Flexible interconnected power distribution network active-reactive optimization method considering power quality demand

Abstract

The invention discloses a flexible interconnected power distribution network active-reactive optimization method considering electric energy quality requirements, and belongs to the technical field of power distribution network optimization operation. The invention aims to provide the active-reactive optimization method for the flexible interconnected power distribution network considering the electric energy quality demand, which simultaneously optimizes the voltage deviation, feeder balance and network loss of the power distribution network through various control means such as soft switching power regulation participation response. The method comprises the following steps: proposing a flexible interconnection system power quality demand subjective and objective weighting evaluation method through calculation of a flexible interconnection system day-ahead index time sequence value, and proposing a flexible interconnection power distribution network active-reactive power optimization model considering a node comprehensive weight through coordinated operation of a traditional network side resource and an intelligent soft switch. According to the invention, the active-reactive optimization method of the flexible interconnection power distribution network considering the power quality demand is analyzed, the effective cooperation of the flexible interconnection device and the controllable network side resource regulation and control is explored, and the influence on the voltage deviation, feeder balance and network loss of the power distribution network is optimized by multiple control means at the same time; and reference is expected to be provided for guiding optimal scheduling of the flexible interconnected power distribution network.

Description

Technical Field

[0001] The invention belongs to the technical field of optimized operation of distribution networks. Background Art

[0002] Due to the constraints of short-circuit capacity and electromagnetic ring network, traditional distribution networks usually adopt closed-loop design and open-loop operation mode, so that the flow is naturally distributed under the influence of network structure parameters and load demand. Only through switch operation can the flow distribution be adjusted to a certain extent. Such a "passive" physical structure and operation mode has low power supply reliability and weak flow control ability, and is prone to problems such as line capacity overload and node voltage exceeding the limit, which greatly limits the access of various types of source-load-storage equipment.

[0003] Flexible interconnected distribution networks are characterized by high complexity and polymorphism in operation, especially the uncertainty of the output of a high proportion of distributed power sources, which easily leads to voltage over-limit and light or heavy load problems on feeders. Traditional flexible distribution network optimization strategies often lack the comprehensive utilization of multiple resources in the entire system, and lack the consideration of scheduling from the perspectives of traditional control equipment and emerging intelligent soft switch equipment, while meeting the differentiated high-quality electricity needs of highly permeable and decentralized sensitive users of distribution networks. However, the individual controllable resources considered in existing studies are not comprehensive enough, or the models used are too simple, which cannot guarantee the reliability and economy of the distribution network. Therefore, effective coordination of flexible interconnected devices and controllable grid-side resource control is required, while considering the interaction and coordination between individual controllable resources in the distribution network to achieve multi-level optimized operation of flexible interconnected distribution networks. Summary of the invention

[0004] The purpose of the present invention is to optimize the active-reactive power optimization method of a flexible interconnected distribution network taking into account power quality requirements by simultaneously optimizing distribution network voltage deviation, feeder balancing and network loss through multiple control means such as soft switch power regulation and participation response.

[0005] The steps of the present invention are: 1. The power quality demand assessment index system includes seven indicators: four indicators of the grid side affecting the load side: voltage deviation rate, voltage fluctuation rate, harmonic voltage distortion rate and three-phase imbalance; three indicators of the load side affecting the grid side: user load level, user satisfaction and power consumption; S1, voltage deviation rate: Where: U i is the real-time voltage of node i in the flexible interconnected distribution network; U N is the nominal voltage value; S2, voltage fluctuation rate: Where: ΔU is the difference between adjacent extreme values ​​on the voltage RMS curve; UN is the nominal voltage value; S3, harmonic voltage distortion rate: Where: U h is the hth harmonic voltage; THD u is the total distortion rate of harmonic voltage; U is the effective value of fundamental voltage; S4, three-phase imbalance: Where: is the negative sequence voltage unbalance; U 1 is the effective value of the voltage positive sequence fundamental component; U 2 is the effective value of the voltage negative sequence fundamental component; S5. User load level: It is classified into three levels according to the user's power consumption, demand and sensitivity to power quality; S6. User satisfaction: It involves users’ perception and feedback on power outage frequency, fault recovery time, power supply quality, etc., and is divided into five levels; S7. Power consumption: The combined weighting method of improved entropy weight method and improved analytic hierarchy process is used to weight each indicator, so that the weight contains comprehensive guidance information of objective attributes and subjective experience; S71. Improved entropy weight method: The entropy value H of each power quality demand evaluation index j : Where: f ij is the relative weight of the ith object on index j; Use anti-entropy to get the initial weight ω 0 : Improved weight ω 3 Defined as: Where s is the average value of all entropy values ​​that are not 1; The final form of the weight ω(j) is: S72. Improved analytic hierarchy process: a ij Indicates the comparison result when the importance of indicator i is higher than that of indicator j: Quantify the difference in importance between two metrics: Quasi-optimal consistent matrix d: The weight vector is obtained by calculating the eigenvector of the quasi-optimal consistent matrix: Finally, normalization is performed so that the total weight sum is equal to 1: S73, combined weighting method to weight each indicator: Second, the optimization model takes the weighted minimum of the voltage deviation index, feeder balance index and distribution network loss index of the partition as the objective function: min f=λ 1 f V +λ 2 f B +λ 3 f Loss (17) Where: 1 ,λ 2 and λ 3 are the weights of voltage deviation index, feeder balance index and distribution network loss index respectively; f V , f B and f Loss They are the voltage deviation index, feeder balancing index and distribution network loss index considering the partition respectively; S1. Voltage deviation index considering partitions Where: α i is the importance coefficient of the i-th node; Ω n is a set of nodes; Ω T is the set of optimization periods; S2, feeder balance index Where: P i,t is the load of the ith feeder; n is the number of feeder lines in the flexible interconnected distribution network; S3, distribution network loss index Where: P 0Z Indicates the transformer no-load loss, kW; P KZ Indicates the rated load power loss of the transformer, kW; F Loss is the total loss of the flexible interconnected distribution network; I ij,t is the current flowing through branch ij at time t; r ijis the equivalent resistance of branch ij; η is the power transfer efficiency of soft switching, which is generally 2%; is the power transmission loss of the i-th two-terminal soft switch at time t; β is the load factor of the transformer, and T is the number of time periods; 3. Model constraints mainly include soft switch constraints, power flow constraints, on-load tap-changing transformer gear constraints, capacitor bank constraints and transformer constraints; S1, soft switching constraints Where: is the capacity of the ith VSC port of the smart soft switch in the distribution network; N is the set of all ports of the SOP; η is the loss coefficient of the VSC port, which is usually 0.02; are the active power and reactive power flowing through the i-th VSC port respectively; S2. Transformer Constraints 0.25≤β i ≤0.75 (25) Where: P t is the active load transmitted by the distribution transformer at time t, S t is the rated capacity of the transformer, β t The load factor of the transformer; S3, On-load tap-changing transformer constraints k ij,t =1+K ij,t Δk oltc K ij,t ∈Z (28) In the formula, r ij and x ij is the resistance and reactance of the branch at the head node i and the terminal node j; P ij and Q ij are the active and reactive powers flowing through branch ij respectively; V i,t,s is the voltage of node i; I ij,t,s is the current flowing through branch ij; k ij,t K is the voltage ratio of the OLTC installed in branch ij; ij,t is the adjustment gear of OLTC; Δk oltc is the OLTC unit adjustment step size; Adjust the maximum value for OLTC gear; S4. Capacitor bank constraints: Where: is the number of CB groups put into use; is the capacity of a single group of CBs; The maximum number of CB groups that can be put into use; S5, power flow constraint Where: X ij is the reactance of branch ij; P t,ij is the active power flowing from node i to node j on branch ij at time t; Q ij is the reactive power flowing from node i to node j on branch ij at time t; P t,j is the active power injected into node j at time t; Q t,j is the reactive power injected into node j at any moment; is the active power injected into the node by the distributed generation connected to node j at time t; is the reactive power injected into node j by the distributed generation connected to node j at time t; is the active power injected into node j by SOP at time t; is the reactive power injected into node j by SOP at time t; is the active power consumed by the load on node j at time t; is the reactive power consumed by the load on node j at time t; U t,i , U t,j are the voltage amplitudes of nodes i and j at time t, U , are the minimum and maximum voltage values ​​of node i under normal conditions, Maximum value of branch current.

[0006] The present invention analyzes the active-reactive optimization method of a flexible interconnected distribution network considering power quality requirements, explores the effective coordination of flexible interconnected devices and controllable grid-side resource regulation, and simultaneously optimizes the impact of multiple control means on distribution network voltage deviation, feeder balance and network loss. It is hoped that this will provide reference and guidance for the optimal scheduling of flexible interconnected distribution networks. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 It is to improve the topology diagram of IEEE33 node flexible interconnected distribution network; Figure 2 It is a distribution network optimization framework diagram considering power quality requirements; Figure 3 It is the voltage deviation rate and voltage fluctuation rate diagram; Figure 4 It is the harmonic voltage distortion rate diagram; Figure 5 It is the distributed power output diagram; Figure 6 is the original feeder load factor diagram of the distribution network; Figure 7 It is the SOP power transfer diagram; Figure 8 This is the result diagram of OLTC and CB gear optimization; Fig. 9 It is a comparison diagram of voltage deviation of distribution network; Fig.10 It is the feeder load rate diagram after distribution network optimization. DETAILED DESCRIPTION

[0008] The present invention proposes an active-reactive optimization method for a flexible interconnected distribution network that takes into account power quality requirements. In order to reduce voltage deviation and achieve load balancing of feeders in a targeted manner, the present invention simultaneously optimizes distribution network voltage deviation, feeder balancing and network loss through multiple control means such as soft switch power regulation and response, which significantly improves the stability and reliability of the flexible interconnected distribution network.

[0009] The method of the present invention fully considers improving the power quality of the distribution network, evaluates the power quality requirements, reasonably allocates voltage resources, ensures system stability, improves power supply reliability, and ultimately meets the user's personalized needs for power quality, thereby improving the operating efficiency of the entire distribution network.

[0010] The problem described in the present invention is achieved through the following technical solutions: (1) Construct an index system for evaluating the power quality demand of sensitive users, and calculate the day-ahead index time series value of the flexible interconnection system.

[0011] (2) Based on the single power quality demand assessment data, a subjective and objective weighted assessment method for the power quality demand of the flexible interconnected system is proposed.

[0012] (3) Through the coordinated operation of traditional grid-side resources and intelligent soft switches, an active-reactive optimization model of a flexible interconnected distribution network considering the comprehensive weight of nodes is proposed.

[0013] The active-reactive optimization method of the flexible interconnected distribution network considering the power quality requirements proposed by the present invention is: In flexible interconnected distribution networks, voltage source converters are usually used in a back-to-back form to form intelligent soft switches, and flexible power transmission is used to achieve load balancing and voltage deviation reduction. The optimization strategy of flexible distribution networks requires the comprehensive utilization of multiple resources in the entire system, and considers scheduling from the perspectives of traditional control resources and intelligent soft switches.

[0014] Ordinary users and sensitive users are mixed together in the distribution network. It is uneconomical and unnecessary to govern the entire area to meet the high-quality electricity demand of sensitive users. Therefore, in view of the problem of sensitive user needs in the flexible interconnected distribution network, the present invention selects four indicators that affect the load side from the grid side: voltage deviation rate, voltage fluctuation rate, harmonic voltage distortion rate and three-phase imbalance, and selects three indicators that affect the grid side from the load side: user load level, user satisfaction and electricity consumption. The index value of each node in the flexible interconnected distribution network is calculated, and the combined weighting method based on the improved entropy weight method and the improved hierarchical analysis method is used to weight each indicator so that the weight contains comprehensive guidance information of objective attributes and subjective experience. The comprehensive weight of the power quality demand of each node is calculated to reflect the requirements of the node for power quality. The importance is partitioned by fuzzy processing to ensure that the power quality demand in the area is accurately defined. Then the node weight is transmitted to the active-reactive optimization model of the flexible interconnected distribution network. According to the day-ahead forecast data of loads and distributed power sources, the flexible interconnected distribution network optimization model takes the weighted minimum of the voltage deviation index, feeder balance index and distribution network loss index (including network loss, soft switch loss and transformer loss, etc.) of the partition as the objective function, and combines the comprehensive weight of the nodes to formulate the voltage regulation equipment switching strategy and intelligent soft switch port output that meet the needs of sensitive users. Optimize the distribution network flow, improve the voltage deviation and feeder balance of the distribution network, and reduce operating losses. Rationally allocate resources to ensure system stability and improve power supply reliability, and ultimately meet users' personalized needs for power quality, thereby improving the operating efficiency and service quality of the entire distribution network.

[0015] The power quality demand evaluation index system considering sensitive users proposed in the present invention is: Power quality refers to the quality of electric energy transmitted by the power system. On a macro level, it includes three aspects: voltage quality, frequency quality, and waveform quality. Most of the existing comprehensive evaluation methods for power quality only consider five commonly used technical indicators, namely voltage deviation, frequency deviation, voltage fluctuation and flicker, harmonics, and three-phase voltage imbalance. Ordinary users and sensitive users are mixed together in the distribution network. It is uneconomical and unnecessary to govern the entire area to meet the high-quality electricity demand of sensitive users. Therefore, in order to address the problem of considering the needs of sensitive users in the flexible interconnected distribution network, this paper selects four indicators that affect the grid side on the load side: voltage deviation rate, voltage fluctuation rate, harmonic voltage distortion rate, and three-phase imbalance, and selects three indicators that affect the load side on the grid side: user load level, user satisfaction, and power consumption.

[0016] The calculation methods of the above 7 evaluation indicators are as follows: 1) Voltage deviation rate. Due to changes in power flow and load, the actual voltage deviates from the rated voltage, causing voltage deviation problems. The voltage deviation rate refers to the percentage of the difference between the actual operating voltage of the system and the standard voltage specified by the system as a percentage of the system standard voltage, that is: Where: U i is the real-time voltage of node i in the flexible interconnected distribution network; U N is the nominal voltage value.

[0017] 2) Voltage fluctuation rate. The voltage in the power system is constantly changing. Voltage fluctuation refers to the phenomenon that the voltage root mean square value deviates from the rated voltage in a continuous period of time. The voltage fluctuation is calculated by taking the difference between the adjacent maximum and minimum values ​​on the voltage root mean square value curve and comparing the difference with the nominal voltage: Where: ΔU is the difference between adjacent extreme values ​​on the voltage RMS curve; U N is the nominal voltage value.

[0018] 3) Harmonic voltage distortion rate. Harmonics in power systems are generally caused by the use of nonlinear loads and nonlinear devices. If the harmonic content of the system is high, the electric energy waveform will be greatly distorted. The distortion of a certain harmonic can be expressed by the ratio of the effective value of the harmonic to the fundamental wave, that is: Where: U h is the hth harmonic voltage; THD u is the total distortion rate of harmonic voltage; U is the effective value of fundamental voltage.

[0019] 4) Three-phase imbalance. When the amplitude and phase of voltage and current do not meet the conditions of three-phase balance, three-phase imbalance occurs. The degree of three-phase voltage imbalance in the power system is expressed by three-phase imbalance. The imbalance is divided into negative sequence and zero sequence, namely: Where: is the negative sequence voltage unbalance; U 1 is the effective value of the voltage positive sequence fundamental component; U 2 It is the effective value of the negative sequence fundamental component of voltage.

[0020] 5) User load level. Different users have different requirements for power quality and different impacts on the power supply system. Load level helps identify which users are "large users" or "sensitive users". Users are classified into three levels according to their power consumption, demand, and sensitivity to power quality. Level 1 load users have extremely high requirements for power quality. Level 2 load users can withstand short-term power quality disturbances, but still need reliable power guarantee. Level 3 load users have relatively low requirements for power quality.

[0021] 6) User satisfaction. User satisfaction involves the user's perception and feedback on power outage frequency, fault recovery time, power supply quality, etc. User satisfaction is used to measure the user's satisfaction with the current power service and is divided into five levels. The larger the satisfaction value, the more satisfied the user is and the lower the demand for power quality regulation. High satisfaction usually means that the user is satisfied with the stability, reliability and power quality of the power supply and believes that their needs have been fully met.

[0022] 7) Electricity consumption. Electricity consumption reflects the actual demand and consumption of electricity by users, which helps to predict demand and detect anomalies. At the same time, electricity consumption is also an important measurement indicator in the differentiated electricity sales of distribution networks.

[0023] The subjective and objective weighted evaluation method for the power quality demand of the flexible interconnection system proposed in the present invention is: The present invention adopts a combined weighting method of improved entropy weight method and improved hierarchical analysis method to weight each indicator, so that the weight contains comprehensive guidance information of objective attributes and subjective experience.

[0024] 71) Improved entropy weight method In the traditional entropy weight method, the calculation of weight mainly depends on the entropy value H j The reflection, that is, through 1-H j To determine the importance of an indicator. However, traditional methods may adjust all indicators equally or too simply in some cases, ignoring the complex relationship between different indicators. Therefore, this paper proposes an improved entropy weight method to more effectively evaluate the importance of each indicator.

[0025] First, f ij is the relative weight of the ith object on the index j, and the entropy value H of each power quality demand evaluation index is calculated. j :

[0026] Then use the anti-entropy value to calculate the initial weight ω 0 , in order to take into account the differences:

[0027] When calculating the index weight of entropy value in traditional entropy weight method, a small change will cause the entropy weight to change exponentially. Considering the differentiated effects of each index, the improved weight ω 3 is defined as: Where s is the average of all entropy values ​​that are not 1.

[0028] The final form of the weight ω(j) is:

[0029] 72) Improved analytic hierarchy process The Analytic Hierarchy Process (AHP) is an effective tool for multi-criteria decision analysis. However, the traditional AHP method has the disadvantages of being highly subjective and susceptible to consistency. In order to solve these problems, this paper proposes an improved AHP method that uses the calculation of a quasi-optimal consensus matrix to improve weight accuracy and decision consistency.

[0030] First, we need to evaluate the relative importance of the indicators. ij Indicates the comparison result when the importance of indicator i is higher than that of indicator j. This step helps to identify the priority of key indicators:

[0031] The difference between the two indicators is processed by logarithmic transformation. This transformation can reduce the impact of extreme values ​​and make the difference smoother and more relative in the calculation process. By calculating the transfer matrix c, the difference in importance between the two indicators can be quantified:

[0032] The quasi-optimal consistent matrix d takes into account all differences in the data set by adjusting the transitivity of each element. This step improves the consistency of the comparison matrix and reduces the deviation caused by subjective judgment. In order to further improve the consistency, the quasi-optimal consistent matrix d is calculated:

[0033] Then, the geometric mean method is used to perform square root processing on the accumulated comparison information, and the result is a eigenvector representing the relative importance of each indicator. The weight vector is obtained by calculating the eigenvector of the quasi-optimal consensus matrix:

[0034] Finally, normalization is performed so that the total weight sum is equal to 1:

[0035] Combination weighting method In order to make up for the shortcomings of a single weighting method, the weighting method that combines the two weighting methods is called the combined weighting method. The combined weighting method that combines the improved entropy weight method and IAHP (i.e. improved analytic hierarchy process) realizes the internal unity of subjective and objective factors:

[0036] Power Quality Needs Assessment The 7 proposed power quality demand assessment indicators can be divided into extremely large and extremely small according to their own attributes. Extremely large means that the larger the value, the more important the evaluation result; extremely small means that the smaller the value, the more important the evaluation result. This paper uses mathematical transformation to transform the indicators into the same type of indicators and converts the indicator value range to [0,1]. The closer the indicator is to 1, the higher the power quality demand value. The objective weight and subjective weight are calculated respectively by the two methods mentioned above, and then the weight of the combined weighting method is calculated. Based on the power quality demand value of the evaluation indicator, the final score of each evaluation indicator is calculated. According to the preset score limit, the importance coefficient of each node is determined.

[0037] The active-reactive optimization model of the flexible interconnected distribution network considering the comprehensive weight of nodes proposed in the present invention is: As a new type of power electronic device that can replace the tie switch, the intelligent soft switch (Soft Open Point, SOP) can be connected to the distribution network to realize the functions of accurate control of feeder power flow, dynamic reactive power compensation and flexible switching of power supply mode. Considering that the current cost of SOP is still high, it is necessary to study the coordinated optimization of SOP and traditional control equipment to achieve economical and safe operation of distribution network.

[0038] The flexible interconnected distribution network optimization model takes the weighted minimum of the voltage deviation index, feeder balance index and distribution network loss index (including network loss, soft switch loss and transformer loss, etc.) of the partition as the objective function, as shown in the following formula: minf=λ 1 f V +λ 2 f B +λ 3 f Loss (17) Where: 1 ,λ 2 and λ 3 are the weights of voltage deviation index, feeder balance index and distribution network loss index respectively; f V , f B and f Loss They are the voltage deviation index, feeder balancing index and distribution network loss index considering the partition respectively.

[0039] 1) Consider the voltage deviation index of the partition Partitioning means that the flexible interconnected distribution network evaluates the power quality demand coefficient of each node, performs fuzzy processing to partition, and divides it into three levels: "important, general, and unimportant". The voltage deviation index considering the partition is expressed as the voltage deviation of the distribution network calculated in combination with the importance of the region. The calculation formula is as follows: Where: α i is the importance coefficient of the i-th node; Ω n is a set of nodes; Ω T It is a set of optimization periods.

[0040] 2) Feeder balance index The feeder balancing index is used to measure the imbalance of the loads of each feeder at a certain time t. The larger the value, the stronger the imbalance of the feeder load. The calculation formula is as follows: Where: P i,t is the load of the i-th feeder; n is the number of feeder lines in the flexible interconnected distribution network.

[0041] 3) Distribution network loss index In order to ensure the economic operation of the distribution network, the distribution network loss index is an indicator used to evaluate and quantify the energy loss in the power system, including network loss, soft switch loss and transformer loss, that is, Where: P 0Z Indicates the transformer no-load loss, kW; P KZ Indicates the rated load power loss of the transformer, kW; F Loss is the total loss of the flexible interconnected distribution network; I ij,t is the current flowing through branch ij at time t; r ij is the equivalent resistance of branch ij; η is the power transfer efficiency of soft switching, which is generally 2%; is the power transmission loss of the ith double-ended soft switch at time t; β is the load factor of the transformer. T is the number of time periods, which is 24.

[0042] The model constraints mainly include soft switch constraints, power flow constraints, on-load tap-changing transformer gear constraints, capacitor bank constraints and transformer constraints, as follows: 1) Soft switching constraints The intelligent soft switch consists of two VSC ports. The sum of the power of each port of the SOP satisfies Kirchhoff's equation, and the transfer power must meet the port capacity constraint: Where: is the capacity of the ith VSC port of the smart soft switch in the distribution network; N is the set of all ports of the SOP; η is the loss coefficient of the VSC port, which is usually 0.02; are the active power and reactive power flowing through the i-th VSC port respectively.

[0043] 2) Transformer constraints In order to avoid the phenomenon of light or heavy load operation of distribution transformers, the upper and lower limits of transformer load rate should not exceed its optimal economic operation range. Therefore, the optimal economic operation range constraint of transformer is: 0.25≤β i ≤0.75 (25) Where: P t is the active load transmitted by the distribution transformer at time t, S t is the rated capacity of the transformer, β t The load factor of the transformer.

[0044] 3) OLTC constraints and CB constraints An On-Load Tap Changer (OLTC) is a device used to adjust the output voltage of a transformer, which can change the tap of the transformer without unloading. The constraints mainly involve the adjustment range of the tap, which can only be adjusted within a specific voltage range. k ij,t =1+K ij,t Δk oltc K ij,t ∈Z (28) In the formula, r ij and x ij is the resistance and reactance of the branch at the head node i and the terminal node j; P ij and Q ij are the active and reactive powers flowing through branch ij respectively; V i,t,s is the voltage of node i; I ij,t,s is the current flowing through branch ij; k ij,t K is the voltage ratio of the OLTC installed in branch ij; ij,t is the adjustment gear of OLTC; Δk oltc is the OLTC unit adjustment step size; Adjust the maximum value for the OLTC gear.

[0045] Capacitor Bank (CB) is a commonly used device in power systems to improve power factor, regulate voltage and reduce reactive power loss in the network. They are usually composed of multiple capacitors that can be connected or removed according to demand to adjust the reactive power of the system. Where: is the number of CB groups put into use; is the capacity of a single group of CBs; The maximum number of CB groups that can be put into use.

[0046] 4) Power flow constraints Where: X ij is the reactance of branch ij; P t,ij is the active power flowing from node i to node j on branch ij at time t; Q ij is the reactive power flowing from node i to node j on branch ij at time t; P t,j is the active power injected into node j at time t; Q t,j is the reactive power injected into node j at any moment; is the active power injected into the node by the distributed generation connected to node j at time t; is the reactive power injected into node j by the distributed generation connected to node j at time t; is the active power injected into node j by SOP at time t; is the reactive power injected into node j by SOP at time t; is the active power consumed by the load on node j at time t; is the reactive power consumed by the load on node j at time t; U t,i , U t,j are the voltage amplitudes of nodes i and j at time t respectively.

[0047] Where: U , are the minimum and maximum voltage values ​​of node i under normal conditions, Maximum value of branch current.

[0048] The simulation analysis is carried out using the improved IEEE33 node flexible interconnected distribution network. Figure 1As shown. The improved IEEE33-node distribution network includes a 110 / 10kV transformer, nodes 12-node 22 and nodes 18-node 33 are connected to the feeder through two double-terminal intelligent soft switches, nodes 13 and 17 are connected to wind power and photovoltaic power respectively, and node 15 is connected to a capacitor bank. Matlab R2023a is used to call the existing commercial solver Cplex to complete the simulation to verify the effectiveness of the strategy proposed in this invention.

[0049] Figure 2 The framework diagram of distribution network optimization considering power quality demand is presented, showing the overall process of the method, including the core steps of power quality demand assessment and optimization strategy. The power quality demand assessment consists of an evaluation index system and subjective and objective weighted assessment. The importance is partitioned through fuzzy processing to ensure that the power quality demand in the region is accurately defined. Then the node weights are transmitted to the active and reactive power optimization model part of the flexible interconnected distribution network. The model considers factors such as load forecasting, node weights, and historical data to formulate a comprehensive strategy to optimize the distribution network flow, improve the voltage deviation and feeder balance of the distribution network, and reduce operating losses. It effectively improves the operating efficiency and power quality of the distribution network, and ensures the reliable and stable operation of the power system.

[0050] Power quality demand assessment indicators such as Figure 3 and Figure 4 As shown, Figure 3 The voltage deviation and voltage fluctuation of each node are calculated. In the figure, the blue line represents the voltage deviation rate, and the red line represents the voltage fluctuation rate. These two indicators clearly reflect the power quality requirements of the distribution network. The voltage deviation gradually increases from node 1, showing a significant upward trend to the peak of node 18, and then starts to increase again at node 19. This shows that the voltage regulation demand is high in the area near node 18 and node 33, and the deviation can be effectively controlled through the optimization strategy. In contrast, the voltage fluctuation value is small, indicating that the overall operation of the system is relatively stable. However, the fluctuation amplitude of individual nodes (such as near node 18 and node 33) is slightly higher, and further adjustment strategies are needed to ensure the stability of the voltage. Figure 4 The harmonic voltage distortion rate change trend under different node numbers and time is shown. The analysis results show that the harmonic voltage distortion rate increases significantly near the nodes where renewable power sources are connected (nodes 13 and 17), especially during the daytime high light hours and nighttime high wind energy hours.

[0051] Figure 5It is a distributed power output diagram. Through the analysis of photovoltaic and wind power output within a day, the red line in the figure represents photovoltaic output and the green line represents wind power output. Photovoltaic output gradually increases during the day (6 to 18 hours), reaches a peak and then gradually decreases in the afternoon. This change is closely related to the intensity of sunshine, reflecting the typical daily periodic fluctuations of photovoltaic power generation. Wind power output is relatively unstable, with relatively large fluctuations and frequent fluctuations throughout the 24 hours. This volatility is mainly due to changes in wind speed and has no obvious time regularity. In the optimization of the distribution network, it is necessary to consider both daily periodicity and random fluctuations, and a flexible scheduling strategy should be adopted to balance load demand and ensure the stability and reliability of the power system.

[0052] After calculating the values ​​of each evaluation index, the subjective and objective weighting method is used to evaluate the comprehensive weight value of each node. Table 1 shows the comprehensive weight of the power quality requirements of each node. Node 30 has the highest comprehensive weight (0.8601), which means that this node has the most urgent requirements for power quality. Similarly, the comprehensive weights of nodes 19, 21, and 22 are relatively low, 0.1379, 0.1545, and 0.1570, respectively, indicating that these nodes may allow larger power quality fluctuations or have relatively low requirements for power quality.

[0053] Table 1

[0054] Figure 6 The load rate changes of three feeders in the flexible interconnected distribution network within a day are shown, among which feeder 1 is from node 1 to node 18, feeder 2 is from node 26 to node 33, and feeder 3 is from node 19 to 25. The load rate of feeder 1 is high throughout the day and shows a downward trend at night; the load rate of feeder 2 is in the middle, with a peak from 7 am to 4 pm; the load of feeder 3 is relatively low, reaching a peak from 11 am to 5 pm. The data in the figure shows the load differences of different feeders over time due to different regional electricity demand. The load rates of some feeders are obviously high, while the loads of other feeders are relatively low, showing an imbalance in load distribution. This uneven distribution indicates that resource allocation needs to be optimized to improve the operating efficiency and reliability of the entire distribution network.

[0055] In order to meet the requirements of feeder balancing, two interconnection devices are installed on feeder 1, feeder 2 and feeder 3. Figure 7The power output of the intelligent soft switch in different time periods is shown. The power transmission from node 33 to 18 is positive, and from node 22 to node 12 is positive. SOP (feeder 1-2) has a higher transmission power in the morning and afternoon, and decreases from midnight to early morning. The power output of SOP (feeder 1-3) fluctuates more violently, especially reaching peaks in the early morning and afternoon. This change is to balance the load demand of the three feeders, and to optimize the operating efficiency and stability of the distribution network by flexibly adjusting the power.

[0056] Figure 8 The operating characteristics of the grid voltage regulation equipment are shown in detail, where the green line represents the gear adjustment of the OLTC and the blue line represents the number of capacitor banks put into operation. The OLTC maintains a higher gear from 0:00 to 7:00 and from 13:00 to 22:00, and makes multiple adjustments during this period to balance the voltage and adapt to load fluctuations by changing the tap switch of the transformer. The number of capacitor banks put into operation remains at a low level of about 2 groups from early morning to morning, and increases to 3 groups from 13:00 to 22:00 as the daytime load increases. These adjustments provide sufficient reactive power compensation, optimize voltage levels, and improve grid stability and power quality.

[0057] By comparison, it can be seen that the optimization strategy adopted by this method has achieved remarkable results in the reliability of power grid operation. Fig. 9 The figure shows the change of voltage deviation over time before and after optimization. Before optimization, the voltage deviation fluctuated greatly, with values ​​between 1.0 and 2.0, especially between 16 and 20 hours. After optimization, the sum of voltage deviations decreased significantly and tended to be stable, ranging from 0.5 to 1.5 overall. Fig.10 The load rate distribution of the distribution network feeders after optimization is shown. From the overall trend, the load rate of each feeder fluctuates slightly over time, but generally remains at a relatively balanced level, indicating that after optimization, the load distribution of the distribution network is more balanced, reducing the risk of overload on a certain feeder.

[0058] The present invention proposes an active-reactive optimization method for a flexible interconnected distribution network taking into account power quality requirements. The comprehensive weight of the flexible interconnected system is included in the objective function of the optimization model. By evaluating the power quality requirements, resources are reasonably allocated to ensure system stability and improve power supply reliability. Ultimately, the user's personalized needs for power quality are met, thereby improving the operating efficiency and service quality of the entire distribution network.

Claims

1. A method for optimizing active and reactive power in a flexible interconnected distribution network considering power quality requirements, characterized in that:

1. The power quality demand assessment index system includes seven indicators: four indicators of the grid side affecting the load side: voltage deviation rate, voltage fluctuation rate, harmonic voltage distortion rate and three-phase imbalance; three indicators of the load side affecting the grid side: user load level, user satisfaction and power consumption; S1, voltage deviation rate: Where: U i is the real-time voltage of node i in the flexible interconnected distribution network; U N is the nominal voltage value; S2, voltage fluctuation rate: Where: ΔU is the difference between adjacent extreme values ​​on the voltage RMS curve; U N is the nominal voltage value; S3, harmonic voltage distortion rate: Where: U h is the hth harmonic voltage; THD u is the total distortion rate of harmonic voltage; U is the effective value of fundamental voltage; S4, three-phase imbalance: Where: is the negative sequence voltage imbalance; U1 is the effective value of the voltage positive sequence fundamental component; U2 is the effective value of the voltage negative sequence fundamental component; S5. User load level: It is classified into three levels according to the user's power consumption, demand and sensitivity to power quality; S6. User satisfaction: It involves users’ perception and feedback on power outage frequency, fault recovery time, power supply quality, etc., and is divided into five levels; S7. Power consumption: The combined weighting method of improved entropy weight method and improved analytic hierarchy process is used to weight each indicator, so that the weight contains comprehensive guidance information of objective attributes and subjective experience; S71. Improved entropy weight method: The entropy value H of each power quality demand evaluation index j : Where: f ij is the relative weight of the ith object on index j; Use the anti-entropy value to get the initial weight ω0: The improved weight ω3 is defined as: Where s is the average value of all entropy values ​​that are not 1; The final form of the weight ω(j) is: S72. Improved analytic hierarchy process: a ij Indicates the comparison result when the importance of indicator i is higher than that of indicator j: Quantify the difference in importance between two metrics: Quasi-optimal consistent matrix d: The weight vector is obtained by calculating the eigenvector of the quasi-optimal consistent matrix: Finally, normalization is performed so that the total weight sum is equal to 1: S73, combined weighting method to weight each indicator: Second, the optimization model takes the weighted minimum of the voltage deviation index, feeder balance index and distribution network loss index of the partition as the objective function: min f=λ1f V +λ2f B +λ3f Loss (17) Where: λ1, λ2 and λ3 are the weights of voltage deviation index, feeder balance index and distribution network loss index respectively; f V , f B and f Loss They are the voltage deviation index, feeder balancing index and distribution network loss index considering the partition respectively; S1. Voltage deviation index considering partitions Where: α i is the importance coefficient of the i-th node; Ω n is a set of nodes; Ω T is the set of optimized time periods; S2, feeder balance index Where: P i,t is the load of the ith feeder; n is the number of feeder lines in the flexible interconnected distribution network; S3, distribution network loss index Where: P 0Z Indicates transformer no-load loss, kW; P KZ Indicates the rated load power loss of the transformer, kW; F Loss is the total loss of the flexible interconnected distribution network; I ij,t is the current flowing through branch ij at time t; r ij is the equivalent resistance of branch ij; η is the power transfer efficiency of soft switching, which is generally 2%; is the power transmission loss of the i-th two-terminal soft switch at time t; β is the load factor of the transformer, and T is the number of time periods; 3. Model constraints mainly include soft switch constraints, power flow constraints, on-load tap-changing transformer gear constraints, capacitor bank constraints and transformer constraints; S1, soft switching constraints Where: is the capacity of the ith VSC port of the smart soft switch in the distribution network; N is the set of all ports of the SOP; η is the loss coefficient of the VSC port, which is usually 0.02; are the active power and reactive power flowing through the i-th VSC port respectively; S2. Transformer Constraints 0.25≤β i ≤0.75 (25) Where: P t is the active load transmitted by the distribution transformer at time t, S t is the rated capacity of the transformer, β t The load factor of the transformer; S3, On-load tap-changing transformer constraints k ij,t =1+K ij,t Δk oltc K ij,t ∈Z (28) In the formula, r ij and x ij is the resistance and reactance of the branch at the head node i and the terminal node j; P ij and Q ij are the active and reactive powers flowing through branch ij respectively; V i,t,s is the voltage of node i; I ij,t,s is the current flowing through branch ij; k ij,t K is the voltage ratio of the OLTC installed in branch ij; ij,t is the adjustment gear of OLTC; Δk oltc is the OLTC unit adjustment step size; Adjust the maximum value for OLTC gear; S4. Capacitor bank constraints: Where: is the number of CB groups put into use; is the capacity of a single group of CBs; is the maximum number of CB groups that can be put into use; S5. Power flow constraints Where: X ij is the reactance of branch ij; P t,ij is the active power flowing from node i to node j on branch ij at time t; Q ij is the reactive power flowing from node i to node j on branch ij at time t; P t,j is the active power injected into node j at time t; Q t,j is the reactive power injected into node j at any moment; is the active power injected into the node by the distributed generation connected to node j at time t; is the reactive power injected into node j by the distributed generation connected to node j at time t; is the active power injected into node j by SOP at time t; is the reactive power injected into node j by SOP at time t; is the active power consumed by the load on node j at time t; is the reactive power consumed by the load on node j at time t; U t,i , U t,j are the voltage amplitudes of nodes i and j at time t, U , are the minimum and maximum voltage values ​​of node i under normal conditions, Maximum value of branch current.

Images