A new fuzzy adjustment method and system for three-phase imbalance in low-voltage distribution areas of power systems

Abstract

This invention belongs to the field of low-voltage distribution network load phase sequence regulation technology, and provides a novel fuzzy adjustment method and system for three-phase imbalance in low-voltage distribution areas of power systems. It includes the following steps: acquiring data from an electricity consumption information collection system and verifying and cleaning the data; calculating the correlation and peak overlap between the three-phase current curves and the phase distribution area outlet current curves; determining whether an event has occurred in the distribution area; determining the type of three-phase load imbalance and the phases of users to be removed; using a genetic optimization algorithm to search for and determine the optimal phase adjustment strategy; screening phase adjustment schemes that improve the overall three-phase imbalance; and the distribution area adjustment scheme. This invention abandons the traditional model of precise adjustment for individual users, instead using the phase quantity to be adjusted as the core parameter, combined with the actual site conditions and a comprehensive evaluation ranking table of users in each phase, allowing maintenance personnel to flexibly select the phase adjustment targets, thereby achieving more efficient and adaptable three-phase load imbalance adjustment.

Description

Technical Field

[0001] This invention belongs to the field of phase sequence regulation technology for low-voltage power distribution network loads, and particularly relates to a new fuzzy adjustment method and system for three-phase imbalance in low-voltage distribution areas of power systems. Background Technology

[0002] The new power system is a modern energy system based on new energy sources, widely integrating distributed power sources, energy storage, electric vehicle charging piles, and smart power terminals. It features a high proportion of renewable energy absorption capacity, deep cyber-physical integration, and coordinated interaction between power generation, grid, load, and storage. With the rapid development and widespread adoption of the new power system, the scale of the power system is continuously expanding, and low-voltage distribution networks are facing profound changes in structure and load characteristics. On the one hand, single-phase loads are continuously increasing, and the spatial and temporal distribution of loads is becoming increasingly uneven. On the other hand, the large-scale integration of new power consumption and generation facilities such as charging piles and distributed photovoltaics makes the load distribution and fluctuations in low-voltage distribution areas more complex. These new scenarios further exacerbate the three-phase imbalance problem in distribution areas. Three-phase imbalance not only leads to the generation of zero-sequence current, causing heating and increased losses in power lines and transformers, but also significantly increases voltage losses in older or remote distribution areas with smaller low-voltage grid wire diameters, resulting in low voltage for users in high-load phases.

[0003] Currently, there are two main solutions to the problem of three-phase load imbalance in transformer substations: one is to use automatic phase-switching switches, which automatically detect three-phase current imbalance and design intelligent optimization algorithms to adjust the phase sequence of the load. However, such equipment is not yet widely adopted, and its economic efficiency and practicality remain to be tested. The second solution is for maintenance personnel to manually adjust the user phases, usually by developing a phase adjustment plan through on-site testing or optimization algorithms. However, in practice, this process faces several limitations: on the one hand, on-site testing is limited by the testing period, making it difficult to fully cover load changes at all times and achieve overall optimization.

[0004] On the other hand, existing optimization algorithms are mostly based on ideal assumptions, such as assuming that the phase information of all users is accurate and can be freely switched, without fully considering the complexity and constraints of actual field operations. For example, some users only have one single-phase downline, making phase adjustment extremely difficult; when multiple users share the same branch, phase needs to be changed simultaneously; in addition, the phase information recorded in the user system may also contain errors. These practical problems reduce the operability and effectiveness of algorithm solutions in field operation and maintenance, making it difficult to implement ideally if relying solely on theoretical optimization results. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention aims to provide a novel fuzzy adjustment method and system for three-phase imbalance in low-voltage power distribution areas of power systems, which offers excellent regulation performance and is applicable to complex distribution area scenarios. This addresses the problems of overly ideal assumptions and difficulties in field implementation found in existing methods.

[0006] On the one hand, the present invention provides a novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system, comprising the following steps:

[0007] S1: Obtain data from the electricity information collection system and verify and clean the data. The data includes: the metering of the switch area to be adjusted and the current data of all users in the past month. The current data includes: electricity consumption, three-phase current curve, user low voltage event record, switch area heavy overload event record and user phase.

[0008] S2: For all users daily, calculate the correlation and peak overlap between the three-phase current curve and the outlet current curve of their respective phase area. Combine this with electricity consumption to perform a multi-indicator weighted comprehensive evaluation and ranking, and obtain a comprehensive evaluation ranking table for each phase user.

[0009] S3: Determine whether any events have occurred in the distribution area within the past month. Events include user low voltage events or heavy overload events, and analyze the correlation between events and three-phase load imbalance; classify event time sets, non-event time sets, high load periods, and normal periods based on their correlation with events.

[0010] S4: Perform stratified sampling on the event time set, high load period and normal period to obtain the sampling time set. Based on the three-phase current and unbalance of the transformer outlet in the sampling time set, determine the type of three-phase load imbalance in the transformer area and the phase of the user to be removed.

[0011] S5: Based on the three-phase load imbalance type of the transformer area and combined with the comprehensive evaluation ranking table, select the top-ranked users from the list of user phases to be removed as candidates, and use the genetic optimization algorithm to search and determine the optimal phase adjustment strategy;

[0012] S6: Enumerate and simulate all subset phase adjustment schemes of the optimal phase adjustment strategy, evaluate the overall three-phase imbalance after the optimization algorithm, and select the top 5 phase adjustment schemes with the best overall three-phase imbalance to determine the range of power adjustment required when the phase of the user to be moved out is transferred to the phase of the user to be moved in.

[0013] S7: Based on the power consumption range, randomly select user combinations from the list of users to be phase-adjusted, and simulate and calculate the overall three-phase imbalance after phase adjustment. If more than 50% of the user combinations reduce the overall three-phase imbalance by more than 20%, the distribution area will determine and adjust the plan according to the power consumption. Otherwise, the distribution area will adjust the plan according to the optimal phase adjustment strategy and the actual situation on site.

[0014] Furthermore, in S1, the data undergoes verification and cleaning, specifically through the following steps:

[0015] S101: Check the integrity of the data. If the amount of current data collected by the user in the past month is less than 70% of the total amount that should be collected, then discard it.

[0016] S102: Use the box plot method to remove outliers from the user's current data, and use the least squares regression method to fill in the missing or removed blank data points.

[0017] S103: Check the remaining user phases. If a user in a certain phase is missing or the number is less than 20% of the total number of users, it is recorded as an anomaly, and the execution of steps S2-S7 is stopped.

[0018] Furthermore, the specific steps of S2 are as follows:

[0019] S201: Analyze all users daily, calculate the Spearman correlation coefficient between the three-phase current curve and the outlet current curve of the corresponding phase area, and calculate the monthly average value to obtain the average Spearman correlation coefficient.

[0020] S202: Analyze all users daily and calculate the overlap between peak electricity consumption times and the peak times of the output current curves of their respective phase transformer areas. If a user and transformer area are simultaneously at times above the 75th percentile for ≥4 days, it is determined to be a peak load overlap day, and the number of peak load overlap days in the past month is obtained.

[0021] S203: Obtain the user's electricity consumption over the past month;

[0022] S204: Calculate the weighted evaluation score for each user in each phase, and rank the users in each phase according to the weighted evaluation score to obtain the comprehensive evaluation ranking table for users in each phase. The specific calculation formula is expressed as follows:

[0023] ;

[0024] In the formula: S is the user's comprehensive evaluation score; i is the indicator number, which includes: average Spearman correlation coefficient, number of peak load overlap days, and electricity consumption; Let i be the value that the user takes on the i-th metric. As the indicator weight; The maximum value of the index for all users in the i-th phase of the current user's phase.

[0025] Furthermore, in S3, the event time set, non-event time set, high-load period and normal period are divided according to their correlation with the event. Specifically, the time associated with the event is recorded as the event time set, and the time not associated with the event is classified as the non-event time set. In the non-event time set, the time when the load of any one phase exceeds the 75th percentile value is classified as the high-load period, and the remaining time is classified as the normal period.

[0026] Furthermore, in S3, it determines whether any events have occurred in the distribution area within the past month. Events include user low voltage events or heavy overload events, and analyzes the correlation between these events and three-phase load imbalance. The specific steps are as follows:

[0027] S301: Check if a user low-voltage event has occurred. If so, obtain the user's phase information. If conditions A, B, and C are met simultaneously, determine that the user low-voltage event is related to three-phase load imbalance.

[0028] A: There are more than 5 low-voltage users, mainly distributed in one or two phases;

[0029] B: At the time of the incident, the three-phase current imbalance at the transformer substation outlet was greater than 0.3;

[0030] C: The outlet current of the low-voltage user's concentrated phase exceeds 25% of the average three-phase current;

[0031] S302: Check whether a heavy overload event has occurred at the outlet of the transformer area. The heavy overload is a non-full-phase heavy overload. If it exists, obtain the three-phase load rate for the corresponding time period. When the three-phase load rate imbalance is greater than 0.3 and the total load rate is less than 100%, it is determined that the heavy overload event is related to the three-phase load imbalance and has adjustability. Otherwise, it is considered that the heavy overload event is not related to the three-phase load imbalance or cannot be improved by phase adjustment.

[0032] Furthermore, in S301, the formula for calculating the three-phase current imbalance is expressed as follows:

[0033] ;

[0034] In the formula: I is the phase current; a, b, and c are the phases in the three phases respectively; max To find the maximum value among the three-phase currents; To find the minimum value of the three-phase current;

[0035] In S302, the formula for calculating the three-phase load factor imbalance is expressed as follows:

[0036] ;

[0037] In the formula: Phase load factor; To find the maximum value among the three-phase load factors; To find the minimum value among the three-phase load rates.

[0038] Furthermore, the specific steps of S5 are as follows:

[0039] S501: For the three-phase current at the transformer substation outlet obtained in S4, calculate the three-phase current imbalance at each time point. For times when the three-phase current imbalance is greater than 0.3, determine the type of three-phase load imbalance in the transformer substation based on D and E:

[0040] D: Determine whether the deviation between the current of the largest phase and the second largest phase exceeds 30%;

[0041] E: Determine whether the deviation between the current of the second largest phase and the smallest phase exceeds 30%;

[0042] The types of three-phase load imbalance are classified as follows: If D and E are satisfied, the three-phase current deviation is large; if D is not satisfied but E is satisfied, two phases are too high and one phase is too low; if D is satisfied but E is not satisfied, one phase is too high and two phases are too low; if D and E are not satisfied, there is no imbalance.

[0043] S502: Statistically analyze the types of three-phase load imbalance that occur at all times, and select the three-phase load imbalance type with the highest frequency as the main three-phase load imbalance type of the transformer area.

[0044] S503: If the main three-phase load imbalance type in the transformer area is large three-phase current deviation, then select the top 10 users with the largest load phase and the top 6 users with the second largest load phase from the comprehensive evaluation ranking table, set the smallest phase as the target to be moved in, and perform optimization calculations.

[0045] S504: If the main three-phase load imbalance type in the transformer area is two phases too high and one phase too low, then select the top 8 users of the largest and second largest phases from the comprehensive evaluation ranking table, set the smallest phase as the target to be moved in, and perform optimization calculations.

[0046] S505: If the main three-phase load imbalance type in the transformer area is one phase is too high and two phases are too low, then select the top 16 users with the largest phase from the comprehensive evaluation ranking table, set the second largest phase and the smallest phase as the target to be moved in, and perform optimization calculations.

[0047] Furthermore, the calculation method for the overall three-phase unbalance of S6 and S7 is as follows: within the sampling time set, the three-phase current unbalance is calculated for each sampling moment of the outlet current of the back-end zone before and after phase adjustment, and then a weighted sum is performed.

[0048] On the other hand, the present invention provides a novel fuzzy adjustment system for three-phase imbalance in low-voltage distribution areas of a power system, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the steps of the novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system according to any embodiment of the present invention.

[0049] The present invention has the following beneficial effects:

[0050] (1) Instead of the traditional model of precise adjustment for individual users, the phase power that needs to be adjusted is used as the core parameter. Combined with the actual situation on site and the comprehensive evaluation ranking table of each phase user, the operation and maintenance personnel can flexibly select the phase adjustment object, thereby achieving more efficient and adaptable three-phase load imbalance adjustment.

[0051] (2) A batch of phase adjustment schemes were selected by using the genetic optimization algorithm and its subset results to provide decision-making reference for operation and maintenance personnel. When sorting users, the power consumption characteristics of users, the similarity with the overall load change of the transformer area, and the power consumption of users were comprehensively considered, thereby improving the adjustment effect while reducing the number of users that need to be adjusted and improving the efficiency of on-site implementation.

[0052] (3) By performing stratified analysis and sampling of event periods and normal periods, and taking into account different time periods such as normal load, peak load and abnormal load, the representativeness and sensitivity of data analysis are greatly improved. The proposed phase adjustment strategy can not only achieve comprehensive optimization throughout the entire time period, but also help solve the problems of low voltage for users and heavy overload of transformer areas caused by three-phase imbalance.

[0053] (4) Through multiple rounds of simulation and phase adjustment strategy screening, the effect of all alternative schemes is pre-evaluated before implementation, which can determine the general applicability of each strategy in advance. If most random user combinations can significantly improve the imbalance, phase adjustment can achieve fuzzy decision-making; in special cases, it can also provide operation suggestions with clear objectives and a clear adjustment range for the operation and maintenance site, taking into account the flexible application of intelligent optimization and manual intervention. Attached Figure Description

[0054] Exemplary embodiments of the present invention can be more fully understood by referring to the following figures:

[0055] Figure 1 A flowchart of a novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system provided in an embodiment of the present invention;

[0056] Figure 2 The three-phase current curves at the outlet of the transformer substation in Embodiment 1 of the present invention are sampled from December 20, 2024 to January 19, 2025.

[0057] Figure 3The three-phase current curves at the outlet of the transformer substation in Embodiment 1 of the present invention, after adjustment according to the preferred scheme from December 20, 2024 to January 19, 2025 (sampled).

[0058] Figure 4 The three-phase current curves at the outlet of the transformer substation in Embodiment 1 of the present invention are obtained after adjustment according to the generated random sampling scheme from December 20, 2024 to January 19, 2025. Detailed Implementation

[0059] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0060] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0061] This invention provides a novel fuzzy adjustment method and system for three-phase imbalance in low-voltage distribution areas of power systems. Using the phase quantity to be adjusted as the core parameter, maintenance personnel flexibly select users whose phase sequence needs adjustment based on actual site conditions and a comprehensive evaluation ranking table of users for each phase. Applicable to complex distribution area environments, this method overcomes the problem of manual phase adjustment, which relies on on-site testing to develop strategies and struggles to achieve full-time optimization. It also solves the difficulties of traditional optimization algorithms, such as overly idealistic strategies that are difficult to implement on-site.

[0062] like Figure 1 As shown, this embodiment of the invention provides a novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system, comprising the following steps:

[0063] S1: Obtain data from the electricity information collection system and verify and clean the data. The data includes: the metering of the switch area to be adjusted and the current data of all users in the past month. The current data includes: electricity consumption, three-phase current curve, user low voltage event record, switch area heavy overload event record and user phase.

[0064] S2: For all users daily, calculate the correlation and peak overlap between the three-phase current curve and the outlet current curve of their respective phase area. Combine this with electricity consumption to perform a multi-indicator weighted comprehensive evaluation and ranking, and obtain a comprehensive evaluation ranking table for each phase user.

[0065] S3: Determine whether any events have occurred in the distribution area within the past month. Events include user low voltage events or heavy overload events, and analyze the correlation between events and three-phase load imbalance. Based on the correlation with events, divide the event time set into Tevent, non-event time set Tnon-event, high load period Thigh, and normal period Tnormal.

[0066] S4: Perform stratified sampling on the event time set Tevent, the high load period Thigh, and the normal period Tnormal to obtain the sampling time set Tsample. Based on the three-phase current and unbalance of the transformer outlet in the sampling time set Tsample, determine the type of three-phase load imbalance in the transformer area and the phases of users to be removed.

[0067] S5: Based on the three-phase load imbalance type of the transformer area and combined with the comprehensive evaluation ranking table, select the top-ranked users from the list of user phases to be removed as candidates. With the goal of minimizing the overall three-phase imbalance, search and determine the optimal phase adjustment strategy through the genetic optimization algorithm.

[0068] S6: Enumerate and simulate all subset phase adjustment schemes of the optimal phase adjustment strategy, evaluate the overall three-phase imbalance after the optimization algorithm, and select the top 5 phase adjustment schemes with the best overall three-phase imbalance to determine the range of power adjustment required when the phase of the user to be moved out is transferred to the phase of the user to be moved in.

[0069] S7: Based on the power consumption range, randomly select user combinations from the list of users to be phase-adjusted, and simulate and calculate the overall three-phase imbalance after phase adjustment. If more than 50% of the user combinations reduce the overall three-phase imbalance by more than 20%, the distribution area will determine and adjust the plan according to the power consumption. Otherwise, the distribution area will adjust the plan according to the optimal phase adjustment strategy and the actual situation on site.

[0070] In some embodiments, in S1, the data is verified and cleaned, and the specific steps are as follows:

[0071] S101: Check the integrity of the data. If the amount of current data collected by the user in the past month is less than 70% of the total amount that should be collected, then discard it.

[0072] S102: Use the box plot method to remove outliers from the user's current data, and use the least squares regression method to fill in the missing or removed blank data points.

[0073] S103: Check the remaining user phases. If a user in a certain phase is missing or the number is less than 20% of the total number of users, it is recorded as an anomaly, and the execution of steps S2-S7 is stopped.

[0074] In some embodiments, in S1, user low voltage event records and transformer area overload event records can be obtained from the power supply command service system, and power consumption and three-phase current curves can be obtained from the power consumption information acquisition system. The event accuracy of the current data is 15 minutes.

[0075] In some embodiments, the specific steps of S2 are as follows:

[0076] S201: Analyze all users daily, calculate the Spearman correlation coefficient between the three-phase current curve and the outlet current curve of the corresponding phase area, and calculate the monthly average value to obtain the average Spearman correlation coefficient.

[0077] S202: Analyze all users daily and calculate the overlap between peak electricity consumption times and the peak times of the output current curves of their respective phase transformer areas. If a user and transformer area are simultaneously at times above the 75th percentile for ≥4 days, it is determined to be a peak load overlap day, and the number of peak load overlap days in the past month is obtained.

[0078] S203: Obtain the user's electricity consumption over the past month;

[0079] S204: Calculate the weighted evaluation score for each user in each phase, and rank the users in each phase according to the weighted evaluation score to obtain the comprehensive evaluation ranking table for users in each phase. The specific calculation formula is expressed as follows:

[0080] ;

[0081] In the formula: S is the user's comprehensive evaluation score; i is the indicator number, which includes: average Spearman correlation coefficient (i=1), peak load overlap days (i=2), and electricity consumption (i=3). Let i be the value that the user takes on the i-th metric. As the indicator weight; The maximum value of the index for all users in the i-th phase of the current user's phase.

[0082] In some embodiments, in S201, the formula for calculating the Spearman correlation coefficient is expressed as:

[0083] ;

[0084] In the formula: r is the Spearman correlation coefficient; d j is the difference in rank between the user current and the transformer outlet current at time j; n is the number of current data points per day, which is 96 in this embodiment;

[0085] In S202, if a user's current on a certain day is greater than its 75th percentile value, and there are ≥4 overlapping times with the times when the transformer outlet current is greater than its 75th percentile value, then it is determined to be a peak load overlapping day.

[0086] In S204, the weights of the three indicators—average Spearman correlation coefficient, peak load overlap day, and electricity consumption—are set to 0.2, 0.2, and 0.6, respectively.

[0087] In some embodiments, in S3, the event time set Tevent, the non-event time set Tnon-event, the high-load period Thigh, and the normal period Tnormal are divided according to their correlation with the event. Specifically, the time associated with the event is recorded as the event time set Tevent, and the time not associated with the event is classified as the non-event time set Tnon-event. In the non-event time set Tnon-event, the time when the load of any one phase exceeds the 75th percentile is classified as the high-load period Thigh, and the remaining time is classified as the normal period Tnormal.

[0088] In some embodiments, in S3, it is determined whether any events have occurred in the transformer area within the past month. Events include user low voltage events or heavy overload events, and the correlation between the events and three-phase load imbalance is analyzed. The specific steps are as follows:

[0089] S301: Check if a user low-voltage event has occurred. If so, obtain the user's phase information. If conditions A, B, and C are met simultaneously, determine that the user low-voltage event is related to three-phase load imbalance.

[0090] A: There are more than 5 low-voltage users, mainly distributed in one or two phases;

[0091] B: At the time of the incident, the three-phase current imbalance at the transformer substation outlet was greater than 0.3;

[0092] C: The outlet current of the low-voltage user's concentrated phase exceeds 25% of the average three-phase current;

[0093] S302: Check whether a heavy overload event has occurred at the outlet of the transformer area. The heavy overload is a non-full-phase heavy overload. If it exists, obtain the three-phase load rate for the corresponding time period. When the three-phase load rate imbalance is greater than 0.3 and the total load rate is less than 100%, it is determined that the heavy overload event is related to the three-phase load imbalance and has adjustability. Otherwise, it is considered that the heavy overload event is not related to the three-phase load imbalance or cannot be improved by phase adjustment.

[0094] In some embodiments, in S301, the formula for calculating the three-phase current imbalance is expressed as:

[0095] ;

[0096] In the formula: I is the phase current; a, b, and c are the phases in the three phases respectively; max To find the maximum value among the three-phase currents; To find the minimum value of the three-phase current;

[0097] In S302, the formula for calculating the three-phase load factor imbalance is expressed as follows:

[0098] ;

[0099] In the formula: Phase load factor; To find the maximum value among the three-phase load factors; To find the minimum value among the three-phase load rates.

[0100] In some embodiments, in S4, if the event time set Tevent exists, the sampling ratios for the event time set Tevent, the high-load period Thigh, and the normal period Tnormal are set to 40%, 30%, and 30%, respectively; if the event time set Tevent does not exist, the sampling ratios for the high-load period Thigh and the normal period Tnormal are set to 50% and 50%, respectively.

[0101] The total sampling period was set to 3 days, and data from 288 time points were sampled from the non-event time set Tnon-event.

[0102] In some embodiments, the specific steps of S5 are as follows:

[0103] S501: For the three-phase current at the transformer substation outlet obtained in S4, calculate the three-phase current imbalance at each time point. For times when the three-phase current imbalance is greater than 0.3, determine the type of three-phase load imbalance in the transformer substation based on D and E:

[0104] D: Determine whether the deviation between the current of the largest phase and the second largest phase exceeds 30%;

[0105] E: Determine whether the deviation between the current of the second largest phase and the smallest phase exceeds 30%;

[0106] The types of three-phase load imbalance are classified as follows: If D and E are satisfied, the three-phase current deviation is large; if D is not satisfied but E is satisfied, two phases are too high and one phase is too low; if D is satisfied but E is not satisfied, one phase is too high and two phases are too low; if D and E are not satisfied, there is no imbalance.

[0107] S502: Statistically analyze the types of three-phase load imbalance that occur at all times, and select the three-phase load imbalance type with the highest frequency as the main three-phase load imbalance type of the transformer area.

[0108] S503: If the main three-phase load imbalance type in the transformer area is large three-phase current deviation, then select the top 10 users with the largest load phase and the top 6 users with the second largest load phase from the comprehensive evaluation ranking table, set the smallest phase as the target to be moved in, and perform optimization calculations.

[0109] S504: If the main three-phase load imbalance type in the transformer area is two phases too high and one phase too low, then select the top 8 users of the largest and second largest phases from the comprehensive evaluation ranking table, set the smallest phase as the target to be moved in, and perform optimization calculations.

[0110] S505: If the main three-phase load imbalance type in the transformer area is one phase is too high and two phases are too low, then select the top 16 users with the largest phase from the comprehensive evaluation ranking table, set the second largest phase and the smallest phase as the target to be moved in, and perform optimization calculations.

[0111] In some embodiments, a genetic optimization algorithm is used to search for and determine the optimal phase modulation strategy, specifically as follows:

[0112] 1) For the user set that needs optimization, each individual is encoded using an encoding method suitable for phase adjustment decision-making. Each individual (chromosome) represents a specific user phase adjustment allocation method. For example, a gene locus can be represented by 0, 1, and 2 to indicate that it is allocated to phases a, b, and c, respectively.

[0113] 2) Population initialization: 50 phase-tuning candidate schemes are randomly generated as the initial population, and each individual is an allocation result to be optimized;

[0114] 3) Objective function design: For each individual, simulate its adjusted phase situation and calculate the overall three-phase imbalance of the optimized three-phase current. Use the overall three-phase imbalance as the objective function to minimize the optimization objective.

[0115] 4) Selection operation: Through methods such as roulette or tournament, individuals with high fitness are selected to form a new parent group, thereby prioritizing the retention of excellent phase adjustment schemes;

[0116] 5) Crossover operation: With a crossover rate of 0.8, single-point or multi-point crossover is used to randomly exchange the phase separation schemes of some users based on the genes in the parent generation, generating new offspring individuals and expanding the search space;

[0117] 6) Mutation operation: Mutation operation is performed on some individual genes with a mutation rate of 0.1, that is, the phase allocation of a single user is randomly changed to avoid the algorithm getting trapped in local optima;

[0118] 7) Iterative evolution: After 100 generations of iteration, the fitness evaluation, selection, crossover, mutation and other steps are repeated to gradually approach the optimal phase adjustment strategy.

[0119] In some embodiments, the overall three-phase unbalance of S6 and S7 is calculated as follows: within the sampling time set Tsample, the three-phase current unbalance is calculated for each sampling time of the outlet current of the backstage area before and after phase adjustment, and then a weighted sum is performed; wherein, the weight of the three-phase current unbalance less than 0.25 can be set to 0, the weight between 0.25 and 0.5 can be set to 1, and the weight greater than 0.5 can be set to 2.

[0120] In some embodiments, in S6, the method for determining the range of electricity consumption to be adjusted when transferring from the user phase to the user phase is as follows: for the monthly electricity consumption of the user in each of the five preferred schemes, the minimum and maximum values ​​are determined as the final recommended adjustment range; for example, if the minimum electricity consumption of the user transferred from phase a to phase b in the five schemes is 100 and the maximum value is 200, then the recommended range of electricity consumption to be adjusted from phase a to phase b is 100 to 200.

[0121] In some embodiments, in S7, user combinations are randomly selected from the list of users to be adjusted, and the adjustment is repeated 10 times. If more than 5 adjustments result in a decrease of more than 20% in the overall three-phase imbalance compared to before the adjustment, it indicates that the distribution area can determine and adjust the scheme according to the electricity consumption.

[0122] In some embodiments, the present invention provides a novel fuzzy adjustment system for three-phase imbalance in low-voltage distribution areas of a power system, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it implements the steps of the novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system according to any embodiment of the present invention.

[0123] Example 1:

[0124] Taking a transformer substation where a three-phase load imbalance event was detected by the power supply command and service system as an example, the calculation period is set from December 20, 2024 to January 19, 2025. The effects before adjustment, after the optimized scheme adjustment, and after random sampling adjustment are respectively as follows: Figure 2 , Figure 3 and Figure 4 As shown in the figure. The results show that after the optimized scheme was adjusted, the overall three-phase imbalance decreased by more than 200%, and the maximum peak current of phase c decreased by 50A, effectively alleviating the low voltage problem of phase c users in this distribution area caused by the three-phase load imbalance.

[0125] In summary, this invention scientifically evaluates the optimality of each phase user's participation in phase adjustment through a multi-index weighted method, and accurately calculates the electricity consumption of each phase user requiring adjustment using an optimization algorithm, with phase adjustment electricity consumption as the core parameter. Maintenance personnel can flexibly select users requiring phase sequence adjustment based on actual site conditions and a comprehensive user evaluation ranking table, and formulate implementation plans. This invention is applicable to the complex and variable distribution area environment under new power systems. It overcomes the limitations of traditional manual phase adjustment, which relies on on-site testing and is difficult to achieve full-time global optimization, and also solves the problems of idealization and practical implementation difficulties of classic optimization algorithms. It provides a practical and feasible technical path and practical basis for the management of load imbalance in low-voltage distribution areas.

[0126] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system, characterized by: Including the following step: S1: Obtain data from the electricity information collection system and verify and clean the data. The data includes: the metering of the switch area to be adjusted and the current data of all users in the past month. The current data includes: electricity consumption, three-phase current curve, user low voltage event record, switch area heavy overload event record and user phase. S2: For all users daily, calculate the correlation and peak overlap between the three-phase current curve and the outlet current curve of their respective phase area. Combine this with electricity consumption to perform a multi-indicator weighted comprehensive evaluation and ranking, and obtain a comprehensive evaluation ranking table for each phase user. S3: Determine whether any events have occurred in the distribution area within the past month. Events include user low voltage events or heavy overload events, and analyze the correlation between events and three-phase load imbalance; classify event time sets, non-event time sets, high load periods, and normal periods based on their correlation with events. S4: Perform stratified sampling on the event time set, high load period and normal period to obtain the sampling time set. Based on the three-phase current and unbalance of the transformer outlet in the sampling time set, determine the type of three-phase load imbalance in the transformer area and the phase of the user to be removed. S5: Based on the three-phase load imbalance type of the transformer area and combined with the comprehensive evaluation ranking table, select the top-ranked users from the list of user phases to be removed as candidates, and use the genetic optimization algorithm to search and determine the optimal phase adjustment strategy; S6: Enumerate and simulate all subset phase adjustment schemes of the optimal phase adjustment strategy, evaluate the overall three-phase imbalance after the optimization algorithm, and select the top 5 phase adjustment schemes with the best overall three-phase imbalance to determine the range of power adjustment required when the phase of the user to be moved out is transferred to the phase of the user to be moved in. S7: Based on the power consumption range, randomly select user combinations from the list of users to be phase-adjusted, and simulate and calculate the overall three-phase imbalance after phase adjustment. If more than 50% of the user combinations reduce the overall three-phase imbalance by more than 20%, the distribution area will determine and adjust the plan according to the power consumption. Otherwise, the distribution area will adjust the plan according to the optimal phase adjustment strategy and the actual situation on site.

2. The novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system as described in claim 1, characterized in that: In S1, the data undergoes verification and cleaning, with the following specific steps: S101: Check the integrity of the data. If the amount of current data collected by the user in the past month is less than 70% of the total amount that should be collected, then discard it. S102: Use the box plot method to remove outliers from the user's current data, and use the least squares regression method to fill in the missing or removed blank data points. S103: Check the remaining user phases. If a user in a certain phase is missing or the number is less than 20% of the total number of users, it is recorded as an anomaly, and the execution of steps S2-S7 is stopped.

3. The novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system as described in claim 2, characterized in that: The specific steps of S2 are as follows: S201: Analyze all users daily, calculate the Spearman correlation coefficient between the three-phase current curve and the outlet current curve of the corresponding phase area, and calculate the monthly average value to obtain the average Spearman correlation coefficient. S202: Analyze all users daily and calculate the overlap between peak electricity consumption times and the peak times of the output current curves of their respective phase transformer areas. If a user and transformer area are simultaneously at times above the 75th percentile for ≥4 days, it is determined to be a peak load overlap day, and the number of peak load overlap days in the past month is obtained. S203: Obtain the user's electricity consumption over the past month; S204: Calculate the weighted evaluation score for each user in each phase, and rank the users in each phase according to the weighted evaluation score to obtain the comprehensive evaluation ranking table for users in each phase. The specific calculation formula is expressed as follows: ； In the formula: S is the user's overall evaluation score; i represents the indicator number, and the indicators include: average Spearman correlation coefficient, number of days with peak load overlap, and electricity consumption. Let i be the value that the user takes on the i-th metric. As the indicator weight; The maximum value of the index for all users in the i-th phase of the current user's phase.

4. The novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system as described in claim 3, characterized in that: In S3, the time periods are divided into event time sets, non-event time sets, high-load periods, and normal periods based on their correlation with events. Specifically, the time associated with an event is recorded as the event time set, and the time not associated with an event is classified as the non-event time set. In the non-event time set, the time when the load of any one phase exceeds the 75th percentile is classified as the high-load period, and the remaining time is classified as the normal period.

5. The novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system as described in claim 4, characterized in that: In S3, it determines whether any events have occurred in the distribution area within the past month. Events include user low voltage events or heavy overload events, and analyzes the correlation between these events and three-phase load imbalance. The specific steps are as follows: S301: Check if a user low-voltage event has occurred. If so, obtain the user's phase information. If conditions A, B, and C are met simultaneously, determine that the user low-voltage event is related to three-phase load imbalance. A: There are more than 5 low-voltage users, mainly distributed in one or two phases; B: At the time of the incident, the three-phase current imbalance at the transformer substation outlet was greater than 0.3; C: The outlet current of the low-voltage user's concentrated phase exceeds 25% of the average three-phase current; S302: Check whether a heavy overload event has occurred at the outlet of the transformer area. The heavy overload is a non-full-phase heavy overload. If it exists, obtain the three-phase load rate for the corresponding time period. When the three-phase load rate imbalance is greater than 0.3 and the total load rate is less than 100%, it is determined that the heavy overload event is related to the three-phase load imbalance and has adjustability. Otherwise, it is considered that the heavy overload event is not related to the three-phase load imbalance or cannot be improved by phase adjustment.

6. The novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system as described in claim 5, characterized in that: In S301, the formula for calculating the three-phase current imbalance is expressed as: ； In the formula: I is the phase current; a, b, and c are the phases in the three phases respectively; max To find the maximum value among the three-phase currents; To find the minimum value of the three-phase current; In S302, the formula for calculating the three-phase load factor imbalance is expressed as follows: ； In the formula: Phase load factor; To find the maximum value among the three-phase load factors; To find the minimum value among the three-phase load rates.

7. The novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system as described in claim 6, characterized in that: The specific steps of S5 are as follows: S501: For the three-phase current at the transformer substation outlet obtained in S4, calculate the three-phase current imbalance at each time point. For times when the three-phase current imbalance is greater than 0.3, determine the type of three-phase load imbalance in the transformer substation based on D and E: D: Determine whether the deviation between the current of the largest phase and the second largest phase exceeds 30%; E: Determine whether the deviation between the current of the second largest phase and the smallest phase exceeds 30%; The types of three-phase load imbalance are classified as follows: If D and E are satisfied, the three-phase current deviation is large; if D is not satisfied but E is satisfied, two phases are too high and one phase is too low; if D is satisfied but E is not satisfied, one phase is too high and two phases are too low; if D and E are not satisfied, there is no imbalance. S502: Statistically analyze the types of three-phase load imbalance that occur at all times, and select the three-phase load imbalance type with the highest frequency as the main three-phase load imbalance type of the transformer area. S503: If the main three-phase load imbalance type in the transformer area is large three-phase current deviation, then select the top 10 users with the largest load phase and the top 6 users with the second largest load phase from the comprehensive evaluation ranking table, set the smallest phase as the target to be moved in, and perform optimization calculations. S504: If the main three-phase load imbalance type in the transformer area is two phases too high and one phase too low, then select the top 8 users of the largest and second largest phases from the comprehensive evaluation ranking table, set the smallest phase as the target to be moved in, and perform optimization calculations. S505: If the main three-phase load imbalance type in the transformer area is one phase is too high and two phases are too low, then select the top 16 users with the largest phase from the comprehensive evaluation ranking table, set the second largest phase and the smallest phase as the target to be moved in, and perform optimization calculations.

8. The novel fuzzy adjustment method for three-phase imbalance in low-voltage distribution areas of a power system as described in claim 7, characterized in that: The calculation method for the overall three-phase unbalance of S6 and S7 is as follows: within the sampling time set, calculate the three-phase current unbalance for each sampling moment of the outlet current of the backstage area before and after phase adjustment, and then perform weighted summation.

9. A novel fuzzy adjustment system for three-phase imbalance in low-voltage distribution areas of a power system, characterized in that: It includes a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, it implements the method described in any one of claims 1 to 8.

Images