A power distribution network restoration strategy optimization method considering decentralized mobile energy storage

Abstract

The application relates to a power distribution network restoration strategy optimization method considering distributed mobile energy storage, which comprises the following steps: S1, considering the application of distributed power supply of the power distribution network, classifying nodes of the power distribution network, and adopting a Distflow model to establish a second-order cone power flow constraint of the power distribution network; S2, considering the dynamic change of the number of islands of the power distribution network, adopting a virtual flow theory to establish a radial topological constraint of the power distribution network; S3, combining the power flow constraint and the topological constraint, taking the maximum total load restored in an optimization time window as an objective function, and establishing a power distribution network restoration optimization model; S4, analyzing the distributed mobile energy storage dispatched by the power distribution network region in the optimization time window, and based on a sensitivity method, proposing a distributed mobile energy storage optimization dispatching model in the optimization time window; and S5, establishing a power distribution network restoration strategy optimization decision model considering the distributed mobile energy storage, and obtaining an optimal power distribution network restoration strategy. The beneficial effect is that the distributed mobile energy storage can be fully dispatched to participate in the power distribution network restoration, and the accuracy is higher.

Description

[Technical Field]

[0001] This invention relates to the field of power system information technology, and specifically to an optimization method for distribution network recovery strategy that considers distributed mobile energy storage. [Background Technology]

[0002] Distribution network recovery strategy optimization falls under the research field of power system recovery and resilience enhancement. In recent years, with the frequent occurrence of natural disasters such as typhoons and snowstorms, seasonal power supply imbalances have become increasingly apparent. To achieve rapid power restoration of the distribution network and ensure the safe and reliable energy use of power users, an effective and accurate distribution network recovery strategy optimization method is urgently needed. After a distribution network accident, an applicable distribution network recovery plan can be formulated using distribution network recovery strategy optimization methods.

[0003] In recent years, distributed mobile energy storage, represented by electric vehicles, has seen rapid development, with the number of electric vehicles in typical large-scale cities reaching hundreds of thousands. As an important supporting measure for distributed mobile energy storage, and with the vigorous development of V2G technology, urban electric vehicle charging stations have become crucial distributed power nodes in power distribution networks. After a power distribution network failure, considering the recovery power provided by distributed mobile energy storage within the region, and using electric vehicle charging stations as interfaces to participate in power distribution network restoration, has gradually become a research hotspot. Therefore, optimization methods for power distribution network restoration strategies that consider distributed mobile energy storage are particularly important.

[0004] Electric vehicle V2G technology refers to the technology of electric vehicles supplying electricity to the power grid. Its core idea is to use the energy storage of a large number of electric vehicles as a buffer between the power grid and renewable energy sources. [Summary of the Invention]

[0005] The purpose of this invention is to propose an optimization method for distribution network restoration strategies that fully utilizes distributed mobile energy storage to participate in distribution network restoration and achieves higher accuracy.

[0006] To achieve the above objectives, the technical solution adopted by this invention is a distribution network restoration strategy optimization method considering distributed mobile energy storage, comprising the following steps:

[0007] Step S1: Considering the application of distributed generation in the distribution network, classify the nodes of the distribution network and use the Distflow model to establish second-order cone power flow constraints for the distribution network.

[0008] Step S2: Considering the dynamic changes in the number of distribution network islands, establish radial topology constraints for the distribution network using virtual flow theory;

[0009] Step S3: Combining power flow constraints and topology constraints, and taking the maximization of the total load restored within the optimization time window as the objective function, establish a distribution network restoration optimization model;

[0010] Step S4: Analyze the dispatchable distributed mobile energy storage in the distribution network area within the optimization time window, and propose an optimized scheduling model for distributed mobile energy storage within the optimization time window based on the sensitivity method.

[0011] Step S5: Combining the distribution network restoration optimization model and the distributed mobile energy storage optimization scheduling model, establish a distribution network restoration strategy optimization decision model that considers distributed mobile energy storage, and solve for the optimal distribution network restoration strategy.

[0012] Preferably, step S1 specifically includes the following sub-steps:

[0013] Step S11: Analyze the source-grid-load-storage resources of the target distribution network, classify the distribution network nodes, and define S. bus For the set of all nodes in the distribution network, For the set of load nodes in the distribution network, S is the set of power source nodes in the distribution network. feeder For distribution network feeder collection;

[0014] Step S12, Define S T For the set of discrete-time sections within the optimal time window T for distribution network restoration, based on the Distflow model, the expression for the second-order cone power flow constraint of the h-th discrete-time section is as follows:

[0015]

[0016]

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023]

[0024]

[0025]

[0026]

[0027]

[0028]

[0029] Equations (1) and (2) represent the active and reactive power balance constraints at distribution network nodes, respectively, where P ij (h) and Q ij (h) represents the active and reactive power variables of the feeder ij at the h-th discrete time section. Let r be the squared variable of the current in feeder ij at the h-th discrete time section. ij and x ij P represents the resistance and reactance parameters of the feeder ij. i DG (h) and Let P be the active and reactive power output variables of the power source at node i in the h-th discrete time section. i load (h) and Let i be the active and reactive load variables of node i at the h-th discrete time section.

[0030] Equation (3) characterizes the constraint on the relationship between feeder node voltage, feeder active power, feeder reactive power, and feeder current in the distribution network, where V i sqr (h) represents the squared voltage variable at node i of the h-th discrete-time section;

[0031] Equation (4) is the feeder capacity constraint after second-order cone relaxation;

[0032] Equations (5)-(12) represent the active and reactive power output and active and reactive load limits of the node power source, where P i DG ,max (h) and P represents the upper limit parameter values ​​for active and reactive power at node i in the h-th discrete time section; i load,max (h) and The active and reactive load upper limit parameters for node i at the h-th discrete time section; v i (h) represents the 0-1 state variable for node i to be restored, and v i (h) = 1 indicates that node i of the h-th discrete-time section has been recovered, v i (h) = 0 indicates that node i of the discrete time section h has not been recovered;

[0033] Equations (13)-(14) represent the current constraints on the node voltage and feeder current of the distribution network. In the equations, and Here are the lower and upper limits of the voltage at node i in the h-th discrete time section; The upper limit parameter value of the power supply of feeder ij at the h-th discrete time section; w ij(h) represents the 0-1 recovery state variable of feeder ij, w ij (h) = 1 indicates that the h-th discrete-time section feeder ij has been restored, w ij (h) = 0 indicates that the h-th discrete-time section feeder ij has not been recovered.

[0034] Preferably, step S2 specifically includes the following sub-steps:

[0035] Step S21: During the distribution network restoration process, the distribution network needs to maintain a radial topology constraint, that is, each restored island also needs to maintain a radial topology:

[0036]

[0037]

[0038] Define a virtual node with the index 0, as shown in equation (15). Virtual node 0 and all nodes of the distribution network constitute the set of virtual network nodes of the distribution network. In the distribution network virtual network, virtual feeders are added to virtual node 0 and the distribution network power source node. As shown in equation (16), the newly added virtual feeder and all feeders of the distribution network constitute the virtual network feeder set of the distribution network.

[0039] Step S22: Considering the dynamic change in the number of distribution network islands during the recovery process, i.e., the number of distribution network islands differs at different discrete time sections, a radial topology constraint for the distribution network is established using virtual flow theory. Firstly, in... Based on this, define the directed graph feeder set of the virtual distribution network.

[0040]

[0041] The radial topology constraints of the h-th discrete-time section distribution network are as follows:

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049] Equation (18) is the virtual power balance constraint of the distribution network virtual network, where H ij (h) represents the virtual power variable of feeder ij in the directed graph of the virtual distribution network at the h-th discrete time section. For node i in the virtual network of the distribution network at the h-th discrete time section, use state 0-1 variables. This indicates that node i in the virtual network of the distribution network at the h-th discrete time section is in use. This indicates that node i in the virtual network of the distribution network at the h-th discrete time section is in an unused state;

[0050] Equation (19) represents the virtual power constraint on the feeders of the directed graph of the distribution network virtual network. In the equation, For the directed graph of the distribution network virtual network, the feeder ij uses state 0-1 variables. This indicates that feeder ij in the directed graph of the virtual distribution network at the h-th discrete-time section is in use. This indicates that the feeder ij of the directed graph of the distribution network virtual network at the h-th discrete time section is in an unused state; M is a large positive integer parameter, and for a distribution network with no more than 100 nodes, M can be 10000;

[0051] Equations (20)-(21) are the consistency constraints for mapping the node states and feeder states in the distribution network and the distribution network virtual network;

[0052] Equation (22) restricts virtual node 0 in the distribution network virtual network from being in use;

[0053] Equation (23) restricts the addition of new virtual feeders in the distribution network virtual network. Usage status;

[0054] Equation (24) restricts the constraint relationship between the state variables used by a node and the state variables used by the feeder connected to the node in the virtual distribution network.

[0055] Preferably, step S3 specifically includes the following sub-steps:

[0056] Step S31: The set of discrete time sections within the distribution network restoration optimization time window T is S. T Let Δt be the discrete time step between adjacent discrete time sections. In the distribution network restoration strategy, the objective is to maximize the total load restored within the restoration optimization time window. The optimization objective of the distribution network restoration optimization model is defined as follows:

[0057]

[0058] In equation (25), The load importance parameter for nodes can be used to classify distribution network load nodes into Class I, Class II, and Class III nodes, with the load importance parameter decreasing sequentially. sum To optimize the total load recovery within the distribution network restoration time window;

[0059] Step S32: Combining the distribution network restoration optimization objective function in equation (25), the power flow constraints and topology constraints in steps S1 and S2 are rewritten and extended. The constraints included in the distribution network restoration optimization model are as follows:

[0060]

[0061]

[0062]

[0063]

[0064]

[0065]

[0066]

[0067]

[0068]

[0069]

[0070]

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] Equations (26)-(50) constitute the constraints of the distribution network restoration optimization model, wherein equations (26)-(39) are the distribution network power flow constraints, the specific definitions of which refer to step S1; equations (40)-(46) are the distribution network radial topology constraints, the specific definitions of which refer to step S2; and equations (47)-(50) are the distribution network restoration persistence constraints, where |S T | is the total number of discrete time sections within the optimal time window for distribution network recovery. Equations (47)-(48) ensure the continuity of the recovery of the distribution network node and feeder status, and equations (49)-(50) ensure the continuity of the recovery of the active and reactive loads of the load nodes.

[0086] Preferably, step S4 specifically includes the following sub-steps:

[0087] Step S41: During the distribution network restoration process, distributed mobile energy storage within the distribution network area can be connected to the distribution network power nodes through spatiotemporal transfer and provide restoration power to the distribution network through V2G technology. The set of distributed mobile energy storage available for dispatch to participate in distribution network restoration within the distribution network restoration optimization time window T is defined as S. EV At the same time, set S EV The total number of distributed mobile energy storage devices included is N. EV According to the order of available scheduling times, the total number of N... EV Distributed mobile energy storage is sorted, i.e., S EV ={1,2,...,N EV};

[0088] Step S42, During the power distribution network restoration process, distributed mobile energy storage EVs i One of the multiple power supply nodes in the distribution network can be selected to connect to the distribution network and provide recovery power to the distribution network.

[0089] Preferably, step S42 proposes a distributed mobile energy storage optimization scheduling model within the distribution network recovery optimization time window based on the sensitivity method, that is, it uses a sequential approach to determine S in turn. EV The total number is N EV The specific steps for selecting appropriate power supply nodes in the distribution network to participate in distribution network restoration for distributed mobile energy storage are as follows:

[0090] Step (1), for S EV The first distributed mobile energy storage device, EV1, is selected based on constraints such as the upper limit on the number of distributed mobile energy storage devices that can be connected within the power nodes of the distribution network. The set of power nodes that EV1 can connect to is defined as follows:

[0091] Step (2), Consider The CCP There are several power supply nodes in the distribution network that can be connected to EV1. Assume that EV1 is connected to the m-th power supply node in the distribution network. Based on the time when EV1 arrives and connects to the power distribution node m, the remaining power battery charge, and the rated power output of the distributed mobile energy storage grid at power distribution node m, update the active and reactive power upper limit parameters of power distribution node m. and Meanwhile, an optimization model for distribution network restoration after EV1 accesses the m-th distribution network power node and participates in distribution network restoration is established using equations (25)-(50). The total load restored by the distribution network within the restoration optimization time window after EV1 accesses the m-th distribution network power node and participates in distribution network restoration is obtained by solving the optimization model. sum (EV1,m); The enumeration method is used to calculate the effect of EV1 participating in distribution network restoration by connecting to different distribution network power nodes, and the total load restored by the distribution network within the restoration optimization time window after EV1 participates in distribution network restoration by connecting to different distribution network power nodes is obtained. sum (EV1,m),

[0092] Step (3): Calculate the total load (Load) restored by the distribution network within the restoration optimization time window after EV1 connects to different distribution network power nodes to participate in distribution network restoration. sum (EV1,m), The data is sorted in descending order. If the total restored load is the same, it is sorted again in ascending order according to the power supply node number of the distribution network. After sorting, it is assumed that EV1 is connected to the power supply node of the distribution network. After participating in the distribution network restoration, the distribution network ranks first in terms of the total load restored within the restoration optimization time window, and the EV1 distribution network power node is considered to be the first in terms of the total load restored. The EV1 power supply node was selected for connection to the distribution network because it exhibits the highest sensitivity in participating in distribution network restoration. Participate in power distribution network restoration;

[0093] Step (4): Complete the determination of the power supply node for EV1 to be connected to the distribution network. After participating in the distribution network restoration, EV1 arrives and connects to the distribution network power node. The time, remaining power battery charge, and power distribution network nodes Distributed mobile energy storage grids deliver rated power and update power supply nodes in the distribution network. Upper limit parameters for active and reactive power and

[0094] Step (5): After completing the update of the power node parameters of the distribution network, repeat steps (1)-(4) to determine the specific power nodes for the distributed mobile energy storage EV2 to participate in the distribution network restoration and update the parameters of the corresponding power nodes; and so on, sequentially determine S EV Internal distributed mobile energy storage The corresponding power supply nodes of the distribution network that participate in the restoration of the distribution network access are also updated with parameters.

[0095] Preferably, step S5 specifically includes the following sub-steps:

[0096] Step S51: Analyze and collect relevant information on the distributed mobile energy storage available for dispatch to participate in distribution network restoration within the distribution network restoration optimization time window T. Based on the sensitivity method proposed in step S4, determine the optimal dispatch model for distributed mobile energy storage within the distribution network restoration optimization time window, and then determine S. EV The total number is N EV Distributed mobile energy storage participates in the corresponding power nodes of the distribution network during distribution network restoration, and simultaneously completes the upper limit parameters of active and reactive power of the corresponding power nodes. and Update parameter values;

[0097] Step S52: Complete the upper limit parameters of active and reactive power for the corresponding power supply nodes in the distribution network. and After the parameter values ​​are updated, the parameter values ​​of the power supply nodes in the power distribution network included in the updated equations (25)-(50) are used to form the optimized decision-making model of the power distribution network restoration strategy considering distributed mobile energy storage. The optimized decision-making model of the power distribution network restoration strategy considering distributed mobile energy storage is a typical mixed integer second-order cone optimization problem. After solving it, the optimal restoration strategy of the power distribution network is obtained.

[0098] Compared with existing technologies, the present invention provides an optimization method for distribution network restoration strategies that considers distributed mobile energy storage. This method has the following advantages: it considers the application of distributed power sources in the distribution network and the dynamic changes in the number of distribution network islands. With the objective function being the maximum total load restored within the optimization time window, it proposes an optimized scheduling model for distributed mobile energy storage within the optimization time window based on the sensitivity method. Solving this model yields the optimal restoration strategy for the distribution network, fully utilizing distributed mobile energy storage to participate in distribution network restoration with higher accuracy. [Attached Image Description]

[0099] Figure 1 This is a step-by-step diagram of a distribution network restoration strategy optimization method that considers distributed mobile energy storage.

[0100] Figure 2 This is an optimization method for distribution network restoration strategies that considers distributed mobile energy storage. It is a schematic diagram of distribution network nodes, feeders, and voltage, current, and power variables.

Detailed Implementation Methods

[0101] The present invention will now be further described with reference to the embodiments and the accompanying drawings.

[0102] Example

[0103] This embodiment implements an interactive game theory method between aggregators and users for flexible optimization.

[0104] This embodiment fully utilizes distributed mobile energy storage to participate in power grid restoration, resulting in higher accuracy.

[0105] Figure 1 This is a flowchart illustrating a distribution network restoration strategy optimization method considering distributed mobile energy storage. As shown in the figure, this embodiment of a distribution network restoration strategy optimization method considering distributed mobile energy storage includes the following steps:

[0106] Step S1: Considering the application of distributed generation in the distribution network, classify the nodes of the distribution network and use the Distflow model to establish second-order cone power flow constraints for the distribution network.

[0107] Step S2: Considering the dynamic changes in the number of distribution network islands, establish radial topology constraints for the distribution network using virtual flow theory;

[0108] Step S3: Combining power flow constraints and topology constraints, and taking the maximization of the total load restored within the optimization time window as the objective function, establish a distribution network restoration optimization model;

[0109] Step S4: Analyze the dispatchable distributed mobile energy storage in the distribution network area within the optimization time window, and propose an optimized scheduling model for distributed mobile energy storage within the optimization time window based on the sensitivity method.

[0110] Step S5: Combining the distribution network restoration optimization model and the distributed mobile energy storage optimization scheduling model, establish a distribution network restoration strategy optimization decision model that considers distributed mobile energy storage, and solve for the optimal distribution network restoration strategy.

[0111] Figure 2 This is an optimization method for distribution network restoration strategies that considers distributed mobile energy storage. A schematic diagram of distribution network nodes, feeders, and voltage, current, and power variables is attached. Figure 2 As shown, in this embodiment, step S1 is specifically implemented as follows:

[0112] Step S11: Analyze the source-grid-load-storage resources of the target distribution network, classify the distribution network nodes, and define S. bus For the set of all nodes in the distribution network, For the set of load nodes in the distribution network, S is the set of power source nodes in the distribution network. feeder This refers to the collection of feeders in the distribution network.

[0113] Step S12, Define S T For the set of discrete-time sections within the optimal time window T for distribution network restoration, based on the Distflow model, the expression for the second-order cone power flow constraint of the h-th discrete-time section is as follows:

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127]

[0128] Equations (1) and (2) represent the active and reactive power balance constraints at distribution network nodes, respectively, where P ij (h) and Q ij (h) represents the active and reactive power variables of the feeder ij at the h-th discrete time section. Let r be the squared variable of the current in feeder ij at the h-th discrete time section. ij and x ij P represents the resistance and reactance parameters of the feeder ij. i DG (h) and Let P be the active and reactive power output variables of the power source at node i in the h-th discrete time section. i load (h) and Let be the active and reactive load variables of node i at the h-th discrete time section.

[0129] Equation (3) characterizes the constraint on the relationship between feeder node voltage, feeder active power, feeder reactive power, and feeder current in the distribution network, where V i sqr (h) represents the squared voltage variable of node i at the h-th discrete time section. Equation (4) is the feeder capacity constraint after second-order cone relaxation. Equations (5)-(12) are the active and reactive power output and active and reactive load limit constraints of the node power supply, where P i DG ,max (h) and P represents the upper limit parameter values ​​for active and reactive power at node i in the h-th discrete time section; i load,max (h) and The active and reactive load upper limit parameters for node i at the h-th discrete time section; v i (h) represents the 0-1 state variable for node i to be restored, and v i (h) = 1 indicates that node i of the h-th discrete-time section has been recovered, v i (h) = 0 indicates that node i of the discrete time section at time h has not been recovered.

[0130] Equations (13)-(14) represent the current constraints on the node voltage and feeder current of the distribution network. In the equations, and Here are the lower and upper limits of the voltage at node i in the h-th discrete time section; The upper limit parameter value of the power supply of feeder ij at the h-th discrete time section; w ij (h) represents the 0-1 recovery state variable of feeder ij, w ij (h) = 1 indicates that the h-th discrete-time section feeder ij has been restored, w ij(h) = 0 indicates that the h-th discrete-time section feeder ij has not been recovered.

[0131] In this embodiment, step S2 is implemented as follows:

[0132] Step S21: During the distribution network restoration process, the distribution network needs to maintain a radial topology constraint, that is, each restored island also needs to maintain a radial topology. Define a virtual node with the index 0, as shown in equation (15). Virtual node 0 and all nodes of the distribution network constitute the set of virtual network nodes of the distribution network. In the distribution network virtual network, virtual feeders are added to virtual node 0 and the distribution network power source node. As shown in equation (16), the newly added virtual feeder and all feeders of the distribution network constitute the virtual network feeder set of the distribution network.

[0133]

[0134]

[0135] Step S22: Considering the dynamic change in the number of distribution network islands during the recovery process, i.e., the number of distribution network islands is different at different discrete time sections, the radial topology constraint of the distribution network is established using virtual flow theory, as shown in Equation (17). First, in Based on this, define the directed graph feeder set of the virtual distribution network.

[0136]

[0137] The radial topology constraints of the h-th discrete-time section distribution network are as follows:

[0138]

[0139]

[0140]

[0141]

[0142]

[0143]

[0144]

[0145] Equation (18) is the virtual power balance constraint of the distribution network virtual network, where H ij (h) represents the virtual power variable of feeder ij in the directed graph of the virtual distribution network at the h-th discrete time section. For node i in the virtual network of the distribution network at the h-th discrete time section, use state 0-1 variables. This indicates that node i in the virtual network of the distribution network at the h-th discrete time section is in use. This indicates that node i in the virtual network of the distribution network at the h-th discrete time section is in an unused state.

[0146] Equation (19) represents the virtual power constraint on the feeders of the directed graph of the distribution network virtual network. In the equation, For the directed graph of the distribution network virtual network, the feeder ij uses state 0-1 variables. This indicates that feeder ij in the directed graph of the virtual distribution network at the h-th discrete-time section is in use. The h-th discrete-time section of the distribution network virtual network feeder ij is in an unused state; M is a large positive integer parameter, which can be 10000 for distribution networks with no more than 100 nodes.

[0147] Equations (20) and (21) represent the consistency constraints on the mapping between node states and feeder states in the distribution network and the distribution network virtual network. Equation (22) restricts virtual node 0 in the distribution network virtual network from being in use, and Equation (23) restricts the addition of new virtual feeders in the distribution network virtual network. The usage status. Equation (24) restricts the constraint relationship between the usage status variables of a node and the usage status variables of the feeders connected to that node in the distribution network virtual network.

[0148] In this embodiment, step S3 is implemented as follows:

[0149] Step S31: The set of discrete time sections within the distribution network restoration optimization time window T is S. T Let Δt be the discrete time step between adjacent discrete time sections. The distribution network restoration strategy aims to maximize the total load restored within the optimal restoration time window. The optimization objective of the distribution network restoration optimization model is defined as follows:

[0150]

[0151] In equation (25), The load importance parameter for nodes can be used to classify distribution network load nodes into Class I, Class II, and Class III nodes, with the load importance parameter decreasing sequentially. sum The total load to be restored within the optimal time window for power distribution network restoration.

[0152] Step S32: Combining the distribution network restoration optimization objective function in equation (25), the power flow constraints and topology constraints in steps S1 and S2 are rewritten and extended. The constraints included in the distribution network restoration optimization model are as follows:

[0153]

[0154]

[0155]

[0156]

[0157]

[0158]

[0159]

[0160]

[0161]

[0162]

[0163]

[0164]

[0165]

[0166]

[0167]

[0168]

[0169]

[0170]

[0171]

[0172]

[0173]

[0174]

[0175]

[0176]

[0177]

[0178] Equations (26)-(50) constitute the constraints of the distribution network restoration optimization model, wherein equations (26)-(39) are the distribution network power flow constraints, the specific definitions of which can be found in step S1; equations (40)-(46) are the distribution network radial topology constraints, the specific definitions of which can be found in step S2; and equations (47)-(50) are the distribution network restoration persistence constraints, where |S T | is the total number of discrete time sections within the optimal time window for distribution network recovery. Equations (47)-(48) ensure the continuity of the recovery of the distribution network node and feeder status, and equations (49)-(50) ensure the continuity of the recovery of the active and reactive loads of the load nodes.

[0179] In this embodiment, step S4 is implemented as follows:

[0180] Step S41: During the distribution network restoration process, distributed mobile energy storage within the distribution network area can be connected to the distribution network power nodes through spatiotemporal transfer and provide restoration power to the distribution network through V2G (distributed mobile energy storage supplying power to the grid) technology. Define the set of distributed mobile energy storage available for dispatch to participate in distribution network restoration within the distribution network restoration optimization time window T as S. EV At the same time, set S EV The total number of distributed mobile energy storage devices included is N. EV According to the order of available scheduling times, the total number of N... EV Distributed mobile energy storage is sorted, i.e., S EV ={1,2,...,N EV}

[0181] Step S42, During the power distribution network restoration process, distributed mobile energy storage EVs i One of the multiple power sources in the distribution network can be selected to connect to the distribution network and provide recovery power. This step proposes a distributed mobile energy storage optimization scheduling model within the distribution network recovery optimization time window based on the sensitivity method, that is, it adopts a sequential approach to determine S in turn. EV The total number is N EV The specific steps for selecting appropriate power supply nodes in the distribution network to participate in distribution network restoration for distributed mobile energy storage are as follows:

[0182] (1) For S EV The first distributed mobile energy storage device, EV1, is selected based on constraints such as the upper limit on the number of distributed mobile energy storage devices that can be connected within the power nodes of the distribution network. The set of power nodes that EV1 can connect to is defined as follows:

[0183] (2) Consider The CCP There are several power supply nodes in the distribution network that can be connected to EV1. Assume that EV1 is connected to the m-th power supply node in the distribution network. Based on the time when EV1 arrives and connects to the power distribution node m, the remaining power battery charge, and the rated power output of the distributed mobile energy storage grid at power distribution node m, update the active and reactive power upper limit parameters of power distribution node m. and Meanwhile, an optimization model for distribution network restoration after EV1 accesses the m-th distribution network power node and participates in distribution network restoration is established using equations (25)-(50). The total load restored by the distribution network within the restoration optimization time window after EV1 accesses the m-th distribution network power node and participates in distribution network restoration is obtained by solving the optimization model. sum (EV1,m). The enumeration method is used to calculate the effect of EV1 participating in distribution network restoration by connecting to different distribution network power nodes, obtaining the total load restored by the distribution network within the optimal restoration time window after EV1 participates in distribution network restoration by connecting to different distribution network power nodes. sum (EV1,m),

[0184] (3) The total load restored by the distribution network within the restoration optimization time window after EV1 connects to different distribution network power nodes and participates in distribution network restoration. sum (EV1,m), The data is sorted in descending order. If the total restored load is the same, it is then sorted again in ascending order by the distribution network power node number. After sorting, assuming EV1 is connected to the distribution network power node... After participating in the distribution network restoration, the distribution network ranks first in terms of the total load restored within the restoration optimization time window, and the EV1 distribution network power node is considered to be the first in terms of the total load restored. The EV1 power supply node was selected for connection to the distribution network because it exhibits the highest sensitivity in participating in distribution network restoration. Participate in the restoration of the power distribution network.

[0185] (4) Complete the determination of the power supply node for EV1 to be connected to the distribution network. After participating in the distribution network restoration, EV1 arrives and connects to the distribution network power node. The time, remaining power battery charge, and power distribution network nodes Distributed mobile energy storage grids deliver rated power and update power supply nodes in the distribution network. Upper limit parameters for active and reactive power and

[0186] (5) After updating the power supply node parameters of the distribution network, repeating steps (1)-(4) can determine the specific power supply nodes for the distributed mobile energy storage EV2 to participate in the distribution network restoration and update the parameters of the corresponding power supply nodes. Similarly, S can be determined sequentially. EV Internal distributed mobile energy storage The corresponding power supply nodes of the distribution network that participate in the restoration of the distribution network access are also updated with parameters.

[0187] In this embodiment, step S5 is implemented as follows:

[0188] Step S51: Analyze and collect relevant information on the distributed mobile energy storage available for dispatch to participate in distribution network restoration within the distribution network restoration optimization time window T. Based on the sensitivity method proposed in step S4, determine the optimal dispatch model for distributed mobile energy storage within the distribution network restoration optimization time window, and then determine S. EV The total number is N EV Distributed mobile energy storage participates in the corresponding power nodes of the distribution network during distribution network restoration, and simultaneously completes the upper limit parameters of active and reactive power of the corresponding power nodes. and Update the parameter values.

[0189] Step S52: Complete the upper limit parameters of active and reactive power for the corresponding power supply nodes in the distribution network. and After updating the parameter values, the parameter values ​​of the power supply nodes in the distribution network included in equations (25)-(50) are updated. The updated objective function (25) and constraint equations (26)-(50) constitute the optimal decision-making model for distribution network restoration strategy considering distributed mobile energy storage. The established optimal decision-making model for distribution network restoration strategy considering distributed mobile energy storage is a typical mixed integer second-order cone optimization problem, which can be effectively solved. After solving the established optimal decision-making model for distribution network restoration strategy considering distributed mobile energy storage, the optimal restoration strategy for the distribution network is obtained.

[0190] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware or by a program instructing related hardware. The program can be stored in a computer-readable storage medium, such as a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM).

[0191] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several improvements and additions without departing from the principle of the present invention, and these improvements and additions should also be considered within the scope of protection of the present invention.

Claims

1. A method for optimizing distribution network restoration strategies considering distributed mobile energy storage, characterized in that... Includes the following steps: Step S1: Considering the application of distributed generation in the distribution network, classify the nodes of the distribution network and use the Distflow model to establish second-order cone power flow constraints for the distribution network. Step S2: Considering the dynamic changes in the number of distribution network islands, establish radial topology constraints for the distribution network using virtual flow theory; Step S3: Combining power flow constraints and topology constraints, establish a distribution network restoration optimization model with the objective function of maximizing the total load restored within the optimization time window; Step S4: Analyze the distributed mobile energy storage available for dispatch in the distribution network area within the optimization time window, and propose an optimized dispatch model for distributed mobile energy storage within the optimization time window based on the sensitivity assessment method of the impact of candidate access nodes on the total load recovery of the distribution network. Step S5: Combining the distribution network restoration optimization model and the distributed mobile energy storage optimization scheduling model, establish a distribution network restoration strategy optimization decision model that considers distributed mobile energy storage, and solve for the optimal distribution network restoration strategy.

2. The method for optimizing distribution network recovery strategies considering distributed mobile energy storage according to claim 1, characterized in that... Step S1 specifically includes the following sub-steps: Step S11: Analyze the source-grid-load-storage resources of the target distribution network, classify the distribution network nodes, and define... For the set of all nodes in the distribution network, For the set of load nodes in the distribution network, For the set of power supply nodes in the distribution network, For distribution network feeder collection; Step S12, Definition Optimize the time window for power distribution network restoration The set of discrete-time sections within the inner region, based on the Distflow model, is the first... The expression for the discrete-time section second-order cone power flow constraint is as follows: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) Equations (1) and (2) represent the active and reactive power balance constraints at distribution network nodes, respectively. and For the first Discrete-time section feeder Active and reactive power variables, For the first Discrete-time section feeder Current square variable, and For feeder Resistance and reactance parameters, and For the first Discrete-time section nodes The active and reactive power output variables of the power source. and For the first Discrete-time section nodes Active and reactive load variables at nodes; Equation (3) characterizes the constraint on the relationship between feeder node voltage, feeder active power, feeder reactive power, and feeder current in the distribution network. For the first Discrete-time section nodes Voltage squared variable; Equation (4) is the feeder capacity constraint after second-order cone relaxation; Equations (5)-(12) represent the active and reactive power output and active and reactive load limits of the node power source, where, and For the first Discrete-time section nodes Upper limit values ​​for active and reactive power; and For the first Discrete-time section nodes Maximum active and reactive load parameters; For nodes Restore the state 0-1 variables. =1 indicates the first Discrete-time section nodes It has been restored. =0 indicates the first Discrete-time section nodes Not recovered; Equations (13)-(14) represent the current constraints on the node voltage and feeder current of the distribution network. In the equations, and For the first Discrete-time section nodes Voltage lower limit and upper limit parameter values; For the first Discrete-time section feeder Battery capacity limit parameter value; For feeder Restore the state 0-1 variables. =1 indicates the first Discrete-time section feeder It has been restored. =0 indicates the first Discrete-time section feeder Not yet restored.

3. The method for optimizing distribution network recovery strategy considering distributed mobile energy storage according to claim 2, characterized in that... Step S2 specifically includes the following sub-steps: Step S21: During the distribution network restoration process, the distribution network needs to maintain a radial topology constraint, that is, each restored island also needs to maintain a radial topology: (15) (16) Define a virtual node with the index 0, as shown in equation (15). Virtual node 0 and all nodes of the distribution network constitute the set of virtual network nodes of the distribution network. In the distribution network virtual network, virtual feeders are added to virtual node 0 and the distribution network power source node. As shown in equation (16), the newly added virtual feeder and all feeders of the distribution network constitute the virtual network feeder set of the distribution network. ; Step S22: Considering the dynamic change in the number of distribution network islands during the recovery process, i.e., the number of distribution network islands differs at different discrete time sections, a radial topology constraint for the distribution network is established using virtual flow theory. Firstly, in... Based on this, define the directed graph feeder set of the virtual distribution network. : (17) No. The radial topology constraints of the discrete-time cross-sectional distribution network are as follows: (18) (19) (20) (21) (22) (23) (24) Equation (18) represents the virtual power balance constraint of the distribution network virtual network, where, For the first Discrete-time cross-section distribution network virtual network directed graph feeder Virtual power variable, For the first Discrete-time cross-section distribution network virtual network node Use state variables 0-1. =1 indicates the first Discrete-time cross-section distribution network virtual network node In use =0 indicates the first Discrete-time cross-section distribution network virtual network node It is not in use; Equation (19) represents the virtual power constraint on the feeders of the directed graph of the distribution network virtual network. In the equation, For the directed graph feeder of the distribution network virtual network Use state variables 0-1. =1 indicates the first Discrete-time cross-section distribution network virtual network directed graph feeder In use =0 indicates the first Discrete-time cross-section distribution network virtual network directed graph feeder It is not in use; For a distribution network with no more than 100 nodes, the parameter should be a large positive integer. Take 10000; Equations (20)-(21) are the consistency constraints for mapping the node states and feeder states in the distribution network and the distribution network virtual network; Equation (22) restricts virtual node 0 in the distribution network virtual network from being in use; Equation (23) restricts the addition of new virtual feeders in the distribution network virtual network. Usage status; Equation (24) restricts the constraint relationship between the state variables used by a node and the state variables used by the feeder connected to the node in the virtual distribution network.

4. The method for optimizing distribution network restoration strategies considering distributed mobile energy storage according to claim 3, characterized in that... Step S3 specifically includes the following sub-steps: Step S31, Optimization time window for distribution network restoration The set of discrete-time sections is We attempt to define the discrete time step between adjacent discrete time sections as follows: The distribution network restoration strategy aims to maximize the total load restored within the restoration optimization time window. The optimization objective of the distribution network restoration optimization model is defined as follows: (25) In equation (25), The load importance parameter is used to classify the distribution network load nodes into Class I, Class II, and Class III nodes, with the load importance parameter decreasing sequentially. To optimize the total load recovery within the distribution network restoration time window; Step S32: Combining the distribution network restoration optimization objective function in equation (25), the power flow constraints and topology constraints in steps S1 and S2 are rewritten and extended. The constraints included in the distribution network restoration optimization model are as follows: (26) (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) (41) (42) (43) (44) (45) (46) (47) (48) (49) (50) Equations (26)-(50) constitute the constraints of the distribution network restoration optimization model, wherein equations (26)-(39) are the distribution network power flow constraints, the specific definitions of which refer to step S1; equations (40)-(46) are the distribution network radial topology constraints, the specific definitions of which refer to step S2; and equations (47)-(50) are the distribution network restoration persistence constraints, wherein... The total number of discrete time sections within the optimal time window for distribution network recovery is determined by equations (47)-(48), which ensure the continuity of the recovery of distribution network node and feeder status, and equations (49)-(50), which ensure the continuity of the recovery of active and reactive loads at load nodes.

5. The method for optimizing distribution network restoration strategies considering distributed mobile energy storage according to claim 4, characterized in that... Step S4 specifically includes the following sub-steps: Step S41: During the distribution network restoration process, distributed mobile energy storage within the distribution network area is connected to the distribution network power nodes via spatiotemporal transfer and provides restoration power to the distribution network through V2G technology, defining the optimal time window for distribution network restoration. The distributed mobile energy storage system that participates in distribution network restoration through internal power dispatch is: At the same time, set The total number of distributed mobile energy storage units included is According to the order of response scheduling, the total quantity is Distributed mobile energy storage is sorted, that is... ; Step S42, During the power distribution network restoration process, distributed mobile energy storage Select one of the multiple power supply nodes in the distribution network to connect to the distribution network and provide recovery power to the distribution network.

6. The method for optimizing distribution network recovery strategy considering distributed mobile energy storage according to claim 5, characterized in that... Step S42 proposes an optimized scheduling model for distributed mobile energy storage within the distribution network recovery optimization time window based on the sensitivity method, that is, it adopts a sequential approach to determine the optimal scheduling model for distributed mobile energy storage. The total number is The specific steps for selecting appropriate power supply nodes in the distribution network to participate in distribution network restoration for distributed mobile energy storage are as follows: Step (1), for The first distributed mobile energy storage vehicle in the project, namely Considering constraints such as the upper limit on the number of distributed mobile energy storage devices connected within the power supply nodes of the distribution network, the following were selected: Accessible power supply nodes in the distribution network, and define them. The set of power supply nodes that can be accessed in the distribution network is as follows ; Step (2), consider The CCP Each power distribution node can provide Access, assuming Access the Each power distribution node ( ),according to Arrive at and connect to the power distribution network node The time, remaining power battery charge, and power distribution network nodes Distributed mobile energy storage grids deliver rated power and update power supply nodes in the distribution network. Upper limit parameters for active and reactive power and Meanwhile, by establishing equations (25)-(50) Access the The distribution network restoration optimization model involves multiple distribution network power nodes and is obtained by solving the optimization model. Access the The total load restored by the distribution network within the optimal restoration time window after each distribution network power node participates in the distribution network restoration. ; Calculate separately using enumeration method The effect of connecting different power supply nodes in the distribution network to participate in distribution network restoration was obtained. The total load restored by the distribution network within the optimal restoration time window after different power supply nodes in the distribution network participate in the restoration. , ; Step (3), for The total load restored by the distribution network within the optimal restoration time window after different power supply nodes in the distribution network participate in the restoration. , Sort in descending order. If the total restored load is the same, sort again in ascending order according to the power supply node number of the distribution network. After sorting, assume Connecting to power distribution network nodes The distribution network ranked first in terms of total load restored within the optimal restoration time window after participating in the distribution network restoration. Distribution network power nodes It has the highest sensitivity in participating in distribution network restoration, therefore it is determined Connecting to power distribution network nodes Participate in power distribution network restoration; Step (4): Confirm completion Connecting to power distribution network nodes After participating in the power distribution network restoration, according to Arrive at and connect to the power distribution network node The time, remaining power battery charge, and power distribution network nodes Distributed mobile energy storage grids deliver rated power and update power supply nodes in the distribution network. Upper limit parameters for active and reactive power and ; Step (5): After completing the update of the power supply node parameters of the distribution network, repeat steps (1)-(4) to determine the distributed mobile energy storage. The specific power nodes involved in the distribution network restoration are identified, and the parameters of the corresponding power nodes are updated; this process is repeated sequentially to determine the relevant power nodes. Internal distributed mobile energy storage - The corresponding power supply nodes of the distribution network that participate in the restoration of the distribution network access are also updated with parameters.

7. The method for optimizing distribution network recovery strategy considering distributed mobile energy storage according to claim 6, characterized in that... Step S5 specifically includes the following sub-steps: Step S51: Optimize the time window for distribution network restoration The analysis and relevant information collection of distributed mobile energy storage participating in distribution network restoration are carried out through internal power supply dispatch. In step S4, an optimal dispatch model for distributed mobile energy storage within the optimal time window for distribution network restoration is proposed based on the sensitivity method. The total number is Distributed mobile energy storage participates in the corresponding power nodes of the distribution network to restore access to the distribution network, and at the same time completes the update of the active power upper limit parameter value and reactive power upper limit parameter value of the corresponding power nodes of the distribution network. Step S52: Simultaneously update the active power upper limit parameter value and reactive power upper limit parameter value of the corresponding power supply node in the distribution network, update the parameter value of the power supply node in the distribution network contained in equation (25)-(50), and the updated optimization objective function (25) and constraint condition equation (26)-(50) constitute the distribution network recovery strategy optimization decision model considering distributed mobile energy storage; the established distribution network recovery strategy optimization decision model considering distributed mobile energy storage is a typical mixed integer second-order cone optimization problem. After solving it, the optimal recovery strategy of the distribution network is obtained.

Images