A power quality multi-objective robust treatment resource in-place distribution method for substation feeder segmented impedance adaptive matching

Abstract

This invention discloses a robust, multi-objective resource allocation method for power quality management in a transformer substation, employing adaptive impedance matching for segmented feeders. The method automatically divides the transformer substation feeder into N segments and identifies the equivalent resistance, reactance, and overall impedance amplitude of each segment in real time through synchronous sampling at both ends. A normalized management weight coefficient is constructed based on the proportion of the overall impedance amplitude of each segment to the total impedance of the entire substation. A four-dimensional sub-objective function is constructed, incorporating voltage offset, harmonic distortion, three-phase imbalance, and line active power loss. A robust disturbance interval is introduced, and a linear weighting method is used to construct a comprehensive minimization objective function. Under constraints such as equipment capacity, voltage safety, and power factor, an improved multi-objective particle swarm optimization algorithm is used to iteratively solve for the optimal output combination of local management equipment for each feeder segment. This invention achieves adaptive matching between management resources and feeder electrical characteristics, significantly improving the overall power quality control effect and operational economy of the transformer substation.

Description

Technical Field

[0001] This invention relates to the field of power quality management technology for low-voltage distribution networks, and in particular to a method for local allocation of multi-objective robust power quality management resources based on adaptive impedance matching of feeder segments in transformer substations. Background Technology

[0002] Low-voltage distribution substations often employ radial branch feeder structures. Due to differences in the outgoing line length, cable cross-sectional area, and installation age of each branch within the same substation, there are significant differences in the equivalent resistance and equivalent reactance of each feeder section. This results in prominent voltage drops and power losses in long-distance, thin-conductor sections, while the power quality disturbance transmission effect is weaker in short-distance, thick-conductor sections. Currently, power quality management in distribution substations generally adopts two extensive approaches: one is a uniform setpoint switching scheme, where the APF, SVG, and three-phase imbalance regulators across the entire substation are set with the same compensation capacity and regulation threshold, without distinguishing between the impedance differences of feeder sections. This leads to insufficient management capacity in high-impedance branches and excessive compensation in low-impedance branches, resulting in wasted equipment capacity and potential resonance risks. The other is a single-objective optimization allocation scheme, which only aims to minimize voltage deviation or only to suppress harmonics, without simultaneously considering multiple constraints such as line loss, three-phase imbalance, and power factor. Under conditions of random load fluctuations, the optimization solution has poor robustness, and management failure may occur even with small changes in operating conditions.

[0003] Existing technologies suffer from the following main shortcomings: First, they lack an online real-time identification model for feeder segment impedance, relying instead on offline ledger parameters. This leads to significant discrepancies between ledger impedance and actual operating impedance due to line aging and joint oxidation. Second, they employ an average allocation model for power management resources, failing to establish an adaptive matching mapping relationship between impedance and compensation capacity, resulting in poor balance of power quality indicators across different sections. Third, their optimization objectives are singular, lacking a multi-objective robust optimization function encompassing harmonics, voltage, three-phase imbalance, and line losses, resulting in insufficient resistance to load disturbances. Fourth, they lack local hierarchical allocation constraints, and the collaborative logic between management devices in different areas is missing, making it difficult to simultaneously achieve optimal economic efficiency and power supply quality for the entire distribution substation. These shortcomings have become key technical bottlenecks restricting the improvement of refined power quality management in distribution substations. Summary of the Invention

[0004] Therefore, the purpose of this invention is to provide a method for local allocation of power quality resources for multi-objective robust governance based on adaptive impedance matching of feeder segments in transformer substations, in order to solve the above-mentioned problems.

[0005] To achieve the above-mentioned objectives, this invention provides a method for local allocation of resources for robust multi-objective power quality management based on adaptive impedance matching of feeder segments in transformer substations. The method includes: Step 1: Adaptive segmentation and global synchronous sampling of transformer feeder topology: Based on the electrical nodes of the transformer area, the transformer feeder is automatically divided into N feeder segments with uniform electrical parameters, and a synchronous sampling terminal is configured at both ends of each feeder segment to collect the three-phase fundamental voltage phasor and three-phase fundamental current phasor of the beginning and end of each segment in real time. Step 2: Construct an online accurate identification model for feeder segment impedance: Based on the three-phase fundamental voltage phasors and three-phase fundamental current phasors at the beginning and end of the feeder, solve for the equivalent resistance of each feeder segment. Equivalent reactance and combined impedance amplitude ; Step 3: Establish an impedance adaptive governance weight normalization configuration model: based on the comprehensive impedance magnitude of each feeder segment. Calculate the proportion of each feeder segment to the total impedance of the entire feeder network, and obtain the governance weight coefficient for each feeder segment. ; Step 4: Construct four-dimensional power quality refinement sub-objective functions: Construct sub-objectives for minimizing voltage offset F1, minimizing harmonic distortion F2, minimizing three-phase imbalance F3, and minimizing line active power loss F4 respectively; Step 5: Construction of a multi-objective robust integrated optimization model: A linear weighted method is adopted, and a robust perturbation interval is introduced. Based on the four sub-objective functions in step four, a comprehensive minimization objective function is constructed. ; Step 6: Set multi-dimensional electrical safety and equipment constraints: Set constraints on the capacity of the treatment equipment, the voltage safety of the power grid, the overall power factor of the distribution area, and the weight normalization constraint; Step 7: On-site resource allocation solution output: Based on the comprehensive minimization objective function constructed in Step 5 and the constraints set in Step 6, the improved multi-objective particle swarm optimization algorithm is used for iterative solution to output the optimal treatment output combination of on-site treatment equipment for each feeder segment; Step 8: Dynamic Iterative Adaptive Control: Repeat steps 1 to 7 at a preset cycle, and iteratively update the model parameters based on the governance effect feedback information of the previous cycle to achieve long-term dynamic adaptive optimization control.

[0006] Furthermore, in step one, the segmentation rule is specifically as follows: Taking the main switch of the power distribution room as the starting node of the feeder, each load branch access point, cable specification change node, line joint node and end load node is used as the segment boundary node, and the entire transformer area feeder is divided into N feeder segments. The sampling terminal is equipped with a Beidou / GPS dual-mode time synchronization module, with a sampling time error of less than or equal to 1 microsecond and a sampling frequency of 10 kHz.

[0007] Furthermore, in step two, for the i-th feeder segment, its starting three-phase voltage phasor is: The phasors of the three-phase current at the beginning are: The three-phase voltage phasors at the end are The phasors of the three-phase currents at the end are Then the voltage drop phasor of that segment ;make Then the equivalent resistance and equivalent reactance The calculation formula is:

[0008]

[0009] The combined impedance magnitude The calculation formula is: .

[0010] Furthermore, in step three, the governance weight coefficient The calculation formula is:

[0011] And the sum of the governance weight coefficients of all feeder segments satisfies .

[0012] Furthermore, in step four, the voltage offset minimization sub-target The expression is:

[0013] in This represents the real-time operating voltage at the end of the i-th feeder segment. This is the rated voltage for the transformer substation.

[0014] Furthermore, in step four, the harmonic distortion minimization sub-objective The expression is:

[0015] in Let be the total harmonic distortion rate of the real-time voltage of the i-th feeder segment, and Satisfy the dynamic correction relationship:

[0016] in To address the original harmonic distortion rate, The equivalent suppression coefficient for harmonic control. Real-time capacity input for harmonic mitigation equipment in the i-th feeder segment.

[0017] Furthermore, in step four, the sub-objective of minimizing the three-phase imbalance is... The expression is:

[0018] in, Let be the real-time negative sequence voltage imbalance of the i-th feeder segment, and Satisfy the dynamic correction relationship: In the formula For the original imbalance, This is the imbalance adjustment coefficient. The real-time active power distribution adjustment amount for the three-phase imbalance adjustment device.

[0019] Furthermore, in step four, the sub-objective of minimizing the active power loss of the line is... The expression is:

[0020] in, This represents the real-time load current of the i-th feeder segment. The equivalent resistance is identified in real time for the i-th feeder segment.

[0021] Furthermore, in step five, the comprehensive minimization objective function The expression is:

[0022] in , , , These are the weight coefficients for each sub-objective, satisfying... Γ is the robust adjustment coefficient, ranging from 0 to 0.3; This represents the maximum deviation of the comprehensive objective function within the load forecast fluctuation range.

[0023] Furthermore, in step seven, the optimal governance output combination is { , , },in The optimal output for reactive power compensation of the SVG in the i-th feeder segment. To determine the optimal capacity for harmonic mitigation of the APF in the i-th feeder segment, The optimal active power regulation for the three-phase imbalance regulation of the i-th feeder segment is defined as follows: During the iterative process of the improved multi-objective particle swarm algorithm, all constraints in step six are verified and illegal solutions are eliminated in each iteration until the convergence accuracy is met or the maximum number of iterations is reached, at which point the global optimal solution is output.

[0024] Compared with the prior art, the beneficial effects of the present invention are: 1. Accurate online impedance identification: Real-time calculation of electrical phasors at the beginning and end of the feeder, unaffected by line aging or connector loss, improves impedance accuracy by more than 85% compared to offline ledgers; 2. Adaptive Differentiated Resource Allocation: By introducing impedance normalization weighting, more governance capacity is automatically allocated to high-impedance long branches, solving the problems of insufficient governance / overcompensation caused by traditional average allocation. The balance of voltage, harmonics, and imbalance at each node in the transformer area is improved by 40%. 3. Multi-objective robust optimization: It simultaneously considers four indicators: voltage, harmonics, three-phase imbalance, and line loss. By incorporating a robust disturbance range, it eliminates the need for frequent re-optimization when the load fluctuates randomly, significantly enhancing the stability of regulation. 4. Localized hierarchical management reduces communication pressure: Optimized computing is deployed to the edge terminals of the distribution area, and each segment performs compensation independently on-site, eliminating the need for the main station to issue control commands at high frequencies, thus reducing communication latency by 70%; Attached Figure Description To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only preferred embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of the overall process of a multi-objective robust power quality management resource allocation method based on adaptive impedance matching of feeder segments in a transformer substation, provided by an embodiment of the present invention.

[0026] Figure 2 This is a schematic diagram of the adaptive segmentation and dual-end synchronous sampling topology principle of the feeder topology of the transformer area provided in this embodiment of the invention.

[0027] Figure 3 This is a complete logical block diagram of the multi-objective robust optimization model based on piecewise impedance weights provided in the embodiments of the present invention. Detailed Implementation

[0028] The principles and features of the present invention are described below with reference to the accompanying drawings. The listed embodiments are only used to explain the present invention and are not intended to limit the scope of the present invention.

[0029] Reference Figure 1-3 This embodiment provides a multi-objective robust power quality management resource allocation method for local allocation based on adaptive impedance matching of feeder segments in a transformer substation. The method includes: Step 1: Adaptive segmentation and global synchronous sampling of transformer feeder topology: Based on the electrical nodes of the transformer area, the transformer feeder is automatically divided into N feeder segments with uniform electrical parameters, and a synchronous sampling terminal is configured at both ends of each feeder segment to collect the three-phase fundamental voltage phasor and three-phase fundamental current phasor of the beginning and end of each segment in real time. Step 2: Construct an online accurate identification model for feeder segment impedance: Based on the three-phase fundamental voltage phasors and three-phase fundamental current phasors at the beginning and end of the feeder, solve for the equivalent resistance of each feeder segment. Equivalent reactance and combined impedance amplitude ; Step 3: Establish an impedance adaptive governance weight normalization configuration model: based on the comprehensive impedance magnitude of each feeder segment. Calculate the proportion of each feeder segment to the total impedance of the entire feeder network, and obtain the governance weight coefficient for each feeder segment. ; Step 4: Construct four-dimensional power quality refinement sub-objective functions: Construct sub-objectives for minimizing voltage offset F1, minimizing harmonic distortion F2, minimizing three-phase imbalance F3, and minimizing line active power loss F4 respectively; Step 5: Construction of a multi-objective robust integrated optimization model: A linear weighted method is adopted, and a robust perturbation interval is introduced. Based on the four sub-objective functions in step four, a comprehensive minimization objective function is constructed. ; Step 6: Set multi-dimensional electrical safety and equipment constraints: Set constraints on the capacity of the treatment equipment, the voltage safety of the power grid, the overall power factor of the distribution area, and the weight normalization constraint; Step 7: On-site resource allocation solution output: Based on the comprehensive minimization objective function constructed in Step 5 and the constraints set in Step 6, the improved multi-objective particle swarm optimization algorithm is used for iterative solution to output the optimal treatment output combination of on-site treatment equipment for each feeder segment; Step 8: Dynamic Iterative Adaptive Control: Repeat steps 1 to 7 at a preset cycle, and iteratively update the model parameters based on the governance effect feedback information of the previous cycle to achieve long-term dynamic adaptive optimization control.

[0030] In step one, the segmentation rule is specifically as follows: Taking the main switch of the power distribution room as the starting node of the feeder, each load branch access point, cable specification change node, line joint node and end load node is used as the segment boundary node, and the entire transformer area feeder is divided into N feeder segments. The sampling terminal is equipped with a Beidou / GPS dual-mode time synchronization module, with a sampling time error of less than or equal to 1 microsecond and a sampling frequency of 10 kHz.

[0031] In step two, for the i-th feeder segment, the three-phase voltage phasor at its starting end is: The phasors of the three-phase current at the beginning are: The three-phase voltage phasors at the end are The phasors of the three-phase currents at the end are Then the voltage drop phasor of that segment ;make Then the equivalent resistance and equivalent reactance The calculation formula is:

[0032]

[0033] The combined impedance magnitude The calculation formula is: .

[0034] In step three, the governance weight coefficient The calculation formula is:

[0035] And the sum of the governance weight coefficients of all feeder segments satisfies .

[0036] In step four, the voltage offset minimization sub-target The expression is:

[0037] in This represents the real-time operating voltage at the end of the i-th feeder segment. This is the rated voltage for the transformer substation.

[0038] In step four, the harmonic distortion minimization sub-objective The expression is:

[0039] in Let be the total harmonic distortion rate of the real-time voltage of the i-th feeder segment, and Satisfy the dynamic correction relationship:

[0040] in To address the original harmonic distortion rate, The equivalent suppression coefficient for harmonic control. Real-time capacity input for harmonic mitigation equipment in the i-th feeder segment.

[0041] In step four, the sub-objective of minimizing the three-phase imbalance is mentioned. The expression is:

[0042] in, Let be the real-time negative sequence voltage imbalance of the i-th feeder segment, and Satisfy the dynamic correction relationship: In the formula For the original imbalance, This is the imbalance adjustment coefficient. The real-time active power distribution adjustment amount for the three-phase imbalance adjustment device.

[0043] In step four, the sub-objective of minimizing the active power loss of the line is mentioned. The expression is:

[0044] in, This represents the real-time load current of the i-th feeder segment. The equivalent resistance is identified in real time for the i-th feeder segment.

[0045] In step five, the comprehensive minimization objective function The expression is:

[0046] in , , , These are the weight coefficients for each sub-objective, satisfying... Γ is the robust adjustment coefficient, ranging from 0 to 0.3; This represents the maximum deviation of the comprehensive objective function within the load forecast fluctuation range.

[0047] In step seven, the optimal governance output combination is { , , },in The optimal output for reactive power compensation of the SVG in the i-th feeder segment. To determine the optimal capacity for harmonic mitigation of the APF in the i-th feeder segment, The optimal active power regulation for the three-phase imbalance regulation of the i-th feeder segment is defined as follows: During the iterative process of the improved multi-objective particle swarm algorithm, all constraints in step six are verified and illegal solutions are eliminated in each iteration until the convergence accuracy is met or the maximum number of iterations is reached, at which point the global optimal solution is output.

[0048] For example, this embodiment takes a typical urban and rural low-voltage distribution substation as an example. The substation includes one main feeder and four branch feeders. The diameter and length of each branch feeder are significantly different. The load types include residential single-phase loads, small industrial and commercial three-phase loads, and a small number of photovoltaic grid-connected inverters. The rated voltage of the substation is 0.4kV.

[0049] Step 1: Adaptive segmentation and global synchronous sampling of feeder topology in the transformer area Using the main switch of the power distribution room as the starting node of the feeder, and each load branch access point, cable specification change point (such as the section transitioning from 120mm² to 70mm²), line joint node, and end load node as the segment boundary node, the feeder of the transformer area is automatically divided into 5 feeder segments, i.e., N=5. The cable material, wire diameter, and laying method within each segment are kept consistent to ensure that the electrical parameters are uniform and without abrupt changes.

[0050] Ten dual-end synchronous sampling terminals were installed at both the beginning and end of each feeder segment. Each sampling terminal integrates a BeiDou / GPS dual-mode time synchronization module, achieving network-wide sampling time synchronization via satellite time synchronization, with time errors controlled within 1 microsecond. The sampling frequency was set to 10kHz. Each terminal collected the beginning-end voltage phasor (including voltage amplitude and voltage phase angle), beginning-end current phasor (including current amplitude and current phase angle), end-end voltage phasor, and end-end current phasor for each phase in real time. The data was aggregated to the edge computing terminal of the distribution area via the communication network.

[0051] Step 2: Construct an online accurate identification model for feeder segment impedance. After receiving all the sampled data, the edge computing terminal calculates the equivalent resistance and equivalent reactance of each feeder segment.

[0052] Taking the first feeder section (from the distribution room outgoing line to the first branch node, with a line length of approximately 120m and a cable cross-section of 95mm²) as an example, let's assume that the sampling time obtains the phasor amplitude of the A-phase voltage at the beginning of this section. =231.5V, phase angle =0.5, amplitude of the first-end current phasor =152.3A, phase angle = 18.6 Terminal voltage amplitude =229.8V, phase angle = 0.3 The terminal current amplitude Ie1 = 150.1A, phase 1= 19.2 .

[0053] The voltage drop phasor is then:

[0054] Substituting the values ​​into the numerical calculation yields the voltage drop. ≈1.82V, imaginary part ≈1.27V. Real part of the current at the beginning. imaginary part .

[0055] Substitute into the equivalent resistance formula:

[0056] Substituting into the equivalent reactance formula:

[0057] This yields the combined impedance magnitude of the first feeder segment:

[0058] Calculate the equivalent resistance of the remaining four feeder segments using the same method. Equivalent reactance and combined impedance amplitude In this embodiment, the following calculations were performed: =0.147Ω (Branch 2, line length 200m, cross-section 70mm²) =0.063Ω (Branch 3, length 60m, cross-section 120mm²) =0.218Ω (4 branches, 280m in length, 50mm² in cross-section) =0.052Ω (Branch 5, length 40m, cross-section 150mm²). The impedance identification result is updated every 30ms to track changes in line parameters in real time.

[0059] Step 3: Establish an impedance adaptive governance weight normalization configuration model Summarize the total impedance of the five segments across the entire domain: =0.091+0.147+0.063+0.218+0.052=0.571Ω Calculate the weighting coefficients for each governance segment:

[0060]

[0061]

[0062]

[0063]

[0064] Verify the global normalization condition: + + + + =0.159+0.258+0.110+0.382+0.091=1.000 satisfies the constraint. It can be seen that the 4th feeder segment (long-distance thin conductor) has the largest weight (0.382), and the 5th feeder segment (short-distance thick conductor) has the smallest weight (0.091). The weight allocation result is consistent with the electrical vulnerability ranking of each segment.

[0065] Step 4: Construct a four-dimensional refined sub-objective function for power quality At the same sampling moment, the edge computing terminal synchronously collects the operating voltage, total harmonic distortion rate, negative sequence voltage imbalance and load current of each feeder segment.

[0066] Voltage offset minimization sub-objective Calculation: Assume the real-time operating voltage at the end of the first feeder section =229.8V, =400V, then the voltage deviation term is This section is about The contribution is ×0.181≈0.029.

[0067] Harmonic distortion minimization sub-objective Calculation: Let the original harmonic distortion rate of the first feeder section before treatment be... =4.8%, this section is equipped with one APF, with a rated harmonic mitigation capacity of 4.8%. =50A, current installed capacity =30A, equivalent suppression coefficient for harmonic mitigation =0.08 (as specified by the equipment manufacturer), then the total harmonic distortion rate after treatment is:

[0068] This section is about The contribution is .

[0069] The sub-objective of minimizing three-phase imbalance Calculation: Let the original three-phase imbalance of the first feeder section before the treatment be... =3.2%, this section is equipped with a three-phase imbalance adjustment device, the current active power adjustment is 3.2%. =12kW, imbalance adjustment coefficient =0.08, then the negative sequence voltage imbalance degree after treatment is:

[0070] This section is about The contribution is

[0071] Sub-objective of minimizing line active power loss Calculation: Assume the real-time load current of the first feeder section =150.1A, Identify resistor =0.082Ω, then the three-phase active power loss is:

[0072] This section is about The contribution is this value. The remaining segments are calculated in the same way.

[0073] It should be noted that the dimensions of each sub-objective function are different. In the specific calculation, per-unit processing can be used to eliminate the influence of the difference in dimensions on the weight allocation (such as using the reference voltage, reference distortion rate, reference current, etc. as the per-unit reference). Those skilled in the art can flexibly select the per-unit reference value according to the actual engineering needs, without affecting the substantive protection scope of the method of the present invention.

[0074] Step 5: Construction of a Multi-Objective Robust Comprehensive Optimization Model The weights of the four sub-objectives are as follows: =0.30 (voltage offset). =0.25 (harmonic distortion), =0.20 (three-phase imbalance). =0.25 (line loss), which meets the requirements. .

[0075] In this embodiment, the photovoltaic penetration rate in the distribution area is low and the load fluctuation is stable, so the robust adjustment coefficient Γ is set to 0.12. Let the maximum deviation of the objective function within the load forecast fluctuation range be... The estimated value is 0.15 (per unit).

[0076] The overall minimized objective function is:

[0077] Step Six: Set up multi-dimensional electrical safety and equipment constraints In this embodiment, the constraints are set as follows: Regarding equipment capacity constraints, the reactive power compensation output of each SVG segment... It must not exceed its rated maximum output. (Section 1 in this embodiment) =120kvar, Section 2 =80kvar, Section 3 =100kvar, Section 4 =60kvar, Section 5 =150kvar); Harmonic mitigation capacity of each segment's APF (Amplifier Power Filter) It must not exceed the rated maximum value. (Section 1 in this embodiment) =50A, paragraph 2 =30A, Section 3 =40A, paragraph 4 =25A, Paragraph 5 =60A); Three-phase imbalance adjustment in each section It must not exceed the rated maximum value. (This embodiment is calculated based on device capacity).

[0078] Grid voltage safety constraints: voltage at the end of each segment Must meet:

[0079] That is, 372V≤ ≤428V.

[0080] Overall power factor constraint for the transformer area: The overall power factor of the transformer area must not be lower than 0.9, that is:

[0081] in The active power of the original load in each section, This is the original reactive power. Provides power for SVG compensation.

[0082] Weight normalization constraint:

[0083] In this embodiment, the original load power of each segment is obtained by sampling terminal measurement and used as known input parameters for the optimization model.

[0084] Step 7: Solve for local resource allocation and output An improved multi-objective particle swarm optimization algorithm was used to solve the above optimization model. The algorithm parameters were set as follows: particle population size was 100, maximum number of iterations was 500, and convergence accuracy was set to less than 1 × 10⁻⁶ for the objective function change. 5 Initialize the particle population, use the output of each governance device as the decision variable, and substitute it into the weight coefficients, objective function and constraints from steps three to six for iterative solution.

[0085] During the iteration process, after each update of particle position and velocity, all constraints in step six are verified, and particles that do not meet the voltage boundary or exceed the capacity limit are eliminated or subjected to boundary constraint processing. After 146 iterations, convergence accuracy is achieved, and the optimal power output combination for each segment is output as follows: Paragraph 1: =68.5kvar =32.4A, =14.2kW; Paragraph 2: =52.3kvar, =21.8A, =18.7kW; Paragraph 3: =41.2kvar =19.6A, =6.8kW; Paragraph 4: =45.8kvar =20.3A, =25.6kW; Paragraph 5: =28.6kvar =12.7A, =3.5kW.

[0086] As can be seen from the power distribution results, the three-phase imbalance regulation amount (25.6kW) obtained by the fourth feeder (with the largest impedance and a weight of 0.382) is significantly higher than that of the fifth feeder (with the smallest impedance and a weight of 0.091, and a regulation amount of 3.5kW), which fully demonstrates the adaptive allocation and control effect of impedance weight on the governance resources.

[0087] Step 8: Dynamic Iterative Adaptive Control The edge computing terminal iterates through steps one through seven every 30 seconds. In each cycle, it re-collects real-time electrical parameters for each segment, updates impedance identification results, adjusts weighting coefficients, optimizes the model, and issues new control commands. Simultaneously, it compares the post-treatment power quality indicators of the current cycle with those of the previous cycle. If harmonic distortion rate or voltage deviation shows a worsening trend, it fine-tunes the convergence accuracy or iteration step size in the model, ensuring the control strategy continuously tracks changes in transformer load and equipment status, achieving long-term dynamic adaptive optimization.

[0088] After the actual transformer substation was put into operation, continuous monitoring was conducted for 72 hours. The results showed that the voltage of each segment was stable within the range of 380V~415V, meeting the national standard requirement of ±7%; the average total harmonic distortion rate of each segment decreased from 4.6% before treatment to 2.1%; the average negative sequence voltage imbalance of each segment decreased from 2.8% before treatment to 1.2%; the total active power loss of the transformer substation decreased from 18.7kW before treatment to 15.2kW, a reduction of approximately 18.7%; and the comprehensive utilization rate of the capacity of each treatment device increased from an average of 54% before treatment to 79%.

[0089] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for local allocation of resources for robust multi-objective power quality management based on adaptive impedance matching of feeder segments in a transformer substation, characterized in that, The method includes: Step 1: Adaptive segmentation and global synchronous sampling of transformer feeder topology: Based on the electrical nodes of the transformer area, the transformer feeder is automatically divided into N feeder segments with uniform electrical parameters, and a synchronous sampling terminal is configured at both ends of each feeder segment to collect the three-phase fundamental voltage phasor and three-phase fundamental current phasor of the beginning and end of each segment in real time. Step 2: Construct an online accurate identification model for feeder segment impedance: Based on the three-phase fundamental voltage phasors and three-phase fundamental current phasors at the beginning and end of the feeder, solve for the equivalent resistance of each feeder segment. Equivalent reactance and combined impedance amplitude ; Step 3: Establish an impedance adaptive governance weight normalization configuration model: based on the comprehensive impedance magnitude of each feeder segment. Calculate the proportion of each feeder segment to the total impedance of the entire feeder network, and obtain the governance weight coefficient for each feeder segment. ; Step 4: Construct four-dimensional power quality refinement sub-objective functions: Construct sub-objectives for minimizing voltage offset F1, minimizing harmonic distortion F2, minimizing three-phase imbalance F3, and minimizing line active power loss F4 respectively; Step 5: Construction of a multi-objective robust integrated optimization model: A linear weighted method is adopted, and a robust perturbation interval is introduced. Based on the four sub-objective functions in step four, a comprehensive minimization objective function is constructed. ; Step 6: Set multi-dimensional electrical safety and equipment constraints: Set constraints on the capacity of the treatment equipment, the voltage safety of the power grid, the overall power factor of the distribution area, and the weight normalization constraint; Step 7: On-site resource allocation solution output: Based on the comprehensive minimization objective function constructed in Step 5 and the constraints set in Step 6, the improved multi-objective particle swarm optimization algorithm is used for iterative solution to output the optimal treatment output combination of on-site treatment equipment for each feeder segment; Step 8: Dynamic Iterative Adaptive Control: Repeat steps 1 to 7 at a preset cycle, and iteratively update the model parameters based on the governance effect feedback information of the previous cycle to achieve long-term dynamic adaptive optimization control.

2. The method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of transformer feeder segments according to claim 1, characterized in that, In step one, the segmentation rule is specifically as follows: Taking the main switch of the power distribution room as the starting node of the feeder, each load branch access point, cable specification change node, line joint node and end load node is used as the segment boundary node, and the entire transformer area feeder is divided into N feeder segments. The sampling terminal is equipped with a Beidou / GPS dual-mode time synchronization module, with a sampling time error of less than or equal to 1 microsecond and a sampling frequency of 10 kHz.

3. The method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of feeder segments in a transformer substation, as described in claim 1, is characterized in that... In step two, for the i-th feeder segment, the three-phase voltage phasor at its starting end is: The phasors of the three-phase current at the beginning are: The three-phase voltage phasors at the end are The phasors of the three-phase currents at the end are Then the voltage drop phasor of that segment ;make Then the equivalent resistance and equivalent reactance The calculation formula is: The combined impedance magnitude The calculation formula is: 。 4. The method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of feeder segments in a transformer substation, as described in claim 3, is characterized in that... In step three, the governance weight coefficient The calculation formula is: The sum of the governance weight coefficients of all feeder segments satisfies .

5. The method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of feeder segments in a transformer substation, as described in claim 4, is characterized in that... In step four, the voltage offset minimization sub-target The expression is: in This represents the real-time operating voltage at the end of the i-th feeder segment. This is the rated voltage for the transformer substation.

6. The method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of transformer feeder segments according to claim 5, characterized in that, In step four, the harmonic distortion minimization sub-objective The expression is: in Let be the total harmonic distortion rate of the real-time voltage of the i-th feeder segment, and Satisfy the dynamic correction relationship: in To address the original harmonic distortion rate, The equivalent suppression coefficient for harmonic control. Real-time capacity input for harmonic mitigation equipment in the i-th feeder segment.

7. The method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of feeder segments in a transformer substation, as described in claim 6, is characterized in that... In step four, the sub-objective of minimizing the three-phase imbalance is mentioned. The expression is: in, Let be the real-time negative sequence voltage imbalance of the i-th feeder segment, and Satisfy the dynamic correction relationship: In the formula For the original imbalance, This is the imbalance adjustment coefficient. The real-time active power distribution adjustment amount for the three-phase imbalance adjustment device.

8. A method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of feeder segments in a transformer substation, as described in claim 7, is characterized in that... In step four, the sub-objective of minimizing the active power loss of the line is mentioned. The expression is: in, This represents the real-time load current of the i-th feeder segment. The equivalent resistance is identified in real time for the i-th feeder segment.

9. A method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of feeder segments in a transformer substation, as described in claim 8, is characterized in that... In step five, the comprehensive minimization objective function The expression is: in , , , These are the weight coefficients for each sub-objective, satisfying... Γ is the robust adjustment coefficient, ranging from 0 to 0.3; This represents the maximum deviation of the comprehensive objective function within the load forecast fluctuation range.

10. A method for local allocation of resources for multi-objective robust power quality management based on adaptive impedance matching of feeder segments in a transformer substation, as described in claim 9, is characterized in that... In step seven, the optimal governance output combination is { , , },in The optimal output for reactive power compensation of the SVG in the i-th feeder segment. To determine the optimal capacity for harmonic mitigation of the APF in the i-th feeder segment, The optimal active power regulation for the three-phase imbalance regulation of the i-th feeder segment is defined as follows: During the iterative process of the improved multi-objective particle swarm algorithm, all constraints in step six are verified and illegal solutions are eliminated in each iteration until the convergence accuracy is met or the maximum number of iterations is reached, at which point the global optimal solution is output.

Images