Power distribution network voltage treatment method considering photovoltaic uncertainty and F-SOP regulation and control

Abstract

The invention provides a power distribution network voltage treatment method considering photovoltaic uncertainty and F-SOP regulation and control, and the method comprises the steps: recognizing adjustable and controllable resources in a system, and determining the output of each adjustable and controllable resource; constructing a day-ahead optimization model and an intra-day rolling optimization model of the power distribution network by taking the minimum weighted sum of the power distribution network loss, the overall voltage deviation degree and the three-phase unbalance degree as a target function based on the output of each adjustable and controllable resource; and performing iterative operation on the intra-day rolling optimization model of the power distribution network based on a solution result of the day-ahead optimization model of the power distribution network, and correcting the output of intra-day photovoltaic, F-SOP and energy storage at corresponding moments. According to the method provided by the invention, on the basis of considering a photovoltaic reactive power compensation device and the like, intra-day collaborative optimization can be realized, electric power resources can be arranged more flexibly, and waste and excessive configuration of the resources are avoided.

Description

Technical Field

[0001] This application relates to the field of power quality management technology for distribution networks, and in particular to a voltage management method for distribution networks that takes into account photovoltaic uncertainty and F-SOP regulation. Background Technology

[0002] To achieve energy conservation and emission reduction, traditional distribution networks are evolving into new distribution networks with multi-functional collaborative capabilities. While large-scale access to distributed resources can increase local power generation capacity and reduce carbon emissions, the fact that most distributed resources are single-phase access leads to problems such as source-load mismatch, voltage exceeding limits, and three-phase imbalance, resulting in excessive power loss and neutral point potential shift. Therefore, improving voltage quality is crucial for the development of new distribution networks.

[0003] Current solutions for voltage exceedance issues in distribution networks containing photovoltaic (PV) systems include installing reactive power compensation devices and restricting distributed energy output, but all have limitations. For example, reactive power compensation devices have limited effectiveness and poor economic efficiency; restricting PV output wastes resources and harms user interests; utilizing the remaining capacity of PV inverters is flexible and economical, but it is affected by the uncertainty of PV output, and over-considering this uncertainty can lead to conservative configuration and resource waste. Back-to-back voltage source converters (VSCs) and VSC-based SOP (soft open point) devices offer new directions for distribution network voltage management, but their existing regulation models lack multi-time-scale coordination mechanisms, and their coordinated optimization with other reactive power compensation devices such as PV remaining capacity, capacitor banks (CBs), and static var generators (SVGs) still requires systematic research. Summary of the Invention

[0004] This application provides a distribution network voltage management method that considers photovoltaic uncertainty and F-SOP regulation. To solve the above-mentioned technical problems, this application adopts the following technical method: This application provides a distribution network voltage management method that considers photovoltaic uncertainties and F-SOP regulation, including: Identify the controllable resources within the system and determine the output of each controllable resource; the controllable resources include uncertain resources and deterministic resources, the uncertain resources include photovoltaics, static var generators, and capacitor banks, and the deterministic resources include F-SOP, energy storage, and flexible loads; Based on the output of each controllable resource, and with the objective function being the minimum weighted sum of distribution network loss, overall voltage deviation, and three-phase imbalance, a day-ahead optimization model and an intraday rolling optimization model for the distribution network are constructed. Based on the solution results of the day-ahead optimization model of the distribution network, the intraday rolling optimization model of the distribution network is iteratively calculated to correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

[0005] Optionally, the constraints on photovoltaics are photovoltaic-related constraints, the constraints on energy storage are energy storage-related constraints, and the constraints on flexible loads are flexible load-related constraints. The photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints are modified through the following steps: Obtain raw data on the output of uncertain resources. Based on the original data of the uncertain resource output, a representative sample set is determined; Based on the representative sample set, the photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints are corrected.

[0006] Optionally, determining a representative sample set based on the original data of the uncertain resource output includes: Data analysis is performed on the raw data of the uncertain resource output, and Latin hypercube sampling is used to generate initial prediction results of the uncertain resource output. The initial prediction results of the uncertain resource output are processed using a backward elimination method based on Euclidean distance to determine a representative sample set.

[0007] Optionally, the step of modifying the photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints based on the representative sample set includes: A fuzzy chance constraint method based on sample mean approximation is used to relax the constraints on the representative sample set, thereby correcting the photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints.

[0008] Optionally, the solution results of the day-ahead optimization model of the distribution network are used to iteratively calculate the intraday rolling optimization model of the distribution network, and correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day, including: Based on the solution results of the day-ahead optimization model of the distribution network, the first action plan of each controllable resource is determined at each first time interval. The first action plan includes the action plan of the static var generator and the action plan of the capacitor bank. Based on the operating schemes of the static var generator and the capacitor bank, the intraday rolling optimization model of the distribution network is iteratively calculated to correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

[0009] Optionally, the iterative calculation of the intraday rolling optimization model of the distribution network based on the operating scheme of the static var generator and the operating scheme of the capacitor bank, and the correction of the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day, includes: Step S201: Using the second time interval as the period, the operating schemes of the static var generator and the capacitor bank are used as known quantities to solve the daily rolling optimization model of the distribution network, and multiple output prediction results of photovoltaic, F-SOP and energy storage every third time interval are obtained in the current period; the third time interval is obtained by shortening the first time interval by a predetermined ratio, the second time interval is twice the first time interval; the third time interval is less than the first time interval. Step S202: Based on the multiple output prediction results, correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day; Step S203: Determine whether the current iteration count has reached the first preset number, where the first preset number is equal to the number of the third time interval within the day; If so, then end the iterative calculation; If not, proceed to step S204.

[0010] Optionally, the step of correcting the output of photovoltaic, F-SOP, and energy storage at corresponding times within the day based on the multiple output prediction results includes: The output prediction result of the first third time interval among the multiple output prediction results is taken as the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

[0011] Optionally, step S204 specifically includes: Set the time step according to the third time interval, shift the current cycle forward by one time step, and re-execute steps S201-S203.

[0012] Optionally, the constraints of the day-ahead dispatch optimization model of the distribution network include three-phase power flow constraints, safe operation constraints, photovoltaic-related constraints, energy storage-related constraints, reactive power compensation device constraints, flexible load-related constraints, power balance constraints, and F-SOP constraints.

[0013] Optionally, the constraints of the intraday rolling optimization model of the distribution network include three-phase power flow constraints, safe operation constraints, photovoltaic-related constraints, energy storage-related constraints, reactive power compensation device constraints, flexible load-related constraints, power balance constraints, F-SOP constraints, and day-ahead-intraday correction adjustment constraints.

[0014] This application has the following beneficial effects: The method proposed in this application can achieve intraday collaborative optimization based on consideration of reactive power compensation devices such as photovoltaics, which can more flexibly allocate power resources and avoid resource waste and over-allocation. Attached Figure Description

[0015] Figure 1 A flowchart illustrating a distribution network voltage management method that considers photovoltaic uncertainties and F-SOP regulation for the implementation of this application; Figure 2 A flowchart illustrating the process of adjusting the output of photovoltaic, F-SOP, and energy storage at corresponding times during the day, provided for the implementation of this application; Figure 3 An example diagram illustrating the output correction process for intraday photovoltaic, F-SOP, and energy storage at corresponding times, provided for embodiments of this application; Figure 4 This application provides an embodiment of an IEEE 33-node distribution network topology diagram. Figure 5 A schematic diagram of the voltage distribution in scenario 1 provided in an embodiment of this application; Figure 5 (a) is a schematic diagram of the voltage distribution of phase A in scenario 1; Figure 5 (b) is a schematic diagram of the voltage distribution of phase B in scenario 1; Figure 5 (c) is a schematic diagram of the voltage distribution of phase C under scenario 1; Figure 6 A schematic diagram of the voltage distribution in scenario 2 provided in the embodiments of this application; Figure 6 (a) is a schematic diagram of the voltage distribution of phase A in scenario 2; Figure 6 (b) is a schematic diagram of the voltage distribution of phase B in scenario 2; Figure 6 (c) is a schematic diagram of the voltage distribution of phase C in scenario 2; Figure 7 A schematic diagram of the voltage three-phase unbalance result for scenario 1 provided in the embodiments of this application; Figure 8 This is a schematic diagram of the voltage three-phase imbalance result in scenario 2 provided in the embodiments of this application. Detailed Implementation

[0016] To facilitate understanding by those skilled in the art, the present application will be further described below in conjunction with embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present application.

[0017] To solve the above technical problems, such as Figure 1 As shown, this application proposes a distribution network voltage management method that considers photovoltaic uncertainties and F-SOP regulation, including: Step S101: Identify the controllable resources within the system and determine the output of each controllable resource; the controllable resources include uncertain resources and deterministic resources, the uncertain resources include photovoltaics, static var generators, and capacitor banks, and the deterministic resources include F-SOP, energy storage, and flexible loads; Before proceeding to this step, it is necessary to determine whether there is a voltage quality problem in the distribution network system. Any existing technology can be used for this determination; specific methods are not required here. Once it is determined that a quality problem has occurred, this step can be performed.

[0018] To ensure the smooth implementation of subsequent regulation, it is necessary to first identify the controllable resources within the system. Based on the predictability and stability of controllable resources, they can be divided into deterministic resources and uncertain resources. Deterministic resources generally include F-SOP (Four-leg Soft Open Point), static var generators, and capacitor banks. Uncertain resources generally include photovoltaics, energy storage, and flexible loads.

[0019] After identifying the aforementioned controllable resources, data modeling is performed on each one, constructing a mathematical model corresponding to each controllable resource. This allows for the construction of F-SOP models, energy storage models, flexible load models, photovoltaic models, static var generator models, and capacitor bank models. Different models have different component compositions, resulting in different power outputs. Therefore, the power output of each controllable resource can be determined using the corresponding model.

[0020] To make the above model more realistic, it is necessary to impose constraints on it based on the actual physical operation. The constraints on the above model are explained as follows: The constraints of F-SOP include F-SOP operation constraints, F-SOP power loss constraints, and F-SOP capacity constraints.

[0021] The operating constraints of F-SOP are as follows: F-SOP has the ability to independently control the three-phase power, enabling it to achieve power transfer between phases at a single node. Therefore, the operating constraints of F-SOP only need to satisfy that the sum of the input power and output power of all phases is zero. The difference from traditional SOP is that it no longer needs to consider the input and output power balance within each phase.

[0022] (1) In the formula: Let VSC of F-SOP access node i at time t be... Active power transmitted in phase; Let VSC of F-SOP access node i at time t be... The power loss generated by the phase It is a set of phase numbers.

[0023] The power loss constraint for F-SOP is: (2) In the formula: The power loss factor of F-SOP is... Let VSC of F-SOP access node i at time t be... The reactive power transmitted in phases.

[0024] The capacity constraint for F-SOP is: (3) In the formula, VSC capacity for F-SOP access node i.

[0025] The photovoltaic (PV) technology in this application is grid-connected distributed PV. While providing active power to the grid, grid-connected distributed PV can also output reactive power. It can utilize the capacity of the grid-connected inverter itself to provide reactive power regulation capabilities to the grid. The power regulation range of the distributed PV inverter is limited by its capacity; therefore, the constraint on PV is generally a PV capacity constraint, specifically: The grid-connected device of the distributed photovoltaic system is equipped with a reactive power compensation device, which means that the photovoltaic device can be used for reactive power compensation. The distribution network can schedule and control both the active and reactive power output of the photovoltaic system. The constraints of the photovoltaic device are photovoltaic-related constraints, which include photovoltaic active power constraints (formula (4)), photovoltaic reactive power constraints (formula (5)) and photovoltaic capacity constraints (formula (6)), as shown in the following formula: (4) (5) (6) In the formula: and The first A photovoltaic device in The active and reactive power output at all times; For the first A photovoltaic in The maximum active power that can be output at any given time is generally taken as the predicted maximum photovoltaic output value; For the first The rated capacity of each photovoltaic unit; and These represent the maximum and minimum power factor values ​​for the photovoltaic device, respectively.

[0026] Distributed energy storage can store excess energy when photovoltaic power generation is at full capacity, and support the power output of the distribution network when photovoltaic output is low. In this application, the voltage problem of the distribution network can be addressed by outputting reactive power, where energy storage is limited by capacity, charging and discharging power, etc. The constraints of energy storage are energy storage-related constraints, including energy storage active power constraints (formula (7)), energy storage energy constraints (formula (8)), and energy storage reactive power constraints (formula (9)), specifically: (7) (8) (9) In the formula: This represents the charging state of the energy storage array. The discharge state of the energy storage array is represented by a 0-1 variable; µ represents the charge / discharge efficiency. and The actual charging active power and actual discharging active power of energy storage; and The actual charging reactive power and actual discharging reactive power of energy storage; and These represent the maximum and minimum discharge power of the energy storage device. and These are the maximum and minimum charging power of the energy storage device. and The charging and discharging efficiency of energy storage devices is considered; to extend the lifespan of energy storage devices and reduce investment costs, full charging and discharging of energy storage devices are not permitted. and These represent the minimum and maximum states of charge of the energy storage device. This refers to the rated capacity of the energy storage device.

[0027] The capacitor bank must meet the following constraints: (10) In the formula: For node j at time t, the CB connected to it is The reactive power compensation of the phase; Let be the number of operational groups at time t, and be a discrete variable value. The compensation power for each CB group is constant; The upper limit of the number of CB groups connected to node j; This represents the maximum number of CB operations. Let T be the number of operational groups at time t-1, and T be the time set that needs to be optimized.

[0028] Furthermore, the absolute value constraint in the above formula can be addressed by adding an auxiliary variable that characterizes the change in CB compensation capacity between adjacent time periods. Then we can obtain: (11) The constraints of the static var generator are: (12) In the formula: The compensation power for each SVG group; , These represent the lower and upper limits of the SVG compensation power, respectively. Considering that during the operation of active power distribution networks, the increasing photovoltaic penetration rate may cause system power flow reversal and overvoltage problems, this application sets the lower limit of the SVG compensation power. .

[0029] Flexible load models include transferable load models and interruptible load models. Therefore, the constraints on flexible loads are flexible load-related constraints, which include transferable load constraints and interruptible load constraints. Transferable loads refer to loads that can flexibly adjust their electricity demand within a specific time period without affecting service quality or core functions. Examples include electric vehicles and industrial equipment with adjustable operating times. These loads have time flexibility and can be adjusted according to power supply conditions and system economics, thereby optimizing energy use or reducing operating costs. The constraints for transferable loads are: (13) (14) In the formula: Let t be the load value (kW) at which the originally planned electrical load transfer occurs; The total number of loads that can be transferred at time t; The participation rate of users with transferable electrical loads.

[0030] Interruptible load constraints are as follows: (15) In the formula, The actual load reduction during period t, in kW; This represents the maximum load reduction that can be achieved during time period t. This is a period that can be interrupted.

[0031] In the construction of the above model, due to the use of photovoltaic equipment to manage power quality, distributed resources such as photovoltaics and energy storage have uncertain output. Furthermore, the stochastic nature of multidimensional parameters such as solar radiation intensity and power load demand significantly increases the complexity of the system optimization solution. To effectively address the impact of these uncertain parameters on the decision-making process, the photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints need to be modified. The modification process is as follows: First, raw power output data from photovoltaic and energy storage resources is acquired. This raw data is then analyzed to examine changes in solar radiation and historical wind speed data, including daily and hourly radiation intensity and wind speed. Subsequently, based on the analyzed data, Latin hypercube sampling is employed, using a normal distribution generator to generate initial predictions of uncertain resource output that conform to the analyzed data. These predictions can represent the power output of distributed generation at any given moment.

[0032] Since the initial predictions generated above constitute a large-scale scenario, directly solving the stochastic programming problem considering all scenarios would be computationally very time-consuming. Therefore, it is necessary to reduce the number of scenarios. Here, a backward elimination method based on Euclidean distance is used to process the initial predictions of uncertain resource output, eliminating N scenarios into n typical output scenarios with the highest probability as a representative sample set.

[0033] By employing the opportunity constraint method based on the approximation of the sample mean, and relaxing the representative sample set, the constraints related to photovoltaics (Equations (4)-(6)), energy storage (Equations (7)-(9)), and flexible load (Equations (13)-(15)) can be corrected. The correction process is as follows: After processing the constraints related to photovoltaics, energy storage, and flexible loads, assuming they all hold at confidence level α, the reliability chance constraint of the system power balance can be expressed as follows: (16) In the formula, To satisfy the confidence level of the above formula, Let J be the injected power at time t. Let i be the total load power at time t. Let i be the line loss at time t. , , , It can be obtained from a representative sample set.

[0034] Using the clear equivalence classes under the trapezoidal fuzzy parameters, the clear equivalence classes for photovoltaic-related constraints, energy storage-related constraints, and flexible load-related constraints are as follows: (17) In the formula, and Let be the fuzzy membership parameter of PV at node j. and Let be the fuzzy membership parameter of ES at node j. and The fuzzy membership parameter for flexible load at node j.

[0035] Since fuzzy parameters cannot be directly applied to formula calculations, the output of photovoltaic, energy storage, and flexible loads in time period i is equivalently processed: (18) In the formula, , , and , where is the fuzzy membership parameter of PV at the node; , , ,and For the fuzzy membership parameter of ES in the node; , , and For flexible loads, fuzzy membership parameters at nodes; (x=PV, ES, Load, y=1, 4) depends on the upper and lower bounds of the parameters of the representative sample set. (x=PV, ES, Load, y=2, 3) depends on the most likely values ​​of the parameters in the representative sample set.

[0036] Step S102: Based on the output of each controllable resource, and with the objective function of minimizing the weighted sum of distribution network loss, overall voltage deviation, and three-phase imbalance, construct a day-ahead optimization model and an intraday rolling optimization model for the distribution network. After obtaining the output of each controllable resource, the output of each controllable resource is used as the particle value (random variable) of the particle swarm optimization algorithm. A corresponding objective function is set, and combined with the relevant constraints of various resources and system power flow constraints, power flow calculation is performed. This yields intermediate variables, namely the voltage of each node in the distribution network and the current of each line. Subsequently, network losses, voltage deviations, and three-phase imbalances in the distribution network can be calculated. It should be noted that the calculation process for obtaining the voltage of each node and the current of each line in the distribution network based on the output of each controllable resource in this application is not limited to any specific calculation method found in existing technologies.

[0037] Therefore, this application can construct a day-ahead optimization model and an intraday rolling optimization model for the distribution network based on the output of each control resource, with the objective function being the minimum weighted sum of distribution network loss, overall voltage deviation, and three-phase imbalance, as shown in the following equation: (19) (20) In the formula: , , These are the weighting factors for the network loss target, the overall voltage deviation target, and the three-phase imbalance target, respectively. , , These are distribution network losses, overall voltage deviation, and three-phase imbalance; It should be noted that both the day-ahead optimization model and the intraday rolling optimization model of the distribution network are based on formula (19), but the calculation methods for the two are different, which will be explained in detail later and will not be explained here.

[0038] Distribution network losses are: (twenty one) In the formula: This refers to the number of nodes in the distribution network. , Between node i and node j respectively The branch current and resistance of the phase; Δt is the duration of a single time interval. This refers to the unit cost of distribution network losses.

[0039] (twenty two) In the formula: For node j The node voltage of the phase; UN is the reference voltage of the distribution network. Penalty cost for voltage deviation in the distribution network.

[0040] The three-phase imbalance is: (twenty three) In the formula: Let be the voltage three-phase unbalance at node j. The penalty cost for three-phase imbalance in the distribution network. According to the existing technical definition, the three-phase imbalance degree of the voltage at node j is defined as shown in equation (24).

[0041] (twenty four) (25) In the formula, (X=A,B,C) is the node voltage of phase X at node j, and Lj is the average value of the three-phase voltage at node j.

[0042] The day-ahead dispatch optimization model of the distribution network needs to consider the three-phase power flow constraints, safe operation constraints, photovoltaic related constraints (Equation (4)-Equation (6)), energy storage related constraints (Equation (7)-Equation (9)), reactive power compensation device constraints (Equation (10)-Equation (12)), flexible load related constraints (Equation (13)-Equation (15)), power balance constraints, and F-SOP constraints (Equation (1)-Equation (3)).

[0043] The three-phase power flow constraints of the distribution network are: (26) In the formula: , For each node j in The active and reactive power injected into the phase; , For node j in The active and reactive power flowing out of the next node k; , For the previous node i in The active and reactive power flowing into node j; , , Between node i and node j respectively The reactance, conductance, and susceptance of the phase. For node i Phase node voltage.

[0044] Safe operation constraints: (27) In the formula: , Let be the upper and lower limits of the node voltage at node j; , These are the upper and lower limits of the branch current between node i and node j; , For the distribution network at The phase interacts with the upstream power grid in terms of active and reactive power; , , , These are its maximum and minimum interaction power values, respectively.

[0045] The constraints of the daily rolling optimization model of the distribution network are the three-phase power flow constraints, safe operation constraints, photovoltaic related constraints (Equation (4)-Equation (6)), energy storage related constraints (Equation (7)-Equation (9)), reactive power compensation device constraints (Equation (10)-Equation (12)), flexible load related constraints (Equation (13)-Equation (15)), power balance constraints, and F-SOP constraints (Equation (1)-Equation (3)).

[0046] In addition to the constraints mentioned above, the intraday rolling optimization model for the distribution network also needs to consider the day-ahead-intraday correction adjustment constraint. F-SOP performs intraday correction based on the day-ahead operating status, and must meet the following day-ahead-intraday correction adjustment constraint for the equipment: (28) In the formula: , The VSCs of F-SOP access node i were located at the same time today and within the day. Active power transmitted in phase; For the VSC of F-SOP access node i in The maximum operable active power transmitted between the previous day and the next day; , The VSCs of F-SOP access node i were located at the same time today and within the day. The reactive power transmitted in phase; For the VSC of F-SOP access node i in The maximum operable reactive power transmitted between the previous day and the next day.

[0047] Step S103: Based on the solution results of the day-ahead optimization model of the distribution network, perform iterative calculations on the intraday rolling optimization model of the distribution network to correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

[0048] By combining the day-ahead optimization model of the distribution network and its corresponding constraints, the day-ahead optimization model of the distribution network is solved, and the solution results are obtained. Based on the solution results, the first action plan of each controllable resource at each first time interval can be obtained. Here, the first action plan of each controllable resource considers the first action plan at each first time interval within 24 hours. This first action plan includes the action plan of the static var generator and the action plan of the capacitor bank.

[0049] After determining the first action plan for each of the aforementioned control resources, since the static var generator (SVR) and capacitor banks are not easily adjustable in a short period of time and cannot achieve real-time rapid adjustment (while the others are easily adjustable and can achieve real-time rapid adjustment), the intraday rolling optimization model of the distribution network can be iteratively calculated based on the action plan of the SVR and capacitor banks. This corrects the output of photovoltaic, F-SOP, and energy storage at the corresponding time points during the day. The specific correction process is as follows: Figure 2 As shown below: Step S201: Using the second time interval as a period, the operating schemes of the static var generator and the capacitor bank are used as known quantities to solve the intraday rolling optimization model of the distribution network, obtaining multiple output prediction results of photovoltaic, F-SOP and energy storage every third time interval within the current period; the third time interval is obtained by shortening the first time interval by a predetermined ratio, the second time interval is twice the first time interval; the third time interval is less than the first time interval; Here, the second time interval is used as the period, and the first time interval is shortened by a predetermined ratio to obtain multiple third time intervals. Generally, to ensure the smooth implementation of the technical solution, the second time interval is set to twice the first time interval. Within the second time interval, the operating schemes of the static var generator (SVM) and the capacitor bank are considered as known quantities, and the intraday rolling optimization model of the distribution network is solved to obtain multiple output prediction results for photovoltaic (PV), F-SOP (F-SOP), and energy storage every third time interval within the current period. Since there are two first time intervals within one second time interval period, it can be understood that the operating schemes of the SVM and the capacitor bank change twice within one period when solving the intraday rolling optimization model of the distribution network. The number of output prediction results for PV, F-SOP, and energy storage is related to the number of third time intervals within the second time interval period.

[0050] Step S202: Based on the multiple output prediction results, correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day; After obtaining multiple output prediction results, the output prediction result of the first third time interval among the multiple output prediction results can be used as the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

[0051] Step S203: Determine whether the current iteration count has reached the first preset count, where the number of iterations is equal to the number of the third time interval within the day; If so, then end the iterative calculation; If not, proceed to step S204.

[0052] To ensure that the output of photovoltaic, F-SOP and energy storage at each moment of the day can be corrected accordingly, the first preset number of iterations must be equal to the number of the third time interval of the day. When the number of iterations reaches the first preset number, it means that all outputs of the day have been corrected and the iteration calculation ends. If it does not reach the first preset number, it means that optimization needs to continue and step 204 can be executed.

[0053] Step S204: Set the time step according to the third time interval, shift the current cycle forward by one time step, and re-execute steps S201-S203.

[0054] The time step is set with the third time interval. Each time the next iteration is performed, the time step is shifted to the next time step, and then steps S201-S203 are re-executed to correct the output of the next third time interval.

[0055] To gain a more concrete understanding of step S103, in conjunction with... Figure 3 Let's take an example to illustrate this: First, taking one hour as the first time interval, we can divide the day-ahead into 24 time periods. Then, by solving the day-ahead optimization model of the distribution network, we can obtain the first action plan for each control resource in the 24 time periods. This leads to the action plans for the static var generators and capacitor banks within the 24 time periods. Next, taking a two-hour time interval as a cycle, and then a 15-minute time interval as a third time interval, we obtain eight third time intervals, resulting in 96 third time intervals per day. Therefore, the number of iterative calculations is 96. Based on the day-ahead optimization model of the distribution network, the operating schemes of the static var generators (SVA) and capacitor banks during the periods of 9:00-11:00 can be obtained. Then, the operating schemes of the SVA and capacitor banks during the periods of 9:00-10:00 and 10:00-11:00 are treated as known quantities. The day-ahead rolling optimization model of the distribution network is then solved to obtain eight output prediction results (0-8) for photovoltaic (PV), F-SOP (F-SOP), and energy storage during the period of 9:00-11:00. Finally, the output prediction results (0-1) for 9:00-9:15 are used as the output of PV, F-SOP, and energy storage during the period of 9:00-9:15. Then, with a time step of 15 minutes, shift one unit to the right, meaning the next cycle is 9:15-11:15, to calculate and update the output of photovoltaics, F-SOP, and energy storage for the next time period from 9:15 to 9:30. This process continues until 96 iterations are completed, thus updating the output of photovoltaics, F-SOP, and energy storage for all times of the day.

[0056] Simulation Analysis The proposed distribution network optimization scheduling method is simulated and verified using an improved IEEE 33-node distribution system. The distribution network structure is as follows: Figure 4 As shown.

[0057] The F-SOP is connected at both ends to nodes 17 and 21 to replace the tie switch at that location. Node 13 is connected to CB, and nodes 21 and 30 are connected to SVG. Each device is numbered with Arabic numerals according to the above connection order.

[0058] To verify the effectiveness and accuracy of the method proposed in this application, two planning methods and scenarios were set up.

[0059] Scenario 1: Without adopting F-SOP, the reactive power optimization scheduling method of the distribution network only considers the reactive power compensation device, directly applying the day-ahead scheduling results, without making corrections during the day, i.e., the uncontrolled method.

[0060] Scenario 2: A distribution network voltage management method considering photovoltaic uncertainties and F-SOP regulation, namely the method proposed in this application.

[0061] Table 1 Optimization results for various scenarios Table 1 Optimization results for each scenario ; Table 1 compares the optimization results for two scenarios under the IEEE 33-node example. Regarding network loss optimization, the method proposed in this application yields the lowest network loss value, a reduction of 19.82% compared to scenario 1; in terms of overall voltage deviation, the method proposed in this application reduces it by 21.39% compared to scenario 1; and in terms of three-phase imbalance, the method proposed in this application reduces it by 10.82% compared to scenario 1. Therefore, the method proposed in this application has significant effects on reducing losses and balancing phase-to-phase voltage.

[0062] Figures 5 and 6 show the specific results of the distribution network voltage distribution in scenarios 1 and 2, respectively. Figure 5 (a) Figure 5 (b) and Figure 5 As shown in (c), in Scenario 1, the maximum three-phase voltage of the distribution network is 1.082 pu, the minimum is 0.998 pu, and the overall voltage deviation is 245.45 pu, all of which do not meet the requirements of the existing standard: the allowable deviation of power supply voltage of 10kV and below is ±7% of the rated voltage. Therefore, Scenario 1 shows some voltage exceeding the limit, and the voltage deviation rate exceeds the allowable range of the existing standard. Figure 6 (a) Figure 6 (b) and Figure 6As shown in (c), no voltage exceedances were observed at any node in Scenario 2. The voltage fully meets the existing ±7% range. This demonstrates that optimizing the distribution network voltage management method using photovoltaic uncertainty and F-SOP control can effectively improve system voltage quality.

[0063] Figure 7 , Figure 8 The results show the three-phase imbalance of the distribution network in scenarios 1 and 2, respectively. It can be seen that after applying the distribution network voltage management method considering photovoltaic uncertainties and F-SOP regulation, the three-phase imbalance at each node is significantly improved, decreasing from 5% before optimization to below 2%. This optimization result not only meets the industry standard requirement that the three-phase imbalance should not exceed 3%, but also further demonstrates the effectiveness of the proposed method. In summary, the method proposed in this application can achieve intraday collaborative optimization based on considering reactive power compensation devices such as photovoltaics, allowing for more flexible allocation of power resources and avoiding resource waste and over-configuration. Furthermore, this application combines fuzzy chance constraint methods with distributed resources such as photovoltaics and energy storage, as well as flexible loads, thereby transforming different uncertainties into deterministic optimization problems under certain confidence levels. This allows for better utilization of distributed resources, while simultaneously addressing power quality issues in the distribution network and improving voltage management effectiveness.

[0064] The above embodiments are preferred implementations of this application. In addition, this application can be implemented in other ways. Any obvious substitutions without departing from the concept of this technical solution are within the protection scope of this application.

[0065] To facilitate understanding by those skilled in the art of the improvements made by this application compared to the prior art, some of the accompanying drawings and descriptions have been simplified, and for clarity, some other elements have been omitted from this application. Those skilled in the art should realize that these omitted elements may also constitute the content of this application.

Claims

1. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation, characterized in that, include: Identify the controllable resources within the system and determine the output of each controllable resource; The adjustable resources include uncertain resources and deterministic resources. The deterministic resources include F-SOP, static var generators, and capacitor banks. The uncertain resources include photovoltaics, energy storage, and flexible loads. Based on the output of each controllable resource, and with the objective function being the minimum weighted sum of distribution network loss, overall voltage deviation, and three-phase imbalance, a day-ahead optimization model and an intraday rolling optimization model for the distribution network are constructed. Based on the solution results of the day-ahead optimization model of the distribution network, the intraday rolling optimization model of the distribution network is iteratively calculated to correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

2. The distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 1, characterized in that, The constraints on photovoltaics are photovoltaic-related constraints, the constraints on energy storage are energy storage-related constraints, and the constraints on flexible loads are flexible load-related constraints. The photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints are corrected through the following steps: Obtain raw data on the output of uncertain resources. Based on the original data of the uncertain resource output, a representative sample set is determined; Based on the representative sample set, the photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints are corrected.

3. The distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 2, characterized in that, The process of determining a representative sample set based on the original data of the uncertain resource output includes: Data analysis is performed on the raw data of the uncertain resource output, and Latin hypercube sampling is used to generate initial prediction results of the uncertain resource output. The initial prediction results of the uncertain resource output are processed using a backward elimination method based on Euclidean distance to determine a representative sample set.

4. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 2, characterized in that, The process of modifying the photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints based on the representative sample set includes: A fuzzy chance constraint method based on sample mean approximation is used to relax the constraints on the representative sample set, thereby correcting the photovoltaic-related constraints, the energy storage-related constraints, and the flexible load-related constraints.

5. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 1, characterized in that, The solution results based on the day-ahead optimization model of the distribution network are used to iteratively calculate the intraday rolling optimization model of the distribution network, correcting the output of photovoltaic, F-SOP, and energy storage at corresponding times during the day, including: Based on the solution results of the day-ahead optimization model of the distribution network, the first action plan of each controllable resource is determined at each first time interval. The first action plan includes the action plan of the static var generator and the action plan of the capacitor bank. Based on the operating schemes of the static var generator and the capacitor bank, the intraday rolling optimization model of the distribution network is iteratively calculated to correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

6. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 5, characterized in that, The operating schemes based on the static var generator and the capacitor bank are used to iteratively calculate the intraday rolling optimization model of the distribution network, correcting the output of photovoltaic, F-SOP, and energy storage at corresponding times during the day, including: Step S201: Using the second time interval as the period, the operating schemes of the static var generator and the capacitor bank are used as known quantities to solve the daily rolling optimization model of the distribution network, and multiple output prediction results of photovoltaic, F-SOP and energy storage every third time interval are obtained in the current period; the third time interval is obtained by shortening the first time interval by a predetermined ratio, the second time interval is twice the first time interval; the third time interval is less than the first time interval. Step S202: Based on the multiple output prediction results, correct the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day; Step S203: Determine whether the current iteration count has reached the first preset number, where the first preset number is equal to the number of the third time interval within the day; If so, then end the iterative calculation; If not, proceed to step S204.

7. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 6, characterized in that, The process of correcting the output of photovoltaic, F-SOP, and energy storage at corresponding times within a day based on the multiple output prediction results includes: The output prediction result of the first third time interval among the multiple output prediction results is taken as the output of photovoltaic, F-SOP and energy storage at the corresponding time of the day.

8. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 6, characterized in that, Step S204 specifically involves: Set the time step according to the third time interval, shift the current cycle forward by one time step, and re-execute steps S201-S203.

9. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 1, characterized in that, The constraints of the day-ahead dispatch optimization model for the distribution network include three-phase power flow constraints, safe operation constraints, photovoltaic-related constraints, energy storage-related constraints, reactive power compensation device constraints, flexible load-related constraints, power balance constraints, and F-SOP constraints.

10. A distribution network voltage management method considering photovoltaic uncertainty and F-SOP regulation according to claim 1, characterized in that, The constraints of the intraday rolling optimization model for the distribution network include three-phase power flow constraints, safe operation constraints, photovoltaic-related constraints, energy storage-related constraints, reactive power compensation device constraints, flexible load-related constraints, power balance constraints, F-SOP constraints, and day-ahead-intraday correction adjustment constraints.

Images