediction. Under good power quality conditions, the system can automatically fine-tune the switching status of reactive power compensation equipment or adjust the operating ratio of linear and nonlinear loads to maintain the overall power factor of the system at a high level (e.g., ≥0).95), further improving the operational efficiency of the distribution network; for example, when the system determines that the power data of the charging load allocation information exceeds the preset constraint range, the system will consider that the equipment in the current charging station is malfunctioning. The system will obtain the harmonic indicators of these power data, which specifically include harmonic distortion rate, harmonic component amplitude, and waveform distortion coefficient. Based on different harmonic indicators, the system will mark the corresponding harmonic over-limit nodes from these power data. Based on the harmonic over-limit nodes, the system will identify the corresponding harmonic source type, which specifically includes rectifier type. Harmonic sources, inverter-type harmonic sources, and unbalanced harmonic sources; the system extracts harmonic indicators (including harmonic distortion rate, harmonic component amplitude, and waveform distortion coefficient) to accurately identify harmonic anomalies from variations in different frequency components and amplitudes. By comparing standard thresholds with measured node data, it automatically marks nodes exceeding harmonic limits, achieving spatial location and hierarchical identification of anomalies. Furthermore, after marking nodes exceeding harmonic limits, it further identifies the type of harmonic source (including rectifier-type, inverter-type, and unbalanced) based on harmonic distribution characteristics. Rectifier-type harmonic sources are often found in fast-charging stations or high-power AC / DC converters. Harmonic sources, caused by conversion devices, typically originate from on-board inverter systems or distributed energy storage grid-connected devices. Unbalanced harmonic sources may be caused by uneven distribution of single-phase charging equipment. Through this classification and identification, the system can achieve an intelligent process of "source identification—classification and management," providing a basis for subsequent implementation of different suppression strategies (such as passive filtering, harmonic power limiting, or load redistribution). This significantly improves the accuracy and efficiency of harmonic management. Furthermore, the system can quickly detect, mark, and classify harmonic sources when anomalies occur. Without affecting the overall power supply continuity, the system can take targeted measures to suppress harmonic propagation or interference. For example, it can prioritize limiting the output power of major harmonic source equipment or activate local filtering devices to stabilize the voltage waveform. This refined source control method not only reduces the scope of intervention on other normal charging equipment but also significantly reduces the negative impact of harmonics on the voltage stability, power factor, and equipment lifespan of the distribution network, thereby maintaining power quality standards and improving the overall reliability and operational safety of the system.

[0167] It should be noted that the process of acquiring the harmonic index of the electrical energy data, marking the corresponding harmonic exceedance nodes from the electrical energy data based on the harmonic index, and identifying the corresponding harmonic source type based on the harmonic exceedance nodes, specifically involves:

[0168] To determine the source of the anomaly, the system first acquires the harmonic index of the electrical energy data; the harmonic index is used to characterize the characteristics of non-fundamental components in the electrical energy waveform, and mainly includes:

[0169] Harmonic distortion rate (THD): represents the ratio of the root mean square value of each harmonic voltage or current to the fundamental component, reflecting the overall degree of waveform distortion;

[0170] Harmonic component amplitude: This represents the specific amplitude of each harmonic (such as the 3rd, 5th, 7th, 11th, etc.), and is used to determine whether the harmonic at a single frequency point exceeds the limit.

[0171] Waveform distortion coefficient: describes the degree to which a voltage or current waveform deviates from a sinusoidal shape, and is used to comprehensively evaluate the nonlinear effects of equipment operation;

[0172] The system performs spectral analysis on the power data using the aforementioned indicators, compares the harmonic frequency components with standard limits, and thus marks the corresponding harmonic exceedance nodes from the power data. The marking methods for these nodes include:

[0173] If the total harmonic distortion (THD) of a node exceeds 5%, it is marked as a first-level over-limit node.

[0174] If the amplitude of a specific harmonic (such as the 5th or 7th) at a certain node exceeds the allowable value (such as 4% of the fundamental frequency), it is marked as a level 2 over-limit node.

[0175] If the waveform distortion coefficient continuously exceeds the set time threshold (e.g., 10 minutes), it is marked as a persistent harmonic source node.

[0176] After node marking is completed, the system further identifies the corresponding harmonic source type based on the harmonic over-limit nodes; the identification of the harmonic source type is based on different frequency components and phase characteristics:

[0177] Rectifier-type harmonic sources: These are mostly characterized by odd harmonics dominating (3rd, 5th, and 7th harmonics being relatively strong), and are usually caused by AC / DC converters or fast charging power supply modules;

[0178] Inverter-type harmonic sources: mainly high-frequency harmonics (11th, 13th, and 17th harmonics), with sharp rising edges of the waveforms, commonly found in DC / AC inverter circuits or energy storage grid-connected devices;

[0179] Unbalanced harmonic sources are mainly characterized by significant differences in the amplitude of the three-phase currents, an increase in the third harmonic current, and an abnormal rise in the neutral line current. They are mostly caused by uneven distribution of single-phase loads.

[0180] Through this feature recognition method, the system can classify and locate the source of harmonic anomalies, providing a precise basis for subsequent implementation of targeted suppression measures (such as putting on filter devices, reducing the power of specific equipment, or adjusting load distribution).

[0181] In this embodiment, step S4, which determines whether the number of risk areas reaches the preset risk limit of the charging station, further includes:

[0182] S41: Based on the spatial location between risk areas, collect the feeder concentration of the risk areas;

[0183] S42: Determine whether the feeder concentration exceeds the preset distribution upper limit;

[0184] S43: If so, identify the clustering risk caused by the feeder concentration, monitor the voltage change direction and amplitude of each node in the risk area according to the clustering risk, and obtain the voltage linkage characteristics corresponding to the clustering risk based on the voltage change direction and amplitude. The clustering risk specifically includes local congestion and voltage linkage.

[0185] In this embodiment, the system collects the feeder concentration of risk areas based on their spatial location. The system then determines whether these feeder concentrations exceed a pre-set distribution upper limit and executes corresponding steps accordingly. For example, if the system determines that the feeder concentration of a risk area does not exceed the pre-set distribution upper limit, the system considers the spatial distribution of risk areas in the current distribution network to be relatively balanced. Since the feeder concentration is still within a safe range, the system does not need to interrupt or delay the charging task and can continue to execute the established charging power allocation plan. Simultaneously, it will perform charging at preset time intervals (such as 5 minutes or 15 seconds). Within minutes, the system recalculates the feeder concentration trend in each risk area. If an area experiences abnormal load growth in a short period (e.g., multiple fast charging piles start simultaneously), the system will trigger an early warning. For example, if the system determines that the feeder concentration in a risk area exceeds a pre-set distribution limit, it considers the spatial distribution of risk areas in the current distribution network to be uneven. The system will identify the clustering risks caused by feeder concentration, which specifically include local congestion and voltage linkage. Based on these clustering risks, the system monitors the voltage change direction and amplitude of each node in the risk area, and obtains the voltage linkage characteristics corresponding to the clustering risks. By judging the feeder concentration in risk areas in real time, the system can detect the uneven spatial distribution of loads in the distribution network. When the feeder concentration exceeds the preset limit, the system automatically identifies the clustering risks caused by excessive centralized access, including local power flow overload and voltage mutual influence. This process breaks away from the traditional method of relying solely on power or current thresholds for risk judgment, and achieves quantifiable identification of spatial dimension risks, thereby discovering potential bottlenecks earlier. By identifying bottleneck nodes and power supply blind spots, the system enhances its adaptability and predictive capabilities to complex power grid structures. Furthermore, based on node data within clustered risk areas, it monitors the direction and magnitude of voltage changes at each node, extracting voltage linkage characteristics corresponding to clustered risks. This process enables the system not only to determine if voltage exceeds limits but also to identify the correlation between voltage changes, such as the linkage effect of a voltage drop in one branch causing voltage fluctuations in neighboring branches. Through this characteristic analysis, the system can dynamically characterize local voltage coupling patterns, providing a scientific basis for subsequent voltage control strategies (such as hierarchical reactive power regulation and flexible load switching). This effectively prevents voltage collapse or reverse linkage phenomena caused by charging clusters. Moreover, by leveraging the analysis results of clustered risks and voltage linkage characteristics, the system can proactively intervene in the scheduling and control process. By limiting charging access power in high-risk areas, distributing loads, or adjusting reactive power compensation strategies, it achieves proactive defense and adaptive optimization control. This method can significantly reduce the risks of local overload, voltage fluctuations, and line congestion caused by concentrated spatial loads, ensuring that the distribution network still has a safety margin and stable power supply capability under high-concurrency charging scenarios.

[0186] It should be noted that, in identifying the aggregation risk caused by the feeder concentration, and based on this aggregation risk, the voltage change direction and amplitude of each node within the risk area are monitored. Based on the voltage change direction and amplitude, the voltage linkage characteristics corresponding to the aggregation risk are obtained, specifically as follows:

[0187] First, the system identifies the types of clustering risks caused by feeder concentration. These clustering risks mainly include:

[0188] Local congestion risk: As multiple feeders simultaneously bear high loads in adjacent areas, the power flow distribution becomes too concentrated, and some lines approach or exceed their rated current, which can easily form "current bottlenecks" or "hot spots".

[0189] Voltage linkage risk: Due to enhanced electrical coupling between nodes, voltage fluctuations in one area may be transmitted to adjacent areas through branches, causing widespread voltage over-limit or linkage drop phenomena;

[0190] Next, based on the aggregation risk, the system monitors the direction and magnitude of voltage changes at each node within the risk area;

[0191] The direction of voltage change is used to determine whether the node voltage is rising (reverse power flow or no reactive power compensation) or falling (concentrated load, excessive power inflow).

[0192] The magnitude of voltage change reflects the degree and rate of change of the node voltage from the reference value (such as the rated 400V or 10kV);

[0193] The system obtains the voltage time series of each node through periodic sampling and historical comparison, and can further obtain the voltage linkage characteristics corresponding to the cluster risk, that is, the voltage influence relationship between each node.

[0194] Typical voltage linkage characteristics include:

[0195] Unidirectional linkage: A change in the voltage of the master node triggers a synchronous change in the downstream nodes. For example, if the voltage of the master feeder drops by 2%, the voltage of the downstream nodes drops by an average of 1.5%.

[0196] Two-way linkage: Multiple branch nodes influence each other. For example, fluctuations in node A cause an increase in node B, and the adjustment of B has a reverse effect on A.

[0197] Regional linkage: Within a certain geographical area, the voltage changes of multiple nodes show similar trends, forming a local voltage coupling zone;

[0198] By analyzing these characteristics, the system can determine the propagation path and impact range of voltage changes in the risk area, providing a basis for decision-making in subsequent scheduling optimization (such as peak shaving, zone compensation, or flexible control).

[0199] In this embodiment, step S1, which identifies the charging load allocation information of a charging station based on the preset architecture layer of the power distribution network, further includes:

[0200] S11: Based on the structural relationship of the architecture layer, obtain the access point information of the charging station, wherein the structural relationship specifically includes the trunk, branches and the end, and the access point information specifically includes the access transformer, feeder number and voltage level;

[0201] S12: Determine whether the access point information matches the distribution network node, wherein the distribution network node specifically includes the node number, the feeder to which it belongs, and the branch line path;

[0202] S13: If so, then according to the mapping relationship of the distribution network nodes, extract the load characteristic information of each node, and generate the power distribution ratio of the charging load based on the load characteristic information. The load characteristic information specifically includes the node charging power, the connected phase, and the node load ratio.

[0203] In this embodiment, the system obtains the charging station's access point information based on the structural relationships at the architecture layer, specifically including the trunk, branches, and terminals. The access point information includes the access transformer, feeder number, and voltage level. The system then determines whether this access point information matches the distribution network nodes, which include the node number, the feeder to which it belongs, and the branch line path, to execute the corresponding steps. For example, if the system determines that the charging station's access point information cannot match the distribution network node, it considers that the actual access location of the charging station in the distribution network topology cannot be accurately located temporarily. Directly entering the subsequent load analysis or prediction model might lead to power allocation errors, voltage calculation distortion, or risk assessment deviations. The system compares the access information reported by the charging station with the node information in the power grid archive database, automatically detecting and marking inconsistencies such as mismatched transformer numbers, missing feeder paths, or mismatched voltage levels. Furthermore, if some access information is missing, the system will use geographic location coordinates (GIS) to identify the missing information. The system performs reverse matching using data or device communication identifiers (such as smart meter ID, charging pile gateway number) to infer possible access nodes for the charging station. If multiple possible nodes exist, the system makes a comprehensive judgment based on parameters such as node voltage level, distance, power flow direction, and phase consistency, and selects the most likely access node as a temporary matching point. For example, when the system determines that the access point information of the charging station can match the distribution network node, the system will consider that it can accurately locate the actual access position of the charging station in the distribution network topology. The system will extract the load characteristic information of each node according to the mapping relationship of the distribution network nodes. The load characteristic information specifically includes the node charging power, access phase, and node load ratio. Based on different load characteristic information, the system generates the power distribution ratio of the charging load.When the system determines that the charging station's access point information matches a distribution network node, it means that the actual access location of the charging station in the power grid topology has been accurately identified. This positioning not only ensures a clear correspondence between each charging station and a distribution network node, but also provides a reliable topological foundation for subsequent power calculation, power flow analysis, and risk assessment. Through precise positioning, the system can effectively avoid load distribution errors or power flow calculation deviations caused by misjudgment of the access point, thereby ensuring the accuracy of power grid analysis and dispatch. Simultaneously, based on the mapping relationship of distribution network nodes, the system extracts load characteristic information from each node, including node charging power, access phase, and node load percentage. This step comprehensively characterizes the power distribution of the charging station at the node level, reflecting the power distribution of each node. By extracting load characteristics, the system can identify high-load nodes or single-phase unbalanced nodes, providing a data foundation for local load optimization and voltage stability analysis. This enables refined management of node operating status. Based on the extracted load characteristics, the system can generate the power distribution ratio of the charging load, clearly defining the load proportion of each node and phase. This allows for a rational allocation of charging power in space and by phase, optimizing overall power flow and reducing the risk of local overload and imbalance. This process not only provides a basis for dynamic load scheduling but also helps identify potential risk areas, enabling proactive measures such as decentralized access or power limiting to improve the safety and operational reliability of the distribution network in centralized charging scenarios.

[0204] It should be noted that the main layer is used to analyze the substation outlet load rate, the branch layer is used to analyze the load concentration of the feeders, and the terminal layer is used to analyze the three-phase balance and overload risk of the low-voltage transformer.

[0205] It should be added that, based on the mapping relationship of the distribution network nodes, load characteristic information of each node is extracted, and based on the load characteristic information, the power distribution ratio of the charging load is generated, specifically as follows:

[0206] Once the system has established a mapping relationship between charging stations and distribution network nodes, the system will extract load characteristic information at the node level one by one to describe the charging load and power distribution pattern of each node.

[0207] 1. Extraction of node load characteristic information. The load characteristic information mainly includes the following:

[0208] Node charging power: The total power (kW) of the charging piles currently carried by each node, used to determine the node load intensity;

[0209] Phase of access: The distribution of node loads in the three-phase power grid (phases A, B, and C), used to analyze the three-phase balance state;

[0210] Node load percentage: The proportion of a node's load power to the total load of its feeder or branch, used to quantify the node's contribution to the overall power flow;

[0211] The system collects power data, node current and voltage information of charging stations, summarizes the above feature information into a node load feature table, and associates it with the distribution network topology model to form a complete load profile of each node in space and phase.

[0212] 2. Generation of charging load power distribution ratio: After obtaining the node load characteristics, the system further calculates the distribution ratio of charging load power among each node and each phase:

[0213] Calculate the percentage of each node in the total charging power based on the node power ratio;

[0214] Based on the access phase, calculate the proportion of each phase in the node load to determine whether the three-phase load is balanced;

[0215] By combining feeder and branch structures, the node power distribution is integrated into a branch-level or feeder-level power distribution map for subsequent load scheduling, power flow calculation and risk identification.

[0216] Through this process, the system can not only clearly define the power distribution of each node and phase, but also identify potential high-load or unbalanced nodes, providing a data foundation for optimizing charging scheduling, reducing local overload, and ensuring voltage stability.

[0217] Reference Appendix Figure 2 A load balancing system for vehicle charging demand in one embodiment of the present invention includes:

[0218] The identification module 10 is used to identify the charging load allocation information of the charging station based on the preset architecture layer of the power distribution network. The architecture layer specifically includes a data perception layer, a prediction and optimization layer, and a control execution layer.

[0219] The judgment module 20 is used to determine whether the charging load allocation information meets the preset safety constraints of the power distribution network;

[0220] The execution module 30 is configured to, if so, construct a power operation model of the charging station based on the charging load allocation information, collect the operating status of the distribution network, generate a predicted power demand of the charging station within a preset time period through the power operation model, and identify risk areas of the distribution network based on the predicted power demand. Specifically, the power operation model includes topology, line impedance, transformer capacity, node voltage level and feeder connection relationship, and the operating status specifically includes charging pile power, node voltage and current and transformer load.

[0221] The second judgment module 40 is used to determine whether the number of risk areas reaches the preset risk limit of the charging station;

[0222] The second execution module 50 is used to, if the risk is reached, acquire the risk factors of the risk area, execute the preset protection measures of the distribution network based on the risk factors, collect the charging indicators of the charging station, and dynamically update the constraint parameters of the distribution network according to the charging indicators. Specifically, the risk factors include low power factor, harmonic over-limit, and three-phase imbalance. The protection measures specifically include configuring reactive power compensation, limiting harmonic source power in scheduling, and adjusting single-phase pile power to balance the load. The charging indicators specifically include voltage over-limit number of times, equipment overload rate, power quality indicators, and user charging completion rate. The constraint parameters specifically include increasing distributed energy storage, upgrading transformers, and optimizing line structure.

[0223] In this embodiment, the identification module 10 identifies the charging load allocation information of the charging station based on the pre-established architecture layer of the distribution network, which specifically includes a data perception layer, a prediction optimization layer, and a control execution layer. Then, the judgment module 20 determines whether the charging load allocation information meets the pre-established safety constraints of the distribution network and executes corresponding steps. For example, when the system determines that the charging load allocation information of the charging station cannot meet the pre-established safety constraints of the distribution network, the system considers that the spatial distribution and power characteristics of the current charging load have violated the safety constraints of the distribution network operation. If no adjustment is made, it may cause local overload, voltage drop, or abnormal power quality, thereby affecting the stability and power supply reliability of the entire distribution system. The system will locate the area causing the constraint violation based on the load allocation information and real-time monitoring data, and simultaneously suspend or delay the scheduling of new charging tasks in that area to prevent... The local current further increases, and the power of the connected charging load is reduced, or time-sharing and zone-limited current operation is implemented to reduce the instantaneous load peak. For example, when the system determines that the charging load allocation information of the charging station can meet the safety constraints set in advance by the distribution network, the execution module 30 will consider that the spatial distribution and power characteristics of the current charging load meet the safety constraints of the distribution network operation. The system will construct the power operation model of the charging station based on the charging load allocation information. The power operation model specifically includes topology, line impedance, transformer capacity, node voltage level and feeder connection relationship. The system will collect the operating status of the distribution network, which specifically includes charging pile power, node voltage and current and transformer load. Through this power operation model, the system will generate the charging predicted power demand of the charging station in the pre-set time period. Based on different charging predicted power demands, the system will identify the risk areas of the distribution network.The system ensures that the spatial distribution and power characteristics of electric vehicle charging loads remain within safe operating ranges by conducting subsequent modeling and prediction under the premise that the charging load allocation information meets the safety constraints of the distribution network. This avoids operational deviations caused by blind modeling or scheduling. This mechanism guarantees that the model construction process is based on real and safe operating conditions, reducing the risks of distribution network overload and voltage exceedances from the source, and making the entire prediction and control process physically feasible and engineering reliable. Simultaneously, a power operation model is constructed based on the load allocation information under safety constraints. This model includes the distribution network topology, line impedance, transformer capacity, node voltage levels, and feeder connections. Combined with real-time collected operating status parameters such as charging pile power, node voltage and current, and transformer load, the system can dynamically calculate the predicted charging power demand within a preset time period. Furthermore, based on generating the predicted charging power demand and identifying risk areas, the system can feed back the prediction results to the subsequent scheduling and control module, enabling adaptive adjustments to the distribution network. For example, by optimizing charging timing, adjusting power allocation, or balancing load structures in advance, adjustments can be made without increasing hardware investment. In this context, the system effectively improves line utilization and power distribution efficiency. This process realizes intelligent operation control of the distribution network in the scenario of large-scale electric vehicle access, ensuring the safety and stability of the power supply system and improving the overall energy utilization and service quality. Then, the second judgment module 40 judges whether the number of these risk areas has reached the risk limit set in advance by the charging station, and executes the corresponding steps. For example, when the system determines that the risk area of ​​the distribution network has not reached the risk limit set in advance by the charging station, the system will consider that the distribution network as a whole is still in a safe operating range. The system will continue to perform prediction, optimization and monitoring work on this basis. The system will maintain the current charging task allocation and power output strategy, and will not actively reduce the charging power to ensure the user charging experience and plan completion rate. At the same time, it will perform real-time monitoring on the few identified risk areas, track the changing trends of their load, voltage and power factor, in order to judge whether there are signs of risk spread or deterioration. Based on the latest operating data, it will use the power operation model to recalculate the predicted power demand for future periods to ensure that the prediction results reflect the latest load dynamics and maintain the timeliness and accuracy of the model.For example, when the system determines that a risk area in the distribution network has reached the pre-set risk limit of the charging station, the second execution module 50 will consider the distribution network to be in an abnormal operating range. The system will collect risk factors for these risk areas, specifically including low power factor, harmonic exceedance, and three-phase imbalance. Based on different risk factors, it will execute pre-set protection measures for the distribution network, specifically including configuring reactive power compensation, limiting harmonic source power during dispatching, and adjusting single-phase pile power to balance the load. It will also collect charging indicators for the charging station, specifically including voltage overruns, equipment overload rate, power quality indicators, and user charging completion rate. Based on these charging indicators, it will dynamically update the constraint parameters of the distribution network, specifically including increasing distributed energy storage, upgrading transformers, and optimizing line structure.

[0224] In this embodiment, it also includes:

[0225] The acquisition module is used to acquire the operating parameters of the power distribution network based on the preset physical structure of the charging station, wherein the physical structure specifically includes the overall configuration of the station, power supply equipment, power distribution lines and charging pile configuration;

[0226] The third judgment module is used to determine whether the operating parameters match the boundary conditions of the power operation model;

[0227] The third execution module is used to identify the branch connection relationship between each node according to the pre-established electrical topology of the charging station and the distribution network if the condition is met, and to dynamically adjust the equipment operation constraints of the charging station according to the branch connection relationship. The branch connection relationship specifically includes the node level, branch direction and single-phase load distribution, and the equipment operation constraints specifically include the overload allowable coefficient, allowable fluctuation range and power qualified range.

[0228] In this embodiment, the system acquires the operating parameters of the power distribution network based on the pre-defined physical configuration of the charging station, specifically including the overall site configuration, power supply equipment, power distribution lines, and charging pile configuration. The system then determines whether these operating parameters match the boundary conditions of the power operation model to execute corresponding steps. For example, if the system determines that the operating parameters of the power distribution network cannot match the boundary conditions of the power operation model, the system considers that the power operation model can no longer accurately reflect the current physical state or operating boundary of the power distribution network. Continuing to predict and schedule based on the original model may lead to errors in risk identification, delayed control response, or even operational risks such as local overload or voltage exceeding limits. The system compares the differences between the operating parameters and the model input boundary to identify the source of the mismatch, such as equipment parameter drift, topology changes, or data acquisition errors, clarifying the source of the deviation, determining subsequent correction strategies, and simultaneously pausing the process. The system uses the current power operation model for risk prediction or load optimization to prevent erroneous operations caused by decisions based on incorrect boundaries. If the mismatch originates from structural changes (such as topology changes or equipment replacement), the system automatically reconstructs the power operation model and regenerates node connection relationships, power distribution, and constraint boundaries. For example, when the system determines that the operating parameters of the distribution network can match the boundary conditions of the power operation model, the system will consider that the power operation model can still accurately reflect the current physical state or operating boundary of the distribution network. The system will identify the branch connection relationships between each node based on the pre-established electrical topology between the charging station and the distribution network. The branch connection relationships specifically include node level, branch direction, and single-phase load distribution. Based on different branch connection relationships, the system dynamically adjusts the equipment operation constraints of the charging station. The equipment operation constraints specifically include overload allowable coefficient, allowable fluctuation range, and power qualified range.

[0229] In this embodiment, the execution module further includes:

[0230] The division unit is used to divide the load concentration area of ​​the distribution network based on the predicted power demand, and to obtain the corresponding load growth trend from the load concentration area;

[0231] The judgment unit is used to determine whether the load growth trend is caused by concentrated charging demand;

[0232] The execution unit is configured to, if so, acquire the node power distribution of the load concentration area, identify abnormal operating indicators of the load concentration area based on the node power distribution, and acquire potential risks of the load concentration area based on the abnormal operating indicators. The abnormal operating indicators specifically include load rate indicators, node voltage level indicators, and power factor indicators. The potential risks specifically include feeders approaching current limits, transformer load rates being too high, node voltages showing a downward trend, and three-phase imbalance in the line.

[0233] In this embodiment, the system divides the power grid into concentrated load areas based on predicted power demand, obtains the corresponding load growth trends from these concentrated load areas, and then determines whether the load growth trend is caused by concentrated charging demand to execute corresponding steps. For example, when the system determines that the load growth trend in a concentrated load area is not caused by concentrated charging demand, the system considers the load change in that area to be non-charging power growth, that is, the load increase is mainly caused by other types of electricity consumption behavior or external factors, rather than the concentrated operation of electric vehicle charging stations. The system will remove this area from the "Concentrated Charging Monitoring Area" label and mark it as a "Regular Load Growth Area" to distinguish it from controllable load growth areas caused by charging. At the same time, it will not adjust the charging pile operating status, power allocation, or scheduling strategy, maintaining the existing power distribution operation mode to prevent interference with normal load power supply due to misjudgment. The system collects voltage, current, and power changes in the area during subsequent monitoring periods. If the growth trend continues and remains unrelated to charging demand, it is confirmed as stable load growth. Otherwise, it triggers a re-judgment process to dynamically track potential charging load intervention. For example, when the system determines that the load growth trend in a concentrated load area is caused by concentrated charging demand, the system considers the load change in that area to be a normal charging power increase. The system acquires the node power distribution of these concentrated load areas and identifies abnormal operating indicators based on the different node power distributions. These abnormal operating indicators include load rate indicators, node voltage level indicators, and power factor indicators. Based on these different abnormal operating indicators, the system identifies potential risks in these concentrated load areas. These potential risks include feeders approaching current limits, transformer overload rates, node voltage showing a downward trend, and three-phase imbalance in the line.

[0234] In this embodiment, the second execution module further includes:

[0235] The acquisition unit is used to acquire the feeder capacity of the power distribution network based on the total charging power pre-collected by the charging station.

[0236] The second judgment unit is used to determine whether the feeder capacity has reached the preset safe load.

[0237] The second execution unit is used to identify the feeder topology of the distribution network if the condition is met, delay the charging access request of the charging station according to the feeder topology, and dynamically adjust the operating charging equipment of the charging station according to the operating status of the risk area. The feeder topology specifically includes lower-level nodes, branches, and charging range.

[0238] In this embodiment, the system obtains the feeder capacity of the distribution network based on the total charging power pre-collected by the charging station. The system then determines whether these feeder capacities have reached a pre-set safe load and executes corresponding steps accordingly. For example, if the system determines that the feeder capacity of the distribution network has not reached the pre-set safe load, the system considers the current operating state of the distribution network to be within a safe margin range, meaning the actual transmission power of the feeder is lower than its rated carrying capacity. The system will maintain the existing charging pile power allocation scheme without forcibly reducing the charging load or delaying the charging plan. If it detects that some nodes are under light load, the system can appropriately guide more charging loads to that area to improve overall power distribution efficiency. Simultaneously, it continuously monitors the feeder current change trend based on time-series data and records the charging... If the load growth rate trend indicates that future power may approach the safe limit, the system can issue a mild early warning signal. Under safe margin conditions, the system can further improve grid energy efficiency through optimization (such as power balancing or electricity price guidance). For example, it can encourage fast charging piles to operate during low-load periods to distribute peak load pressure, reduce line losses, and improve power supply economy. For example, when the system determines that the feeder capacity of the distribution network has reached the preset safe load, the system will consider the current operating state of the distribution network to be in an abnormal range. The system will identify the feeder topology of the distribution network, which specifically includes lower-level nodes, branches, and charging range. Based on these feeder topologies, the system will delay charging access requests from charging stations and dynamically adjust the operating charging equipment of charging stations according to the operating status of risk areas.

[0239] In this embodiment, the determination module further includes:

[0240] The detection unit is used to detect the power data of the charging load allocation information based on the pre-divided equipment types of the charging station, wherein the equipment types specifically include linear equipment and nonlinear equipment, and the power data specifically includes harmonic content and neutral line current.

[0241] The third judgment unit is used to determine whether the electrical energy data exceeds the preset constraint range;

[0242] The third execution unit is configured to, if so, acquire the harmonic index of the power data, mark the corresponding harmonic over-limit node from the power data according to the harmonic index, and identify the corresponding harmonic source type based on the harmonic over-limit node. The harmonic index specifically includes harmonic distortion rate, harmonic component amplitude and waveform distortion coefficient, and the harmonic source type specifically includes rectifier harmonic source, inverter harmonic source and unbalanced harmonic source.

[0243] In this embodiment, the system detects the electrical energy data of the charging load allocation information based on the pre-defined equipment types of the charging station, specifically linear and nonlinear equipment. This electrical energy data includes harmonic content and neutral current. The system then determines whether this electrical energy data exceeds a pre-set constraint range and executes corresponding steps accordingly. For example, if the system determines that the electrical energy data of the charging load allocation information does not exceed the pre-set constraint range, the system considers that the harmonics, current imbalances, and neutral current generated by all equipment in the current charging station are within acceptable limits. The system maintains the existing power allocation and scheduling plan, without adjusting the output or access sequence of charging piles to ensure continuous charging for users. Simultaneously, it continues to periodically collect harmonic content and neutral current data, performs waveform analysis on key nodes (such as the main feeder and transformer neutral point), and establishes an electrical... The energy quality time series model is used for subsequent risk prediction. Under good power quality conditions, the system can automatically fine-tune the switching status of reactive power compensation equipment or adjust the operating ratio of linear and nonlinear loads to maintain the overall power factor of the system at a high level (e.g., ≥0.95), further improving the operating efficiency of the distribution network. For example, when the system determines that the power data of the charging load allocation information exceeds the preset constraint range, the system will consider that the equipment in the current charging station is abnormal during operation. The system will obtain the harmonic indicators of these power data. The harmonic indicators specifically include harmonic distortion rate, harmonic component amplitude, and waveform distortion coefficient. Based on different harmonic indicators, the corresponding harmonic over-limit nodes are marked from these power data. Based on the harmonic over-limit nodes, the corresponding harmonic source type is identified. The harmonic source type specifically includes rectifier harmonic sources, inverter harmonic sources, and unbalanced harmonic sources.

[0244] In this embodiment, the second determination module further includes:

[0245] The acquisition unit is used to acquire the feeder concentration of the risk areas based on their spatial location.

[0246] The fourth judgment unit is used to determine whether the feeder concentration exceeds a preset distribution upper limit;

[0247] The fourth execution unit is used to identify the clustering risk caused by the feeder concentration if the condition is met, monitor the voltage change direction and amplitude of each node in the risk area according to the clustering risk, and obtain the voltage linkage characteristics corresponding to the clustering risk based on the voltage change direction and amplitude. Specifically, the clustering risk includes local congestion and voltage linkage.

[0248] In this embodiment, the system collects the feeder concentration of risk areas based on their spatial location. The system then determines whether these feeder concentrations exceed a pre-set distribution upper limit and executes corresponding steps accordingly. For example, if the system determines that the feeder concentration of a risk area does not exceed the pre-set distribution upper limit, the system considers the spatial distribution of risk areas in the current distribution network to be relatively balanced. Since the feeder concentration is still within a safe range, the system does not need to interrupt or delay the charging task and can continue to execute the established charging power allocation plan. Simultaneously, it will perform charging at preset time intervals (such as 5 minutes or 15 seconds). Within minutes, the feeder concentration trend of each risk area is recalculated. If an area experiences abnormal load growth in a short period of time (such as multiple fast charging piles starting up at the same time), the system will trigger an early warning. For example, when the system determines that the feeder concentration of a risk area exceeds the preset distribution limit, the system will consider that the spatial distribution of each risk area in the current distribution network is not balanced. The system will identify the clustering risk caused by the feeder concentration. The clustering risk specifically includes local congestion and voltage linkage. Based on these clustering risks, the system monitors the voltage change direction and voltage change amplitude of each node in the risk area. Based on the voltage change direction and voltage change amplitude, the system obtains the voltage linkage characteristics corresponding to the clustering risk.

[0249] In this embodiment, the identification module further includes:

[0250] The second acquisition unit is used to acquire the access point information of the charging station based on the structural relationship of the architecture layer, wherein the structural relationship specifically includes the trunk, branches and the end, and the access point information specifically includes the access transformer, feeder number and voltage level;

[0251] The fifth judgment unit is used to determine whether the access point information matches the distribution network node, wherein the distribution network node specifically includes the node number, the feeder to which it belongs, and the branch line path;

[0252] The fifth execution unit is used to extract load characteristic information on each node according to the mapping relationship of the distribution network nodes, and generate the power distribution ratio of the charging load based on the load characteristic information. The load characteristic information specifically includes node charging power, connected phase and node load ratio.

[0253] In this embodiment, the system obtains the charging station's access point information based on the structural relationships at the architecture layer, specifically including the trunk, branches, and terminals. The access point information includes the access transformer, feeder number, and voltage level. The system then determines whether this access point information matches the distribution network nodes, which include the node number, the feeder to which it belongs, and the branch line path, to execute the corresponding steps. For example, if the system determines that the charging station's access point information cannot match the distribution network node, it considers that the actual access location of the charging station in the distribution network topology cannot be accurately located temporarily. Directly entering the subsequent load analysis or prediction model might lead to power allocation errors, voltage calculation distortion, or risk assessment deviations. The system compares the access information reported by the charging station with the node information in the power grid archive database, automatically detecting and marking inconsistencies such as mismatched transformer numbers, missing feeder paths, or mismatched voltage levels. Furthermore, if some access information is missing, the system will use geographic location coordinates (GIS) to identify the missing information. The system performs reverse matching using data or device communication identifiers (such as smart meter IDs and charging pile gateway numbers) to infer possible access nodes for the charging station. If multiple possible nodes exist, the system makes a comprehensive judgment based on parameters such as node voltage level, distance, power flow direction, and phase consistency, and selects the most likely access node as a temporary matching point. For example, when the system determines that the access point information of the charging station can match the distribution network node, the system will consider that it can accurately locate the actual access position of the charging station in the distribution network topology. The system will extract the load characteristic information of each node according to the mapping relationship of the distribution network nodes. The load characteristic information specifically includes the node charging power, access phase, and node load ratio. Based on different load characteristic information, the system generates the power distribution ratio of the charging load.

[0254] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A load balancing method for vehicle charging demand, characterized in that, Includes the following steps: Based on the pre-defined architecture layer of the power distribution network, the charging load allocation information of the charging station is identified. Specifically, the architecture layer includes a data perception layer, a prediction and optimization layer, and a control execution layer. Determine whether the charging load allocation information meets the preset safety constraints of the power distribution network; If so, then based on the charging load allocation information, a power operation model of the charging station is constructed, the operating status of the distribution network is collected, and the predicted power demand of the charging station within a preset time period is generated through the power operation model. Based on the predicted power demand, the risk areas of the distribution network are identified. The power operation model specifically includes topology, line impedance, transformer capacity, node voltage level and feeder connection relationship, and the operating status specifically includes charging pile power, node voltage and current and transformer load. Determine whether the number of risk areas reaches the preset risk limit of the charging station; If the risk is reached, the risk factors of the risk area are obtained. Based on the risk factors, the preset protection measures of the distribution network are executed, the charging indicators of the charging station are collected, and the constraint parameters of the distribution network are dynamically updated according to the charging indicators. Specifically, the risk factors include low power factor, harmonic over-limit, and three-phase imbalance. The protection measures specifically include configuring reactive power compensation, limiting harmonic source power in scheduling, and adjusting single-phase pile power to balance the load. The charging indicators specifically include voltage over-limit number, equipment overload rate, power quality indicators, and user charging completion rate. The constraint parameters specifically include increasing distributed energy storage, upgrading transformers, and optimizing line structure. The step of constructing the power operation model of the charging station based on the charging load allocation information further includes: Based on the preset physical configuration of the charging station, the operating parameters of the power distribution network are obtained, wherein the physical configuration specifically includes the overall configuration of the station, power supply equipment, power distribution lines and charging pile configuration; Determine whether the operating parameters match the boundary conditions of the power operation model; If so, then based on the pre-established electrical topology between the charging station and the power distribution network, the branch connection relationship between each node is identified, and the equipment operation constraints of the charging station are dynamically adjusted according to the branch connection relationship. Specifically, the branch connection relationship includes node level, branch direction and single-phase load distribution, and the equipment operation constraints include overload allowable coefficient, allowable fluctuation range and power qualified range. The step of identifying risk areas of the distribution network based on the predicted power demand further includes: Based on the predicted power demand, the load concentration areas of the distribution network are divided, and the corresponding load growth trends are obtained from the load concentration areas. Determine whether the load growth trend is caused by concentrated charging demand; If so, the node power distribution of the load concentration area is obtained, and based on the node power distribution, abnormal operating indicators of the load concentration area are identified. Based on the abnormal operating indicators, potential risks of the load concentration area are obtained. The abnormal operating indicators specifically include load rate indicators, node voltage level indicators, and power factor indicators. The potential risks specifically include feeders approaching current limits, transformer load rates being too high, node voltages showing a downward trend, and three-phase imbalance in the line.

2. The load balancing method for vehicle charging demand according to claim 1, characterized in that, The step of collecting the charging indicators of the charging station and dynamically updating the constraint parameters of the distribution network based on the charging indicators further includes: Based on the total charging power pre-collected by the charging station, the feeder capacity of the distribution network is obtained; Determine whether the feeder capacity has reached the preset safe load; If so, the feeder topology of the distribution network is identified, and the charging access request of the charging station is delayed according to the feeder topology. Based on the operating status of the risk area, the operating charging equipment of the charging station is dynamically adjusted. The feeder topology specifically includes lower-level nodes, branches, and charging range.

3. The load balancing method for vehicle charging demand according to claim 1, characterized in that, The step of determining whether the charging load allocation information meets the preset safety constraints of the distribution network further includes: Based on the pre-classified equipment types of the charging station, the power data of the charging load allocation information is detected, wherein the equipment types specifically include linear equipment and nonlinear equipment, and the power data specifically includes harmonic content and neutral line current; Determine whether the electrical energy data exceeds a preset constraint range; If so, the harmonic index of the power data is obtained, and the corresponding harmonic over-limit node is marked from the power data according to the harmonic index. Based on the harmonic over-limit node, the corresponding harmonic source type is identified. The harmonic index specifically includes harmonic distortion rate, harmonic component amplitude and waveform distortion coefficient, and the harmonic source type specifically includes rectifier harmonic source, inverter harmonic source and unbalanced harmonic source.

4. The load balancing method for vehicle charging demand according to claim 1, characterized in that, The step of determining whether the number of risk areas reaches the preset risk limit of the charging station further includes: Based on the spatial location between risk areas, the feeder concentration of the risk areas is collected; Determine whether the feeder concentration exceeds a preset upper limit for distribution; If so, the clustering risk caused by the concentration of the feeder is identified. Based on the clustering risk, the voltage change direction and amplitude of each node in the risk area are monitored. Based on the voltage change direction and amplitude, the voltage linkage characteristics corresponding to the clustering risk are obtained. Specifically, the clustering risk includes local congestion and voltage linkage.

5. The load balancing method for vehicle charging demand according to claim 1, characterized in that, The step of identifying the charging load allocation information of charging stations based on the pre-set architecture layer of the distribution network also includes: Based on the structural relationship of the architecture layer, the access point information of the charging station is obtained, wherein the structural relationship specifically includes the trunk, branches and the end, and the access point information specifically includes the access transformer, feeder number and voltage level; Determine whether the access point information matches the distribution network node, wherein the distribution network node specifically includes the node number, the feeder to which it belongs, and the branch line path; If so, then based on the mapping relationship of the distribution network nodes, the load characteristic information of each node is extracted, and based on the load characteristic information, the power distribution ratio of the charging load is generated. Specifically, the load characteristic information includes the node charging power, the connected phase, and the node load ratio.

6. A load balancing system for automobile charging demand, characterized in that, include: The identification module is used to identify the charging load allocation information of the charging station based on the preset architecture layer of the power distribution network. The architecture layer specifically includes a data perception layer, a prediction and optimization layer, and a control execution layer. The judgment module is used to determine whether the charging load allocation information meets the preset safety constraints of the power distribution network; An execution module is configured to, if so, construct a power operation model for the charging station based on the charging load allocation information, collect the operating status of the distribution network, generate a predicted power demand for the charging station within a preset time period using the power operation model, and identify risk areas of the distribution network based on the predicted power demand. Specifically, the power operation model includes topology, line impedance, transformer capacity, node voltage level, and feeder connection relationship, and the operating status specifically includes charging pile power, node voltage and current, and transformer load. The second judgment module is used to determine whether the number of risk areas reaches the preset risk limit of the charging station; The second execution module is used to, if the risk is reached, acquire the risk factors of the risk area, execute the preset protection measures of the distribution network based on the risk factors, collect the charging indicators of the charging station, and dynamically update the constraint parameters of the distribution network according to the charging indicators. Specifically, the risk factors include low power factor, harmonic over-limit, and three-phase imbalance. The protection measures specifically include configuring reactive power compensation, limiting harmonic source power in scheduling, and adjusting single-phase pile power to balance the load. The charging indicators specifically include voltage over-limit number of times, equipment overload rate, power quality indicators, and user charging completion rate. The constraint parameters specifically include increasing distributed energy storage, upgrading transformers, and optimizing line structure. This also includes: The acquisition module is used to acquire the operating parameters of the power distribution network based on the preset physical structure of the charging station, wherein the physical structure specifically includes the overall configuration of the station, power supply equipment, power distribution lines and charging pile configuration; The third judgment module is used to determine whether the operating parameters match the boundary conditions of the power operation model; The third execution module is used to identify the branch connection relationship between each node according to the pre-established electrical topology of the charging station and the distribution network if the condition is met, and to dynamically adjust the equipment operation constraints of the charging station according to the branch connection relationship. The branch connection relationship specifically includes the node level, branch direction and single-phase load distribution, and the equipment operation constraints specifically include the overload allowable coefficient, allowable fluctuation range and power qualified range. The execution module further includes: The division unit is used to divide the load concentration area of ​​the distribution network based on the predicted power demand, and to obtain the corresponding load growth trend from the load concentration area; The judgment unit is used to determine whether the load growth trend is caused by concentrated charging demand; The execution unit is configured to, if so, acquire the node power distribution of the load concentration area, identify abnormal operating indicators of the load concentration area based on the node power distribution, and acquire potential risks of the load concentration area based on the abnormal operating indicators. The abnormal operating indicators specifically include load rate indicators, node voltage level indicators, and power factor indicators. The potential risks specifically include feeders approaching current limits, transformer load rates being too high, node voltages showing a downward trend, and three-phase imbalance in the line.

Images