Feeder timing prediction method based on power distribution network operation and device thereof

Abstract

This invention provides a method and apparatus for feeder timing prediction based on distribution network operation. The method includes: determining the load power sequence and load flow direction sequence based on feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches; identifying reverse overload states based on the load power sequence and load flow direction sequence of each target feeder to obtain reverse overload feeders and non-reverse overload feeders; analyzing the potential evolution path of each non-reverse overload feeder in the next prediction period based on the load change synchronicity of each feeder pair between the non-reverse overload feeders and the reverse overload feeders, and constructing a reverse overload influence propagation chain; and analyzing the load evolution trend of each target feeder in the next prediction period based on the reverse overload influence propagation chain to obtain the load timing prediction result of each target feeder. This invention avoids the scheduling lag and equipment overload risk caused by prediction blind spots.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method and apparatus for predicting feeder timing based on the operation of a power distribution network. Background Technology

[0002] In distribution network operation and management, feeder load forecasting is a crucial aspect of ensuring the safe and stable operation of the power grid. Currently, feeder load time-series forecasting commonly employs time-series modeling methods based on historical load data of individual feeders. These methods typically assume that the operating states of each feeder are independent, extrapolating future load periods solely based on its own historical load trends.

[0003] However, in actual urban power distribution networks, multiple feeders within the same power supply area often exhibit significant dynamic load coupling due to factors such as sharing substation buses, similar load characteristics of adjacent users, proximity of distributed generation access locations, or frequent operation of tie switches. When one of the feeders experiences reverse overload (i.e., the power flow direction is opposite to the design direction and the load rate exceeds the safety threshold) due to excessive reverse power from distributed generation or sudden load changes, its abnormal operating state will rapidly affect neighboring feeders through electrical connections or operation scheduling strategies within the area, triggering a chain reaction of load transfers or voltage overruns.

[0004] Because existing prediction methods do not consider the coupling and propagation mechanism of operating states between feeders, especially neglecting the disturbance effect of reverse heavy overload events on the load evolution paths of surrounding feeders, they cannot identify the risk of normally functioning feeders entering reverse heavy overload due to the influence of abnormal feeders when predicting the load at the next time moment. This prediction blind spot makes it difficult for the dispatching system to take intervention measures such as power transfer, load shedding, or adjustment of distributed power sources before the event occurs, resulting in long-term equipment overload, protection malfunctions, or even partial power outages. Therefore, in the feeder timing prediction process, based on the reverse heavy overload state that has already occurred in the area, it is impossible to dynamically predict its disturbance impact on the load evolution paths of other feeders. Summary of the Invention

[0005] This invention provides a method and apparatus for predicting feeder timing based on distribution network operation, aiming to solve the problem of being unable to dynamically predict the impact of reverse heavy overload disturbances on other feeders, realize the early perception of potential new reverse heavy overload feeders, and avoid scheduling delays and equipment overload risks caused by prediction blind spots.

[0006] In a first aspect, the present invention provides a feeder timing prediction method based on distribution network operation, comprising: Based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches, the load power sequence and load flow direction sequence of each target feeder in the current time period are determined. Based on the load power sequence and load flow direction sequence of each target feeder, reverse heavy overload status identification is performed to obtain reverse heavy overload feeders and non-reverse heavy overload feeders. Based on the synchronicity of load changes of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous period, the potential evolution path of each non-reverse heavy overload feeder affected by the reverse heavy overload in the next prediction period is analyzed, and the propagation chain of the reverse heavy overload influence is constructed. Based on the analysis of the reverse overload influence propagation chain, the load evolution trend of each target feeder in the next prediction period is obtained, and the load time-series prediction results of each target feeder are obtained.

[0007] Optionally, the step of analyzing the load evolution trend of each target feeder in the next prediction period based on the reverse overload influence propagation chain to obtain the load time-series prediction result of each target feeder includes: Based on the positional relationship of each target feeder in the reverse heavy overload influence propagation chain, the load evolution role of each target feeder in the next prediction period is determined; The evolution mode is determined based on the load evolution role of each target feeder, and the load power change direction and load flow direction are determined based on the evolution mode of each target feeder in the next forecast period. Based on the load power change direction and load flow direction of each target feeder, and combined with the end value of the load power sequence and the end state of the load flow direction sequence in the current period, the estimated load power and estimated load flow direction at the first time point of the next forecast period are determined. Based on the estimated load power and load flow direction of each target feeder and the temporal evolution of their evolution patterns, the extension trajectories of the load power sequence and load flow direction sequence at each subsequent time point in the next prediction period are analyzed to obtain the load time series prediction results for each target feeder.

[0008] Optionally, the load evolution roles include reverse overload feeders belonging to the reverse overload influence propagation chain and non-reverse overload feeders with potential state transition conditions, as well as feeders not belonging to the reverse overload influence propagation chain; the determination of the evolution mode based on the load evolution role of each target feeder includes: For a target feeder whose load evolution role belongs to the reverse heavy overload feeder in the reverse heavy overload influence propagation chain, the load continuation mode in which the target feeder maintains the reverse heavy overload state in the next prediction period is determined as the evolution mode based on its load power sequence and load flow direction sequence in the current period. For a target feeder whose load evolution role belongs to a non-reverse heavy overload feeder in the reverse heavy overload influence propagation chain, based on its power flow disturbance response characteristics, the joint evolution mode of load power jump and load flow direction reversal that occurs in the next prediction period of the target feeder is determined as the evolution mode. For target feeders whose load evolution role does not belong to the reverse heavy overload influence propagation chain, the load stability evolution mode of the target feeder in the next prediction period is obtained based on its load power sequence and load flow direction sequence in the current period.

[0009] Optionally, the step of analyzing the potential evolution path of each non-reverse heavy overload feeder affected by reverse heavy overload in the next predicted period based on the load change synchronicity of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous time period, and constructing the reverse heavy overload influence propagation chain, includes: The load power sequence of each feeder pair within a preset continuous time period is paired point by point according to the same time point alignment relationship to obtain the synchronous time sequence pair of each feeder pair; For each feeder pair's synchronization time series pair, the direction consistency judgment is determined based on the first load power change direction of the non-reverse heavy overload feeder and the second load power change direction of the reverse heavy overload feeder within the same time interval; the direction consistency judgment result is determined; the load power change direction represents the increase or decrease trend of the load power value at the current time point relative to the load power value at the previous time point. Based on the direction consistency determination results of each feeder pair, the total number of time intervals in which the change direction of each feeder pair is consistent within a preset continuous period is determined, and based on the ratio of the total number of time intervals to the total number of time intervals within the preset continuous period, the load change direction synchronization rate of each feeder pair is determined. If the load change direction synchronization rate is greater than or equal to the preset synchronization rate threshold, then the load change direction synchronization rate is determined as the load coupling strength of each feeder pair; otherwise, the load coupling strength is determined to be zero. Based on the load coupling strength analysis of each feeder pair, the potential evolution path of each non-reverse heavy overload feeder affected by reverse heavy overload in the next prediction period is analyzed, and the propagation chain of reverse heavy overload impact is constructed.

[0010] Optionally, the step of analyzing the potential evolution path of each non-reverse overload feeder affected by reverse overload in the next prediction period based on the load coupling strength analysis of each feeder pair, and constructing the reverse overload influence propagation chain, includes: Iterate through the load coupling strength of each feeder pair, identify the feeder pairs with load coupling strength greater than zero as load-coupled feeder pairs, and determine the unidirectional coupling connection relationship with the reverse heavy overload feeder as the source end and the non-reverse heavy overload feeder as the receiving end based on the pairing structure between the reverse heavy overload feeder and the non-reverse heavy overload feeder in each load-coupled feeder pair. Based on the operating status characteristics of the reverse overload feeder in each unidirectional coupling connection during the current time period, the initial action mode of the reverse overload feeder influencing the non-reverse overload feeder coupled with it is determined; the operating status characteristics characterize the load power change trend direction and load flow direction stability of the reverse overload feeder during the current time period. Based on the unidirectional coupling connection between each non-reverse overload feeder and all its coupled reverse overload feeders and the corresponding initial operating mode, the set of multi-source influence inputs on the non-reverse overload feeder in the next prediction period is determined. Based on the persistence characteristics of each reverse heavy overload feeder in the multi-source influence input set, the reverse heavy overload influence propagation chain is constructed; the persistence characteristics represent the monotonic continuation of the load power sequence and the consistent continuation of the load flow direction sequence of the reverse heavy overload feeder within a preset continuous time period.

[0011] Optionally, constructing the reverse overload influence propagation chain based on the persistence characteristics of each reverse overload feeder in the multi-source influence input set includes: Based on the persistence characteristics analysis of each reverse overload feeder in the multi-source influence input set, the persistence influence of each reverse overload feeder on the non-reverse overload feeder is analyzed, and the persistence judgment result is obtained. Based on the continuity judgment result as a continuous impact input, and combined with the load power sequence and load flow direction sequence of the corresponding non-reverse heavy overload feeder in the current period, the power flow disturbance response characteristics of the non-reverse heavy overload feeder in the next prediction period are determined. Based on the joint analysis of the load power growth direction and the reversal possibility of the load power flow direction in the power flow disturbance response characteristics, it is determined whether a non-reverse heavy overload feeder has the potential state transition conditions to evolve into a reverse heavy overload feeder. Based on the potential state transition conditions of all non-reverse heavy overload feeders and their unidirectional coupling connection with the upstream reverse heavy overload feeder, the non-reverse heavy overload feeder with potential state transition conditions is used as the downstream transmission node. The reverse heavy overload influence propagation chain is constructed by stepping along the unidirectional coupling connection, starting from the reverse heavy overload feeder, and propagating through the unidirectional coupling connection to cover all non-reverse heavy overload feeders with potential state transition conditions.

[0012] Optionally, the step of identifying reverse overload conditions based on the load power sequence and load flow direction sequence of each target feeder to obtain reverse overload feeders and non-reverse overload feeders includes: The first feeder with reverse power flow behavior is identified based on the load flow direction sequence of each target feeder in the current time period; having reverse power flow behavior indicates that there is at least one time point in time where the load flow direction is from the end of the feeder to the substation outgoing switch. The absolute value of the maximum active power is determined based on the active power value corresponding to the time point when the load flow direction is reversed in the load power sequence of the first feeder. The first feeder whose maximum active power absolute value is greater than or equal to the active power threshold corresponding to the rated current carrying capacity of the feeder is determined as the second feeder that meets the reverse heavy load requirement. The total duration during which the second feeder meets the requirement that the load flow direction is reverse and the absolute value of active power is not less than the active power threshold during the current time period is determined, as well as the proportion of the total duration to the total duration of the current time period. The second feeder with a total duration not less than the preset minimum duration threshold and a duration ratio not less than the preset time ratio threshold is identified as the reverse overload feeder, and the feeders other than the reverse overload feeder among the target feeders are identified as non-reverse overload feeders.

[0013] In a second aspect, the present invention also provides a feeder timing prediction device based on distribution network operation, for implementing the feeder timing prediction method based on distribution network operation as described in the first aspect; the feeder timing prediction device based on distribution network operation includes: The power grid operation monitoring module is used to determine the load power sequence and load flow direction sequence of each target feeder in the current time period based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches. The heavy overload condition identification module is used to identify the reverse heavy overload condition based on the load power sequence and load flow direction sequence of each target feeder, and to obtain the reverse heavy overload feeder and the non-reverse heavy overload feeder. The evolution path analysis module is used to analyze the potential evolution path of each non-reverse overload feeder affected by reverse overload in the next prediction period based on the load change synchronicity of each feeder pair of the non-reverse overload feeder and the reverse overload feeder in a preset continuous period, and to construct the reverse overload influence propagation chain. The feeder timing prediction module is used to analyze the load evolution trend of each target feeder in the next prediction period based on the reverse heavy overload influence propagation chain, and obtain the load timing prediction result of each target feeder.

[0014] Thirdly, the present invention also provides an electronic device, comprising: a memory for storing computer software programs; and a processor for reading and executing the computer software programs, thereby realizing the feeder timing prediction method based on the operation of the distribution network as described above.

[0015] Fourthly, the present invention also provides a non-transitory computer-readable storage medium storing a computer software program, which, when executed by a processor, implements the feeder timing prediction method based on the operation of a power distribution network as described above.

[0016] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the feeder timing prediction method based on the operation of a power distribution network as described above.

[0017] The feeder timing prediction method based on distribution network operation provided in this invention identifies reverse heavy overload feeders and non-reverse heavy overload feeders by identifying reverse heavy overload status. Then, based on the classification results, it analyzes the load change synchronicity of each feeder pair of non-reverse heavy overload feeders and reverse heavy overload feeders within a preset continuous time period. It clarifies the potential evolution path of each non-reverse heavy overload feeder affected by reverse heavy overload in the next prediction period and constructs the reverse heavy overload influence propagation chain. Finally, based on the constructed influence propagation chain, it analyzes the load evolution trend of each target feeder in the next prediction period and obtains the load timing prediction result. Therefore, the embodiments of the present invention achieve accurate identification of reverse overload feeders, dynamic tracing and propagation chain construction of their influence paths on surrounding non-reverse overload feeders, and time-series prediction of all-target feeder loads combined with the influence of propagation chain disturbances. This fills the prediction blind spot that ignores the propagation mechanism of reverse overload disturbances between feeders, and can dynamically predict the disturbance impact of reverse overload on the load evolution path of other feeders. It enables early detection of potential new reverse overload feeders and avoids scheduling delays and long-term equipment overload risks. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the feeder timing prediction method based on distribution network operation provided in an embodiment of the present invention. Figure 2 This is a schematic diagram of the feeder timing prediction device based on distribution network operation provided in an embodiment of the present invention; Figure 3 An embodiment diagram of the electronic device provided in this invention; Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] In the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0021] In the description of this invention, the term "for example" is used to mean "used as an example, illustration, or description." Any embodiment described as "for example" in this invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use the invention. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that the invention can be made without using these specific details. In other instances, well-known structures and processes will not be described in detail to avoid obscuring the description of the invention with unnecessary detail. Therefore, the invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed herein.

[0022] Optionally, see Figure 1 , Figure 1 This is a flowchart illustrating the feeder timing prediction method based on distribution network operation provided by the present invention. In this embodiment of the invention, the execution entity of the feeder timing prediction method based on distribution network operation is the feeder timing prediction device.

[0023] Therefore, feeder timing prediction methods based on distribution network operation include: Step 10: Based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches, determine the load power sequence and load flow direction sequence of each target feeder in the current time period.

[0024] Optionally, the feeder timing prediction device acquires feeder operation data for each target feeder within the power supply area, as well as power flow direction data for the outgoing switches of the corresponding substation. The target feeder refers to the feeder within the power supply area selected as the object of analysis and prediction.

[0025] Optionally, feeder operation data refers to data reflecting the feeder's operating status, including active power data, reactive power data, and current phase angle data. Active power data refers to the power used for external work within the electrical energy consumed or output by the feeder per unit time. Reactive power data refers to the power exchanged between the feeder and the power system per unit time that does not perform external work. Current phase angle data refers to the phase angle data of the current signal in the feeder as it changes over time.

[0026] Optionally, a substation outgoing line switch refers to a switching device installed at the outgoing end of a substation to control the connection and disconnection of the power supply circuit from the substation to the feeder. Power flow direction refers to the direction in which electricity flows within a power system. For a substation outgoing line switch, the direction of power flow from the substation to the feeder is the forward power flow direction, and the direction of power flow from the feeder to the substation is the reverse power flow direction.

[0027] Furthermore, the feeder timing prediction device preprocesses the feeder operation data acquired for each target feeder. The preprocessing process includes data cleaning and data sampling: data cleaning refers to removing abnormal data from the feeder operation data. Abnormal data includes data that significantly deviates from the normal operating range, duplicate data, and missing data. For missing data, valid data from adjacent time periods is used for interpolation to supplement it. Adjacent time periods refer to several normal time periods immediately before and after the time period corresponding to the missing data. Specifically, valid data from one time period before and after the missing data can be selected for average interpolation to supplement it.

[0028] Optionally, data sampling refers to extracting the cleaned feeder operation data at preset time intervals to obtain the active power data sequence and reactive power data sequence arranged in chronological order within the current time period. The preset time interval refers to the time interval for data collection that is set in advance and can be set according to the actual prediction accuracy requirements, for example, 5 minutes.

[0029] Furthermore, the feeder timing prediction device determines the load power sequence of each target feeder based on the preprocessed active power data sequence and reactive power data sequence.

[0030] The load power sequence refers to the sequence of load power of the target feeder at each sampling time in the current period, arranged in chronological order. The load power is calculated from active power data and reactive power data. Specifically, the load power at each sampling time is equal to the square root of the sum of the squares of the active power data and the reactive power data at that time.

[0031] Furthermore, the feeder timing prediction device, in conjunction with the power flow direction of the substation outgoing switches, determines the load power flow direction sequence for each target feeder. The load power flow direction sequence refers to the sequence of load power flow directions of the target feeder at each sampling time within the current time period, arranged chronologically. Specifically, the device compares the current phase angle data at each sampling time with the reference phase angle corresponding to the power flow direction of the substation outgoing switches. The reference phase angle is a pre-set standard value used to determine the power flow direction. If the difference between the current phase angle data and the reference phase angle is within a preset positive range, the load power flow direction at that sampling time is determined to be positive. The preset positive range refers to the pre-set range of phase angle differences representing a positive power flow direction. If the difference between the current phase angle data and the reference phase angle is within a preset negative range, the load power flow direction at that sampling time is determined to be negative. The preset negative range refers to the pre-set range of phase angle differences representing a negative power flow direction. By arranging the load power flow directions determined at each sampling time in chronological order, the load power flow direction sequence of the target feeder can be obtained.

[0032] In one embodiment, three target feeders are selected within the power supply area, namely target feeder 1, target feeder 2, and target feeder 3. The current time period is set to 14:00-14:20, and the preset time interval is 5 minutes. Therefore, the sampling times within the current time period are 14:00, 14:05, 14:10, 14:15, and 14:20, for a total of 5 sampling times. The preset reference phase angle is 0 degrees, the preset positive range is -30 degrees to 30 degrees, and the preset negative range is 150 degrees to 210 degrees.

[0033] The feeder timing prediction device acquires feeder operation data (including active power data, reactive power data, and current phase angle data) of three target feeders, as well as power flow direction data of the substation outgoing switch (here, the power flow direction reference of the substation outgoing switch is set to positive from the substation to the feeder).

[0034] Furthermore, the feeder timing prediction device preprocesses the acquired feeder operation data: For target feeder 1, it is found that its active power data at 14:05 is missing. The active power data of 80 kW at 14:00 and 82 kW at 14:10 are selected for average interpolation to supplement it. The supplemented active power data at 14:05 is (80+82) / 2=81 kW.

[0035] Furthermore, the reactive power data repeatedly recorded at 14:15 by target feeder 2 was removed, and only one valid data point was retained; the feeder operation data of target feeder 3 showed no abnormalities and was directly retained. After data cleaning, the cleaned data of the three target feeders were sampled at 5-minute intervals to obtain the active power data sequence and reactive power data sequence of each target feeder at 5 sampling times.

[0036] Furthermore, the feeder timing prediction device calculates the load power sequence for each target feeder: taking target feeder 1 as an example, its active power data sequence at the sampling time is 80 kW, 81 kW, 82 kW, 83 kW, and 84 kW, and its reactive power data sequence is 30 kV, 31 kV, 32 kV, 33 kV, and 34 kV. The load power at each sampling time is calculated as the square root of the sum of the squares of the active power data and the reactive power data. For example, the load power at 14:00 is the square root of (80 squared + 30 squared), which is approximately 85.44 kW. The load power of target feeder 1 at the five sampling times is calculated sequentially and arranged in chronological order to form the load power sequence of target feeder 1.

[0037] Furthermore, the feeder timing prediction device determines the load flow direction sequence for each target feeder: taking target feeder 2 as an example, its current phase angle data at five sampling times are 10 degrees, 25 degrees, 160 degrees, 190 degrees, and 25 degrees. Comparing the current phase angle data at each time with the reference phase angle of 0 degrees, 10 degrees, 25 degrees, and 25 degrees are all within the preset positive range of -30 degrees to 30 degrees, thus the load flow direction at these three times is determined to be positive; 160 degrees and 190 degrees are both within the preset negative range of 150 degrees to 210 degrees, thus the load flow direction at these two times is determined to be negative. Arranging the above determination results in the time sequence of 14:00, 14:05, 14:10, 14:15, and 14:20, the load flow direction sequence for target feeder 2 is obtained as positive, positive, negative, negative, positive. Similarly, the feeder timing prediction device completes the determination of the load flow direction sequences for target feeders 1 and 3.

[0038] Step 20: Based on the load power sequence and load flow direction sequence of each target feeder, reverse heavy overload state identification is performed to obtain reverse heavy overload feeders and non-reverse heavy overload feeders.

[0039] Optionally, the feeder timing prediction device identifies the reverse overload state based on the load power sequence and load flow direction sequence of each target feeder, distinguishing between reverse overload feeders and non-reverse overload feeders, as described in steps 201 to 204. Among them, the reverse heavy overload state refers to the operating state of the target feeder that simultaneously meets the conditions of reverse power flow direction and load power exceeding a certain proportion of the feeder's rated load power.

[0040] Optionally, a reverse overload feeder refers to a target feeder that is in a reverse overload state; a non-reverse overload feeder refers to a target feeder that is not in a reverse overload state.

[0041] Optionally, the rated load power refers to the maximum load power that the feeder is designed to operate stably for a long period of time, and is an inherent parameter of the feeder; the specific ratio refers to the ratio threshold set in advance according to the operating safety requirements and design standards of the feeder. For example, it is set to 100%, which means that the load power exceeds the rated load power and is considered overloaded, and set to 80%, which means that the load power exceeds 80% of the rated load power and is considered overloaded.

[0042] Step 30: Based on the load change synchronicity of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous time period, analyze the potential evolution path of each non-reverse heavy overload feeder affected by the reverse heavy overload in the next prediction time period, and construct the reverse heavy overload influence propagation chain.

[0043] Optionally, the feeder timing prediction device analyzes the potential evolution path of each non-reverse overload feeder affected by reverse overload in the next prediction period based on the load change synchronicity of each feeder pair of the non-reverse overload feeder and the reverse overload feeder in a preset continuous period, and constructs the reverse overload influence propagation chain, as in steps 301 to 304.

[0044] Here, a feeder pair refers to a combination consisting of one non-reverse heavy overload feeder and one reverse heavy overload feeder. A preset continuous time period refers to a pre-defined continuous time interval used to analyze the synchronicity of load changes. This time interval must include multiple consecutive sampling moments; for example, if set to 1 hour, it corresponds to 12 five-minute sampling moments. Load change synchronicity refers to the degree of consistency in the trend of load power changes over time between the two feeders within the preset continuous time period; the more consistent the trend, the higher the synchronicity. The next forecast period refers to the time interval after the current period where load forecasting needs to be performed; its duration is the same as the current period. For example, if the current period is 20 minutes, the next forecast period is also 20 minutes. Optionally, the potential evolution path refers to the direction and process in which the load state of a non-reverse overload feeder may change under the influence of a reverse overload feeder.

[0045] Optionally, the reverse overload impact propagation chain refers to a chain structure formed by connecting feeders affected by the reverse overload in sequence according to the order and degree of impact, starting from the reverse overload feeder. It is used to characterize the propagation path and range of the reverse overload impact between feeders.

[0046] Step 40: Based on the reverse overload influence propagation chain analysis, analyze the load evolution trend of each target feeder in the next prediction period to obtain the load time series prediction results of each target feeder.

[0047] Optionally, the feeder timing prediction device analyzes the load evolution trend of each target feeder in the next prediction period based on the reverse heavy overload influence propagation chain, and obtains the load timing prediction result of each target feeder, as in steps 401 to 404.

[0048] Among them, the load evolution trend refers to the direction and magnitude of the change in the load power of the target feeder in the time dimension, such as the load power gradually increasing, gradually decreasing, or remaining stable.

[0049] Optionally, the load time-series prediction result refers to the data sequence obtained by predicting the load power of the target feeder at each sampling time in the next prediction period according to the time sequence. This sequence can reflect the load change of the target feeder in the next prediction period.

[0050] This invention achieves accurate identification of reverse overload feeders, dynamic tracing and propagation chain construction of their impact paths on surrounding non-reverse overload feeders, and time-series prediction of all-target feeder loads by combining the propagation chain disturbance effects. It fills the prediction blind spot that ignores the propagation mechanism of reverse overload disturbances between feeders, and can dynamically predict the disturbance impact of reverse overload on the load evolution path of other feeders. It enables early detection of potential new reverse overload feeders and avoids scheduling delays and long-term equipment overload risks.

[0051] Optionally, the processes of steps 201 to 204 include: Step 201: Identify the first feeder with reverse power flow behavior based on the load flow direction sequence of each target feeder in the current time period.

[0052] Optionally, the feeder timing prediction device acquires the load flow direction sequence for each target feeder. Here, reverse flow behavior indicates that the load flow direction at at least one time point is from the feeder end to the substation outgoing switch; the time point refers to each sampling moment in the load flow direction sequence; the feeder end refers to the end of the feeder furthest from the substation outgoing switch, which is the terminal part of the feeder supplying power to the user side; the substation outgoing switch refers to the switching equipment installed at the substation outgoing end, used to control the on / off of the substation power supply circuit to the feeder.

[0053] Furthermore, the feeder timing prediction device traverses and checks the load flow direction sequence of each target feeder one by one to determine whether there is at least one time point in the sequence where the load flow direction conforms to the definition of reverse flow behavior, that is, the load flow direction is from the end of the feeder to the substation outgoing switch.

[0054] Furthermore, the target feeders that meet the time points of reverse power flow behavior after verification are identified as the first feeders with reverse power flow behavior; the target feeders whose load power flow direction at all time points in the load power flow direction sequence is not from the end of the feeder to the substation outgoing switch are excluded from the first feeders.

[0055] In one embodiment, there are three target feeders (target feeder 1, target feeder 2, and target feeder 3) within the power supply area. The current time period is 14:00-14:20, with a preset time interval of 5 minutes. The sampling times (time points) within the current time period are 14:00, 14:05, 14:10, 14:15, and 14:20, for a total of 5 time points. The load flow direction sequence of each target feeder has been determined by step 10, as follows: The load flow direction sequence of target feeder 1 is forward, forward, forward, forward, forward (forward refers to flow from the substation outgoing switch to the end of the feeder); the load flow direction sequence of target feeder 2 is forward, forward, reverse, reverse, forward (reverse refers to flow from the end of the feeder to the substation outgoing switch); the load flow direction sequence of target feeder 3 is forward, reverse, forward, reverse, forward.

[0056] After obtaining the load flow direction sequences of the three target feeders, the feeder timing prediction device checks them one by one: In the sequence of target feeder 1, all time points are in the forward direction, with no time points in the reverse flow direction, therefore it is not included as the first feeder; in the sequence of target feeder 2, the time points 14:10 and 14:15 are in the reverse direction, indicating at least one time point in the reverse flow direction, therefore it is determined to be the first feeder; in the sequence of target feeder 3, the time points 14:05 and 14:15 are in the reverse direction, indicating at least one time point in the reverse flow direction, therefore it is determined to be the first feeder. The final first feeders are target feeders 2 and 3.

[0057] Step 202: Determine the absolute value of the maximum active power based on the active power value corresponding to the time point when the load flow direction is reversed in the load power sequence of the first feeder.

[0058] Optionally, the feeder timing prediction device acquires the load power sequence of each first feeder and the reverse power flow time points in the load power flow direction sequence of each first feeder. Here, the active power value refers to the active power data at the corresponding time point in the load power sequence, and the active power data refers to the power data of the electrical energy consumed or output by the feeder per unit time that performs work externally; the maximum absolute value of active power refers to the largest value selected after taking the absolute value of each value from all the extracted active power values ​​corresponding to the reverse power flow time points.

[0059] Furthermore, for each first feeder, the feeder timing prediction device accurately extracts the active power value at the corresponding time point from the load power sequence of that first feeder based on the identified reverse power flow time point.

[0060] Furthermore, the feeder timing prediction device calculates the absolute value for each extracted active power value. The calculation method for the absolute value is as follows: if the active power value is positive, its absolute value is the value itself; if the active power value is negative, its absolute value is the opposite of the value; if the active power value is 0, its absolute value is 0.

[0061] Furthermore, the feeder timing prediction device compares all the calculated absolute values ​​and selects the largest value, which is the maximum active power absolute value of the first feeder.

[0062] In one embodiment, the first feeders are target feeder 2 and target feeder 3; the reverse power flow time points of each first feeder have been determined, and the load power sequence (containing active power values) has been obtained in step 10, as follows: the active power values ​​at each time point in the load power sequence of target feeder 2 are 82 kW, 81 kW, 85 kW, 86 kW, and 83 kW, and the reverse power flow time points are 14:10 and 14:15; the active power values ​​at each time point in the load power sequence of target feeder 3 are 78 kW, 79 kW, 80 kW, 81 kW, and 82 kW, and the reverse power flow time points are 14:05 and 14:15.

[0063] For target feeder 2, the active power values ​​corresponding to its reverse power flow time points of 14:10 and 14:15 are extracted as 85 kW and 86 kW, respectively, that is, the absolute values ​​are 85 kW and 86 kW. Comparing the two absolute values, the maximum value is 86 kW. Therefore, the maximum absolute value of the active power of target feeder 2 is 86 kW.

[0064] For target feeder 3, the active power values ​​corresponding to its reverse power flow time points of 14:05 and 14:15 are extracted as 79 kW and 81 kW, respectively, that is, the absolute values ​​are 79 kW and 81 kW. Comparing the two absolute values, the maximum value is 81 kW. Therefore, the maximum absolute value of the active power of target feeder 3 is 81 kW.

[0065] Step 203: The first feeder whose maximum active power absolute value is greater than or equal to the active power threshold corresponding to the rated current carrying capacity of the feeder is identified as the second feeder that satisfies the reverse heavy load condition. The total duration during which the second feeder satisfies the load flow direction being reversed and the absolute value of active power not less than the active power threshold during the current time period is determined, as well as the proportion of the total duration to the total duration of the current time period.

[0066] Optionally, the feeder timing prediction device acquires the absolute value of the maximum active power of each first feeder, as well as the pre-set active power threshold and the total duration of the current period corresponding to each first feeder. The active power threshold corresponding to the rated current carrying capacity of the feeder refers to the critical value of active power calculated based on the rated current carrying capacity of the feeder. The rated current carrying capacity refers to the maximum allowable current capacity specified in the design of the feeder for long-term stable operation.

[0067] Optionally, the active power threshold of this embodiment of the invention is determined by: obtaining it through a power calculation method based on the rated voltage, rated current carrying capacity and power factor of the feeder. The power calculation method is that the active power is equal to the rated voltage multiplied by the rated current carrying capacity multiplied by the power factor.

[0068] Optionally, the reverse overload condition refers to the maximum absolute value of the active power of the first feeder being greater than or equal to the active power threshold. The total duration refers to the sum of the durations of all continuous or discontinuous time intervals within the current period where the load flow direction is reversed and the absolute value of the active power is not less than the active power threshold. The total duration of the current period refers to the overall length of the current analysis period. The duration ratio refers to the ratio of the total duration to the total duration of the current period.

[0069] Furthermore, the feeder timing prediction device compares the absolute value of the maximum active power of each first feeder with the corresponding active power threshold to determine whether the absolute value of the maximum active power is greater than or equal to the active power threshold. Further, the feeder timing prediction device identifies the first feeder that satisfies the condition of having a maximum active power absolute value greater than or equal to the active power threshold as the second feeder that satisfies the reverse overload condition.

[0070] Furthermore, for each second feeder, the feeder timing prediction device re-traverses its load flow direction sequence and load power sequence, identifies all time points that simultaneously satisfy "the load flow direction is reversed" and "the absolute value of active power is not less than the active power threshold", integrates the time intervals corresponding to these time points, and calculates the total duration of all time intervals that meet the conditions. This total duration is the total duration.

[0071] Furthermore, the feeder timing prediction device calculates the duration ratio, which is calculated as follows: the duration ratio equals the total duration divided by the total duration of the current time period.

[0072] In one embodiment, the first feeders are target feeder 2 and target feeder 3; the maximum active power absolute value of target feeder 2 is 86 kW, and the maximum active power absolute value of target feeder 3 is 81 kW; the preset active power threshold of target feeder 2 is 80 kW, and the active power threshold of target feeder 3 is 75 kW; the current time period is 14:00-14:20, and the total duration of the current time period is 20 minutes; the load flow direction sequence and load power sequence of each second feeder remain unchanged.

[0073] The feeder timing prediction device makes a comparison and judgment: the target feeder 2's 86 kW is greater than 80 kW, which meets the reverse heavy load condition and is determined to be the second feeder; the target feeder 3's 81 kW is greater than 75 kW, which meets the reverse heavy load condition and is determined to be the second feeder.

[0074] For target feeder 2, traverse its load flow direction sequence and load power sequence to identify time points that meet the conditions: 14:10 (reverse, active power 85 kW, absolute value 85 kW > 80 kW) and 14:15 (reverse, active power 86 kW, absolute value 86 kW > 80 kW). The time intervals corresponding to the two time points are both 5 minutes (the preset time interval is 5 minutes), and the total duration is 5 minutes + 5 minutes = 10 minutes; the duration ratio = 10 minutes / 20 minutes = 0.50.

[0075] For target feeder 3, traverse its load flow direction sequence and load power sequence to identify time points that meet the conditions: 14:05 (reverse, active power 79 kW, absolute value 79 kW > 75 kW) and 14:15 (reverse, active power 81 kW, absolute value 81 kW > 75 kW). The time intervals corresponding to the two time points are both 5 minutes, and the total duration is 5 + 5 minutes = 10 minutes; the duration ratio = 10 / 20 = 0.50.

[0076] Step 204: The second feeder with a total duration not less than the preset minimum duration threshold and a duration ratio not less than the preset time ratio threshold is identified as the reverse overload feeder, and the feeders in the target feeders other than the reverse overload feeder are identified as non-reverse overload feeders.

[0077] Optionally, the feeder timing prediction device acquires the total duration and duration ratio of each second feeder, as well as preset minimum duration thresholds and preset time percentage thresholds. The preset minimum duration threshold, in minutes, is the minimum duration standard set in advance according to feeder operation safety requirements for determining whether a reverse overload has occurred; the preset time percentage threshold, in decimal form, is the minimum time percentage standard set in advance according to feeder operation safety requirements for determining whether a reverse overload has occurred.

[0078] Furthermore, the feeder timing prediction device performs a dual comparison of the total duration of each second feeder with the preset minimum duration threshold and the duration ratio with the preset time proportion threshold to determine whether the two conditions of "total duration is not less than the preset minimum duration threshold" and "duration ratio is not less than the preset time proportion threshold" are met simultaneously.

[0079] Furthermore, the feeder timing prediction device identifies the second feeder that simultaneously meets the above two conditions as a reverse overload feeder.

[0080] Furthermore, the feeder timing prediction device sorts through all target feeders and identifies the other feeders, except for the reverse heavy overload feeder, as non-reverse heavy overload feeders.

[0081] In one embodiment, the second feeders are target feeder 2 and target feeder 3; the total duration of target feeder 2 is 10 minutes, and the duration ratio is 0.50; the total duration of target feeder 3 is 10 minutes, and the duration ratio is 0.50; the preset minimum duration threshold is 8 minutes, and the preset time ratio threshold is 0.40; there are 3 initial target feeders (target feeder 1, target feeder 2, and target feeder 3).

[0082] The feeder timing prediction device performs a dual comparison judgment: the 10-minute interval of target feeder 2 is greater than 8 minutes and the 0.50 interval is greater than 0.40. Both conditions are met, and it is determined to be a reverse heavy overload feeder; the 10-minute interval of target feeder 3 is greater than 8 minutes and the 0.50 interval is greater than 0.40. Both conditions are met, and it is determined to be a reverse heavy overload feeder.

[0083] After reviewing all target feeders, except for the reverse heavy overload feeders (target feeder 2 and target feeder 3), the remaining target feeder 1 is determined to be a non-reverse heavy overload feeder.

[0084] The embodiments of the present invention can accurately identify reverse heavy overload feeders and non-reverse heavy overload feeders, providing accurate data for the subsequent construction of the reverse heavy overload impact propagation chain. This enables the dynamic prediction of the disturbance impact of reverse heavy overload on the load evolution path of other feeders based on the reverse heavy overload impact propagation chain, achieving early detection of potential new reverse heavy overload feeders and avoiding scheduling lag and long-term equipment overload risks.

[0085] Optionally, the processes of steps 301 to 305 include: Step 301: Pair the load power sequence of each feeder pair in the preset continuous time period point by point according to the same time point alignment relationship to obtain the synchronous time sequence pair of each feeder pair.

[0086] Optionally, the feeder timing prediction device acquires the identification information of all non-reverse heavy overload feeders and reverse heavy overload feeders, as well as the load power sequence of each feeder within a preset continuous time period. The identification information refers to information used to uniquely distinguish different feeders, such as the feeder number.

[0087] Optionally, the same time point alignment relationship refers to matching the sampling times of the two load power sequences in the feeder pair one-to-one to ensure that the two paired load power values ​​correspond to the same time point. The synchronous time series pair refers to the paired sequence formed by combining the two load power sequences in the feeder pair point by point after aligning them at the same time point. Each element in the sequence is a combination of the non-reverse heavy overload feeder load power value and the reverse heavy overload feeder load power value corresponding to the same time point.

[0088] Furthermore, the feeder timing prediction device constructs all possible feeder pairs based on the identification information. The construction rule is that each feeder pair contains only one non-reverse heavy overload feeder and one reverse heavy overload feeder, and feeder pairs containing the same two feeders are not constructed repeatedly.

[0089] Furthermore, for each feeder pair, the feeder timing prediction device extracts the load power sequence of the non-reverse heavy overload feeder and the load power sequence of the reverse heavy overload feeder, and checks whether the sampling times of the two sequences are completely consistent. If there are missing sampling times, the load power values ​​of adjacent valid sampling times are used for interpolation to supplement them, ensuring that the number of sampling times and the specific time points of the two sequences are completely matched.

[0090] Furthermore, the feeder timing prediction device pairs the load power values ​​corresponding to the same time point in the two sequences in chronological order to form a synchronous time sequence pair, wherein the length of the synchronous time sequence pair is consistent with the number of sampling times of the two load power sequences.

[0091] In one embodiment, the non-reverse heavy overload feeder is target feeder 1, and the reverse heavy overload feeders are target feeder 2 and target feeder 3. The preset continuous time period is set to 14:00-14:20, and the preset time interval is 5 minutes. Therefore, the sampling times within the preset continuous time period are 14:00, 14:05, 14:10, 14:15, and 14:20, for a total of 5 sampling times. The total duration of the current time period is 20 minutes. The load power sequence of each feeder within the preset continuous time period has been obtained by step 10, as follows: the load power sequence of target feeder 1 is 80 kW, 81 kW, 82 kW, 83 kW, and 84 kW; the load power sequence of target feeder 2 is 82 kW, 81 kW, 85 kW, 86 kW, and 83 kW; and the load power sequence of target feeder 3 is 78 kW, 79 kW, 80 kW, 81 kW, and 82 kW.

[0092] The feeder timing prediction device first constructs feeder pairs: according to the construction rules, two feeder pairs are formed, namely feeder pair 1 (target feeder 1 + target feeder 2) and feeder pair 2 (target feeder 1 + target feeder 3).

[0093] The sampling times of the load power sequence for each feeder pair were checked and found to be the five time points mentioned above, with no missing times, requiring no interpolation. Then, the feeder pairs were matched point-by-point according to the same time points: the synchronization time sequence pairs for feeder pair 1 were (80 kW, 82 kW), (81 kW, 81 kW), (82 kW, 85 kW), (83 kW, 86 kW), and (84 kW, 83 kW); the synchronization time sequence pairs for feeder pair 2 were (80 kW, 78 kW), (81 kW, 79 kW), (82 kW, 80 kW), (83 kW, 81 kW), and (84 kW, 82 kW).

[0094] Step 302: For the synchronization time series pair of each feeder pair, determine the direction consistency based on the first load power change direction of the non-reverse heavy overload feeder and the second load power change direction of the reverse heavy overload feeder within the same time interval, and determine the direction consistency determination result.

[0095] Optionally, the load power change direction represents the trend of increase or decrease of the load power value at the current time point relative to the load power value at the previous time point; the first load power change direction refers to the load power change direction of the non-reverse heavy overload feeder in the feeder pair; the second load power change direction refers to the load power change direction of the reverse heavy overload feeder in the feeder pair; the same time interval refers to selecting the same time interval composed of two adjacent consecutive sampling times for two load power sequences, that is, for two sequences, the nth time interval is the interval composed of the nth sampling time and the (n+1)th sampling time (n is a positive integer starting from 1); the direction consistency judgment refers to comparing whether the first load power change direction and the second load power change direction are the same within the same time interval; the direction consistency judgment result refers to the judgment result for each same time interval.

[0096] Furthermore, for each feeder pair, the feeder timing prediction device extracts the load power sequence of the non-reverse heavy overload feeder and the load power sequence of the reverse heavy overload feeder from its synchronous time series pair, and sequentially traverses each time interval to calculate the load power change direction of the two sequences within each time interval.

[0097] Optionally, the calculation and judgment rules for the load power change direction in this embodiment of the invention are as follows: if the load power value at the current time point is greater than the load power value at the previous time point, then the load power change direction is "increase"; if the load power value at the current time point is less than the load power value at the previous time point, then the load power change direction is "decrease"; if the load power value at the current time point is equal to the load power value at the previous time point, then the load power change direction is "unchanged".

[0098] Furthermore, the feeder timing prediction device compares the direction of change of the first load power and the direction of change of the second load power within the same time interval: If both are "increase", both are "decrease", or both are "unchanged", then the result of the direction consistency determination for the time interval is "consistent"; if the two change in different directions (such as one "increase" and the other "decrease", one "increase" and the other "unchanged", one "decrease" and the other "unchanged"), then the result of the direction consistency determination for the time interval is "inconsistent".

[0099] Furthermore, the feeder timing prediction device records the direction consistency determination results of each feeder pair across all time intervals in chronological order, thereby obtaining the direction consistency determination result of the feeder pair.

[0100] In one embodiment, feeder pair 1 is (target feeder 1 + target feeder 2), and its synchronous time series pairs correspond to the following two load power sequences: target feeder 1: 80 kW, 81 kW, 82 kW, 83 kW, 84 kW; target feeder 2: 82 kW, 81 kW, 85 kW, 86 kW, 83 kW; feeder pair 2 is (target feeder 1 + target feeder 3), and its two load power sequences are: target feeder 1: 80 kW, 81 kW, 82 kW, 83 kW, 84 kW; target feeder 2: 82 kW, 81 kW, 85 kW, 86 kW, 83 kW; and feeder pair 2 is (target feeder 1 + target feeder 3), and its two load power sequences are: target feeder 1: 80 kW, 81 kW, 82 kW, 83 kW, 84 kW; target feeder 2: 82 kW, 81 kW, 85 kW, 86 kW, 83 kW. 1 kW, 82 kW, 83 kW, 84 kW; Target feeder 3: 78 kW, 79 kW, 80 kW, 81 kW, 82 kW; The preset sampling time within the continuous period is 5 times, so the number of time intervals is 4, namely interval 1 (14:00-14:05), interval 2 (14:05-14:10), interval 3 (14:10-14:15), and interval 4 (14:15-14:20).

[0101] For feeder pair 1, the direction consistency is determined by traversing each time interval: In interval 1, the power of target feeder 1 increases from 80 kW to 81 kW (increase), while the power of target feeder 2 decreases from 82 kW to 81 kW (decrease). The directions of change are different, so the determination result is "inconsistent"; In interval 2, target feeder 1 increases from 81 kW to 82 kW (increase), while target feeder 2 increases from 81 kW to 85 kW (increase). The directions of change are the same, so the determination result is "consistent"; In interval 3, target feeder 1 increases from 82 kW to 83 kW (increase), while target feeder 2 increases from 85 kW to 86 kW (increase). The directions of change are the same, so the determination result is "consistent"; In interval 4, target feeder 1 increases from 83 kW to 84 kW (increase), while target feeder 2 decreases from 86 kW to 83 kW (decrease). The directions of change are different, so the determination result is "inconsistent". The direction consistency determination result for feeder pair 1 is [inconsistent, consistent, consistent, inconsistent].

[0102] For feeder pair 2, the directional consistency is determined by iterating through each time interval: In interval 1, target feeder 1 increases from 80 kW to 81 kW (an increase), and target feeder 3 increases from 78 kW to 79 kW (an increase), both are determined to be "consistent"; In interval 2, target feeder 1 increases from 81 kW to 82 kW (an increase), and target feeder 3 increases from 79 kW to 80 kW (an increase), both are determined to be "consistent"; In interval 3, target feeder 1 increases from 82 kW to 83 kW (an increase), and target feeder 3 increases from 80 kW to 81 kW (an increase), both are determined to be "consistent"; In interval 4, target feeder 1 increases from 83 kW to 84 kW (an increase), and target feeder 3 increases from 81 kW to 82 kW (an increase), both are determined to be "consistent". Therefore, the set of directional consistency determination results for feeder pair 2 is [consistent, consistent, consistent, consistent].

[0103] Step 303: Based on the direction consistency determination result of each feeder pair, determine the total number of time intervals in which the change direction of each feeder pair is consistent within a preset continuous period, and based on the ratio of the total number of time intervals to the total number of time intervals within the preset continuous period, determine the load change direction synchronization rate of each feeder pair.

[0104] Optionally, the feeder timing prediction device acquires a set of direction consistency determination results for each feeder pair, as well as the total number of time intervals within a preset continuous period. The total number of time intervals with consistent direction of change refers to the number of time intervals in the direction consistency determination result set that are judged as "consistent"; the total number of time intervals refers to the total number of time intervals contained within the preset continuous period, calculated as the total number of sampling times within the preset continuous period minus 1; the load change direction synchronization rate refers to the ratio of the total number of time intervals with consistent direction of change to the total number of time intervals, used to quantitatively characterize the degree of synchronization of the load change directions of the two feeders in the feeder pair, with a value ranging from 0 to 1, the closer the value is to 1, the higher the degree of synchronization.

[0105] Furthermore, the feeder timing prediction device iterates through and counts the set of direction consistency determination results for each feeder pair, counts the number of time intervals with a determination result of "consistent", and obtains the total number of time intervals with consistent change direction for that feeder pair.

[0106] Furthermore, the feeder timing prediction device confirms the total number of time intervals within the corresponding preset continuous period of the feeder pair and calculates the load change direction synchronization rate. The calculation method is as follows: the load change direction synchronization rate is equal to the total number of time intervals with the same change direction divided by the total number of time intervals.

[0107] In one embodiment, the set of direction consistency determination results for feeder pair 1 is [inconsistent, consistent, consistent, inconsistent], and the set of direction consistency determination results for feeder pair 2 is [consistent, consistent, consistent, consistent]; the preset total number of sampling times within a continuous time period is 5, so the total number of time intervals is 5-1=4.

[0108] For feeder pair 1, the total number of time intervals with consistent load change direction is calculated: intervals 2 and 3 are judged as "consistent," totaling 2 intervals. Therefore, the total number of time intervals with consistent load change direction is 2. The load change direction synchronization rate is calculated as: 2 / 4 = 0.5. For feeder pair 2, the total number of time intervals with consistent load change direction is calculated: all 4 intervals are judged as "consistent," therefore, the total number of time intervals with consistent load change direction is 4. The load change direction synchronization rate is calculated as: 4 / 4 = 1.

[0109] Step 304: If the load change direction synchronization rate is greater than or equal to the preset synchronization rate threshold, then the load change direction synchronization rate is determined as the load coupling strength of each feeder pair; otherwise, the load coupling strength is determined to be zero.

[0110] Optionally, the feeder timing prediction device acquires the load change direction synchronization rate of each feeder pair, as well as a preset synchronization rate threshold. The preset synchronization rate threshold is a critical value pre-set based on feeder operating characteristics and load correlation analysis requirements, used to determine whether there is a significant load coupling relationship between the two feeders in the feeder pair; its value ranges from 0 to 1. The load coupling strength is a quantitative indicator characterizing the degree of load correlation between the reverse overloaded feeder and the non-reverse overloaded feeder in the feeder pair. Its value is either the load change direction synchronization rate or 0. A higher value indicates a closer load correlation between the two, and a greater potential impact of the reverse overloaded feeder on the non-reverse overloaded feeder.

[0111] Optionally, the feeder timing prediction device compares the load change direction synchronization rate of each feeder pair with a preset synchronization rate threshold. Further, the feeder timing prediction device determines the load coupling strength based on the comparison results: if the load change direction synchronization rate of the feeder pair is greater than or equal to the preset synchronization rate threshold, it indicates a significant load coupling relationship between the two feeders, and the load change direction synchronization rate is directly determined as the load coupling strength of the feeder pair; if the load change direction synchronization rate of the feeder pair is less than the preset synchronization rate threshold, it indicates that the load coupling relationship between the two feeders is not significant, and the potential impact of the reverse overloaded feeder on the non-reverse overloaded feeder is negligible, and the load coupling strength of the feeder pair is determined to be 0.

[0112] In one embodiment, the load change direction synchronization rate of feeder pair 1 is 0.5, and the load change direction synchronization rate of feeder pair 2 is 1. Based on the feeder operating characteristics and load correlation analysis requirements, the preset synchronization rate threshold is set to 0.6. The feeder timing prediction device makes a comparison judgment: the synchronization rate of feeder pair 1 (0.50) is less than the preset synchronization rate threshold (0.60), therefore the load coupling strength of feeder pair 1 is determined to be 0; the synchronization rate of feeder pair 2 (1) is greater than the preset synchronization rate threshold (0.60), therefore the load coupling strength of feeder pair 2 is determined to be 1. The final set of load coupling strengths is [feeder pair 1: 0; feeder pair 2: 1].

[0113] Step 305: Based on the load coupling strength of each feeder pair, analyze the potential evolution path of each non-reverse heavy overload feeder affected by reverse heavy overload in the next prediction period, and construct the reverse heavy overload influence propagation chain.

[0114] Optionally, the potential evolution path refers to the direction and process of the load state of a non-reverse heavy overload feeder changing under the influence of a reverse heavy overload feeder. The feeder timing prediction device analyzes the potential evolution path of each non-reverse heavy overload feeder affected by the reverse heavy overload in the next prediction period based on the load coupling strength of each feeder pair, and constructs the reverse heavy overload influence propagation chain, as detailed in steps 3051 to 3054.

[0115] The embodiments of the present invention can construct a reverse overload impact propagation chain based on the potential evolution path of non-reverse overload feeders affected by reverse overload. This reverse overload impact propagation chain clearly characterizes the propagation trajectory and degree of impact of reverse overload in the feeder network. Therefore, the disturbance impact of reverse overload on the load evolution path of other feeders can be dynamically predicted based on the reverse overload impact propagation chain, enabling early detection of potential new reverse overload feeders and avoiding scheduling delays and long-term equipment overload risks.

[0116] Optionally, the processes of steps 3051 to 3054 include: Step 3051: Iterate through the load coupling strength of each feeder pair, identify the feeder pairs with load coupling strength greater than zero as load-coupled feeder pairs, and determine the unidirectional coupling connection relationship with the reverse heavy overload feeder as the source end and the non-reverse heavy overload feeder as the receiving end based on the pairing structure between the reverse heavy overload feeder and the non-reverse heavy overload feeder in each load-coupled feeder pair.

[0117] Optionally, the feeder timing prediction device acquires the load coupling strength of all feeder pairs, as well as the composition information of each feeder pair (i.e., the non-reverse heavy overload feeder identifier and the reverse heavy overload feeder identifier included in each feeder pair). Here, a load-coupled feeder pair refers to a feeder pair with a load coupling strength greater than zero, indicating a significant load coupling relationship between the reverse heavy overload feeder and the non-reverse heavy overload feeder in this type of feeder pair, where the reverse heavy overload feeder may affect the non-reverse heavy overload feeder.

[0118] Among them, the unidirectional coupling connection relationship refers to the relationship in which the influence is transmitted only from the reverse heavy overload feeder to the non-reverse heavy overload feeder, and does not include the reverse influence transmission from the non-reverse heavy overload feeder to the reverse heavy overload feeder.

[0119] Optionally, the source end refers to the initiating end that affects the transmission, which is the reverse heavy overload feeder in this case; the receiving end refers to the receiving end that affects the transmission, which is the non-reverse heavy overload feeder in this case; the feeder identifier refers to information used to uniquely distinguish different feeders, such as the feeder number.

[0120] Furthermore, the feeder timing prediction device iterates through the load coupling strength of each feeder pair one by one to determine whether its load coupling strength is greater than zero.

[0121] Furthermore, the feeder timing prediction device identifies feeder pairs with load coupling strength greater than zero as load-coupled feeder pairs and records the composition information of each load-coupled feeder pair.

[0122] Furthermore, based on the composition information of each load-coupled feeder pair, its pairing structure (i.e., the correspondence between the reverse heavy overload feeder and the non-reverse heavy overload feeder) is clarified, and the unidirectional coupling connection relationship is determined accordingly: the reverse heavy overload feeder in the load-coupled feeder pair is set as the source end, and the non-reverse heavy overload feeder is set as the receiving end, forming a unidirectional coupling connection relationship of "source end (reverse heavy overload feeder) → receiving end (non-reverse heavy overload feeder)", and the load coupling strength corresponding to this unidirectional coupling connection relationship is recorded at the same time.

[0123] In one embodiment, all feeder pairs include feeder pair 1 (target feeder 1 + target feeder 2) and feeder pair 2 (target feeder 1 + target feeder 3); the load coupling strength of feeder pair 1 is 0, and the load coupling strength of feeder pair 2 is 1; the type corresponding to each feeder identifier is: target feeder 1 is a non-reverse heavy overload feeder, and target feeder 2 and target feeder 3 are reverse heavy overload feeders.

[0124] The feeder timing prediction device iterates through the load coupling strength of each feeder pair: the load coupling strength of feeder pair 1 is not greater than 0 and is not included in the load coupling feeder pair; the load coupling strength of feeder pair 2 is greater than 0 and is determined to be a load coupling feeder pair.

[0125] Based on the composition information of feeder pair 2 (non-reverse heavy overload feeder: target feeder 1; reverse heavy overload feeder: target feeder 3), the unidirectional coupling connection relationship is determined to be "source end (target feeder 3) → receiver end (target feeder 1)", and the corresponding load coupling strength is 1.

[0126] Step 3052: Based on the operating status characteristics of the reverse overload feeder in each unidirectional coupling connection during the current time period, determine the initial action mode in which the reverse overload feeder affects the non-reverse overload feeder coupled with it.

[0127] Optionally, the feeder timing prediction device acquires all unidirectional coupling connections and the operating status characteristics of the source end (reverse heavy overload feeder) in each unidirectional coupling connection during the current time period. The operating status characteristics characterize the load power change trend direction and load flow direction stability of the reverse heavy overload feeder during the current time period. The load power change trend direction refers to the overall increase or decrease direction of the load power of the reverse heavy overload feeder during the current time period, rather than the local change direction of a single time interval. The stability of the load flow direction refers to the degree to which the load flow direction of the reverse heavy overload feeder remains reversed during the current time period. The initial action mode refers to the initial way in which the reverse heavy overload feeder exerts its influence on the non-reverse heavy overload feeder; its type is determined by the operating status characteristics, and may include, for example, "reverse stability - power increase influence mode," "reverse stability - power decrease influence mode," and "reverse instability - power fluctuation influence mode."

[0128] Furthermore, the feeder timing prediction device analyzes the operating state characteristics of each reverse overloaded feeder, specifically including two dimensions: First, determine the direction of load power change trend. The analysis method in this embodiment of the invention is as follows: count the load power change direction of the reverse overload feeder in all time intervals within the current period. If the number of time intervals in the "increasing" direction accounts for more than 50% of the total number of time intervals, then the overall load power change trend direction is "increasing"; if the number of time intervals in the "decreasing" direction accounts for more than 50% of the total number of time intervals, then the overall load power change trend direction is "decreasing"; if the proportion of time intervals in both the "increasing" and "decreasing" directions is not greater than 50%, then the overall load power change trend direction is "fluctuating".

[0129] Second, the stability of the load flow direction is determined. The determination method in this embodiment of the invention is as follows: the proportion of the number of time points in which the load flow direction of the reverse overload feeder is reversed in the current time period to the total number of time points. If the proportion is greater than or equal to the preset power flow stability threshold (the preset power flow stability threshold is a critical value set in advance according to the feeder operating characteristics, such as 80%), then the stability of the load flow direction is "stable"; if the proportion is less than the preset power flow stability threshold, then the stability of the load flow direction is "unstable".

[0130] Furthermore, based on the analysis results of the above two dimensions, the feeder timing prediction device determines the initial operating mode: if the load flow direction stability is "stable" and the load power change trend is "increasing", then the initial operating mode is "reverse stability - increasing power influence mode"; if the load flow direction stability is "stable" and the load power change trend is "decreasing", then the initial operating mode is "reverse stability - decreasing power influence mode"; if the load flow direction stability is "unstable", regardless of the load power change trend, the initial operating mode is "reverse instability - power fluctuation influence mode".

[0131] In one embodiment, the unidirectional coupling connection relationship is "source end (target feeder 3) → receiver end (target feeder 1)"; the current time period is 14:00-14:20, the total number of time points is 5, and the total number of time intervals is 4; the load power sequence of target feeder 3 (reverse heavy overload feeder) in the current time period is 78 kW, 79 kW, 80 kW, 81 kW, 82 kW, and the load flow direction sequence is forward, reverse, forward, reverse, forward; the preset power flow stability threshold is set to 80%.

[0132] Furthermore, the feeder timing prediction device analyzes the operating status characteristics of the target feeder 3: First, determine the direction of load power change trend by statistically analyzing the power change direction for each time interval: 14:00-14:05 (78→79, increase), 14:05-14:10 (79→80, increase), 14:10-14:15 (80→81, increase), and 14:15-14:20 (81→82, increase). All four time intervals show an "increase" direction, accounting for 100% (greater than 50%). Therefore, the load power change trend is "increasing".

[0133] Second, the stability of the load flow direction is determined by counting the number of time points where the load flow direction is reversed: 14:05 and 14:15, a total of 2, accounting for 2 / 5 = 40% of the total number of time points. 40% is less than the preset load flow stability threshold of 80%, so the stability of the load flow direction is "unstable".

[0134] Based on the analysis results, the initial operating mode of target feeder 3 is "reverse instability-power fluctuation influence mode", that is, the initial operating mode in which target feeder 3 exerts influence on target feeder 1 is this mode.

[0135] Step 3053: Based on the unidirectional coupling connection relationship between each non-reverse heavy overload feeder and all its coupled reverse heavy overload feeders and the corresponding initial operating mode, determine the set of multi-source influence inputs that the non-reverse heavy overload feeder will be affected by in the next prediction period.

[0136] Optionally, the feeder timing prediction device acquires all unidirectional coupling connections, the initial operating mode corresponding to each unidirectional coupling connection, and the identification information of all non-reverse overload feeders.

[0137] Among them, the multi-source influence input set refers to the set of information related to the influence of multiple different reverse heavy overload feeders on a single non-reverse heavy overload feeder at the same time. Each element in the set corresponds to the influence information exerted on it by a reverse heavy overload feeder, including the identifier of the reverse heavy overload feeder, the corresponding unidirectional coupling connection relationship and the initial action mode.

[0138] Optionally, multi-source impact refers to a single non-reverse heavy overload feeder being simultaneously affected by two or more reverse heavy overload feeders.

[0139] Furthermore, the feeder timing prediction device uses non-reverse heavy overload feeders as the classification basis to group all unidirectional coupling connection relationships and their corresponding initial operating modes: all unidirectional coupling connection relationships with the same non-reverse heavy overload feeder at the receiving end, as well as the initial operating mode and source end (reverse heavy overload feeder) identifier corresponding to each unidirectional coupling connection relationship are grouped together.

[0140] Furthermore, it is determined whether the group corresponding to each non-reverse overload feeder contains two or more unidirectional coupling connections: if it contains two or more, the non-reverse overload feeder has multi-source influence, and all information in the group (reverse overload feeder identifier, unidirectional coupling connection, initial operating mode) is integrated to form the multi-source influence input set of the non-reverse overload feeder; if it contains only one unidirectional coupling connection, the non-reverse overload feeder has single-source influence, and the information in the group is still integrated to form the multi-source influence input set (the set contains only one influence element); if it does not contain any unidirectional coupling connection, the non-reverse overload feeder has no reverse overload influence, and its multi-source influence input set is an empty set.

[0141] In one embodiment, the non-reverse heavy overload feeder includes a target feeder 1, and a new non-reverse heavy overload feeder target feeder 4 is added; the unidirectional coupling connection relationships include: relationship 1 (source end: target feeder 3 → receiver end: target feeder 1), relationship 2 (source end: target feeder 2 → receiver end: target feeder 1), and relationship 3 (source end: target feeder 2 → receiver end: target feeder 4); the initial operating modes corresponding to each relationship are: relationship 1 corresponds to "reverse instability - power fluctuation influence mode", relationship 2 corresponds to "reverse stability - power increase influence mode", and relationship 3 corresponds to "reverse stability - power decrease influence mode".

[0142] Furthermore, the feeder timing prediction device groups feeders based on non-reverse overload feeders: the group corresponding to target feeder 1 includes relation 1, relation 2, and the corresponding initial operating mode and source end identifier (target feeder 3, target feeder 2); the group corresponding to target feeder 4 includes relation 3, and the corresponding initial operating mode and source end identifier (target feeder 2).

[0143] Integrating a set of multi-source influence inputs: The multi-source influence input set of target feeder 1 is {[Source end: target feeder 3, connection relationship: relationship 1, initial action mode: reverse instability - power fluctuation influence mode], [Source end: target feeder 2, connection relationship: relationship 2, initial action mode: reverse stability - power increasing influence mode]}; the multi-source influence input set of target feeder 4 is {[Source end: target feeder 2, connection relationship: relationship 3, initial action mode: reverse stability - power decreasing influence mode]}.

[0144] Step 3054: Based on the persistence characteristics of each reverse overload feeder in the multi-source influence input set, construct the reverse overload influence propagation chain.

[0145] Optionally, the persistence characteristic characterizes the monotonic continuation of the load power sequence and the consistent continuation of the load flow direction sequence of the reverse overload feeder within a preset continuous period. The monotonic continuation of the load power sequence refers to the degree to which the load power change trend of the reverse overload feeder remains unchanged within the preset continuous period. The consistent continuation of the load flow direction sequence refers to the degree to which the load flow direction of the reverse overload feeder remains reversed within the preset continuous period. Optionally, the feeder timing prediction device constructs the reverse overload influence propagation chain based on the persistence characteristics of each reverse overload feeder in the multi-source influence input set, specifically as described in steps 30541 to 30544.

[0146] The embodiments of the present invention can construct a reverse overload impact propagation chain that accurately reflects the propagation trajectory, degree of impact, and sequence of impact. Therefore, based on the reverse overload impact propagation chain, the disturbance impact of reverse overload on the load evolution path of other feeders can be dynamically predicted, realizing the early detection of potential new reverse overload feeders and avoiding scheduling lag and long-term equipment overload risks.

[0147] Optionally, the processes of steps 30541 to 30544 include: Step 30541: Based on the persistence characteristics of each reverse overload feeder in the multi-source influence input set, analyze the persistence influence of each reverse overload feeder on the non-reverse overload feeder to obtain the persistence judgment result.

[0148] Optionally, the feeder timing prediction device acquires the multi-source influence input set for each non-reverse heavy overload feeder, as well as the persistence characteristics of each reverse heavy overload feeder within the set. Here, persistence influence refers to the ability of the reverse heavy overload feeder to extend its influence on the non-reverse heavy overload feeder from the current time period to the next prediction time period.

[0149] Optionally, the continuity judgment result refers to the result of judging whether each reverse heavy overload feeder has continuity in relation to the non-reverse heavy overload feeder, including two cases: "has continuity" and "does not have continuity"; the next forecast period refers to the time interval after the current period that needs to be forecasted for the load, and its duration is consistent with the duration of the current period.

[0150] Furthermore, the feeder timing prediction device decomposes and analyzes the continuity characteristics of each reverse overload feeder, breaking it down into two dimensions: the monotonic continuity of the load power sequence and the consistent continuity of the load flow direction sequence. Corresponding continuity judgment thresholds are set for each dimension: First, a monotonic continuity threshold is set, a critical proportion, pre-defined based on the feeder load change characteristics, to determine whether the load power change trend direction continues. For example, 80% represents the minimum proportion of time intervals in the last N time intervals (N is a preset number, such as 3) within a preset continuous period that the reverse overload feeder must maintain the same direction as the overall load power change trend. Second, a consistent continuity threshold is set, a critical proportion, pre-defined based on the feeder flow operation characteristics, to determine whether the load flow direction continues in the reverse direction. For example, 80% represents the minimum proportion of time points in the last M time points (M is a preset number, such as 3) within a preset continuous period where the load flow direction is reversed.

[0151] Furthermore, the feeder timing prediction device performs continuity determination on two dimensions respectively: For the monotonicity of the load power sequence, the number of time intervals in the last N time intervals of the reverse overload feeder that are consistent with the overall load power change trend is counted, and the proportion is calculated. If the proportion is greater than or equal to the monotonicity continuity threshold, the dimension is determined to "satisfy continuity"; if the proportion is less than the monotonicity continuity threshold, the dimension is determined to "not satisfy continuity".

[0152] For the consistency of load flow direction sequence, the number of time points in the last M time points of the reverse heavy overload feeder where the load flow direction is reversed within the preset continuous period is counted, and the proportion is calculated. If the proportion is greater than or equal to the consistency continuity threshold, the dimension is determined to "satisfy continuity"; if the proportion is less than the consistency continuity threshold, the dimension is determined to "not satisfy continuity".

[0153] Furthermore, the feeder timing prediction device performs a joint determination: only when both the monotonicity of the load power sequence and the consistency of the load flow direction sequence of the reverse overload feeder "satisfy continuity" are the influence of the reverse overload feeder on the non-reverse overload feeder determined to "have continuity"; if either dimension "does not satisfy continuity," it is determined to "not have continuity."

[0154] In one embodiment, the non-reverse heavy overload feeders are target feeder 1 and target feeder 4; the multi-source influence input set of target feeder 1 includes two influence elements, namely source-end target feeder 2 (reverse heavy overload feeder) and source-end target feeder 3 (reverse heavy overload feeder); the preset continuous time period is 14:00-14:20, the number of time points is 5 (14:00, 14:05, 14:10, 14:15, 14:20), and the number of time intervals is 4 (14:00-14:05, 14:05-14:10, 14:10-14:15, 14:15-14:20); N=3 (last 3 time intervals), monotonicity continuation threshold=80%, M=3 (last 3 time points), and consistency continuation threshold=80%.

[0155] For target feeder 2 (reverse heavy overload feeder): Its overall load power change trend is known to be "increasing," and the power change direction in the last three time intervals (14:05-14:10, 14:10-14:15, 14:15-14:20) is also "increasing," accounting for 3 / 3 = 100% > 80%, thus satisfying the monotonicity continuity dimension of the load power sequence. Its load flow direction sequence is forward, forward, reverse, reverse, forward, with two reverse time points in the last three time points (14:10, 14:15, 14:20), accounting for 2 / 3 ≈ 66.67% < 80%, thus not satisfying the consistency continuity dimension of the load flow direction sequence. Joint judgment: The influence of target feeder 2 on target feeder 1 "does not have continuity."

[0156] For target feeder 3 (reverse heavy overload feeder): Its overall load power change trend is known to be "increasing," with the power change direction in the last three time intervals all showing an "increasing" trend, accounting for 100% > 80%. The monotonicity of the load power sequence "satisfies continuity." Its load flow direction sequence is forward, reverse, forward, reverse, forward, with only one reverse time point in the last three time points, accounting for 1 / 3 ≈ 33.33% < 80%. The consistency of the load flow direction sequence "does not satisfy continuity." Joint judgment: The impact of target feeder 3 on target feeder 1 "does not possess continuity."

[0157] Step 30542: Based on the continuity judgment result being an input with continuity, and combining the load power sequence and load flow direction sequence of the corresponding non-reverse heavy overload feeder in the current period, determine the power flow disturbance response characteristics of the non-reverse heavy overload feeder in the next prediction period.

[0158] Optionally, the feeder timing prediction device acquires the continuity judgment result of each non-reverse heavy overload feeder, filters out the influence inputs that "have continuity" (i.e. the influence of the corresponding reverse heavy overload feeder), and acquires the load power sequence and load flow direction sequence of the non-reverse heavy overload feeder in the current time period.

[0159] Among them, the power flow disturbance response characteristics refer to the possible changes in load power and load flow direction of a non-reverse heavy overload feeder in the next forecast period when it is affected by a continuous reverse heavy overload. This includes three core dimensions: load power response amplitude, load power response direction, and load flow direction response trend. The load power response amplitude refers to the range of possible changes in load power of the non-reverse heavy overload feeder in the next forecast period. The load power response direction refers to the direction of increase or decrease in load power of the non-reverse heavy overload feeder in the next forecast period. The load flow direction response trend refers to whether the load flow direction of the non-reverse heavy overload feeder tends to reverse in the next forecast period.

[0160] Furthermore, the feeder timing prediction device calculates the load power response amplitude: based on the load power sequence of the non-reverse heavy overload feeder in the current period, it calculates the average change in load power within a preset continuous period (the calculation method is: subtract the load power value of the first time point from the load power value of the last time point, and then divide by the total number of time intervals); at the same time, it calculates the average change in load power of the reverse heavy overload feeder in the current period with continuity; the average changes of the two are added together to obtain the reference value of the load power response amplitude, and then the fluctuation range is calculated according to the preset fluctuation coefficient (the preset fluctuation coefficient is a coefficient set in advance according to the fluctuation characteristics of feeder operation, such as 0.1). The final load power response amplitude is the reference value ± (reference value * preset fluctuation coefficient).

[0161] Furthermore, the feeder timing prediction device determines the load power response direction: if the overall load power change trend of the continuous reverse overload feeder is "increasing", then the load power response direction of the non-reverse overload feeder is "increasing"; if the overall load power change trend of the reverse overload feeder is "decreasing", then the load power response direction of the non-reverse overload feeder is "decreasing"; if the overall load power change trend of the reverse overload feeder is "fluctuating", then the overall load power change trend direction of the non-reverse overload feeder in the current period is taken as the load power response direction.

[0162] Furthermore, the feeder timing prediction device analyzes the load flow direction response trend: it statistically analyzes the proportion of reverse time points in the current period load flow direction sequence of the non-reverse heavy overload feeder. If the proportion is greater than 0 and less than the preset reverse trend threshold (the preset reverse trend threshold is a pre-set critical proportion, such as 30%), then the load flow direction response trend is "there is a possibility of a weak reverse change"; if the proportion is equal to 0, then the response trend is "there is no possibility of a reverse change"; if the proportion is greater than or equal to the preset reverse trend threshold, then the response trend is "there is a possibility of a strong reverse change".

[0163] Furthermore, the feeder timing prediction device integrates the results of the above three dimensions to form the power flow disturbance response characteristics of the non-reverse heavy overload feeder in the next prediction period.

[0164] In one embodiment, a new scenario with continuous influence is added: a non-reverse heavy overload feeder target feeder 5, whose multi-source influence input set includes the influence of the source-end target feeder 6 (reverse heavy overload feeder), and the continuity judgment result is "has continuity"; the preset fluctuation coefficient = 0.1, the preset reverse trend threshold = 30%; the load power sequence of the target feeder 5 in the current period (14:00-14:20) is 90 kW, 91 kW, 92 kW, 93 kW, 94 kW, and the load flow direction sequence is all positive; the overall load power change trend direction of the target feeder 6 in the current period is "increasing", and its average load power change = (power at the last time point - power at the first time point) / total number of time intervals = (85-80) / 4 = 1.25 kW / interval.

[0165] Furthermore, the feeder timing prediction device calculates the power flow disturbance response characteristics of the target feeder 5: First, the load power response amplitude: the average change in load power of target feeder 5 during the current period is (94-90) / 4 = 1 kW / interval, the baseline value is 1 + 1.25 = 2.25 kW / interval, and the fluctuation range is 2.25 * 0.1 = 0.225 kW. Therefore, the load power response amplitude is 2.25 ± 0.225 kW (i.e., 2.025 kW to 2.475 kW). Second, the load power response direction: since the trend of target feeder 6 is "increasing", the response direction of target feeder 5 is also "increasing". Third, the load flow direction response trend: the proportion of reverse time points in the current period of target feeder 5 is 0, so the response trend is "no possibility of reverse change". Integrating these characteristics, the load power response amplitude of target feeder 5 is 2.025-2.475 kW, the load power response direction is increasing, and the load flow direction response trend has no possibility of reverse change.

[0166] Step 30543: Based on the power flow disturbance response characteristics, a joint analysis is conducted on the reversal probability of the load power growth direction and the load flow direction to determine whether a non-reverse heavy overload feeder has the potential state transition conditions to evolve into a reverse heavy overload feeder.

[0167] Optionally, the feeder timing prediction device acquires the power flow disturbance response characteristics of each non-reverse heavy overload feeder, extracts the load power response direction (i.e., the load power growth direction) and the load power flow direction response trend (i.e., the possibility of load power flow direction reversal), and simultaneously acquires the active power threshold corresponding to the rated current carrying capacity of the non-reverse heavy overload feeder. Here, the potential state transition condition refers to the preconditions for the non-reverse heavy overload feeder to evolve from the current non-reverse heavy overload state to the reverse heavy overload state in the next prediction period; the load power growth direction is the load power response direction, including three types: increasing, decreasing, and fluctuating; the possibility of load power flow direction reversal is the load power flow direction response trend, including three types: strong reverse reversal possibility, weak reverse reversal possibility, and no reverse reversal possibility; the active power threshold corresponding to the rated current carrying capacity refers to the critical value of active power calculated based on the rated current carrying capacity of the feeder, where the rated current carrying capacity refers to the maximum allowable current capacity specified in the feeder's design for long-term stable operation.

[0168] Optionally, the conditions for determining potential state transitions are set as follows: Condition 1, the load power growth direction is "increasing", and the maximum possible load power value for the next forecast period calculated based on the load power response amplitude in the power flow disturbance response characteristics (i.e., the upper limit of the load power response amplitude plus the load power value at the last time point of the current period) is greater than or equal to the active power threshold; Condition 2, the possibility of load flow direction reversal is "there is a strong possibility of reverse transition" or "there is a weak possibility of reverse transition".

[0169] Optionally, the feeder timing prediction device makes a joint analysis based on the above conditions: if a non-reverse heavy overload feeder simultaneously satisfies both conditions one and two, it is determined that it "has potential state transition conditions"; if it only satisfies one of the conditions or neither of the two conditions is satisfied, it is determined that it "does not have potential state transition conditions".

[0170] In one embodiment, the non-reverse heavy overload feeder is the target feeder 5, whose power flow disturbance response characteristics are a load power response amplitude of 2.025-2.475 kW, an increasing load power response direction, and no possibility of a reverse change in the load flow direction response trend; the load power value at the last time point of the current period (14:20) is 94 kW; the active power threshold of the target feeder 5 is 98 kW; the newly added non-reverse heavy overload feeder target feeder 6 has a power flow disturbance response characteristics of a load power response amplitude of 3.0-3.5 kW, an increasing load power response direction, and a strong possibility of a reverse change in the load flow direction response trend; the load power value at the last time point of the current period is 95 kW, and the active power threshold is 99 kW.

[0171] For target feeder 5: the maximum possible load power value for the next forecast period = 94 + 2.475 = 96.475 kW < 98 kW, which does not meet condition one; the load flow direction response trend has no possibility of reversal, which does not meet condition two. Joint judgment: no potential state transition conditions are met.

[0172] For target feeder 6: the maximum possible load power value for the next forecast period is 95 + 3.5 = 98.5 kW < 99 kW, which does not meet condition one; although condition two is met, because condition one is not met, the combined judgment is: there is no potential state transition condition.

[0173] The newly added non-reverse heavy overload feeder target feeder 7 has the following characteristics: load power response amplitude of 4.0-4.5 kW, increasing load power response direction, and a strong possibility of a reverse reversal in the load power flow direction response trend; the load power value at the last time point of the current period is 96 kW, and the active power threshold is 100 kW. The maximum possible load power value for the next forecast period = 96 + 4.5 = 100.5 kW ≥ 100 kW, satisfying condition one; and also satisfying condition two. Joint judgment: It possesses the potential state transition condition.

[0174] Step 30544: Based on the potential state transition conditions of all non-reverse heavy overload feeders and their unidirectional coupling connection with the upstream reverse heavy overload feeder, take the non-reverse heavy overload feeder with potential state transition conditions as the downstream transmission node, and recursively build a reverse heavy overload influence propagation chain that starts from the reverse heavy overload feeder, is transmitted through the unidirectional coupling connection, and covers all non-reverse heavy overload feeders with potential state transition conditions.

[0175] Optionally, the feeder timing prediction device acquires the potential state transition condition determination results of all non-reverse heavy overload feeders, the unidirectional coupling connection relationship between each non-reverse heavy overload feeder and the reverse heavy overload feeder, and the load coupling strength corresponding to each unidirectional coupling connection relationship. Here, the upstream reverse heavy overload feeder refers to the reverse heavy overload feeder that acts as the source end in the unidirectional coupling connection relationship, i.e., the reverse heavy overload feeder that influences the non-reverse heavy overload feeders; the downstream transmission node refers to a non-reverse heavy overload feeder with potential state transition conditions. This type of feeder may evolve into a reverse heavy overload feeder under the influence of reverse heavy overload, thus becoming a new source of influence transmitting its influence to the next level feeder; the step-by-step recursion refers to starting from the initial reverse heavy overload feeder, first connecting the non-reverse heavy overload feeders directly affected by it and possessing potential state transition conditions, then using these non-reverse heavy overload feeders as new transmission nodes to connect other non-reverse heavy overload feeders affected by them and possessing potential state transition conditions, and so on, until no new transmission nodes can be added.

[0176] Optionally, the feeder timing prediction device performs node layering: the initial reverse overload feeder is classified as the first layer node (source node); the non-reverse overload feeder that has a unidirectional coupling connection with the first layer node and has potential state transition conditions is classified as the second layer node (directly affected node); the non-reverse overload feeder that has a unidirectional coupling connection with the second layer node, has potential state transition conditions, and has not been layered is classified as the third layer node, and so on, to complete the layering of all non-reverse overload feeders with potential state transition conditions.

[0177] Furthermore, the feeder timing prediction device constructs a propagation chain based on the hierarchical results and unidirectional coupling connection relationships: starting from the first-level node, it points to the second-level node through a unidirectional coupling connection relationship, forming the first segment of the propagation chain; the second-level node points to the third-level node through a unidirectional coupling connection relationship, forming the second segment of the propagation chain; the nodes of each level are connected in sequence to ensure that each downstream node is connected to the propagation chain only through the corresponding upstream node, and the load coupling strength corresponding to each segment connection is used as an indicator of the degree of influence and marked on the corresponding connection segment of the propagation chain.

[0178] Furthermore, the feeder timing prediction device verifies the constructed propagation chain: it checks whether all non-reverse heavy overload feeders with potential state transition conditions have been included in the propagation chain, and whether the connection between each node conforms to the unidirectional coupling connection relationship. If there are any nodes that have not been included or connection errors, they are corrected and adjusted in time to finally form a complete reverse heavy overload impact propagation chain.

[0179] In one embodiment, the initial reverse overload feeders (first-layer nodes) are target feeders 8 and 9; the non-reverse overload feeders with potential state transition conditions are target feeders 7, 10, and 11; the unidirectional coupling connection relationships and load coupling strengths are: target feeder 8 → target feeder 7 (coupling strength 0.9), target feeder 8 → target feeder 10 (coupling strength 0.7), target feeder 7 → target feeder 11 (coupling strength 0.8), and target feeder 9 → target feeder 10 (coupling strength 0.6); other non-reverse overload feeders do not have potential state transition conditions.

[0180] Node hierarchy: Level 1 nodes (source nodes): Target feeder 8, Target feeder 9; Level 2 nodes (directly affected by Level 1 and capable of transformation): Target feeder 7 (affected by Target feeder 8), Target feeder 10 (affected by Target feeder 8 and Target feeder 9); Level 3 nodes (directly affected by Level 2 and capable of transformation): Target feeder 11 (affected by Target feeder 7).

[0181] Constructing propagation chains: Starting with target feeders 8 and 9, target feeder 8 points to target feeder 7 through a connection with coupling strength of 0.9, and to target feeder 10 through a connection with coupling strength of 0.7; target feeder 9 points to target feeder 10 through a connection with coupling strength of 0.6; target feeder 7 points to target feeder 11 through a connection with coupling strength of 0.8. This ultimately forms two parallel and intersecting propagation chains: Propagation Chain 1: Target Feeder 8 → Target Feeder 7 → Target Feeder 11; Propagation Chain 2: Target Feeder 8 → Target Feeder 10, Target Feeder 9 → Target Feeder 10.

[0182] Verification: All feeders with potential state transition conditions (target feeders 7, 10, and 11) have been included, and all connections conform to unidirectional coupling relationships. The reverse heavy overload impact propagation chain is complete.

[0183] This invention constructs a reverse overload impact propagation chain through node layering and step-by-step recursion, presenting the propagation trajectory of the reverse overload impact from the source feeder to downstream feeders at all levels. It can dynamically predict the disturbance impact of reverse overload on the load evolution path of other feeders based on the reverse overload impact propagation chain, realize the early detection of potential new reverse overload feeders, and avoid scheduling delays and long-term equipment overload risks.

[0184] Optionally, the processes of steps 401 to 404 include: Step 401: Based on the positional relationship of each target feeder in the reverse heavy overload influence propagation chain, determine the load evolution role of each target feeder in the next prediction period.

[0185] Optionally, the feeder timing prediction device acquires the complete structural information of the constructed reverse heavy overload impact propagation chain, as well as the identification information of all target feeders. Here, positional relationship refers to the existence, specific node level, and connection relationship of the target feeder in the reverse heavy overload impact propagation chain; load evolution role refers to the classification identifier used to define the degree of impact of the target feeder on the reverse heavy overload and its potential for state evolution in the next prediction period. Positional relationships include three categories: the first category is reverse heavy overload feeders belonging to the reverse heavy overload impact propagation chain (i.e., the source node of the propagation chain or a transmission node that has evolved into a reverse heavy overload state); the second category is non-reverse heavy overload feeders belonging to the reverse heavy overload impact propagation chain with potential state transition conditions (i.e., downstream transmission nodes of the propagation chain, with the potential to evolve into reverse heavy overload feeders); and the third category is target feeders not belonging to the reverse heavy overload impact propagation chain (i.e., feeders unaffected by or negligibly affected by the reverse heavy overload); feeder identifier refers to information used to uniquely distinguish different feeders, such as feeder number.

[0186] Furthermore, the feeder timing prediction device matches and verifies the identification information of each target feeder with the node information of the reverse overload impact propagation chain one by one to determine whether the target feeder exists in the propagation chain.

[0187] Furthermore, for the target feeder existing in the propagation chain, its node attributes are checked: if the node is a reverse heavy overload feeder (source node or transformed transmission node), its load evolution role is determined as "reverse heavy overload affects the reverse heavy overload feeder in the propagation chain"; if the node is a non-reverse heavy overload feeder with potential state transition conditions (downstream transmission node), its load evolution role is determined as "reverse heavy overload affects the non-reverse heavy overload feeder with potential state transition conditions in the propagation chain".

[0188] Furthermore, for target feeders that have not been matched with propagation chain node information, their load evolution role is determined as "not belonging to the reverse heavy overload impact propagation chain".

[0189] In one embodiment, the propagation chain structure is propagation chain 1 (target feeder 8 → target feeder 7 → target feeder 11) and propagation chain 2 (target feeder 8 → target feeder 10, target feeder 9 → target feeder 10); wherein, target feeders 8 and 9 are reverse overload feeders (source nodes), and target feeders 7, 10, and 11 are non-reverse overload feeders (conduction nodes) with potential state transition conditions; all target feeders include target feeders 7, 8, 9, 10, 11, and 12 (target feeder 12 is added to reflect the role of "not belonging to the propagation chain").

[0190] Furthermore, the feeder timing prediction device performs matching verification and role determination: target feeders 8 and 9 are matched as propagation chain source nodes (reverse heavy overload feeders), and their role is "reverse heavy overload feeders in the reverse heavy overload influence propagation chain"; target feeders 7, 10, and 11 are matched as propagation chain transmission nodes (non-reverse heavy overload feeders with potential state transition conditions), and their role is "non-reverse heavy overload feeders with potential state transition conditions in the reverse heavy overload influence propagation chain"; target feeder 12 is not matched with a propagation chain node, and its role is "not belonging to the reverse heavy overload influence propagation chain".

[0191] Step 402: Determine the evolution mode based on the load evolution role of each target feeder, and determine the load power change direction and load flow direction in the next forecast period based on the evolution mode of each target feeder.

[0192] Optionally, the feeder timing prediction device acquires the load evolution role of each target feeder. Here, the evolution mode refers to the model used to characterize the load state evolution of the target feeder in the next prediction period, specifically including load continuation mode, joint evolution mode, and load stability evolution mode; the load power change direction value refers to the specific definition of the increase or decrease direction of the load power of the target feeder in the next prediction period, including "increasing," "decreasing," and "stable." The load flow direction value refers to the specific definition of the load flow direction of the target feeder in the next prediction period, including "forward," "reverse," and "fluctuating." Forward means the load flow direction is from the substation outgoing switch to the feeder end; reverse means the load flow direction is from the feeder end to the substation outgoing switch; fluctuating means the load flow direction alternates between forward and reverse.

[0193] Optionally, the current time period refers to the time interval in which load data collection and analysis have been completed. Power flow disturbance response characteristics refer to the possible changes in load power and load flow direction in the next forecast period when a non-reverse heavy overload feeder is affected by a continuous reverse heavy overload, including the load power response amplitude, load power response direction, and load flow direction response trend.

[0194] Secondly, the specific evolutionary pattern is determined through steps 4021 to 4023: Step 4021: For the target feeder whose load evolution role belongs to the reverse heavy overload influence propagation chain, based on its load power sequence and load flow direction sequence in the current time period, determine the load continuation mode in which the target feeder maintains the reverse heavy overload state in the next prediction period as the evolution mode. Specifically, the analysis is as follows: obtain the load power sequence and load flow direction sequence of the target feeder in the current time period, and analyze the stability of its reverse heavy overload state (stability judgment criteria: the proportion of time points in the current time period where the load flow direction is reversed is ≥ the preset stable proportion threshold, and the load power is continuously greater than or equal to the value of the load power). If the load is determined to be a stable reverse overload state, the evolution mode is determined to be the load continuation mode. This mode indicates that the target feeder will continue the current reverse overload operation state in the next prediction period, the load power change trend is consistent with the current period, and the load flow direction is continuously reversed. If the load is determined to be an unstable reverse overload state (the above proportion is less than the preset stable proportion threshold or the load power is lower than the active power threshold), a fluctuation correction coefficient is superimposed on the load continuation mode. The correction coefficient ranges from 0.05 to 0.1 and is used to characterize the power fluctuation impact under unstable conditions.

[0195] Step 4022: For a target feeder whose load evolution role belongs to the non-reverse heavy overload feeder in the reverse heavy overload influence propagation chain, based on its power flow disturbance response characteristics, the joint evolution mode of load power jump and load flow direction reversal in the next forecast period is determined as the evolution mode. Specifically, the analysis is as follows: First, obtain the power flow disturbance response characteristics of the target feeder and extract the load power response amplitude (jump amplitude), load power response direction (jump direction), and load flow direction response trend (reversal probability). If the load flow direction response trend in the power flow disturbance response characteristics is "there is a strong possibility of reversal" or "there is a weak possibility of reversal", and the load power response direction is "increasing", then the evolution mode is determined to be a joint evolution mode. This mode represents that the target feeder will simultaneously experience a load power jump (the jump amplitude is the load power response amplitude in the power flow disturbance response characteristics) and a load flow direction reversal (from the current positive or fluctuating state to the reverse) in the next forecast period. If the load power response direction is "decreasing", then the joint evolution mode represents the joint change of load power decrease and power flow direction reversal.

[0196] Step 4023: For target feeders whose load evolution role does not belong to the reverse heavy overload influence propagation chain, based on their load power sequence and load flow direction sequence in the current period, the load stability evolution mode that allows the target feeder to maintain its original operating state in the next prediction period is obtained as the evolution mode. The specific analysis is as follows: First, obtain the load power sequence and load flow direction sequence of the target feeder in the current period, and calculate the load power fluctuation coefficient in the current period (fluctuation coefficient = difference between the maximum and minimum power values ​​in the current period / average power value in the current period); if the fluctuation coefficient ≤ the preset stability threshold (e.g., 0.03), then the evolution mode is determined to be the load stability evolution mode. This mode indicates that the target feeder will maintain its current operating state in the next prediction period, the load power will be basically stable (fluctuation range ≤ 3% of the average power in the current period), and the load flow direction will be consistent with the state at the end of the current period; if the fluctuation coefficient > the preset stability threshold, then a small fluctuation correction needs to be included in the load stability evolution mode, with a fluctuation range of 3% to 5% of the average power in the current period.

[0197] Furthermore, based on the evolution mode determined above, the feeder timing prediction device clarifies the load power change direction and load flow direction of each target feeder in the next prediction period: if the evolution mode is a load continuation mode (corresponding to the role of a reverse heavy overload feeder in the propagation chain), the load power change direction follows the overall load power change trend direction of its current period, and the load flow direction is "reverse"; if the evolution mode is a joint evolution mode (corresponding to the role of a non-reverse heavy overload feeder in the propagation chain), the load power change direction is the load power response direction in the power flow disturbance response characteristics, and the load flow direction is "reverse" (corresponding to the possibility of power flow reversal); if the evolution mode is a load stable evolution mode (corresponding to the role of a feeder not belonging to the propagation chain), the load power change direction is "stable", and the load flow direction follows the load flow direction of the last time point of its current period.

[0198] For the embodiment of step 4021: Target feeder 8 (role: reverse overload feeder in the propagation chain), the current load power sequence is 80 kW, 81 kW, 82 kW, 83 kW, 85 kW (all > active power threshold 80 kW), and the load flow direction sequence is reverse, reverse, reverse, reverse, reverse (reverse time point percentage = 5 / 5 = 100% > 80%, stable reverse overload state). The feeder timing prediction device determines its evolution mode as load continuation mode, the load power change direction value follows the current time period's "increasing" trend, and the load flow direction value is "reverse".

[0199] For the embodiment of step 4022: Target feeder 7 (role: a non-reverse heavy overload feeder in the propagation chain with potential state transition conditions), its power flow disturbance response characteristics are: load power response amplitude of 2.025-2.475 kW, increasing load power response direction, and a strong possibility of a reverse change in the load flow direction response trend. The feeder timing prediction device determines its evolution mode as a joint evolution mode, and further determines that the load power change direction is "increasing" in the power flow disturbance response characteristics, and the load flow direction is "reverse" (corresponding to the possibility of power flow reversal).

[0200] For the embodiment of step 4023: Target feeder 12 (role: not part of the propagation chain), the current load power sequence is 70 kW, 70.1 kW, 69.9 kW, 70.2 kW, 70 kW, the current average power is 70.04 kW, the difference between the maximum and minimum power is 0.3 kW, and the fluctuation coefficient is 0.3 / 70.04 ≈ 0.004 < 0.03 (preset stability threshold). The feeder timing prediction device determines its evolution mode as a load stability evolution mode, further determines the load power change direction as "stable", and the load flow direction as "positive" following the state at the end of the current period.

[0201] Step 403: Based on the load power change direction value and load flow direction value of each target feeder, and combined with the end value of the load power sequence and the end state of the load flow direction sequence in the current period, determine the estimated load power value and the estimated load flow direction value at the first time point of the next prediction period.

[0202] Optionally, the feeder timing prediction device acquires the load power change direction value and load flow direction value for each target feeder, and simultaneously acquires the last value of the load power sequence and the last state of the load flow direction sequence for each target feeder in the current time period. Here, the last value of the load power sequence in the current time period refers to the load power data corresponding to the last sampling moment of the current time period; the last state of the load flow direction sequence in the current time period refers to the load flow direction corresponding to the last sampling moment of the current time period; the first time point of the next prediction period refers to the first sampling moment of the next prediction period, and the time interval between it and the last sampling moment of the current time period is equal to a preset sampling interval (e.g., 5 minutes); the load power estimate refers to the predicted estimated value of the load power at the first time point of the next prediction period; and the load flow direction estimate refers to the predicted estimated result of the load flow direction at the first time point of the next prediction period.

[0203] Furthermore, the feeder timing prediction device calculates the load power estimate, and the calculation rule is determined according to the direction of load power change: if the value is "increasing", then the load power estimate = the last value of the load power sequence in the current time period + the average change in load power in the current time period (average change in load power in the current time period = (power at the last time point of the current time period - power at the first time point of the current time period) / the total number of time intervals in the current time period); if the value is "decreasing", then the load power estimate = the last value of the load power sequence in the current time period - the average change in load power in the current time period; if the value is "stable", then the load power estimate = the last value of the load power sequence in the current time period.

[0204] Furthermore, the feeder timing prediction device determines the load flow direction estimate, and the determination rule is based on the load flow direction value: if the value is "forward" or "reverse", the load flow direction estimate is directly taken as that value; if the value is "fluctuating", the load flow direction estimate is opposite to the state at the end of the current time period; if the value follows the state at the end of the current time period, the load flow direction estimate = the state at the end of the load flow direction sequence of the current time period.

[0205] In one embodiment, the current time period is 14:00-14:20, the preset sampling interval is 5 minutes, and the total number of time intervals in the current time period is 4; the relevant data of each target feeder are as follows: Target feeder 8 (role: reverse overload feeder in the propagation chain), the evolution mode corresponds to the load power change direction value of "increasing" and the load flow direction value of "reverse", the last value of the load power sequence in the current time period (14:20) is 85 kW, the power at the first time point of the current time period (14:00) is 80 kW, and the average change in load power in the current time period = (85-80) / 4 = 1.25 kW; Target feeder 12 (role: not belonging to the propagation chain) The evolution mode corresponds to the load power change direction value being "stable" and the load flow direction value following the current end state. The end value of the load power sequence in the current period is 70 kW, and the end state of the load flow direction sequence in the current period is "positive". Target feeder 7 (role: non-reverse heavy overload feeder with potential state transition conditions in the propagation chain), the evolution mode corresponds to the load power change direction value being "increasing" and the load flow direction value being "reverse". The end value of the load power sequence in the current period is 94 kW, the power at the first time point of the current period is 90 kW, and the average change in load power in the current period = (94-90) / 4 = 1 kW.

[0206] Furthermore, the estimated values ​​for the first time point (14:25) of the next forecast period for each target feeder are calculated: the estimated load power of target feeder 8 = 85 + 1.25 = 86.25 kW, and the estimated load flow direction = "reverse"; the estimated load power of target feeder 12 = 70 kW (stable value), and the estimated load flow direction = "forward" (using the final state); the estimated load power of target feeder 7 = 94 + 1 = 95 kW, and the estimated load flow direction = "reverse".

[0207] Step 404: Based on the estimated load power and load flow direction of each target feeder and the time evolution law of their evolution patterns, analyze the extension trajectory of the load power sequence and the extension trajectory of the load flow direction sequence at each subsequent time point in the next prediction period, and obtain the load time series prediction result of each target feeder.

[0208] Optionally, the feeder timing prediction device acquires the estimated load power and load flow direction of each target feeder at the first time point of the next prediction period, and simultaneously acquires the time evolution pattern corresponding to the evolution mode of each target feeder. The time evolution pattern refers to the fixed pattern of load power and load flow direction changes over time in the evolution mode, including the stability of the load power change amplitude (e.g., continuous change of a fixed average change) and the duration requirement of the load flow direction (e.g., continuing in the opposite direction throughout the entire prediction period). Subsequent time points in the next prediction period refer to other sampling times within the next prediction period besides the first time point, with the number consistent with the number of sampling times in the current period (e.g., 4, corresponding to 14:30, 14:35, 14:40, and 14:45). The load power sequence extension trajectory refers to the load power data sequence at all time points in the next prediction period, extending from the estimated load power at the first time point according to the time evolution pattern. The load flow direction sequence extension trajectory refers to the load flow direction data sequence at all time points in the next prediction period, extending from the estimated load flow direction at the first time point according to the time evolution pattern.

[0209] Furthermore, the feeder timing prediction device constructs a load power sequence extension trajectory: according to the time evolution law of the evolution mode, if the load power change amplitude is stable (such as a fixed average change), then starting from the load power estimate at the first time point, the load power prediction value at each subsequent time point = the load power prediction value at the previous time point + (or -) the corresponding change ("increasing" takes "+", "decreasing" takes "-", and "stable" takes 0); if the change amplitude fluctuates, then on the basis of the fixed average change, a random fluctuation value within a preset fluctuation range (such as ±0.2 kW) is superimposed (the fluctuation value is calculated by a preset fluctuation coefficient, which is 0.05, and the fluctuation value = the power at the previous time point * 0.05).

[0210] Furthermore, the feeder timing prediction device constructs the load flow direction sequence extension trajectory: according to the time evolution law of the evolution mode, if the load flow direction is "forward" or "reverse", then the predicted load flow direction value of all subsequent time points is consistent with the estimated load flow direction value of the first time point; if the value is "fluctuation", then the predicted value of subsequent time points is determined according to the alternating law of "forward-reverse", and the alternation period is twice the preset sampling interval (such as 10 minutes).

[0211] Furthermore, the feeder timing prediction device integrates the constructed load power sequence extension trajectory and load flow direction sequence extension trajectory to obtain the load timing prediction result for each target feeder. The load timing prediction result includes the load power prediction sequence (the load power prediction values ​​at all time points in the next prediction period arranged in chronological order) and the load flow direction prediction sequence (the load flow direction prediction values ​​at all time points in the next prediction period arranged in chronological order).

[0212] In one embodiment, the next prediction period is 14:25-14:45, and the sampling times are 14:25, 14:30, 14:35, 14:40, and 14:45 (a total of 5 time points). The preset fluctuation coefficient is 0.05. The time evolution of each target feeder is as follows: Target feeder 8 (reverse heavy overload continuous evolution mode): the load power increases by a fixed average amount of 1.25 kW, and the load flow direction continues to reverse; Target feeder 12 (no impact stable evolution mode): the load power is stable and does not change, and the load flow direction continues to be positive; Target feeder 7 (potential reverse heavy overload transition evolution mode): the load power increases by a fixed average amount of 1 kW, and the load flow direction continues to reverse.

[0213] Construct the sequence extension trajectory and prediction results for each target feeder: the load power prediction sequence for target feeder 8 = 86.25 kW (14:25), 87.5 kW (14:30, 86.25+1.25), 88.75 kW (14:35), 90 kW (14:40), 91.25 kW (14:45); the load flow direction prediction sequence = reverse, reverse, reverse, reverse, reverse; integrate them into the load time sequence prediction result for target feeder 8.

[0214] The load power prediction sequence for target feeder 12 is 70 kW (14:25), 70 kW (14:30), 70 kW (14:35), 70 kW (14:40), and 70 kW (14:45); the load flow direction prediction sequence is forward, forward, forward, forward, and forward; the load timing prediction results of target feeder 12 are integrated.

[0215] The load power prediction sequence for target feeder 7 is 95 kW (14:25), 96 kW (14:30, 95+1), 97 kW (14:35), 98 kW (14:40), and 99 kW (14:45); the load flow direction prediction sequence is reverse, reverse, reverse, reverse, reverse; these are integrated into the load time sequence prediction result for target feeder 7.

[0216] Embodiments of the present invention.

[0217] Furthermore, the feeder timing prediction device based on distribution network operation provided by the present invention will be described below. The feeder timing prediction device based on distribution network operation described below and the feeder timing prediction method based on distribution network operation described above can be referred to in correspondence with each other.

[0218] Optionally, refer to Figure 2 , Figure 2 This is a schematic diagram of the feeder timing prediction device based on distribution network operation provided by the present invention. The feeder timing prediction device based on distribution network operation includes: The power grid operation monitoring module 210 is used to determine the load power sequence and load flow direction sequence of each target feeder in the current time period based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switch. The heavy overload state identification module 220 is used to identify the reverse heavy overload state based on the load power sequence and load flow direction sequence of each target feeder, and to obtain the reverse heavy overload feeder and the non-reverse heavy overload feeder. The evolution path analysis module 230 is used to analyze the potential evolution path of each non-reverse overload feeder affected by reverse overload in the next prediction period based on the load change synchronicity of each feeder pair of the non-reverse overload feeder and the reverse overload feeder in a preset continuous period, and to construct the reverse overload influence propagation chain. The feeder timing prediction module 240 is used to analyze the load evolution trend of each target feeder in the next prediction period based on the reverse heavy overload influence propagation chain, and obtain the load timing prediction results of each target feeder.

[0219] This invention achieves accurate identification of reverse overload feeders, dynamic tracing and propagation chain construction of their impact paths on surrounding non-reverse overload feeders, and time-series prediction of all-target feeder loads by combining the propagation chain disturbance effects. It fills the prediction blind spot that ignores the propagation mechanism of reverse overload disturbances between feeders, and can dynamically predict the disturbance impact of reverse overload on the load evolution path of other feeders. It enables early detection of potential new reverse overload feeders and avoids scheduling delays and long-term equipment overload risks.

[0220] Please see Figure 3 , Figure 3An embodiment diagram of an electronic device provided in accordance with the present invention. For example... Figure 3 As shown, an embodiment of the present invention provides an electronic device 300, including a memory 310, a processor 320, and a computer program 311 stored in the memory 310 and executable on the processor 320. When the processor 320 executes the computer program 311, it performs the following steps: Based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches, the load power sequence and load flow direction sequence of each target feeder in the current time period are determined. Based on the load power sequence and load flow direction sequence of each target feeder, reverse heavy overload status identification is performed to obtain reverse heavy overload feeders and non-reverse heavy overload feeders. Based on the synchronicity of load changes of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous period, the potential evolution path of each non-reverse heavy overload feeder affected by the reverse heavy overload in the next prediction period is analyzed, and the propagation chain of the reverse heavy overload influence is constructed. Based on the reverse overload effect propagation chain analysis, the load evolution trend of each target feeder in the next prediction period is obtained, and the load time series prediction results of each target feeder are obtained.

[0221] Please see Figure 4 , Figure 4 An embodiment diagram of a computer-readable storage medium provided in accordance with an embodiment of the present invention is shown. Figure 4 As shown, this embodiment provides a computer-readable storage medium 400 on which a computer program 311 is stored. When the computer program 311 is executed by a processor, it performs the following steps: Based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches, the load power sequence and load flow direction sequence of each target feeder in the current time period are determined. Based on the load power sequence and load flow direction sequence of each target feeder, reverse heavy overload status identification is performed to obtain reverse heavy overload feeders and non-reverse heavy overload feeders. Based on the synchronicity of load changes of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous period, the potential evolution path of each non-reverse heavy overload feeder affected by the reverse heavy overload in the next prediction period is analyzed, and the propagation chain of the reverse heavy overload influence is constructed. Based on the reverse overload effect propagation chain analysis, the load evolution trend of each target feeder in the next prediction period is obtained, and the load time series prediction results of each target feeder are obtained.

[0222] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the feeder timing prediction method based on distribution network operation provided by the above methods, the method including: Based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches, the load power sequence and load flow direction sequence of each target feeder in the current time period are determined. Based on the load power sequence and load flow direction sequence of each target feeder, reverse heavy overload status identification is performed to obtain reverse heavy overload feeders and non-reverse heavy overload feeders. Based on the synchronicity of load changes of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous period, the potential evolution path of each non-reverse heavy overload feeder affected by the reverse heavy overload in the next prediction period is analyzed, and the propagation chain of the reverse heavy overload influence is constructed. Based on the reverse overload effect propagation chain analysis, the load evolution trend of each target feeder in the next prediction period is obtained, and the load time series prediction results of each target feeder are obtained.

[0223] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0224] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0225] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A feeder timing prediction method based on distribution network operation, characterized in that, include: Based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches, the load power sequence and load flow direction sequence of each target feeder in the current time period are determined. Based on the load power sequence and load flow direction sequence of each target feeder, reverse heavy overload status identification is performed to obtain reverse heavy overload feeders and non-reverse heavy overload feeders. Based on the synchronicity of load changes of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous period, the potential evolution path of each non-reverse heavy overload feeder affected by the reverse heavy overload in the next prediction period is analyzed, and the propagation chain of the reverse heavy overload influence is constructed. Based on the analysis of the reverse overload influence propagation chain, the load evolution trend of each target feeder in the next prediction period is obtained, and the load time-series prediction results of each target feeder are obtained.

2. The feeder timing prediction method based on distribution network operation according to claim 1, characterized in that, The analysis of the load evolution trend of each target feeder in the next forecast period based on the reverse overload influence propagation chain yields the load time-series forecast results for each target feeder, including: Based on the positional relationship of each target feeder in the reverse heavy overload influence propagation chain, the load evolution role of each target feeder in the next prediction period is determined; The evolution mode is determined based on the load evolution role of each target feeder, and the load power change direction and load flow direction are determined based on the evolution mode of each target feeder in the next forecast period. Based on the load power change direction and load flow direction of each target feeder, and combined with the end value of the load power sequence and the end state of the load flow direction sequence in the current period, the estimated load power and estimated load flow direction at the first time point of the next forecast period are determined. Based on the estimated load power and load flow direction of each target feeder and the temporal evolution of their evolution patterns, the extension trajectories of the load power sequence and load flow direction sequence at each subsequent time point in the next prediction period are analyzed to obtain the load time series prediction results for each target feeder.

3. The feeder timing prediction method based on distribution network operation according to claim 2, characterized in that, The load evolution roles include reverse overload feeders that belong to the reverse overload influence propagation chain and non-reverse overload feeders that have potential state transition conditions, as well as those that do not belong to the reverse overload influence propagation chain. The determination of the evolution mode based on the load evolution role of each target feeder includes: For a target feeder whose load evolution role belongs to the reverse heavy overload feeder in the reverse heavy overload influence propagation chain, the load continuation mode in which the target feeder maintains the reverse heavy overload state in the next prediction period is determined as the evolution mode based on its load power sequence and load flow direction sequence in the current period. For a target feeder whose load evolution role belongs to a non-reverse heavy overload feeder in the reverse heavy overload influence propagation chain, based on its power flow disturbance response characteristics, the joint evolution mode of load power jump and load flow direction reversal that occurs in the next prediction period of the target feeder is determined as the evolution mode. For target feeders whose load evolution role does not belong to the reverse heavy overload influence propagation chain, the load stability evolution mode of the target feeder in the next prediction period is obtained based on its load power sequence and load flow direction sequence in the current period.

4. The feeder timing prediction method based on distribution network operation according to claim 1, characterized in that, The method analyzes the potential evolution path of each non-reverse heavy overload feeder affected by reverse heavy overload in the next predicted period based on the load change synchronicity of each feeder pair of the non-reverse heavy overload feeder and the reverse heavy overload feeder within a preset continuous time period, and constructs the reverse heavy overload impact propagation chain, including: The load power sequence of each feeder pair within a preset continuous time period is paired point by point according to the same time point alignment relationship to obtain the synchronous time sequence pair of each feeder pair; For each feeder pair's synchronization time series pair, the direction consistency judgment is determined based on the first load power change direction of the non-reverse heavy overload feeder and the second load power change direction of the reverse heavy overload feeder within the same time interval; the direction consistency judgment result is determined; the load power change direction represents the increase or decrease trend of the load power value at the current time point relative to the load power value at the previous time point. Based on the direction consistency determination results of each feeder pair, the total number of time intervals in which the change direction of each feeder pair is consistent within a preset continuous period is determined, and based on the ratio of the total number of time intervals to the total number of time intervals within the preset continuous period, the load change direction synchronization rate of each feeder pair is determined. If the load change direction synchronization rate is greater than or equal to the preset synchronization rate threshold, then the load change direction synchronization rate is determined as the load coupling strength of each feeder pair; otherwise, the load coupling strength is determined to be zero. Based on the load coupling strength analysis of each feeder pair, the potential evolution path of each non-reverse heavy overload feeder affected by reverse heavy overload in the next prediction period is analyzed, and the propagation chain of reverse heavy overload impact is constructed.

5. The feeder timing prediction method based on distribution network operation according to claim 4, characterized in that, The analysis of the load coupling strength of each feeder pair reveals the potential evolution path of each non-reverse overload feeder affected by the reverse overload in the next prediction period, constructing a reverse overload impact propagation chain, including: Iterate through the load coupling strength of each feeder pair, identify the feeder pairs with load coupling strength greater than zero as load-coupled feeder pairs, and determine the unidirectional coupling connection relationship with the reverse heavy overload feeder as the source end and the non-reverse heavy overload feeder as the receiving end based on the pairing structure between the reverse heavy overload feeder and the non-reverse heavy overload feeder in each load-coupled feeder pair. Based on the operating status characteristics of the reverse overload feeder in each unidirectional coupling connection during the current time period, the initial action mode of the reverse overload feeder influencing the non-reverse overload feeder coupled with it is determined; the operating status characteristics characterize the load power change trend direction and load flow direction stability of the reverse overload feeder during the current time period. Based on the unidirectional coupling connection between each non-reverse overload feeder and all its coupled reverse overload feeders and the corresponding initial operating mode, the set of multi-source influence inputs on the non-reverse overload feeder in the next prediction period is determined. Based on the persistence characteristics of each reverse heavy overload feeder in the multi-source influence input set, the reverse heavy overload influence propagation chain is constructed; the persistence characteristics represent the monotonic continuation of the load power sequence and the consistent continuation of the load flow direction sequence of the reverse heavy overload feeder within a preset continuous time period.

6. The feeder timing prediction method based on distribution network operation according to claim 5, characterized in that, The construction of the reverse overload influence propagation chain based on the persistence characteristics of each reverse overload feeder in the multi-source influence input set includes: Based on the persistence characteristics analysis of each reverse overload feeder in the multi-source influence input set, the persistence influence of each reverse overload feeder on the non-reverse overload feeder is analyzed, and the persistence judgment result is obtained. Based on the continuity judgment result as a continuous impact input, and combined with the load power sequence and load flow direction sequence of the corresponding non-reverse heavy overload feeder in the current period, the power flow disturbance response characteristics of the non-reverse heavy overload feeder in the next prediction period are determined. Based on the joint analysis of the load power growth direction and the reversal possibility of the load power flow direction in the power flow disturbance response characteristics, it is determined whether a non-reverse heavy overload feeder has the potential state transition conditions to evolve into a reverse heavy overload feeder. Based on the potential state transition conditions of all non-reverse heavy overload feeders and their unidirectional coupling connection with the upstream reverse heavy overload feeder, the non-reverse heavy overload feeder with potential state transition conditions is used as the downstream transmission node. The reverse heavy overload influence propagation chain is constructed by stepping along the unidirectional coupling connection, starting from the reverse heavy overload feeder, and propagating through the unidirectional coupling connection to cover all non-reverse heavy overload feeders with potential state transition conditions.

7. The feeder timing prediction method based on distribution network operation according to any one of claims 1 to 6, characterized in that, The reverse overload state identification is performed based on the load power sequence and load flow direction sequence of each target feeder to obtain reverse overload feeders and non-reverse overload feeders, including: The first feeder with reverse power flow behavior is identified based on the load flow direction sequence of each target feeder in the current time period; having reverse power flow behavior indicates that there is at least one time point in time where the load flow direction is from the end of the feeder to the substation outgoing switch. The absolute value of the maximum active power is determined based on the active power value corresponding to the time point when the load flow direction is reversed in the load power sequence of the first feeder. The first feeder whose maximum active power absolute value is greater than or equal to the active power threshold corresponding to the rated current carrying capacity of the feeder is determined as the second feeder that meets the reverse heavy load requirement. The total duration during which the second feeder meets the requirement that the load flow direction is reverse and the absolute value of active power is not less than the active power threshold during the current time period is determined, as well as the proportion of the total duration to the total duration of the current time period. The second feeder with a total duration not less than the preset minimum duration threshold and a duration ratio not less than the preset time ratio threshold is identified as the reverse overload feeder, and the feeders other than the reverse overload feeder among the target feeders are identified as non-reverse overload feeders.

8. A feeder timing prediction device based on distribution network operation, characterized in that, Used to implement the feeder timing prediction method based on distribution network operation as described in any one of claims 1 to 7; The feeder timing prediction device based on distribution network operation includes: The power grid operation monitoring module is used to determine the load power sequence and load flow direction sequence of each target feeder in the current time period based on the feeder operation data of each target feeder in the power supply area and the power flow direction of the substation outgoing switches. The heavy overload condition identification module is used to identify the reverse heavy overload condition based on the load power sequence and load flow direction sequence of each target feeder, and to obtain the reverse heavy overload feeder and the non-reverse heavy overload feeder. The evolution path analysis module is used to analyze the potential evolution path of each non-reverse overload feeder affected by reverse overload in the next prediction period based on the load change synchronicity of each feeder pair of the non-reverse overload feeder and the reverse overload feeder in a preset continuous period, and to construct the reverse overload influence propagation chain. The feeder timing prediction module is used to analyze the load evolution trend of each target feeder in the next prediction period based on the reverse heavy overload influence propagation chain, and obtain the load timing prediction result of each target feeder.

9. An electronic device, comprising: Memory, used to store computer software programs; A processor for reading and executing the computer software program, characterized in that, when the processor executes the computer software program, it implements the feeder timing prediction method based on distribution network operation as described in any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium, wherein a computer software program is stored therein, characterized in that, When the computer software program is executed by the processor, it implements the feeder timing prediction method based on the operation of the distribution network as described in any one of claims 1 to 7.

11. A computer program product, characterized in that, The system includes a computer program, which, when executed by a processor, implements the feeder timing prediction method based on distribution network operation as described in any one of claims 1 to 7.

Images