Self-balancing control method and system of distribution line

Abstract

The invention discloses a self-balancing control method and system for a distribution line, and relates to the technical field of self-balancing control, a plurality of unbalance factors are determined according to the identification of an unbalance event, and a balance adjustment system is determined according to the plurality of unbalance factors, the corresponding unbalance degree and the branch point position of a line load branch point. And the accuracy of the balance adjustment system is improved. Determining a corresponding balance adjustment amount based on detection of a balance adjustment system, and determining a corresponding self-balance adjustment event according to the balance adjustment amount, a corresponding adjustment direction and a state distribution diagram of the distribution diagram of the distribution line; a plurality of self-balancing adjustment items are determined based on the recognition of the self-balancing adjustment event, and a corresponding cooperative control system is determined according to each self-balancing adjustment item, the load data combination of each line load branch point and the corresponding branch point combination, so that the accuracy of the cooperative control system is improved.

Description

Technical Field

[0001] This invention relates to the technical field of self-balancing control, and more particularly to a self-balancing control method and system for power distribution lines. Background Technology

[0002] With the development of technology, power distribution lines are gradually being applied to people's lives and serve as components of the power grid. Power distribution lines can be laid in corresponding residential areas, housing areas, or industrial areas. Current is transmitted along the power distribution lines via a preset path. In existing technologies, the distribution map of the power distribution lines is collected, and each line area is determined based on the identification of the distribution map. The corresponding line load branch points are marked, and the line load branch points are monitored in real time. The corresponding balance state is marked based on the changes in current. However, the instantaneous active power and average power are ignored, which affects the accuracy of the imbalance of each line load branch point and results in low accuracy of the coordinated control system of power distribution lines. Summary of the Invention

[0003] The purpose of this invention is to overcome the shortcomings of the prior art. This invention provides a self-balancing control method and system for power distribution lines.

[0004] This invention provides a self-balancing control method for power distribution lines, comprising: Collect the distribution map of the power distribution line, determine multiple line load branch points based on the distribution map and corresponding multiple working data, and determine the corresponding electrical parameter combination based on the detection of each line load branch point; Based on the identification of the electrical parameter combination, the corresponding instantaneous active power and average power are determined. Based on the matching of the instantaneous active power, average power and the corresponding voltage data, the corresponding unbalance is determined to mark the unbalance state of the line load branch point. Based on the tracing of the unbalanced state of the load branch point of the line, the corresponding unbalanced event is determined. Based on the identification of the unbalanced event, multiple unbalanced factors are determined. Based on the multiple unbalanced factors, the corresponding unbalance degree, and the branch point location of the load branch point of the line, a balance adjustment system is determined. A balance adjustment system includes the following elements: system ID and target branch point, adjustment target, core adjustment principle, and detailed execution plan. In the balance regulation system, the corresponding balance regulation amount is determined based on the detection of the balance regulation system, and the corresponding self-balancing regulation event is determined based on the balance regulation amount, the corresponding regulation direction and the state distribution diagram of the power distribution line. Based on the identification of this self-balancing adjustment event, multiple self-balancing adjustment items are determined. According to the combination of load data of each self-balancing adjustment item, each line load branch point, and the corresponding branch point combination, a corresponding collaborative control system is determined to trigger the collaborative control of each line load branch point in the balance dimension.

[0005] This invention provides a self-balancing control system for power distribution lines, which is applied to the aforementioned self-balancing control method for power distribution lines.

[0006] Compared with the prior art, the beneficial effects of the present invention are: (1) Collect the distribution map of the power distribution line, determine multiple line load branch points based on the distribution map of the power distribution line and the corresponding multiple working data, and determine the corresponding electrical parameter combination based on the detection of each line load branch point; determine the corresponding instantaneous active power and average power based on the identification of the electrical parameter combination, and determine the corresponding unbalance based on the matching of the instantaneous active power, average power and corresponding voltage data, so as to mark the unbalance state of the line load branch point. The introduction of instantaneous active power and average power, and the consideration of the instantaneous active power, average power and corresponding voltage data, improve the accuracy of the unbalance of each line load branch point.

[0007] (2) Based on the traceability of the unbalanced state of the load branch point of the line, the corresponding unbalanced event is determined. Based on the identification of the unbalanced event, multiple unbalanced factors are determined. Based on the multiple unbalanced factors, the corresponding unbalance degree and the branch point location of the load branch point of the line, the balance adjustment system is determined. Multiple unbalanced factors are controlled, and the consideration of multiple unbalanced factors, the corresponding unbalance degree and the branch point location of the load branch point of the line is realized, which improves the accuracy of the balance adjustment system.

[0008] (3) In the balance regulation system, the corresponding balance regulation amount is determined based on the detection of the balance regulation system. The corresponding self-balance regulation event is determined based on the balance regulation amount, the corresponding regulation direction and the state distribution diagram of the distribution line. Multiple self-balance regulation items are determined based on the identification of the self-balance regulation event. The corresponding collaborative control system is determined based on the load data combination of each self-balance regulation item, each line load branch point and the corresponding branch point combination. The self-balance regulation event is further controlled, realizing the overall consideration of each self-balance regulation item, each line load branch point and the corresponding branch point combination, improving the accuracy of the collaborative control system, so as to trigger the collaborative control of each line load branch point in the balance dimension. Attached Figure Description

[0009] Figure 1This is a flowchart illustrating the self-balancing control method for power distribution lines in an embodiment of the present invention. Figure 2 This is a flowchart illustrating step S11 in the self-balancing control method for power distribution lines in this embodiment of the invention. Figure 3 This is a flowchart illustrating step S12 in the self-balancing control method for power distribution lines in an embodiment of the present invention. Figure 4 This is a flowchart illustrating step S13 in the self-balancing control method for power distribution lines in this embodiment of the invention. Figure 5 This is a flowchart illustrating step S14 in the self-balancing control method for power distribution lines in this embodiment of the invention. Figure 6 This is a flowchart illustrating step S15 in the self-balancing control method for power distribution lines in an embodiment of the present invention. Figure 7 This is a schematic diagram of the structure of the self-balancing control system for power distribution lines in an embodiment of the present invention. Detailed Implementation

[0010] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0011] Please see Figures 1 to 7 A self-balancing control method for power distribution lines, applied to self-balancing control scenarios; the self-balancing control method for power distribution lines includes: Step S11: Collect the distribution map of the power distribution line, determine multiple line load branch points based on the distribution map of the power distribution line and the corresponding multiple working data, and determine the corresponding electrical parameter combination based on the detection of each line load branch point; Step S12: Determine the corresponding instantaneous active power and average power based on the identification of the electrical parameter combination, and determine the corresponding unbalance based on the matching of the instantaneous active power, average power and corresponding voltage data, so as to mark the unbalance state of the line load branch point; Step S13: Based on the tracing of the unbalanced state of the load branch point of the line, determine the corresponding unbalanced event, determine multiple unbalanced factors based on the identification of the unbalanced event, and determine the balance adjustment system based on the multiple unbalanced factors, the corresponding unbalance degree and the branch point location of the load branch point of the line. Step S14: In the balance adjustment system, the corresponding balance adjustment amount is determined based on the detection of the balance adjustment system, and the corresponding self-balancing adjustment event is determined according to the balance adjustment amount, the corresponding adjustment direction and the state distribution diagram of the distribution line. Step S15: Based on the identification of the self-balancing adjustment event, determine multiple self-balancing adjustment items, and determine the corresponding collaborative control system according to the combination of load data of each self-balancing adjustment item, each line load branch point, and the corresponding branch point combination, so as to trigger the collaborative control of each line load branch point in the balance dimension.

[0012] refer to Figure 2 In step S11, the specific steps are as follows: S111: Mark the location of the power distribution line, determine the distribution map of the power line based on the tracing of the location of the power distribution line, determine multiple line areas based on the identification of the distribution map of the power line, and determine multiple line load branch points based on the area location of each line area, the corresponding area load and the corresponding multiple working data. S112: In multiple line load branch points, multiple electrical parameters are determined based on the dynamic detection of each line load branch point. The multiple electrical parameters include three-phase voltage, three-phase current and neutral current; the corresponding electrical parameter combination is determined based on the multiple electrical parameters and the corresponding line load branch points.

[0013] In the embodiments of this application, a precise and system-identifiable geographic coordinate and topological reference system is established for the power distribution network by integrating a geographic information system (GIS) and an equipment management system (such as DMS or AMI). The marker is not just a point or a line on the map, but includes the latitude and longitude coordinates of the line's starting point (such as a substation outgoing switch cabinet), ending point, key nodes (such as poles and cable wells), and the line's physical attributes (such as overhead line / cable, model, and length).

[0014] After obtaining the location markers, the system uses algorithms to trace and reconstruct the complete electrical topology of the line. This process is not just about drawing a diagram, but about generating a structured data model that describes the line's connection relationships, phase allocation (how the three phases A, B, and C are allocated along the line), and the hierarchical relationships of each node (main line, branch line, secondary branch line). The tracing process is linked to the asset database to obtain the rated parameters of each line segment (such as impedance and current carrying capacity), providing an accurate power grid model for subsequent power flow calculations and imbalance analysis.

[0015] Intelligent zoning of power lines based on topology and operational characteristics is not a simple geographical division, but an electrical regional division. Common zoning strategies include: topology zoning, which uses main tie switches or T-junctions as boundaries to naturally divide the line into several electrically independent power supply areas; load density zoning, which divides the line into high-density load areas (such as commercial areas), medium-density load areas (such as mixed areas), and low-density load areas (such as rural areas) based on historical load data; and imbalance-sensitive areas, which identify line segments that have historically experienced frequent three-phase imbalances due to the connection of a large number of single-phase loads (such as charging piles and air conditioners) and classify them as key monitoring areas. Each area is assigned a unique identifier and records all the line segments and nodes it contains.

[0016] From numerous potential nodes, the most representative "line load branch points" are selected through a multi-dimensional weighted evaluation process: In terms of regional location, branch points located at the end of a region typically have a more direct impact on downstream imbalances than those closer to the power source, thus receiving higher weights; in terms of regional load, the three-phase imbalance at branch points in regions with high loads has a greater impact on the entire line, and the system calculates the average active and reactive power of each region as weighting factors; multiple operating data are crucial for dynamic evaluation, including real-time three-phase imbalance, neutral current amplitude, load volatility, and user type; considering all these factors, the system uses a scoring algorithm (e.g., Score = w1 × location weight + w2 × load weight + w3 × imbalance weight + w4 × volatility weight) to score each candidate branch point, ultimately selecting the highest-scoring points as the final "line load branch points".

[0017] Specifically, the power distribution line is a 5-kilometer-long 10kV hybrid line that originates from a 110kV substation in the suburbs. The first 2 kilometers are overhead lines, and the last 3 kilometers are cables. It mainly supplies power to an area that includes residential areas, small commercial areas, and a newly built electric vehicle charging station. In recent years, due to summer air conditioning loads and the use of charging piles, the line has frequently experienced three-phase imbalance problems, resulting in increased line losses and unstable voltage at the end.

[0018] The system retrieves the coordinate data of the power distribution lines from the GIS database; the starting point is the substation (longitude X1, latitude Y1), and the ending point is the last pole (longitude X2, latitude Y2); the locations of all poles, cable wells, and T-junctions are accurately marked and associated with their asset IDs; based on the location markings, the system reconstructs the electrical topology of the power distribution lines; the diagram clearly shows that: the main line starts from the substation, and at pole #15, a branch line (A1 branch) is formed by a T-junction leading to the residential area, at cable well #25, a branch line (A2 branch) is formed by a T-junction leading to the commercial area, and at cable well #30, a branch line (A3 branch) is formed by a T-junction dedicated to the electric vehicle charging station.

[0019] System analysis of topology and load characteristics divides the distribution lines into three regions: Region 1 (upstream mixed zone), from the substation to tower #15, includes some industrial and residential loads, with relatively stable loads; Region 2 (residential and commercial zone), including branches A1 and A2, is a typical single-phase load concentration area, with obvious peaks and valleys between residential loads at night and commercial loads during the day, resulting in a prominent imbalance problem; Region 3 (charging station sensitive zone), including branch A3, where the high-power DC fast charging piles connected are huge random single-phase loads, with a strong impact on the power grid.

[0020] The system evaluated all T-junctions in three areas: Branch A1 is located in Area 2, with a large load (the entire residential area), and operational data shows that its three-phase current imbalance often exceeds 15% at night, with a high neutral current; Branch A2 is also located in Area 2, with commercial loads causing high daytime imbalance; Branch A3 is located in Area 3, and although its total load is not as large as A1, its load fluctuation rate is extremely high, and the charging piles can cause a phase-to-phase imbalance of more than 25% instantaneously when they start up; The end point of the main line is located after the #30 cable well, which is the convergence point of all unbalanced currents, but adjustment at this point cannot solve the source problem; Final decision: After comprehensive scoring, the system determined that Branch A1 and Branch A3 are the key "line load branch points" for this self-balancing control; Point A1 represents a typical, continuous residential load imbalance, while Point A3 represents an impulsive, highly sensitive new type of load; Deploying control devices at these two points can solve the overall imbalance problem of the distribution line most effectively at the lowest cost.

[0021] Furthermore, load imbalance control devices or dedicated monitoring terminals integrating high-precision sensors are deployed at each selected line load branch point. These sensors include Hall effect current sensors or Rogowski coils for non-invasive and accurate measurement of three-phase and neutral currents, and resistive voltage divider networks or voltage transformers for acquiring three-phase voltages. Data synchronization is crucial, typically achieved at the microsecond level through GPS pulse-per-second (PPS) synchronization or the IEEE 1588 Precision Time Protocol (PTP) to ensure that data from all branch points are strictly aligned in time. The sampling frequency needs to reach several kilohertz to capture high-order harmonics, and the analog-to-digital converter (ADC) resolution must be at least 16 bits to ensure accuracy.

[0022] The system integrates scattered electrical parameters collected from different physical locations into a data model. A unique logical ID is created for each "line load branch point," and all electrical parameters (Ia, Ib, Ic, In, Ua, Ub, Uc) collected from that point are strongly bound to this ID. The electrical parameter combination includes not only the original instantaneous values ​​but also derived key indicators calculated in real time from these original values, such as the instantaneous active power of each phase, the total active power of the three phases, the average active power of the three phases, the three-phase current imbalance, and the effective value of the neutral current. All the above parameters (original and derived values), along with a precise timestamp and branch point ID, are encapsulated into a data frame and uploaded to the master station or area controller via the communication network. This complete data frame constitutes an "electrical parameter combination."

[0023] Specifically, two key "line load branch points" have been identified for the power distribution lines: the A1 branch point (residential area) and the A3 branch point (electric vehicle charging station). A load imbalance control device with a built-in GPS module is installed in the cable distribution box at the A1 branch point and the switch cabinet at the A3 branch point for network-wide synchronization. During the peak residential air conditioning load at 8 PM in summer, the device at point A1 dynamically detects the load at a sampling rate of 6.4kHz. The sensor array collects real-time data showing that the effective value of the current in phase A is 150A, phase B is 80A, phase C is 75A, and the effective value of the neutral current is as high as 70A, indicating a severe imbalance. At the same time, when an electric vehicle connects to a DC fast charging pile at point A3, the device at point A3 simultaneously detects that the charging pile is connected to phase B, causing the phase B current to jump instantaneously from 20A to 120A. The currents in phases A and C remain around 20A, while the neutral current jumps from near 0A to approximately 100A, demonstrating a very strong impact imbalance.

[0024] The MCU / DSP at point A1 processes the collected data in real time, calculates derived indicators, and generates a structured data packet containing BranchPoint_ID, precise timestamp, effective values ​​of current in each phase, active power in each phase, unbalance index, and status label (such as "Severe_Imbalance"). At the same time, the device at point A3 also generates a data packet with a similar structure, showing a surge in phase B current, an unbalance index as high as 0.83, and a status label of "Critical_Imbalance". These two structured "electrical parameter combinations" are sent to the collaborative control master station in real time. After receiving them, the master station can immediately identify that the distribution line has experienced severe unbalance at two different locations simultaneously, and the nature of the imbalance is different (A1 is continuous, A3 is impulsive).

[0025] refer to Figure 3 In step S12, the specific steps are as follows: S121: Dynamically identify the electrical parameter combination, determine the first power data combination and the second power data combination based on the dynamic identification of the electrical parameter combination, determine the corresponding instantaneous active power based on the identification of the first power data combination, and determine the corresponding average power based on the identification of the second power data combination; S122: Determine the first unbalance coefficient based on the instantaneous active power and the corresponding voltage data, determine the second unbalance coefficient based on the average power and the corresponding voltage data, determine the corresponding unbalance degree based on the mapping relationship between the first unbalance coefficient, the second unbalance coefficient and the unbalance degree, and determine the corresponding unbalance state based on the matching of each load branch point of the line.

[0026] In the embodiments of this application, two datasets for different purposes are constructed from a continuous data stream according to different time scales; the system continuously and at high speed parses the data stream uploaded in real time from the branch point, identifies the order of each data packet by timestamp, and groups and slices the data according to a preset time window, which slides and updates continuously as the data stream arrives.

[0027] The first power data set typically consists of all sampling points within one or several power frequency cycles (e.g., 10ms or 20ms). It preserves the details of load fluctuations and is mainly used to calculate instantaneous active power to capture rapid load changes and impact characteristics. The second power data set, on the other hand, is a long-term, low-frequency dataset, usually aggregated from multiple "first power data sets" over time. The time scale is on the order of seconds or minutes. It filters out high-frequency fluctuations through smoothing and reflects the average load level over a longer period of time. It is mainly used to calculate average power.

[0028] At each sampling point t, the system calculates the instantaneous active power of phases A, B, and C based on the instantaneous values ​​of the three-phase voltage ua(t), ub(t), uc(t) and the instantaneous values ​​of the three-phase current ia(t), ib(t), ic(t) at that point, using instantaneous power theory: pa(t) = ua(t) × ia(t), pb(t) = ub(t) × ib(t), pc(t) = uc(t) × ic(t). This calculation process is usually completed by the MCU / DSP of the local device at the branch point or the real-time calculation engine of the master station. Since it is a multiplication of instantaneous values, the calculation result is a curve that fluctuates rapidly with time, accurately reflecting the instantaneous rate at which energy flows from the grid to the load at each moment.

[0029] Within a defined time interval T (i.e., the time span of the "second power data combination"), the instantaneous active power is integrated or discretely summed, and then averaged over time. Taking phase A as an example, its average active power Pa = (1 / T) × ∫[pa(t)dt], which is expressed as Pa = (1 / N) × Σ[pa(n)] in a discrete system, where N is the total number of sampling points within the time interval. The system performs an arithmetic average of all instantaneous active power data points included in the "second power data combination". This process effectively filters out high-frequency noise caused by motor start-up and shutdown, switch on / off, etc., and obtains a stable value that can represent the average energy consumption level of the load within this time period. The three-phase average power P_avg is the arithmetic mean of Pa, Pb, and Pc.

[0030] Specifically, the control system is processing the real-time data stream uploaded from the A3 branch point (electric vehicle charging station). This data stream contains high-frequency three-phase voltage and current sampling values. The control system receives the data stream from point A3, with timestamps accurate to microseconds, and sets two time windows: a short window of 20ms and a long window of 1s. The system captures all sampling points from t=0.000s to t=0.020s (assuming a sampling rate of 6.4kHz, there are 128 points). The voltage and current data of these 128 points constitute a "first power data combination". The system continuously collects 50 such "first power data combinations" (i.e., data for 1 second), performs preliminary processing on them, and forms a "second power data combination" containing all power information within 1 second.

[0031] The system processes the "first power data combination" from t=0.000s to t=0.020s. At the sampling point of t=0.005s, ub(0.005s)=220V and ib(0.005s)=150A are measured. Therefore, the instantaneous active power of phase B at this moment is pb(0.005s)=220V×150A=33kW. At t=0.010s, due to the PWM rectification of the charging pile, ib(0.010s) fluctuates and becomes 145A. Therefore, pb(0.010s)=220V×145A=31.9kW. In this way, the system obtains a detailed fluctuation curve of the instantaneous active power of phase B within 20ms.

[0032] The system processes the "second power data combination" within 1 second. This dataset contains all instantaneous active power values ​​within 50 20ms windows. The system integrates and averages all instantaneous values ​​of pa(t), pb(t), and pc(t) within this 1 second. The calculation results show that in the past 1 second, the average active power of phase A is Pa=4.6kW, the average active power of phase B is Pb=27.6kW, and the average active power of phase C is Pc=4.6kW. This average power value of Pb=27.6kW is a stable indicator that can be used for subsequent imbalance assessment. It eliminates millisecond-level ripple interference and reflects the average operating power of the charging pile within 1 second.

[0033] Furthermore, a first imbalance coefficient is determined based on the instantaneous active power and the corresponding voltage data, a second imbalance coefficient is determined based on the average power and the corresponding voltage data, the corresponding imbalance degree is determined based on the mapping relationship between the first imbalance coefficient, the second imbalance coefficient, and the imbalance degree, and the corresponding imbalance state is determined based on the matching of each load branch point of the line. This approach takes into account the overall matching of each load branch point of the line, ensuring the accuracy of the corresponding imbalance state. At the same time, the introduction of instantaneous active power and average power takes into account the consideration of instantaneous active power, average power, and the corresponding voltage data, improving the accuracy of the imbalance degree of each load branch point of the line.

[0034] At this point, based on the instantaneous active power pa(t), pb(t), pc(t) calculated by S121, within a very short time window (such as half a power frequency cycle, 10ms), the system calculates the peak, valley, or effective value of the instantaneous active power of each phase. The imbalance is quantified by an index reflecting the severity of the fluctuation. A common algorithm is to calculate the ratio of the maximum fluctuation amplitude of the three-phase instantaneous active power within this window to the average value. An example formula is: First imbalance coefficient = max(|pa(t)-P_avg|,|pb(t)-P_avg|,|pc(t)-P_avg|) / P_avg. The higher this value, the more severe the impact, indicating that at least one phase load has deviated drastically from its average level in a short period of time. During this process, voltage data is mainly used to ensure the accuracy of the calculation. When a voltage dip or rise occurs, the system will correct or mark the power data in conjunction with the rate of change of voltage amplitude to improve the reliability of the imbalance coefficient judgment.

[0035] The three-phase average active power Pa, Pb, and Pc calculated based on S121 are the most classic methods for calculating unbalance, directly reflecting the symmetry of the three-phase power distribution. A commonly used formula is: Second Unbalance Coefficient = (P_max - P_min) / P_avg, where P_max and P_min are the maximum and minimum values ​​of the three-phase average power, respectively, and P_avg is the average value of the three-phase average power. This indicator is directly related to the magnitude of the neutral current and line losses. Voltage data is used for compensation here. In the case of three-phase voltage asymmetry, even if the three-phase impedance is symmetrical, unbalanced current will still be generated. Therefore, the system will use three-phase voltage data (such as three-phase voltage unbalance) to compensate for the power calculation results. For example, a method based on symmetrical components can be used to separate the unbalance components caused by voltage asymmetry and load asymmetry, so that the second unbalance coefficient can more accurately reflect the degree of imbalance of the load itself.

[0036] The first and second imbalance coefficients are integrated into a unified "imbalance degree" rating for decision-making. This mapping relationship model can be a logic table based on expert rules; for example, a rule table model can set: IF (second imbalance coefficient < 5%) AND (first imbalance coefficient < 10%) THEN imbalance degree = "normal"; IF (second imbalance coefficient > 20%) OR (first imbalance coefficient > 50%) THEN imbalance degree = "severe"; ELSE imbalance degree = "moderate"; or a weighted fusion model can be used: comprehensive imbalance index = w1 × second imbalance coefficient + w2 × first imbalance coefficient, and the imbalance degree level is divided according to the threshold range of the comprehensive index. The design of the mapping relationship reflects the power grid operation strategy. For example, for areas such as electric vehicle charging stations, the system pays more attention to dynamic impacts, so the first imbalance coefficient will be given a higher weight.

[0037] The system creates a state variable for each line load branch point. When the calculated "imbalance" is "severe", the state variable of that branch point is set to Imbalance_State="Severe_Imbalance". This state label is a direct trigger for subsequent steps. For example, in S13, only branch points with a state of "severe imbalance" or "moderate imbalance" will enter the "imbalance event tracing" process; branch points with a state of "normal" are ignored, thereby optimizing system resources and achieving precise control.

[0038] Specifically, the system has obtained the instantaneous active power and average power at two branch points, A1 (residential area) and A3 (charging station). Regarding the first imbalance coefficient, at point A1: the system analyzed the instantaneous active power curve within 20ms at point A1 and found that due to the random start and stop of multiple air conditioner compressors, the power of phase A fluctuated between 30kW and 40kW, while phases B and C were relatively stable. The maximum fluctuation amplitude was calculated to be approximately 6.5kW. Compared to P_avg=23.4kW, the first imbalance coefficient was approximately 27.8%, indicating a certain degree of dynamic fluctuation. At point A3: the system analyzed the instantaneous active power curve within 20ms at point A3 and found that the power of phase B exhibited high-frequency and large-amplitude pulsations due to the PWM rectification of the charging pile, with its instantaneous peak value being much higher than the average value. The first imbalance coefficient was calculated to be approximately 65%, indicating a strong dynamic impact.

[0039] For the second imbalance coefficient, at point A1: use average power Pa=34.5kW, Pb=18.4kW, Pc=17.3kW; Where P_max=34.5kW, P_min=17.3kW, P_avg=23.4kW; the calculated second imbalance coefficient is (34.5-17.3) / 23.4≈73.5%, which indicates a serious structural imbalance; point A3: using average power Pa=4.6kW, Pb=27.6kW, Pc=4.6kW; Where P_max=27.6kW, P_min=4.6kW, P_avg=12.3kW; the calculated second imbalance coefficient is (27.6-4.6) / 12.3≈187%, which indicates the existence of extreme structural imbalance.

[0040] The system adopts a weighted fusion model, assigning higher weight to steady-state imbalance (w1=0.7) and lower weight to dynamic imbalance (w2=0.3); the comprehensive index of point A1 = 0.7×73.5%+0.3×27.8%≈59.7%, and according to the mapping relationship (e.g., >40% is severe), the imbalance degree = "severe"; the comprehensive index of point A3 = 0.7×187%+0.3×65%≈150.4%, and according to the mapping relationship, the imbalance degree = "severe".

[0041] Point A1: Because its imbalance is "severe", the system marks it with Imbalance_State="Severe_Imbalance"; Point A3: Because its imbalance is also "severe", the system also marks it with Imbalance_State="Severe_Imbalance"; Through step S122, the system quantifies and classifies the complex electrical phenomena of points A1 and A3 into the "severe imbalance" state.

[0042] refer to Figure 4 In step S13, the specific steps are as follows: S131: Trace the unbalanced state of the load branch point of the line, and mark multiple unbalanced features during the tracing process. Determine the corresponding unbalanced event based on the feature location, corresponding feature shape and branch point location of the load branch point of the line. S132: Based on the detection of the imbalance event, multiple imbalance items are identified, and corresponding imbalance factors are determined according to the identification of each imbalance item, so as to collect multiple imbalance factors; the first level of balance adjustment content is determined according to the multiple imbalance factors and the corresponding imbalance degree. S133: Determine the second level of balance adjustment content based on multiple imbalance factors and the branch point location of the load branch point of the line, and determine the balance adjustment system based on the first level of balance adjustment content and the second level of balance adjustment content.

[0043] In the embodiments of this application, the data source for tracing includes all relevant data within a certain period of time (e.g., the first 5 minutes) before the branch point is marked as being in an "unbalanced state," such as time-series data of electrical quantities like three-phase voltage, current, active / reactive power, and neutral current at high sampling rates; records of power quality events such as voltage sags, swells, harmonics, and flicker; equipment status data such as switch status, protection action alarms, and equipment switching records from SCADA systems or IoT platforms; and external system data such as charging start / stop records from electric vehicle charging pile management systems and the operating status of air conditioning units in building automation systems. The tracing algorithm typically employs a sliding window technique to analyze the time-series data segment by segment, searching for abnormal patterns or abrupt changes. Simultaneously, through an event association algorithm, the abrupt changes in electrical quantities are matched with the state changes of external systems using timestamps to establish a preliminary causal relationship.

[0044] During the tracing process, the system utilizes preset or machine learning-trained feature extractors to identify key features related to imbalance from massive amounts of data. These features include precise timestamps (accurate to milliseconds) and specific topological locations (such as "A-phase line" and "B-phase CT measurement point"). More importantly, the system categorizes the waveform, amplitude, frequency, and other characteristics into several typical patterns: step type, which refers to a large unidirectional change in electrical quantity within a very short time, usually corresponding to the switching of large-capacity loads; gradual type, which refers to a continuous unidirectional change in electrical quantity over a longer time scale, usually corresponding to the natural growth of regional loads; pulsating type, which refers to periodic or non-periodic fluctuations in electrical quantity around a certain average value at a specific frequency, usually corresponding to nonlinear loads such as frequency converters and charging piles; and random type, which refers to irregular and violent random fluctuations in electrical quantity, usually corresponding to impact loads such as electric arc furnaces and spot welding machines.

[0045] The system fuses and correlates multiple marked features. For example, a "step-type" current increase in phase B is accompanied by a "pulsating" power fluctuation, and they highly overlap in time. The system matches the fused feature combination with an internally maintained "imbalance event knowledge base," which contains feature templates for various typical imbalance events, such as "single-phase motor starting," "capacitor bank switching," and "EV fast charging pile access." When the matching degree exceeds a set threshold, a specific "imbalance event" is identified. If a precise match cannot be found, it is classified as a general event such as "unknown single-phase high-power load access."

[0046] Specifically, the system has marked the A3 branch point (electric vehicle charging station) as "severely unbalanced". The control system master station immediately started the tracing program to retrieve all data of the A3 point in the 5 minutes before 15:30:05 (the time the status was marked). The data sources include: electrical waveform data with a sampling rate of 6.4kHz uploaded by the A3 point control device, the "closed" status record of the A3 point switch in the SCADA system, and the business log of "charging request approved" provided by the charging pile management system.

[0047] In the data at t=15:30:05.120, the system detected a significant "step-type" characteristic: its time position is 15:30:05.120, its topological position is the B-phase current channel, and its morphology is that the effective value of the B-phase current jumps from 20A to 120A within 20ms, with an amplitude change of 100A. At the same time, in the data after t=15:30:05.150, the system detected a "pulsating" characteristic: its time position is after 15:30:05.150, its topological position is the calculated value of the B-phase active power, and its morphology is that the B-phase active power fluctuates at a frequency of about ±2kW around the average value of 27kW at a frequency of 100Hz. This is a typical characteristic of a six-pulse rectifier.

[0048] The system merges two time-related features, "step increase in phase B current" and "high-frequency pulsation in phase B power," and compares them with the knowledge base. It does not match the "single-phase motor start" template (which usually has no high-frequency pulsation), but it highly matches the "single-phase rectifier-type high-power load access" template. To further confirm, the system queries external data sources and finds that the charging pile management system recorded a log at 15:30:05.000 stating "charging pile No. 2 started charging," and this charging pile is configured in phase B. Combining the electrical features and the external log, the system determines with a very high degree of confidence that the event causing the severe imbalance at point A3 is "start-up and operation of the phase B electric vehicle DC fast charging pile."

[0049] Furthermore, based on the detection of this imbalance event, multiple imbalance items are identified, and corresponding imbalance factors are determined according to the identification of each imbalance item, so as to collect multiple imbalance factors; the first-level balance adjustment content is determined based on multiple imbalance factors and their corresponding imbalance degrees, which takes into account the overall consideration of multiple imbalance factors and their corresponding imbalance degrees, and ensures the accuracy of the first-level balance adjustment content.

[0050] At this point, the system queries the corresponding electrical impact model based on the identified imbalance event (such as "EV fast charging pile access"). This model defines the typical electrical changes that the event will cause. The system lists these changes one by one to form an "imbalance item" list, and verifies whether these items have actually occurred through real-time calculations and records their quantitative values. For example, for the "single-phase high-power load access" event, typical imbalance items include: a significant increase in the active and reactive power of the access phase, the active and reactive power of the non-access phase remaining basically unchanged, a severe asymmetry in the amplitude and phase of the three-phase current, a sharp increase in the effective value of the neutral current, and the injection of harmonic currents of a specific frequency.

[0051] The system performs cluster analysis on multiple related "imbalance items" to identify their common physical roots, relying on a predefined "item-factor" mapping knowledge base. For example, if items such as "increased power in connected phases, unchanged power in unconnected phases, asymmetrical three-phase currents, and increased neutral current" occur simultaneously, they can be summarized as an imbalance factor: "uneven distribution of active / reactive power between phases". If a harmonic current item of a specific frequency is detected, it can be summarized as an imbalance factor: "additional imbalance and loss caused by harmonic pollution". The system matches the detected imbalance items with the factor definitions in the knowledge base through a rule engine or simple logical judgment, and finally extracts a few core imbalance factors.

[0052] The system will create a list or data structure to store all identified imbalance factors and their severity (which can be represented by the quantitative values ​​of the corresponding items), for example: [{Factor:“Active Power Imbalance”,Severity:“High”},{Factor:“Harmonic Pollution”,Severity:“Medium”}]. This structured set of factors will be used as key input to the subsequent S133 step to determine the adjustment content.

[0053] The system has an internal "factor-strategy" mapping table that defines the fundamental adjustment strategy to be adopted for each imbalance factor. For example, for the "active power imbalance" factor, the first level of adjustment is: "implementing phase-to-phase active power transfer"; for the "reactive power imbalance" factor, the first level of adjustment is: "performing phase-by-phase reactive power compensation"; for the "harmonic pollution" factor, the first level of adjustment is: "performing active or passive filtering". When there are multiple imbalance factors, the first level of adjustment is a combination of these strategies. The degree of imbalance determines the "priority" and "intensity" of the adjustment. For example, for a "severe imbalance" state, the active power transfer strategy has the highest priority; for a "slight imbalance" state, only a small-scale compensation is needed, and the degree of imbalance also affects the target value of the adjustment, such as "reducing the degree of imbalance to below 5%".

[0054] Specifically, the system has determined that the imbalance event at branch point A3 (electric vehicle charging station) is "B-phase electric vehicle DC fast charging pile starts operation"; based on the "EV fast charging pile access" event model, the system detects the current electrical data at point A3 and confirms the following items have occurred: Item 1, the active power of phase B jumps from 4.6kW to 27.6kW, an increase of 23kW; Item 2, the active power of phases A and C remains unchanged at 4.6kW; Item 3, the effective value of the neutral current jumps from close to 0A to about 100A; Item 4, the phase B current waveform shows obvious 6kHz switching frequency ripple.

[0055] The system summarizes the above items: "Item 1" and "Item 2" directly point to a core factor: "severe imbalance in the distribution of active power between phases"; "Item 3" is a direct consequence of "severe imbalance in the distribution of active power between phases"; "Item 4" points to another factor: "injection of high-frequency harmonic current"; the system finally collected two main imbalance factors: "uneven active power" and "harmonic pollution".

[0056] The system structurally summarizes the two identified factors and their severity into a factor set. Based on the collected imbalance factors and the "severe imbalance" of 187% at point A3, the system queries the "factor-strategy" mapping table. For the "active power imbalance" factor, with an imbalance level of "severe," the core strategy determined by the system is "implementing rapid, large-capacity phase-to-phase active power transfer." For the "harmonic pollution" factor, although it exists, the current main problem is the huge active power imbalance leading to excessive neutral current, so the priority of harmonic mitigation is temporarily postponed. The final determined first-level balance adjustment content is: "Using phase-to-phase active power transfer as the core adjustment means to prioritize solving the severe active power imbalance problem."

[0057] Therefore, the second level of balance adjustment is determined based on multiple imbalance factors and the location of the load branch points of the line. The balance adjustment system is determined based on the first and second level of balance adjustment, which takes into account the overall consideration of the first and second level of balance adjustment, ensuring the accuracy of the balance adjustment system. At the same time, multiple imbalance factors are controlled, realizing the consideration of multiple imbalance factors, their corresponding imbalance degrees, and the location of the load branch points of the line, thus improving the accuracy of the balance adjustment system.

[0058] At this point, the "physical strategy" (first layer of content) determined in S132 is transformed into a specific and executable "execution strategy" (second layer of content). The imbalance factors determine the "specific parameters" and "constraints" of the adjustment. For example, the "active power imbalance" factor clarifies that the physical quantity that needs to be adjusted is active power, and the adjustment amount needs to be calculated based on the current power gap or surplus. The branch point location provides the "context environment" required for execution of the adjustment, including topology constraints (whether the point is at the end of the line or on the main line), equipment capacity (rated capacity of the installed control device), load characteristics (whether the downstream is a sensitive load or a normal load), and communication conditions (whether there is a reliable real-time communication link). Combining the above information, the second layer of balancing content constitutes a detailed execution plan, including the executing entity (which device), execution topology (back-to-back converters, etc.), adjustment direction (from which phase to which phase), adjustment amount (specific kW value), adjustment rate (kW / s), and safety constraints (not exceeding capacity, voltage sag rate, etc.).

[0059] The physical strategy and execution strategy are integrated into a complete, structured, and directly executable "balance regulation system." This system is a comprehensive set of action instructions; it is not a simple superposition of the two parts, but an organic integration. A balance regulation system includes the following elements: system ID and target branch point for unique identification; regulation target (derived from the first level), such as "reduce the imbalance from 187% to below 10%"; core regulation principle (derived from the first level), such as "balance the three-phase load through phase-to-phase active power transfer"; and detailed execution plan (derived from the second level), including execution device, regulation topology, regulation direction, target regulation amount, control strategy (such as using PWM-based closed-loop control and setting the power rise slope) and safety constraints (such as the apparent power of the device must not exceed the rated value and the voltage deviation must not be lower than the threshold). In addition, it also includes start-up conditions and priorities to ensure execution at the right time and in the right order.

[0060] Specifically, the system has determined the first level of regulation for the A3 branch point: "Using the transfer of active power between phases as the core regulation method to prioritize the resolution of severe active power imbalance problems"; the system comprehensively analyzes the imbalance factors and the location of the branch point; the core factor is "uneven active power", which requires the transfer of approximately 11.5kW of active power; the A3 point is the end of the line, connected to an electric vehicle charging station with high power quality requirements, and a 50kVA load imbalance control device is installed at this point.

[0061] Based on this, the system makes a comprehensive decision: the executing entity must be the 50kVA control device at point A3, because end-point regulation is the fastest and most effective; the regulation topology adopts the device's built-in back-to-back converter, which is the optimal topology for achieving precise and rapid phase-to-phase power transfer; the regulation direction, based on real-time data, is determined to transfer power from phase B (overloaded phase) to phases A and C (underloaded phases); regarding the regulation rate, since the downstream is a charging pile (power electronic load, sensitive to voltage fluctuations), smooth regulation must be adopted, and the system sets the regulation rate to 1.5kW / s to ensure that the regulation is completed within about 8 seconds, with a smooth process; regarding safety constraints, the regulation amount of 11.5kW is much smaller than the device capacity of 50kVA, meeting the capacity protection requirements; the final determined second-level balance regulation content is: "At point A3, through the 50kVA back-to-back converter, smoothly transfer approximately 11.5kW of active power from phase B to phases A and C at a rate of 1.5kW / s."

[0062] The system integrates the first and second layers of content to generate a complete and immediately executable instruction set, named the "A3-Charging Pile Impact Imbalance Local Rapid Balancing System." Its adjustment objective is to reduce the imbalance at point A3 from 187% to below 10%, and the adjustment principle is phase-to-phase active power transfer. The execution scheme is clearly defined: the device is a 50kVA control device at point A3, the topology is a back-to-back converter, the direction is B→A,C, the adjustment amount is 11.5kW, the rate is 1.5kW / s (smooth adjustment), the constraint is total power <50kVA, and the start command is immediate execution. Through step S133, the system generates a complete "balance adjustment system" for the imbalance problem at point A3, encompassing both macroscopic objectives and microscopic control details. This system can be directly invoked by subsequent steps, ultimately transforming into precise PWM control signals for power electronic devices to achieve intelligent, rapid, and safe self-balancing adjustment.

[0063] refer to Figure 5 In step S14, the specific steps are as follows: S141: Real-time monitoring of the balance adjustment system, detection of the balance adjustment system, marking multiple balance adjustment contents during the detection process, determining the corresponding balance adjustment amount based on each balance adjustment content, the corresponding balance dimension and the current load level of the line load branch point, and marking the corresponding adjustment direction; S142: Collect the distribution map of the power distribution line, determine the state distribution map of the power distribution line based on the distribution map of the power distribution line and the real-time status of multiple line load branch points, determine the self-balancing adjustment framework based on the state distribution map of the power distribution line and the balance adjustment amount, and determine the corresponding self-balancing adjustment event based on the self-balancing adjustment framework and the adjustment direction corresponding to the balance adjustment amount.

[0064] In the embodiments of this application, the balance adjustment system is monitored in real time, the balance adjustment system is detected, and multiple balance adjustment contents are marked during the detection process. Based on each balance adjustment contents, the corresponding balance dimension and the current load level of the line load branch point, the corresponding balance adjustment amount is determined, and the corresponding adjustment direction is marked. This approach takes into account the overall consideration of each balance adjustment contents, the corresponding balance dimension and the current load level of the line load branch point, ensuring the accuracy of the corresponding balance adjustment amount.

[0065] At this point, the system continuously subscribes to and receives real-time electrical data streams uploaded by the target branch point. This means that the regulation decision is not based on historical data, but on the "latest state" of the power grid. At the same time, the system will check whether the various preconditions of the regulation system are met, such as whether the control device is online, whether the communication is normal, and whether the grid voltage is within the allowable range. The system parses the structured text of the "balance regulation system" and decomposes it into independent, executable "balance regulation content" task items, such as "phase-to-phase active power transfer" and "harmonic current suppression". Each marked content will be associated with one or more "balance dimensions", such as "active power" or "harmonic spectrum". This dimension indicates the electrical quantities that need to be considered in subsequent calculations.

[0066] Based on real-time load conditions, a precise and dynamic "balance adjustment amount" is calculated for each "balance adjustment content". For each marked content, the system calls its corresponding calculation module. For example, for the "phase-to-phase active power transfer" content, the system will activate its "active power calculation" module. The calculation module uses the latest electrical parameter combination of the branch point as input. The load level is usually quantified by the three-phase average active power P_avg and the three-phase power Pa,Pb,Pc.

[0067] The calculation of the adjustment amount is not a simple application of a formula, but a closed-loop control process to ensure the smoothness and accuracy of the adjustment. The system calculates the adjustment amount under ideal conditions through a digital controller (such as a PI controller). The input is the error between the "target imbalance" and the "current real-time imbalance", and the output is the final "balance adjustment amount". This closed-loop control can dynamically adjust the adjustment rate and amplitude according to the real-time changes in load. The calculated adjustment amount must be verified against the rated capacity of the control device. If it exceeds the upper limit of the capacity, the system will limit it to the maximum allowable value to ensure the safe operation of the device.

[0068] The system determines the power flow direction based on the real-time ranking of the current three-phase loads. The phase with the heaviest load is marked as the "energy output phase" (or "excess phase"), and the phase with the lightest load is marked as the "energy input phase" (or "insufficient phase"). The adjustment direction is precisely marked, for example: "transfer power from phase B (excess phase) to phases A and C (insufficient phases)". In the case of multi-phase insufficiency, the system will further refine the specific power value allocated to each phase according to the load differences of each phase.

[0069] Specifically, the system has generated a "balance regulation system" for the A3 branch point (electric vehicle charging station); the system continuously receives data uploaded by the A3 point control device to confirm that the device is online and the grid voltage is normal; the system analyzes the "balance regulation system" of the A3 point, identifies the core task, and marks it as: regulation content 1 - interphase active power transfer, the balance dimension of which is active power.

[0070] The system obtains the latest data at point A3: Pa=4.8kW, Pb=28.0kW, Pc=4.8kW, and calculates P_avg=12.53kW. The system's PI controller calculates the current imbalance error and performs calculations based on its internal parameters. Assuming that the calculation is successful, the controller outputs a smooth adjustment command, which is quantified into a specific power value: ΔP_actual=11.6kW. The device capacity at point A3 is 50kVA, and the adjustment amount of 11.6kW is much smaller than its capacity limit, so the verification passes. Therefore, the system determines the balance adjustment amount for this adjustment to be 11.6kW.

[0071] The system compares the three-phase power: Pb (28.0kW) is the maximum value, and Pa (4.8kW) and Pc (4.8kW) are the minimum values. Therefore, the system clearly marks the adjustment direction as "transferring power from phase B to phases A and C". Since the loads of phases A and C are equal, the system will further set the allocation strategy, such as allocating ΔP_actual / 2 = 5.8kW to each phase. Through step S141, the system successfully transforms a macroscopic "balanced adjustment system" into a dynamic command that contains a precise adjustment amount (11.6kW) and a clear direction (B→A,C) and can be executed immediately, thus making full preparations for subsequent global optimization and final physical execution.

[0072] Furthermore, the distribution map of the power distribution lines is collected. Based on the distribution map of the power distribution lines and the real-time status of multiple line load branch points, the state distribution map of the power distribution lines is determined. According to the state distribution map of the power distribution lines and the balance adjustment amount, the self-balancing adjustment framework is determined. According to the self-balancing adjustment framework and the adjustment direction corresponding to the balance adjustment amount, the corresponding self-balancing adjustment event is determined. This takes into account the overall consideration of the self-balancing adjustment framework and the adjustment direction corresponding to the balance adjustment amount, ensuring the accuracy of the corresponding self-balancing adjustment event.

[0073] At this point, the system retrieves the static model of the power distribution line from a GIS (Geographic Information System) or network topology database. This model includes the physical connections and geographical locations of substations, feeders, switches, transformers, and all line load branch points. The system obtains the real-time operating status of all key branch points through a real-time communication network. This status data includes imbalance status labels such as "normal," "slight imbalance," and "severe imbalance" obtained from S12, the operating status (online / offline) and current available capacity of each branch point control device, and key electrical quantities such as total active power, power factor, and voltage level. The system uses the above real-time status data as a dynamic layer and renders it onto the static distribution map to generate a visualized and information-rich "status distribution map."

[0074] The system analyzes the "state distribution map" to identify all branch points that require adjustment and their severity. Based on the situation analysis results, the system selects an optimal adjustment framework: when only one branch point needs adjustment, or when the imbalance at that point is localized and does not affect other nodes, the system selects a "single-point local adjustment framework," which has the advantages of fast response speed, simple control logic, and low dependence on communication. When multiple branch points are unbalanced simultaneously, or when adjusting a certain point will significantly affect the power flow distribution of other points, the system selects a "multi-branch point collaborative adjustment framework." This framework is more complex and requires global coordination, but it can achieve network-wide optimization. The "balance adjustment amount" calculated by S141 is an important basis for framework selection. A large adjustment amount means that the point is the main contradiction in the entire network and should be prioritized (single-point priority); while multiple smaller adjustment amounts require collaborative optimization.

[0075] A "self-balancing adjustment event" is a highly structured set of instructions or messages containing all the information required to perform a complete adjustment. Its components include: a globally unique event ID; a frame type that clearly indicates whether it is a "single-point local adjustment" or a "multi-branch point collaborative adjustment"; a list of target nodes, which is a single node ID in a single-point frame or multiple node IDs in a collaborative frame; and precise instructions obtained from S141 for each target node, such as adjustment amount, adjustment direction, adjustment rate, and execution timestamp. This complete "self-balancing adjustment event" is sent to S15, where the collaborative control system of S15 parses it and ultimately triggers the physical devices at each branch point to execute it.

[0076] Specifically, the system has calculated the precise regulation amount (11.6kW) and direction (B→A,C) for branch point A3; the system retrieves the topology diagram of the power distribution line, showing that after the substation outgoing line, it passes through two main branch points, A1 (residential area) and A3 (charging station); the real-time status is superimposed: the status of point A3 is "severe imbalance", the control device is "online", and the available capacity is "50kVA"; the status of point A1 is "moderate imbalance", the control device is "online", and the available capacity is "30kVA"; the system generates a dynamic diagram, in which point A3 is marked in red (severe) and point A1 is marked in yellow (moderate), and both points show available regulation resources.

[0077] System analysis of the state diagram revealed two imbalance points, but the imbalance at point A3 (187%) was significantly higher than that at point A1 (60%), making it the primary contradiction in the current power grid. Applying the principle of minimum cost, the system assessment found that prioritizing the resolution of point A3 would require only 11.6kW of regulation to significantly reduce the overall neutral current and losses of the line, yielding the most significant effect. Therefore, the system decided to adopt a "single-point priority local regulation framework," which means first concentrating resources on resolving point A3, and then reassessing the overall network status after it stabilizes to determine whether regulation of point A1 is necessary.

[0078] The system combines the "single-point priority local adjustment framework" with the specific instructions for point A3 calculated in S141, encapsulating them into a complete "self-balancing adjustment event". The final event instructions include: event ID (Event_AutoBalance_20231027_001), framework type, target node, and detailed adjustment instructions, including adjustment action (active power transfer), adjustment amount (11.6kW), direction (from phase B to phases A and C), adjustment rate (1.5kW / s), and execution time (immediate execution). Through step S142, the system elevates an isolated adjustment calculation for point A3 to the level of network-wide optimization, ultimately generating a "self-balancing adjustment event" containing global optimization strategies and precise execution instructions, paving the way for the final execution in S15.

[0079] refer to Figure 6 In step S15, the specific steps are as follows: S151: Collect self-balancing adjustment events, determine multiple self-balancing adjustment items based on the identification of self-balancing adjustment events, determine the item priority of each self-balancing adjustment item based on the matching of each self-balancing adjustment item, and mark the item content of each self-balancing adjustment item. S152: Collect load data combinations from each line load branch point, and determine the collaborative control content based on the project priority, corresponding project content, and load data combinations of each line load branch point for each self-balancing adjustment project. S153: Mark the branch point combination corresponding to each line load branch point, determine the corresponding collaborative control system based on the collaborative control content and the branch point combination corresponding to each line load branch point, assign the corresponding collaborative control item based on the identification of the collaborative control system, and trigger the collaborative control of each line load branch point in the balance dimension based on the collaborative control item.

[0080] In the embodiments of this application, self-balancing adjustment events are collected, and multiple self-balancing adjustment items are determined based on the identification of the self-balancing adjustment events. The item priority of each self-balancing adjustment item is determined based on the matching of each self-balancing adjustment item, and the item content of each self-balancing adjustment item is marked. This approach takes into account the overall matching of each self-balancing adjustment item and ensures the accuracy of the item priority of each self-balancing adjustment item.

[0081] At this point, the event subscriber or message queue of the control system receives the "self-balancing adjustment event" from S14. The system parses the event's metadata, such as event ID, timestamp, and event type, to confirm its validity and timeliness. The system then analyzes the event body in depth and decomposes the event into specific "self-balancing adjustment projects" based on the "framework type" and "target node list" defined in the event. For example, if the event framework is "single-point local adjustment," the event is decomposed into a single project; if the event framework is "multi-branch point collaborative adjustment," the event is decomposed into multiple projects, each project being clearly associated with an execution subject (branch point) and a core task.

[0082] An execution order is established for the decomposed adjustment projects to ensure that the system can prioritize the most critical and urgent tasks, thereby achieving optimal resource allocation and minimizing risks. The system has a built-in multi-dimensional priority evaluation model, which calculates a priority score for each project based on a series of rules and weights.

[0083] The assessment dimensions (matching criteria) include: the severity of imbalance, with the imbalance at the corresponding branch point of the project being the primary assessment dimension, and projects with severe imbalance having higher priority than those with moderate or slight imbalance; the type of regulation, with different regulation types having different priorities. Typically, "active power transfer" is used to address urgent issues such as excessive neutral current and line losses, and has the highest priority, followed by "reactive power compensation," and then "harmonic mitigation"; the impact on grid security, assessing the impact of project implementation on grid security. For example, if the implementation of a project can prevent voltage overruns, its priority will be significantly increased; and economic considerations, with projects that bring the greatest loss reduction benefits under certain strategies receiving higher priority. The calculated priorities are ultimately labeled as "high," "medium," "low," or a specific numerical value and attached to each regulation project.

[0084] The system creates a detailed list of contents for each adjustment item to ensure that the execution unit (branch point control device) can understand the task unambiguously when it receives instructions. This item list usually includes: item ID (unique identifier), target node (clearly indicating the branch point to execute the item), adjustment type (clearly indicating the physical quantity to be adjusted), target adjustment amount (precise value), adjustment direction (clear vector), execution constraints (including adjustment rate, maximum allowable voltage deviation, etc.), and the item priority determined in the previous step.

[0085] Specifically, the system has generated a "self-balancing adjustment event" for point A3; the system receives the event Event_AutoBalance_20231027_001, identifies its frame as Single_Point_Priority_Local, and the target node as A3_Feeder; since it is a single-point frame, the event is directly decomposed into a single item: self-balancing adjustment item 1 - power transfer at point A3.

[0086] The system prioritizes "Project 1"; regarding the severity of imbalance, the imbalance at point A3 is 187%, which is considered "severe," and this item receives the highest score; regarding the type of adjustment, the project type is "active power transfer," which is used to resolve severe active power imbalance, and this item receives the highest score; regarding the impact on grid security, this adjustment can significantly reduce neutral current and improve grid security, and this item receives a high score; after comprehensive evaluation, the system marks "Project 1" as having the highest priority.

[0087] The system creates a detailed task description for "Project 1", marking all its contents as follows: Project ID: Proj_A3_ActiveP_001; Target Node: A3_Feeder_Node; Adjustment Type: Active_Power_Transfer; Target Adjustment Amount: 11.665kW (assuming the value after subsequent fine-tuning); Adjustment Direction: From_Phase_BTo_Phase_A_and_C; Execution Constraint: Ramp_Rate=1.5kW / s; Project Priority: Highest; Through step S151, the system successfully transforms an abstract "event" into a "adjustment project" with clear priority and detailed content, laying a solid foundation for subsequent final decision-making and execution based on real-time operating conditions.

[0088] Furthermore, load data combinations from various line load branch points are collected. Based on the project priorities, corresponding project contents of each self-balancing adjustment project, and the load data combinations of the line load branch points, the coordinated control content is determined. This approach takes into account the overall considerations of the project priorities, corresponding project contents of each self-balancing adjustment project, and the load data combinations of the line load branch points, ensuring the accuracy of the coordinated control content.

[0089] At this point, the system sends data requests to all line load branches (i.e., target nodes) corresponding to the "self-balancing regulation items" determined in S151 via high-speed communication networks (such as IEC61850 GOOSE or MQTT). The collected "load data combination" is a comprehensive dataset, which typically includes steady-state electrical quantities (three-phase voltage, current, active / reactive power, neutral current), dynamic electrical quantities (harmonic components, voltage fluctuations and flicker values), and equipment status quantities (operating status of control devices, DC bus voltage, switching device temperature, etc.). To ensure the accuracy of decision-making, the data acquisition of all target nodes must be time-synchronized, usually using GPS or IEEE1588 Precision Time Protocol (PTP) for time synchronization, ensuring that all data reflect the power grid conditions at the same moment.

[0090] The decision inputs include the project priorities that determine the processing order, the project content that provides the initial target value, and the load data combination that provides the current operating conditions and constraints. The dynamic decision process performs parameter fine-tuning. The system uses the latest load data to perform a final precise calculation on the "target adjustment amount" in S151 to ensure that the ideal balance can be achieved after adjustment.

[0091] The system will use the latest load data to conduct a comprehensive "pre-regulation safety rehearsal". Through grid model or power flow calculation, it will simulate the state of the grid after the regulation operation and check voltage constraints (whether the voltage of each phase after regulation is still within the allowable deviation), current / capacity constraints (whether the line current and apparent power of the control device exceed the rated value), and power quality constraints (whether the regulation process will introduce new harmonics or cause flicker to exceed the standard). Only when all safety checks are passed will the system finally determine the "cooperative control content". This content is the low-level instructions that directly drive the hardware, such as the active power reference value and power change rate instruction sent to the converter controller.

[0092] Specifically, the system has generated a regulation item with the highest priority for point A3; the system sends a high-priority data request to the load imbalance control device at point A3; the device at point A3 responds to the request and uploads the latest data with the timestamp t=15:30:10.000; the data combination includes: three-phase voltage (Ua=220.1V, Ub=219.5V, Uc=220.3V), three-phase active power (Pa=4.82kW, Pb=27.95kW, Pc=4.81kW), three-phase current (Ia=22.1A, Ib=128.5A, Ic=22.0A, In=98.2A) and device status (operating, temperature: 45°C).

[0093] The system's decision inputs are: project priority "highest"; project content (target adjustment amount 11.665kW, direction B→A,C, speed 1.5kW / s); and the latest load data combination. During the dynamic decision-making process, the system performs parameter fine-tuning, recalculates the adjustment amount to make the three-phase power completely equal based on the latest Pa, Pb, and Pc values, and obtains a more accurate ΔP=11.565kW, which is then used as the final target. Next, a safety verification is performed: voltage pre-simulation shows that the three-phase voltage will stabilize at around 220V after adjustment, which is completely within the ±5% safety range; capacity pre-simulation shows that the apparent power of the device after adjustment is approximately 12kVA, far less than its rated capacity of 50kVA.

[0094] After all verifications are passed, the system finally determines the "cooperative control content" as follows: send an active power reference value command to the PWM controller of the A3 point control device, requiring it to transfer 5.7825kW of active power from phase B to phases A and C at a rate of 1.5kW / s; through step S152, the system accurately adjusts and rigorously verifies a preset adjustment plan based on the latest grid operating conditions, and finally generates a low-level control command that can be executed safely and reliably, making full preparations for the final triggering.

[0095] Therefore, the branch point combinations corresponding to each line load branch point are marked. Based on the collaborative control content and the branch point combinations corresponding to each line load branch point, the corresponding collaborative control system is determined. Based on the identification of the collaborative control system, corresponding collaborative control items are assigned, and collaborative control of each line load branch point in the balance dimension is triggered based on the collaborative control items. This approach is compatible with the overall consideration of the collaborative control content and the branch point combinations corresponding to each line load branch point, ensuring the accuracy of the corresponding collaborative control system. At the same time, it further controls the self-balancing adjustment events, realizing the overall consideration of each self-balancing adjustment item, the load data combination of each line load branch point, and the corresponding branch point combination, thereby improving the accuracy of the collaborative control system and triggering collaborative control of each line load branch point in the balance dimension.

[0096] At this point, the system iterates through all the "self-balancing adjustment projects" that were determined in S151 and whose final decisions were made in S152, and extracts the "target node" field from the "project content" of each project. The system summarizes and removes duplicates from all the extracted target nodes to form a unique "branch point combination" list. This list clearly defines the scope of the control command to be issued, and can be a single-point combination or a multi-point combination.

[0097] The "cooperative control content" generated in S152 for a single node is combined with the "branch point combination" defined in S153 to construct a "cooperative control system" containing global execution logic. The cooperative control system is a structured, self-contained execution scheme that integrates all necessary information to ensure that each node can work together accurately and without error. Its components include: a globally unique system ID for tracking and auditing; a branch point combination that clearly defines all nodes involved in this control; a control content mapping table that maps each branch point node to its corresponding "cooperative control content"; timing logic that defines the execution order and time relationship of the control content of each node (such as "immediate execution" or "synchronous execution"); and communication and security policies that specify the protocols and channels for communication with each node, as well as security mechanisms such as instruction confirmation, timeout retry, and failure rollback.

[0098] The system identifies all elements in the "cooperative control system," especially the "control content mapping table" and "timing logic," as one or more "cooperative control projects." Each project is essentially a complete low-level control instruction package with a target address. When cooperative control is triggered, the system sends the "cooperative control project" (i.e., the instruction package) to the corresponding control device in the "branch point combination" via the communication network. After receiving the instruction, the control device at each branch point parses the instruction using its local MCU / DSP and directly drives its internal power electronic devices (such as IGBTs or MOSFETs). For example, by adjusting the duty cycle and phase of the PWM drive signal, the back-to-back converter is precisely controlled to achieve the transfer of active power between phases. During execution, the device continuously monitors the actual transferred power and compares it with the reference value in the instruction, forming a dynamic closed-loop control to ensure that the actual adjustment is consistent with the target value. Thus, the cooperative control of the line load branch points in a specific balance dimension is successfully triggered and executed.

[0099] Specifically, the system has generated the final "cooperative control content" for point A3; the system checks the currently unique "self-balancing adjustment project" (project 1), whose target node is A3_Feeder_Node; the system generates the "branch point combination" as: [A3_Feeder_Node]; the system combines the "cooperative control content" of S152 with the above "branch point combination" to construct a "cooperative control system", the elements of which include: system ID (Ctrl_Sys_AutoBalance_20231027_001), branch point combination ([A3_Feeder_Node]), control content mapping table (mapping A3_Feeder_Node to specific power reference values, directions and rates, etc.), timing logic (Execute_Immediately), and communication strategy (such as using the IEC61850 GOOSE protocol and setting a timeout time).

[0100] The system solidifies the above structure into a "cooperative control project," which is a command packet containing all the above information, ready to be sent via a GOOSE message. When control is triggered, the system sends this GOOSE message to the load imbalance control device at point A3 via the fiber optic network. The IED (Intelligent Electronic Device) of the device at point A3 receives the message, parses the active power reference value as 11.565kW, the direction as B→A,C, and the slope as 1.5kW / s. The device's DSP immediately executes the algorithm, generates a new PWM drive signal, controls the switching action of the IGBT, activates the back-to-back converter, and begins to smoothly transfer power from phase B to phases A and C. At the same time, the device's sensors continuously monitor the transferred power and compare it with the reference value to form a closed-loop feedback, ensuring the accuracy and stability of the adjustment process. Thus, the self-balancing control at point A3 is successfully triggered.

[0101] Please see Figure 7 , Figure 7 This is a schematic diagram of the structural composition of the self-balancing control system for power distribution lines in an embodiment of the present invention; the self-balancing control system for power distribution lines includes: The electrical parameter combination module 21 is used to collect the distribution map of the power distribution line, determine multiple line load branch points based on the distribution map of the power distribution line and the corresponding multiple working data, and determine the corresponding electrical parameter combination based on the detection of each line load branch point. The unbalanced state module 22 is used to determine the corresponding instantaneous active power and average power based on the identification of the electrical parameter combination, and to determine the corresponding unbalance degree based on the matching of the instantaneous active power, average power and corresponding voltage data, so as to mark the unbalanced state of the line load branch point. The balance adjustment system module 23 is used to determine the corresponding unbalance event based on the tracing of the unbalance state of the load branch point of the line, determine multiple unbalance factors based on the identification of the unbalance event, and determine the balance adjustment system based on the multiple unbalance factors, the corresponding unbalance degree and the branch point location of the load branch point of the line. The self-balancing adjustment event module 24 is used to determine the corresponding balance adjustment amount based on the detection of the balance adjustment system in the balance adjustment system, and to determine the corresponding self-balancing adjustment event based on the balance adjustment amount, the corresponding adjustment direction and the state distribution diagram of the distribution line. The collaborative control system module 25 is used to determine multiple self-balancing adjustment items based on the identification of the self-balancing adjustment event, and to determine the corresponding collaborative control system according to the combination of load data of each self-balancing adjustment item, each line load branch point and the corresponding branch point combination, so as to trigger the collaborative control of each line load branch point in the balance dimension.

[0102] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A self-balancing control method for power distribution lines, characterized in that, include: Collect the distribution map of the power distribution line, determine multiple line load branch points based on the distribution map and corresponding multiple working data, and determine the corresponding electrical parameter combination based on the detection of each line load branch point; Based on the identification of the electrical parameter combination, the corresponding instantaneous active power and average power are determined. Based on the matching of the instantaneous active power, average power and the corresponding voltage data, the corresponding unbalance is determined to mark the unbalance state of the line load branch point. Based on the tracing of the unbalanced state of the load branch point of the line, the corresponding unbalanced event is determined. Based on the identification of the unbalanced event, multiple unbalanced factors are determined. Based on the multiple unbalanced factors, the corresponding unbalance degree, and the branch point location of the load branch point of the line, a balance adjustment system is determined. A balance adjustment system includes the following elements: system ID and target branch point, adjustment target, core adjustment principle, and detailed execution plan. In the balance regulation system, the corresponding balance regulation amount is determined based on the detection of the balance regulation system, and the corresponding self-balancing regulation event is determined based on the balance regulation amount, the corresponding regulation direction and the state distribution diagram of the power distribution line. Based on the identification of this self-balancing adjustment event, multiple self-balancing adjustment items are determined. According to the combination of load data of each self-balancing adjustment item, each line load branch point, and the corresponding branch point combination, a corresponding collaborative control system is determined to trigger the collaborative control of each line load branch point in the balance dimension.

2. The self-balancing control method for power distribution lines according to claim 1, characterized in that, The process involves collecting a distribution map of the power distribution lines, determining multiple line load branch points based on the distribution map and corresponding working data, and determining corresponding electrical parameter combinations based on the detection of each line load branch point, including: The location of the power distribution line is marked, and the distribution map of the power distribution line is determined based on the tracing of the location of the power distribution line. Multiple line areas are determined based on the identification of the distribution map of the power distribution line. Multiple line load branch points are determined based on the regional location of each line area, the corresponding regional load, and the corresponding multiple working data. In multiple line load branch points, multiple electrical parameters are determined based on the dynamic detection of each line load branch point. These multiple electrical parameters include three-phase voltage, three-phase current, and neutral current. Based on these multiple electrical parameters and the corresponding line load branch points, a corresponding combination of electrical parameters is determined.

3. The self-balancing control method for power distribution lines according to claim 1, characterized in that, The process of determining the corresponding instantaneous active power and average power based on the identification of the electrical parameter combination, and determining the corresponding unbalance degree based on the matching of the instantaneous active power, average power, and corresponding voltage data to mark the unbalance state of the line load branch point includes: The electrical parameter combination is dynamically identified, and a first power data combination and a second power data combination are determined based on the dynamic identification of the electrical parameter combination. The corresponding instantaneous active power is determined based on the identification of the first power data combination, and the corresponding average power is determined based on the identification of the second power data combination. The first imbalance coefficient is determined based on the instantaneous active power and the corresponding voltage data. The second imbalance coefficient is determined based on the average power and the corresponding voltage data. The corresponding imbalance degree is determined based on the mapping relationship between the first imbalance coefficient, the second imbalance coefficient and the imbalance degree. The corresponding imbalance state is determined based on the matching of each load branch point of the line.

4. The self-balancing control method for power distribution lines according to claim 1, characterized in that, The process of determining the corresponding imbalance event based on the tracing of the imbalance state of the line load branch point, identifying multiple imbalance factors based on the identification of the imbalance event, and determining the balance adjustment system based on the multiple imbalance factors, the corresponding imbalance degree, and the branch point location of the line load branch point includes: The unbalanced state of the line load branch point is traced, and multiple unbalanced features are marked during the tracing process. The corresponding unbalanced events are determined based on the feature location, corresponding feature shape, and branch point location of the line load branch point.

5. The self-balancing control method for power distribution lines according to claim 4, characterized in that, The method of determining the corresponding imbalance event based on the tracing of the imbalance state of the line load branch point, determining multiple imbalance factors based on the identification of the imbalance event, and determining the balance adjustment system based on the multiple imbalance factors, the corresponding imbalance degree, and the branch point location of the line load branch point, further includes: Based on the detection of this imbalance event, multiple imbalance items are identified, and corresponding imbalance factors are determined according to the identification of each imbalance item, so as to collect multiple imbalance factors; the first level of balance adjustment content is determined based on multiple imbalance factors and corresponding imbalance degree. The second level of balancing adjustment is determined based on multiple imbalance factors and the location of the branch points of the line load branch points. The balancing adjustment system is then determined based on the first and second level of balancing adjustment.

6. The self-balancing control method for power distribution lines according to claim 1, characterized in that, In the balancing control system, the corresponding balancing control amount is determined based on the detection of the balancing control system, and the corresponding self-balancing control event is determined based on the balancing control amount, the corresponding control direction, and the state distribution diagram of the power distribution line, including: The system monitors the balance adjustment system in real time, detects the balance adjustment system, and marks multiple balance adjustment contents during the detection process. Based on each balance adjustment content, the corresponding balance dimension and the current load level of the line load branch point, the corresponding balance adjustment amount is determined and the corresponding adjustment direction is marked.

7. The self-balancing control method for power distribution lines according to claim 6, characterized in that, In the balancing control system, the corresponding balancing control amount is determined based on the detection of the balancing control system, and the corresponding self-balancing control event is determined based on the balancing control amount, the corresponding control direction, and the state distribution diagram of the power distribution line. This also includes: The distribution map of the power distribution lines is collected. Based on the distribution map of the power distribution lines and the real-time status of multiple line load branch points, the state distribution map of the power distribution lines is determined. The self-balancing adjustment framework is determined according to the state distribution map of the power distribution lines and the balance adjustment amount. The corresponding self-balancing adjustment event is determined according to the self-balancing adjustment framework and the adjustment direction corresponding to the balance adjustment amount.

8. The self-balancing control method for power distribution lines according to claim 1, characterized in that, The process involves identifying multiple self-balancing adjustment items based on the recognition of the self-balancing adjustment event, and determining a corresponding coordinated control system based on the combination of load data from each self-balancing adjustment item, each line load branch point, and the corresponding branch point combination. This system triggers coordinated control of each line load branch point in the balance dimension, including: Collect self-balancing adjustment events, identify multiple self-balancing adjustment items based on the identification of self-balancing adjustment events, determine the project priority of each self-balancing adjustment item based on the matching of each self-balancing adjustment item, and mark the project content of each self-balancing adjustment item. The load data of each line load branch point is collected and combined. Based on the project priority, corresponding project content and load data combination of each line load branch point, the coordinated control content is determined.

9. The self-balancing control method for power distribution lines according to claim 8, characterized in that, The process of identifying multiple self-balancing adjustment items based on the recognition of the self-balancing adjustment event, and determining a corresponding coordinated control system based on the combination of load data of each self-balancing adjustment item, each line load branch point, and the corresponding branch point combination to trigger coordinated control of each line load branch point in the balance dimension, further includes: The branch point combination corresponding to each line load branch point is marked. Based on the collaborative control content and the branch point combination corresponding to each line load branch point, the corresponding collaborative control system is determined. Based on the identification of the collaborative control system, the corresponding collaborative control items are rated, and the collaborative control of each line load branch point in the balance dimension is triggered based on the collaborative control items.

10. A self-balancing control system for a power distribution line, characterized in that, The self-balancing control system of the power distribution line is applied to the self-balancing control method of the power distribution line as described in any one of claims 1-9.

Images