Active power distribution network cooperative control method based on power flow element model

Abstract

The invention discloses an active power distribution network cooperative control method based on a power flow element model, and the method comprises the steps: building a comprehensive optimization objective function of emergency voltage control of a power distribution network, and building a high-dimensional linear state space model between node voltage and an input variable in an offline manner; and establishing a mapping relation meta-model between the power distribution network topology filial generation matrix and a power flow matrix in the incomplete dimension rising power flow model, and on the basis of the new power flow matrix and the meta-model, carrying out online collaborative optimization on a topology structure and VAR equipment to realize emergency voltage control of the power distribution network. According to the method, under the extreme weather condition, the voltage safety of the power distribution network is realized through collaborative optimization of the topological structure and the economic cost of the VAR.

Description

Technical Field

[0001] This invention belongs to the field of power system operation technology, and in particular relates to a method and system for emergency topology reconfiguration and VAR collaborative control of active distribution networks based on power flow element model. Background Technology

[0002] The high proportion of distributed generation significantly increases the sensitivity of distribution network safety to weather conditions, forming a distribution system with deep weather-power coupling. Sudden weather changes and extreme weather events typically affect the safe operation of distribution systems in two ways: 1) causing large-scale power fluctuations in distribution networks with a high proportion of distributed generation, leading to power flow reversals and voltage exceedances; 2) extreme weather causes active or passive adjustments to the distribution network topology, triggering a redistribution of power flow, resulting in line overloads or node overvoltage / undervoltage problems. Therefore, distribution network voltage control under weather changes has attracted considerable attention, requiring more flexible, adaptive, and reliable methods. With the occurrence of distribution network line faults and voltage exceedances under extreme weather events, it is necessary to adjust the distribution network topology to ensure the integrity of the power supply topology and voltage safety. However, the reconfigured topology may cause a redistribution of power flow in the distribution network, making the distribution network voltage control problem more complex. Therefore, voltage control through topology reconfiguration and reactive power (VAR) co-optimization is crucial for distribution networks with a high proportion of distributed generation.

[0003] Research on distribution network reconfiguration mainly focuses on dynamic reconfiguration over time intervals, typically using a certain time period as the switching operation cycle. By adjusting the flexible resources of the distribution network, the optimal voltage distribution within that time period is achieved. However, the distribution network reconfiguration problem exhibits significant discreteness and nonlinearity. Numerous studies have proposed analytical and heuristic algorithms for topology-voltage optimization. However, heuristic algorithms suffer from slow convergence speeds and susceptibility to local optima, resulting in slow solution times. Furthermore, both analytical and heuristic algorithms rely on feeder parameters to construct the optimization model. Accurate feeder parameters are difficult to obtain in practical medium- and low-voltage distribution networks, making the aforementioned topology optimization methods inapplicable. Therefore, it is necessary to study topology-voltage collaborative optimization control strategies for distribution network parameters that are incomplete, in order to address voltage variations caused by distributed power fluctuations under extreme weather conditions.

[0004] When physical parameters are incomplete, data-driven methods can utilize historical distribution network operation data to derive optimized operation strategies. Researchers have used convolutional neural networks to construct a mapping between load patterns and optimal topology reconfiguration schemes. Others have used graph convolutional networks based on measurements and network topology to achieve accurate voltage prediction. However, these black-box models lack an explanation of the physical mechanisms and are difficult to generalize to different scenarios. In particular, after topology adjustments, the voltage optimization model adjusts accordingly, and existing data-driven methods struggle to achieve adaptive model transfer. Summary of the Invention

[0005] This invention proposes an emergency topology reconfiguration and VAR collaborative control method for active distribution networks based on a power flow element model. A comprehensive optimization model for topology reconfiguration and VAR is constructed based on a topology adaptive power flow element model, realizing predictive optimization based on distributed generation and load power. The method is performed in a rolling manner. Considering the operating frequency limitations of line switches, it ensures that voltage overruns, power flow reversals, and power supply safety issues caused by a high proportion of distributed generation access to the distribution network under extreme weather conditions are prevented.

[0006] To achieve the above-mentioned objectives, the present invention utilizes the following technical solution:

[0007] Firstly, an active distribution network cooperative control method based on a power flow element model includes the following steps:

[0008] Step 1: Construct a comprehensive optimization objective function for emergency voltage control of the distribution network to minimize the costs of switching operations, distributed generation active power, energy storage output power, load shedding, and voltage deviation. Its expression is:

[0009] ;

[0010] And meet the operating constraints of the distribution network;

[0011] Step 2: Based on historical operating data of the distribution network, a high-dimensional linear state-space model between node voltages and input variables is constructed offline through state-space mapping and least squares training. This includes:

[0012] Step 2-1: Constructing Input Samples and output samples , , , Let represent the equilibrium voltage of the reference node under the g-th sample, and the active power and reactive power column vectors composed of each node of the distribution network under the g-th sample, respectively.

[0013] Step 2-2: Upgrade the state space of the input samples to obtain the upgraded variables. ;

[0014] Steps 2-3: Based on the upgraded input sample set X and output sample set Y, obtain the data-driven power flow matrix through least squares training. ;

[0015] Steps 2-4: Constructing an incomplete up-dimensional power flow model: ,in, and These are the power flow variable coefficient matrices, Represents state variables, Represents decision variables, Functions for increasing the dimensionality of state variables;

[0016] Step 3: Establish the distribution network topology sub-matrix based on the topology adaptive power flow meta-model method. The power flow matrix in the incomplete up-dimensional power flow model Metamodel of mapping relationships between Specifically, it includes:

[0017] Step 3-1: Construct the sub-matrix of the distribution network topology Its elements represent the connection state between nodes;

[0018] Step 3-2: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context.] and Convert to column vector form and And establish mapping relationships ;

[0019] Step 3-3: Let and Let represent the column vector samples of the offspring matrix and the column vector samples of the power flow matrix, respectively. Therefore, the meta-model can be represented as: ;

[0020] Steps 3-4: Obtain the updated power flow matrix based on the current topology. : ( Represents the column vector of the child matrix under the current topology. This is the power flow matrix corresponding to this topology. and Represented as power flow matrix The number of rows and columns;

[0021] Step 4: Based on the aforementioned trend matrix Heyuan model Online collaborative optimization of topology and VAR equipment enables emergency voltage control of the distribution network.

[0022] In some implementations, the dimensionality-upgrading function The dimension of is E, and its e-th dimension function is . = ,in, Indicates and Basis vectors of the same dimension represent The e-th dimension-increasing function, This represents the e-th Euclidean distance of the g-th data sample.

[0023] In some implementations, the decision variable u includes the active power of the distributed generation. and reactive power Active power of energy storage and reactive power and the active power of flexible loads. .

[0024] In some implementations, the state variable The active and reactive power of the uncontrollable load.

[0025] In some implementations, the metamodel The following was obtained through least squares regression training:

[0026] ;

[0027] in, and These represent column vector samples of the offspring matrix and column vector samples of the power flow matrix, respectively.

[0028] In some implementations, the online optimization phase employs a rolling optimization method, updating the VAR optimization results every 15 minutes and the topology every 6 hours.

[0029] In some implementations, the penalty coefficient in the comprehensive optimization objective function , , , and These represent the penalty coefficients for voltage deviation, DG abandonment, ES output power, load shedding, and switching cost, respectively.

[0030] In some implementations, the power distribution network operation constraints include:

[0031] Voltage amplitude upper and lower limit constraints;

[0032] Flexible load reduction constraints;

[0033] Distributed power generation / energy storage operation constraints;

[0034] Topological connectivity constraints.

[0035] Thirdly, the present invention proposes an electronic device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the active distribution network coordinated control method based on the power flow element model.

[0036] Secondly, the present invention proposes a computer-readable storage medium storing a computer program thereon, characterized in that the program, when executed by a processor, implements the active distribution network coordinated control method based on the power flow element model.

[0037] Compared with the prior art, the beneficial effects and advantages of the present invention are as follows:

[0038] 1) The IDL-PF model, built based on state space dimensionality increase and least squares training, can accurately capture the nonlinear characteristics of power flow in distribution networks and is suitable for various operating scenarios;

[0039] 2) Through meta-model It can quickly map the relationship between topology and power flow matrix, support real-time model updates after online topology changes, avoid retraining, and improve system robustness;

[0040] 3) By comprehensively considering the costs of switching actions, power curtailment, load shedding, and voltage deviation through a multi-objective optimization function, operating costs are significantly reduced while ensuring safety.

[0041] 4) The topology and VAR are optimized in a rolling manner. The VAR setting value is updated every 15 minutes and the topology structure is optimized every 6 hours, which balances real-time performance and computational efficiency and is suitable for emergency voltage control in extreme weather conditions.

[0042] 5) Utilizing offline training based on historical operating data under the existing topology, the online part only needs to predict data and topology status, which is easy to integrate into the existing distribution network control platform. Attached Figure Description

[0043] Figure 1 This is a flowchart of the active distribution network collaborative control method based on the power flow element model of the present invention.

[0044] Figure 2 These are detailed flowcharts of steps S2 and S3 of the present invention, (2a) detailed flowchart of step S2, (2b) detailed flowchart of step S3.

[0045] Figure 3 This is a diagram illustrating the technical implementation of the present invention.

[0046] Figure 4 This is a power distribution network diagram after a fault occurs, serving as a verification example of the present invention.

[0047] Figure 5 This is a voltage distribution diagram of the distribution network under different voltage control schemes during a typical time period. Detailed Implementation

[0048] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0049] Example 1: As Figure 1 The diagram shows the overall process of the multi-source water transfer project analysis method based on the scheduling model proposed in this invention, which specifically includes the following steps:

[0050] To achieve topology-VAR coordinated optimization, this invention proposes a method for emergency topology reconfiguration and VAR coordinated control of active distribution networks based on a power flow meta-model. In the offline training phase, the traditional nonlinear power flow (PF) model is transformed into a linear power flow model independent of impedance parameters based on Koopman operator theory. Power flow matrices under different topologies are used to train the meta-model to establish the mapping between graph data and the linear power flow matrix. In the online optimization phase, a topology-VAR coordinated optimization model is constructed based on the power flow meta-model to achieve effective voltage control under emergency conditions.

[0051] Step 1: Construct a comprehensive optimization objective function for emergency voltage control of the distribution network to minimize the switching action cost, distributed generation active power cost, energy storage output power cost, load shedding cost, and voltage deviation cost. Its expression is: In emergency voltage control of the distribution network, the comprehensive objective of ADN is to ensure voltage safety at the lowest cost under extreme weather conditions, including minimizing switching action, abandoned DG active power, energy storage (ES) output power, load shedding, and voltage deviation costs. The decision variables are switching state, DG, ES, and flexible load power, which can be expressed as Equation (1):

[0052] ; (1)

[0053] In the formula, This indicates the scheduling optimization period. In this invention, the optimization period is 12 hours, the topology is optimized every 6 hours, and the VAR optimization results are updated every 15 minutes. This represents the voltage at node i at time t. Indicates the voltage reference value. and This represents the reduction in active power of DG and flexible load at node i at time t. Let represent the active power of the energy storage ES at node i at time t. and This indicates the switching state of line ij in the original topology / reconstructed topology. , and These represent the sets of DG, ES, and flexible loads in the distribution network, respectively. It represents the set of all lines in a power distribution network. , , , and These represent the penalty coefficients for voltage deviation, DG abandonment, ES output power, load shedding, and switching cost, respectively.

[0054] In the topology reconfiguration and VAR collaborative optimization, the operating constraints of the distribution network include upper and lower limits of voltage amplitude (2)-(3), flexible load reduction constraints (4), operating constraints of distributed generation / energy storage (DG / ES) (5)-(6), and topology connectivity constraints (7). Among them, the DG / ES operating constraints include active and reactive power output range constraints of DG and ES.

[0055] (2)

[0056] (3)

[0057] (4)

[0058] (5)

[0059] (6)

[0060] (7)

[0061] In the formula, and This represents the active / reactive power at node i at time t. and Indicates the line Resistance and reactance, This indicates the mapping relationship between the above variables and node voltages under different topologies. and These are the lower and upper limits of the voltage amplitude. and Let represent the DG and the active power of the load at node i at time t, respectively. , and Let DG, ES, and reactive power of the load at node i at time t be represented respectively. and For nodes The initial value of the load power injection at the location, and Let represent the active power and reactive power of the load reduction at node i at time t, respectively. This represents the power factor angle at load node i. This represents the predicted active power of DG at time t. This indicates the rated capacity of distributed generation (DG). This represents the charge / discharge loss rate of energy storage (ES) at node i. This represents the charging state of energy storage (ES) at node i at time t. This represents the rated capacity of energy storage (ES) at node i at time t. This represents the rated capacity of energy storage (ES) at node i at time t. Let represent the minimum and maximum values ​​of the charge state at node i at time t, respectively.

[0062] Step 2: Based on the operational data in the historical database of the distribution network control center, a high-dimensional linear state-space model between node voltages and input variables is constructed offline through state-space mapping and least squares training. The output variables in each distribution network include node voltages. Input variables include the slack node voltage. Active power column vector of distribution network nodes reactive power column vector A high-dimensional linear state-space mapping model between them; specifically including:

[0063] Step 2-1: Collect G data samples based on the data, construct input and output samples; construct the distribution network voltage through least squares data-driven training. Active power injected into nodes Node-injected reactive power In a high-dimensional linear model between these two parameters, the g-th input sample is defined as:

[0064] (8)

[0065] The output sample is defined as:

[0066] (9)

[0067] in, , , Let represent the equilibrium voltage of the reference node under the g-th sample, and the active power and reactive power column vectors composed of each node of the distribution network under the g-th sample, respectively.

[0068] Step 2-2: Upscale the state space of the input samples to obtain the upscaled variables, which can be represented as:

[0069] (10)

[0070] In the formula, This represents the g-th input sample in the distribution network. This represents the up-dimensional function of the g-th input sample in the distribution network. Let represent the upgraded variable of the g-th input sample in the distribution network. Assume the dimension of the upgraded function is . ,but It can be defined as:

[0071] (11)

[0072] Defined as:

[0073] (12)

[0074] in, Indicates and Basis vectors of the same dimension represent The e-th dimension-increasing function, This represents the e-th Euclidean distance of the g-th data sample.

[0075] Steps 2-3: Based on the upgraded input sample set X and output sample set Y, obtain the data-driven power flow matrix through least squares training. The data-driven PF matrix is ​​obtained through data-driven training based on the aforementioned historical samples, where the input sample set after the distribution network dimensionality upgrade is defined as:

[0076] (13)

[0077] The distribution network output sample set is defined as:

[0078] (14)

[0079] Steps 2-4: Construct an incomplete dimensional upscaling power flow model, specifically including:

[0080] Distribution network data-driven power flow (PF) matrix The formula is as follows:

[0081] (15)

[0082] in, represent matrix transpose, represent The matrix pseudoinverse is given; therefore, the linear power distribution (PF) relationship of the distribution network is defined as:

[0083] (16)

[0084] Considering the strong nonlinearity of the dimension-upgrading function in equation (12), the input variables are divided into decision variables and state variables. Specifically, only the state variables are dimension-upgraded to adapt to the nonlinearity of PF. Therefore, the incomplete dimension-upgrading power flow model, i.e., the IDL-PF model, can be expressed as:

[0085] (17)

[0086] in, and These are the variable coefficient matrices. In this invention, the active power of distributed generation (DG) is used as the variable coefficient matrix. and reactive power Energy storage (ES) active power and reactive power Active power of flexible loads For decision variables, i.e. , The state variables are represented, including the active and reactive power of uncontrollable loads, thus transforming the original nonlinear PF model into a linear model.

[0087] In summary, the power flow (PF) equations and operating constraints of the distribution network are as follows:

[0088] (18)

[0089] Step 3: Establish the distribution network topology sub-matrix based on the topology adaptive power flow meta-model method. The power flow matrix in the incomplete up-dimensional power flow model Metamodel of mapping relationships between Specifically, it includes:

[0090] Step 3-1: Construct the sub-matrix of the distribution network topology Its elements represent the connection status between nodes; in this invention, the topology data of the distribution network can be simplified into a sub-matrix. , is represented as:

[0091] (19)

[0092] in, This indicates the connection state from node i to node j. Its value is 1 if node j is downstream of node i, and 0 otherwise. This represents the set of downstream nodes of node i.

[0093] Step 3-2: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context.] and Convert to column vector form and And establish a child matrix and IDL-PF

[0094] matrix The mapping relationship between them can be represented as:

[0095] (20)

[0096] in, Representative metamodel, and These are column vector samples of the offspring matrix and the linearized power flow matrix, respectively:

[0097] ; (twenty one)

[0098] in, The representative will Convert to A column vector of the form.

[0099] Step 3-3: Train the meta-model based on historical samples under topology Its sample set is defined as:

[0100] ; (twenty two)

[0101] in, and Let S represent the column vector samples of the offspring matrix and the linearized power flow matrix, respectively, and S be the number of retraining samples.

[0102] Step 3-3: Metamodel It can be determined using the least squares method:

[0103] ; (twenty three)

[0104] Metamodel using Moore-Penrose pseudoinverse It can be represented as:

[0105] ; (twenty four)

[0106] Steps 3-4: Obtain the updated power flow matrix based on the current topology, using the meta-model. The corresponding IDL-PF matrix It can be updated to:

[0107] (25)

[0108] in, This represents the column vector of the child matrix, composed of switch variables. Therefore, the corresponding IDL-PF model is:

[0109] (26)

[0110] In summary, equations (1), (3) to (7), and (26) constitute a complete distribution network emergency topology reconfiguration-VAR collaborative optimization control model.

[0111] Step 4: Based on the power flow matrix Heyuan model Online collaborative optimization of topology and VAR equipment enables emergency voltage control of the distribution network.

[0112] Example 3: A non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements a transient / steady-state calculation method for AC-side faults in an actively supported flexible DC-DC converter according to Example 1 of the present invention. The present invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate methods for implementing the transient / steady-state calculation method for AC-side faults in an actively supported flexible DC-DC converter. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0113] Example 4: An electronic device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements a transient / steady-state calculation method for AC-side faults of an actively supported flexible DC-DC converter according to Example 1 of the present invention. These computer program instructions may also be stored in a computer-readable storage medium capable of directing a computer or other programmable data processing device to operate in a specific manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process... Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0114] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0115] Verification example:

[0116] First, scenario setting. In extreme weather conditions, such as... Figure 4 The diagram shows the distribution network lines after a fault, including photovoltaic 1 and energy storage 2. Four different control methods were used for case analysis and verification.

[0117] Scenario 1: Natural voltage distribution without optimization;

[0118] Scenario 2: Independent VAR optimization;

[0119] Scenario 3: Particle Swarm Optimization Algorithm;

[0120] Scenario 4: Voltage optimization method based on topology adaptive power flow meta-model.

[0121] Secondly, a comparison of optimization results:

[0122] Figure 5 The voltage distribution of the distribution network under different voltage control schemes during a typical time period is shown. When a fault occurs on lines 19-20, the terminal bus voltage exceeds the limit under the no-voltage optimization mode, mainly due to the supply-demand imbalance caused by the line fault. Under the VAR independent optimization method, the voltage distribution is improved by adjusting flexible resources; however, due to the lack of topology optimization, the voltage exceeding the limit still occurs. In contrast, the particle swarm optimization algorithm and the voltage optimization method based on the topology adaptive power flow element model achieve voltage security and improve the overall voltage distribution by coordinating the network topology and flexible resources in the distribution network, and the proposed method outperforms the particle swarm optimization method.

[0123] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0124] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for coordinated control of active distribution networks based on a power flow meta-model, characterized in that, Includes the following steps: Step 1: Construct a comprehensive optimization objective function for emergency voltage control of the distribution network to minimize the costs of switching operations, distributed generation active power, energy storage output power, load shedding, and voltage deviation. Its expression is: ; And meet the operating constraints of the distribution network; Step 2: Based on historical operating data of the distribution network, a high-dimensional linear state-space model between node voltages and input variables is constructed offline through state-space mapping and least squares training. This includes: Step 2-1: Constructing Input Samples and output samples , , , Let represent the equilibrium voltage of the reference node under the g-th sample, and the active power and reactive power column vectors composed of each node of the distribution network under the g-th sample, respectively. This represents the g-th sample; Step 2-2: Upgrade the state space of the input samples to obtain the upgraded variables. ; Steps 2-3: Based on the upgraded input sample set X and output sample set Y, obtain the data-driven power flow matrix through least squares training. ; Steps 2-4: Constructing an incomplete dimensional power flow model: ,in, and These are the power flow variable coefficient matrices, Represents state variables, Represents decision variables, Functions for increasing the dimensionality of state variables; Step 3: Establish the distribution network topology sub-matrix based on the topology adaptive power flow meta-model method. The power flow matrix in the incomplete up-dimensional power flow model Metamodel of mapping relationships between Specifically, it includes: Step 3-1: Construct the sub-matrix of the distribution network topology Its elements represent the connection state between nodes; Step 3-2: [The text appears to be incomplete and contains several grammatical errors. A more accurate translation would require the full context.] and Convert to column vector form and And establish mapping relationships ; Step 3-3: Let and The meta-model represents the column vector samples of the offspring matrix and the power flow matrix, respectively, as follows: ; Steps 3-4: Obtain the updated power flow matrix based on the current topology. : ,in, Represents the column vector of the child matrix under the current topology. This is the power flow matrix corresponding to this topology. and Represented as power flow matrix The number of rows and columns; Step 4: Based on the aforementioned trend matrix Heyuan model Online collaborative optimization of topology and VAR equipment enables emergency voltage control of the distribution network.

2. The method according to claim 1, characterized in that, The dimensionality-upgrading function The dimension of is E, and its e-th dimension function is . = ,in, Indicates and Basis vectors of the same dimension represent The e-th dimension-increasing function, This represents the e-th Euclidean distance of the g-th data sample.

3. The method according to claim 1, characterized in that, The decision variable u includes the active power of the distributed generation. and reactive power Active power of energy storage and reactive power and the active power of flexible loads. .

4. The method according to claim 1, characterized in that, The state variable The active and reactive power of the uncontrollable load.

5. The method according to claim 1, characterized in that, The metamodel The following was obtained through least squares regression training: ; in, and These represent column vector samples of the offspring matrix and column vector samples of the power flow matrix, respectively.

6. The method according to claim 1, characterized in that, The online optimization phase adopts a rolling optimization method, updating the VAR optimization results every 15 minutes and the topology structure every 6 hours.

7. The method according to claim 1, characterized in that, The penalty coefficient in the comprehensive optimization objective function , , , and These represent the penalty coefficients for voltage deviation, DG abandonment, ES output power, load shedding, and switching cost, respectively.

8. The method according to claim 1, characterized in that, The power distribution network operation constraints include: Voltage amplitude upper and lower limit constraints; Flexible load reduction constraints; Distributed power generation / energy storage operation constraints; Topological connectivity constraints.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method according to any one of claims 1 to 8.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the method according to any one of claims 1 to 8.

Images