A fault recovery method for AC / DC hybrid distribution network under extreme disasters

Abstract

The invention discloses a method for restoring an AC / DC hybrid distribution network under extreme disasters, that is, taking the minimum system load loss and the minimum line active power network loss as objective functions, establishing a timing process considering fault repair, and quickly deploying a mobile emergency power supply after an accident, and performing a model for real-time coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network. Under given constraints, a mobile emergency power supply deployment and switch action timing scheme that minimizes the system load loss and the line active power network loss is solved; the method divides the process after the AC / DC hybrid distribution network is affected by an extreme event into multiple time periods, and realizes the restoration of the AC / DC hybrid distribution network under extreme disasters by simulating the timing repair of faults and the timing reconstruction of the network.

Description

Technical Field

[0001] The invention belongs to the field of power system analysis, and in particular relates to a fault recovery method for an AC / DC hybrid distribution network under extreme disasters. Background Art

[0002] AC / DC hybrid distribution network is an important development direction of active distribution network in the future. The increasingly frequent extreme disasters have posed a severe challenge to the recovery capacity of AC / DC hybrid distribution network under severe faults. However, the current related research mainly focuses on the economic dispatch and N-1 fault recovery of AC / DC hybrid distribution network, and almost no research focuses on the improvement of the recovery capacity of AC / DC hybrid distribution network when it is affected by extreme disasters and severe faults occur. Therefore, it is urgent to make full use of the existing resources of the distribution network and propose methods that are conducive to improving the recovery capacity of AC / DC hybrid distribution network. The severe fault recovery method of AC / DC hybrid distribution network considering fault repair, mobile emergency power supply deployment and remote control switch dynamic network reconstruction is of great help to improve the recovery capacity of AC / DC hybrid distribution network under extreme disasters. After the fault occurs, MEG is quickly deployed to form a recovery island. Considering the fault repair process, the coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network is implemented to ensure the rapid recovery of critical loads. Summary of the invention

[0003] In order to further improve the recovery capability of AC / DC hybrid distribution network under extreme disasters and improve the efficiency of resource allocation and scheduling of distribution system, the present invention provides a method for recovering AC / DC hybrid distribution network under extreme disasters. After a fault occurs, a mobile emergency power supply is quickly deployed to form a recovery island. Considering the fault repair process, the coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network is implemented to ensure the rapid recovery of critical loads. The specific scheme includes the following steps:

[0004] A method for restoring an AC / DC hybrid distribution network under extreme disasters, wherein the method divides the process of the AC / DC hybrid distribution network being affected by an extreme event into multiple time periods to simulate the timing repair of the fault and the timing reconstruction of the network, thereby realizing the restoration of the distribution network, and comprises the following steps:

[0005] Input the initial data information of AC / DC hybrid distribution network system;

[0006] According to the initial data information, the objective function is to minimize the loss load of the AC / DC hybrid distribution network system and the line active power network loss. The model is established to consider the timing process of fault repair, quickly deploy mobile emergency power supply after the accident, and perform real-time coordinated reconstruction of the AC / DC hybrid distribution network AC / DC part. Its representation is:

[0007]

[0008] Under given constraints, a mobile emergency power supply deployment and switch action timing scheme is solved to minimize the loss load of the AC / DC hybrid distribution network system and the line active power network loss; the given constraints of the model for collaborative reconstruction of the AC / DC part of the AC / DC hybrid distribution network include: topology constraints, fault area constraints, AC branch power flow constraints, DC branch power flow constraints, VSC converter station branch constraints, mobile emergency power supply scheduling constraints, power output constraints, and safety constraints;

[0009] Solve the model of AC / DC part collaborative reconstruction to obtain an AC / DC hybrid distribution network extreme disaster recovery plan that minimizes the AC / DC hybrid distribution network system loss load and the line active power network loss. Further, the AC / DC hybrid distribution network system loss load and the line active power network loss are minimized as the objective function form:

[0010]

[0011] Among them, ρ1 and ρ2 represent the weights of the load loss term and the network active power loss term in the objective function respectively; T represents the time period set; B represents the set of all nodes in the distribution network; E represents the set of all branches in the distribution network; P S,j,t represents the amount of active load loss of node j in time period t; represents the square of the current of line ij in time period t; r ij Represents the resistance of line ij.

[0012] Furthermore, the topological constraints include:

[0013]

[0014]

[0015] Among them, E A is the set of AC branches in the distribution network; L is the set of loops consisting of all AC branches in the distribution network; Ω RCS A collection of branches of the distribution network equipped with remote control switches; L k,ij is the state of branch ij in the ijth loop. If branch ij is closed, then L k,ij =1, otherwise L k,ij =0;s ij,t is the state of branch ij. If branch ij is closed, then s ij,t =1, otherwise s ij,t =0; Constraints (2) to (3) indicate that the radial constraints should be maintained before the AC part of the distribution network fails, and the branches with configured switches are in the closed state.

[0016] Furthermore, the fault area constraints:

[0017]

[0018]

[0019]

[0020]

[0021]

[0022]

[0023] Among them, n i,t is the fault region status of node i. If node i is in the fault region, then n i,t =1, otherwise n i,t =0; f ij,t Indicates whether branch ij is short-circuited during period t. If branch ij is short-circuited, then f ij,t =1, otherwise f ij,t =0;P L,j and Q L,j Respectively represent the active load and reactive load of node j; P S,j,t and Q S,j,t They represent the active load loss and reactive load loss of node j respectively; constraints (4) to (5) indicate that the nodes connected to the short-circuit branch are in the fault area; constraints (6) to (7) indicate that the fault area states of the nodes on both sides of the closed branch are the same; constraints (8) to (9) are load loss constraints.

[0024] Furthermore, the AC branch power flow constraints are:

[0025]

[0026]

[0027]

[0028]

[0029] Among them, δ(j) is the set of child nodes of node j; π(j) is the set of parent nodes of node j; H ij,t and G ij,t are the active power and reactive power of branch ij respectively; x ij is the resistance and reactance of branch ij; P SUB,j,t and Q SUB,j,t are the active power output and reactive power output of the substation at node j respectively; P DG,j and Q DG,j are the active output and reactive output of distributed generation at node j respectively; M is a maximum positive constant; is the square of the voltage at node j; P MEG,j and Q MEG,j are the active output and reactive output of the mobile emergency power supply at node j respectively; constraints (10) to (11) are the active power flow balance constraints and reactive power flow balance constraints of some nodes in the AC branch of the distribution network; constraint (12) is the voltage balance constraint of some nodes in the AC branch of the distribution network; constraint (13) is the second-order cone relaxation constraint of some nodes in the AC branch of the distribution network.

[0030] Furthermore, the DC branch power flow constraint is:

[0031]

[0032]

[0033]

[0034] Among them, E D is the set of DC branches of the distribution network; constraint (14) is the active power flow balance constraint of some nodes of the DC branches of the distribution network; constraint (15) is the voltage balance constraint of some nodes of the DC branches of the distribution network; constraint (16) is the second-order cone relaxation constraint of some DC branches of the distribution network.

[0035] Furthermore, the VSC converter station branch constraints are:

[0036]

[0037]

[0038]

[0039]

[0040] Among them, E VSC B is the VSC branch set; VSC,A B is the node set connected to the AC side of the VSC branch; VSC,D It is the node set connected to the DC side of the VSC branch; is the square of the potential in the VSC converter station of branch ij; and They are the upper and lower limits of reactive power compensation of the VSC converter station of branch ij respectively; is the per unit value of the square of the potential in the VSC converter station of branch ij; is the per-unit value of the square of the voltage at node j; constraints (17) to (18) are the DISTFLOW branch constraints of the VSC converter station equivalent circuit; constraint (19) is the second-order cone voltage constraint of the VSC converter station; constraint (20) is the reactive power compensation constraint of the VSC converter station.

[0041] Furthermore, the mobile emergency power supply scheduling constraints:

[0042]

[0043]

[0044]

[0045] Among them, Ω MEG MEG set; B M is the set of candidate nodes that can be connected to MEG; m k,i,t Indicates whether the mobile emergency power supply k is connected to node i. If so, then m k,i,t =1, otherwise m k,i,t =0; Constraint (21) indicates that each MEG can only be deployed to one location; Constraint (22) indicates that the MEG can only be deployed to a connectable candidate node; Constraint (23) indicates that the MEG will not be transferred after deployment.

[0046] Furthermore, the power output constraint is:

[0047]

[0048]

[0049]

[0050] Among them, g k,i Indicates whether substation k is located at node i. If so, g k,i =1, otherwise g k,i =0;d k,i Indicates whether distributed power source k is connected to node i. If yes, then d k,i =1, otherwise d k,i =0; represents the output upper limit of substation k; represents the output upper limit of distributed generation k; represents the output upper limit of the kth MEG; constraint (24) is the substation output constraint; constraint (25) is the distributed power output constraint; constraint (26) is the mobile emergency power output constraint.

[0051] Furthermore, the security constraints:

[0052]

[0053]

[0054]

[0055] in, and are the upper and lower limits of the voltage at node j respectively; is the upper limit of the current of branch ij; is the power upper limit of branch ij; constraint (27) is the node voltage range constraint; constraint (28) is the branch current capacity constraint; constraint (29) is the branch flow capacity constraint.

[0056] Beneficial Effects

[0057] Compared with the prior art, the present invention has the following beneficial effects:

[0058] First, existing research on various resources in distribution network resilience improvement measures is relatively comprehensive, but it mainly focuses on AC distribution networks, and almost no research has been conducted on resilience improvement measures and extreme disaster recovery methods for AC / DC hybrid distribution networks. Compared with the existing technology, the present invention fills this gap and proposes an AC / DC hybrid distribution network recovery method under extreme disasters.

[0059] At the same time, existing research on AC / DC hybrid distribution networks mainly focuses on safety analysis, economic dispatch, etc. Some work has studied the recovery strategy of AC / DC hybrid distribution network N-1 fault, and there is almost no literature on the response measures to severe faults of AC / DC hybrid distribution networks under extreme disasters. Compared with the prior art, the present invention proposes a model that considers the timing process of fault repair, quickly deploys mobile emergency power supplies after an accident, and performs real-time coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network, which can effectively deal with the Nk fault of the AC / DC hybrid distribution network.

[0060] In summary, the method proposed in the present invention can achieve efficient recovery of AC / DC hybrid distribution networks after extreme disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A schematic flow chart of an implementation method of an AC / DC hybrid distribution network restoration method under extreme disasters provided by the present invention;

[0062] Figure 2 The present invention relates to a modified IEEE 33-node power distribution system;

[0063] Figure 3 The present invention discloses an AC / DC hybrid distribution network after a fault. DETAILED DESCRIPTION

[0064] The technical solution of the present invention is further described in detail below with reference to the accompanying drawings and specific implementation examples.

[0065] The AC / DC hybrid distribution network restoration method under extreme disasters provided by the present invention has an implementation flow chart as shown in FIG1 , and is described in detail as follows:

[0066] Step 1: Given fault information, substation location information, distributed power source location information, load demand information, network topology information, mobile emergency power source information, switch installation information and other information.

[0067] Step 2 is to take the minimum system load loss and the minimum line active power network loss as the objective function, establish a time sequence process considering fault repair, quickly deploy mobile emergency power after the accident, and perform a model for real-time coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network. Under given constraints, solve the deployment of mobile emergency power and the timing scheme of switch action that minimizes the system load loss and the line active power network loss. This method divides the process of the AC / DC hybrid distribution network after being affected by extreme events into multiple time periods to simulate the timing repair of faults and the timing reconstruction of the network. The constraints included in this process are: topology constraints, fault area constraints, AC branch flow constraints, DC branch flow constraints, VSC converter station branch constraints, mobile emergency power dispatch constraints, power output constraints, and safety constraints.

[0068] Step 3 solves the model to obtain an extreme disaster recovery plan for the AC / DC hybrid distribution network that minimizes system load loss and line active power network loss.

[0069] In step 2, the objective function is to minimize the system loss load and the line active power network loss, which is expressed as:

[0070]

[0071] Among them, ρ1 and ρ2 represent the weights of the load loss term and the network active power loss term in the objective function respectively; T represents the time period set; B represents the set of all nodes in the distribution network; E represents the set of all branches in the distribution network; P S,j,t represents the amount of active load loss of node j in time period t; represents the square of the current of line ij in time period t; r ij Represents the resistance of line ij.

[0072] In step 2, the constraints include:

[0073] Topological constraints:

[0074]

[0075]

[0076] Among them, E A is the set of AC branches in the distribution network; L is the set of loops consisting of all AC branches in the distribution network; Ω RCS A collection of branches of the distribution network equipped with remote control switches; L k,ijis the state of branch ij in the ijth loop. If branch ij is closed, then L k,ij =1, otherwise L k,ij =0;s ij,t is the state of branch ij. If branch ij is closed, then s ij,t =1, otherwise s ij,t =0.

[0077] Constraints (2) to (3) indicate that the radial constraints must be maintained before an accident occurs in the AC part of the distribution network, and the branches with configured switches are in a closed state.

[0078] Fault region constraints:

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085] Among them, n i,t is the fault region status of node i. If node i is in the fault region, then n i,t =1, otherwise n i,t =0; f ij,t Indicates whether branch ij is short-circuited during period t. If branch ij is short-circuited, then f ij,t =1, otherwise f ij,t =0;P L,j and Q L,j Respectively represent the active load and reactive load of node j; P S,j,t and Q S,j,t They represent the active load loss and reactive load loss of node j respectively.

[0086] Constraints (4) to (5) indicate that the nodes connected to the short-circuited branch are in the fault region; constraints (6) to (7) indicate that the fault region states of the nodes on both sides of the closed branch are the same; constraints (8) to (9) are load loss constraints.

[0087] AC branch power flow constraints:

[0088]

[0089]

[0090]

[0091]

[0092] Among them, δ(j) is the set of child nodes of node j; π(j) is the set of parent nodes of node j; H ij,t and G ij,t are the active power and reactive power of branch ij respectively; x ij is the resistance and reactance of branch ij; P SUB,j,t and Q SUB,j,t are the active power output and reactive power output of the substation at node j respectively; P DG,j and Q DG,j are the active output and reactive output of distributed generation at node j respectively; M is a maximum positive constant; is the square of the voltage at node j; P MEG,j and Q MEG,j are the active output and reactive output of the mobile emergency power supply at node j respectively.

[0093] Constraints (10) to (11) are active power flow balance constraints and reactive power flow balance constraints for some nodes in the AC branch of the distribution network; constraint (12) is the voltage balance constraint for some nodes in the AC branch of the distribution network; constraint (13) is the second-order cone relaxation constraint for some nodes in the AC branch of the distribution network.

[0094] DC branch power flow constraints:

[0095]

[0096]

[0097]

[0098] Among them, E D It is a collection of DC branches in the distribution network.

[0099] Constraint (14) is the active power flow balance constraint of some nodes in the DC branch of the distribution network; constraint (15) is the voltage balance constraint of some nodes in the DC branch of the distribution network; constraint (16) is the second-order cone relaxation constraint of some nodes in the DC branch of the distribution network.

[0100] VSC converter station branch constraints:

[0101]

[0102]

[0103]

[0104]

[0105] Among them, EVSC B is the VSC branch set; VSC,A B is the node set connected to the AC side of the VSC branch; VSC,D It is the node set connected to the DC side of the VSC branch; is the square of the potential in the VSC converter station of branch ij; and They are the upper and lower limits of reactive power compensation of the VSC converter station of branch ij respectively; is the per unit value of the square of the potential in the VSC converter station of branch ij; is the per unit value of the square of the voltage at node j.

[0106] Constraints (17) to (18) are the DISTFLOW branch constraints of the VSC converter station equivalent circuit; constraint (19) is the second-order cone voltage constraint of the VSC converter station; and constraint (20) is the reactive power compensation constraint of the VSC converter station.

[0107] Mobile emergency power dispatch constraints:

[0108]

[0109]

[0110]

[0111] Among them, Ω MEG MEG set; B M is the set of candidate nodes that can be connected to MEG; m k,i,t Indicates whether the mobile emergency power supply k is connected to node i. If so, then m k,i,t =1, otherwise m k,i,t =0.

[0112] Constraint (21) indicates that each MEG can only be deployed to one location; constraint (22) indicates that the MEG can only be deployed to a connectable candidate node; constraint (23) indicates that the MEG will not be transferred after deployment.

[0113] Power output constraints:

[0114]

[0115]

[0116]

[0117] Among them, g k,i Indicates whether substation k is located at node i. If so, g k,i =1, otherwise g k,i =0;d k,iIndicates whether distributed power source k is connected to node i. If yes, then d k,i =1, otherwise d k,i =0; represents the output upper limit of substation k; represents the output upper limit of distributed generation k; Indicates the output upper limit of the kth MEG.

[0118] Constraint (24) is the substation output constraint; constraint (25) is the distributed power output constraint; constraint (26) is the mobile emergency power output constraint.

[0119] Security constraints:

[0120]

[0121]

[0122]

[0123] in, and are the upper and lower limits of the voltage at node j respectively; is the upper limit of the current of branch ij; is the upper power limit of branch ij.

[0124] Constraint (27) is the node voltage range constraint; constraint (28) is the branch current capacity constraint; constraint (29) is the branch flow capacity constraint.

[0125] In step 2, the temporal process of fault repair is considered, and the model of rapid deployment of mobile emergency power supply after the accident and real-time coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network is expressed as follows:

[0126]

[0127] The present invention adopts Figure 2 The modified IEEE 33-node power distribution system shown verifies the effectiveness and correctness of the method proposed in the present invention:

[0128] Step 1: Node 1 is a substation node with a capacity of 5MVA. Six DGs are connected to nodes 7, 28, 37, 41, 46, and 48, respectively, and the capacity of each DG is 0.7MVA. The number of MEGs available in the distribution network is 2, and the capacity of each MEG is 0.5MVA. The load is divided into level 1 load, level 2 load, and general load, with weight factors of 10, 3, and 1, respectively. Branches L38, L42, L47, and L51 are VSC converter station branches, the remaining black branches are AC branches, and the blue branches are DC branches.

[0129] Step 2: Taking the minimum system load loss and the minimum line active power network loss as the objective function, a time sequence process considering fault repair is established, and a model is established for the rapid deployment of mobile emergency power after the accident and the coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network in real time. Under the given constraints, the deployment of mobile emergency power and the timing scheme of switch action that minimize the system load loss and the line active power network loss are solved. This method divides the process of the AC / DC hybrid distribution network after being affected by extreme events into multiple time periods to simulate the timing repair of faults and the timing reconstruction of the network. The constraints included in this process are: topology constraints, fault area constraints, AC branch power flow constraints, DC branch power flow constraints, VSC converter station branch constraints, mobile emergency power dispatch constraints, power output constraints, and safety constraints.

[0130] Step 3: By solving the time sequence process of fault repair, the model of rapid deployment of mobile emergency power supply after the accident and real-time coordinated reconstruction of the AC and DC parts of the AC / DC hybrid distribution network is obtained. Figure 3 The fault repair timing and switch action timing are shown in Table 1.

[0131] Table 1 Fault repair and RCS dynamic reconstruction timing

[0132]

[0133] Among them, in the 3rd period, the L40 fault was repaired, and L55 was closed to make up for the power shortage in the area where node 19 was located; in the 5th period, the L30 fault was repaired, and L30 was closed to reduce the power shortage in the area where node 25 was located (this area still has a power shortage); in the 6th period, the faults of L1 and L34 were repaired, L22 was closed to make up for the power shortage in the area where node 25 was located, and L34 was closed to restore power supply to the area where node 9 was located. Since the MEG located only at node 15 was not enough to fully restore the power supply in the area, L39 was closed to make up for the power supply shortage; in the 8th period, the L33 and L49 faults were repaired, and the area where node 45 was located restored to normal power supply; in the 9th period, the L11 fault was repaired, and L10 was closed to restore power supply to the area where node 12 was located; in the 11th period, L17 was repaired, and L36 was closed to restore power supply to the area where node 18 was located; in the 12th period, the L6 fault was repaired, and the area where node 6 was located restored to normal power supply.

Claims

1. A method for fault recovery of an AC / DC hybrid distribution network under extreme disasters, characterized in that: The method divides the process after the AC / DC hybrid distribution network is affected by an extreme event into multiple time periods to simulate the timing repair of the fault and the timing reconstruction of the network, thereby realizing the distribution network recovery, and includes the following steps: Input the initial data information of AC / DC hybrid distribution network system; According to the initial data information, the objective function is to minimize the loss load of the AC / DC hybrid distribution network system and the line active power network loss. The model is established to consider the timing process of fault repair, quickly deploy mobile emergency power supply after the accident, and perform real-time coordinated reconstruction of the AC / DC hybrid distribution network AC / DC part. Its representation is: Where: ρ1 and ρ2 represent the weights of the load loss term and the network active power loss term in the objective function respectively; T represents the time period set; B represents the set of all nodes in the distribution network; E represents the set of all branches in the distribution network; P S,j,t represents the amount of active load loss of node j in time period t; represents the square of the current of line ij in time period t; r ij represents the resistance of line ij; Under given constraints, solve and obtain the mobile emergency power supply deployment and switch action timing scheme that minimizes the loss load of the AC / DC hybrid distribution network system and the line active power network loss; wherein: the given constraints include: topology constraints, fault area constraints, AC branch power flow constraints, DC branch power flow constraints, VSC converter station branch constraints, mobile emergency power supply scheduling constraints, power output constraints, and safety constraints; The model of coordinated reconstruction of the AC and DC parts is solved to obtain an extreme disaster recovery plan for the AC / DC hybrid distribution network that minimizes the loss of load in the AC / DC hybrid distribution network system and the active power network loss of the lines.

2. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 1, characterized in that: The topological constraints include: Among them, E A is the set of AC branches in the distribution network; L is the set of loops consisting of all AC branches in the distribution network; Ω RCS A collection of branches of the distribution network equipped with remote control switches; L k,ij is the state of branch ij in the ijth loop. If branch ij is closed, then L k,ij =1, otherwise L k,ij =0;s ij,t is the state of branch ij. If branch ij is closed, then s ij,t =1, otherwise s ij,t =0; Constraints (2) to (3) indicate that the radial constraints should be maintained before the AC part of the distribution network fails, and the branches with configured switches are in the closed state.

3. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 1, characterized in that: The fault region constraints: Among them, n i,t is the fault region status of node i. If node i is in the fault region, then n i,t =1, otherwise n i,t =0; f ij,t Indicates whether branch ij is short-circuited during period t. If branch ij is short-circuited, then f ij,t =1, otherwise f ij,t =0;P L,j and Q L,j Respectively represent the active load and reactive load of node j; P S,j,t and Q S,j,t They represent the active load loss and reactive load loss of node j respectively; constraints (4) to (5) indicate that the nodes connected to the short-circuit branch are in the fault area; constraints (6) to (7) indicate that the fault area states of the nodes on both sides of the closed branch are the same; constraints (8) to (9) are load loss constraints.

4. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 1, characterized in that: AC branch power flow constraints: Among them, δ(j) is the set of child nodes of node j; π(j) is the set of parent nodes of node j; H ij,t and G ij,t are the active power and reactive power of branch ij respectively; x ij is the resistance and reactance of branch ij; P SUB,j,t and Q SUB,j,t are the active power output and reactive power output of the substation at node j respectively; P DG,j and Q DG,j are the active output and reactive output of distributed generation at node j respectively; M is a maximum positive constant; is the square of the voltage at node j; P MEG,j and Q MEG,j are the active output and reactive output of the mobile emergency power supply at node j respectively; constraints (10) to (11) are the active power flow balance constraints and reactive power flow balance constraints of some nodes in the AC branch of the distribution network; constraint (12) is the voltage balance constraint of some nodes in the AC branch of the distribution network; constraint (13) is the second-order cone relaxation constraint of some nodes in the AC branch of the distribution network.

5. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 1, characterized in that: The DC branch power flow constraint is: Among them, E D is the set of DC branches of the distribution network; constraint (14) is the active power flow balance constraint of some nodes of the DC branches of the distribution network; constraint (15) is the voltage balance constraint of some nodes of the DC branches of the distribution network; constraint (16) is the second-order cone relaxation constraint of some DC branches of the distribution network.

6. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 1, characterized in that: The VSC converter station branch constraints: Among them, E VSC B is the VSC branch set; VSC,A B is the node set connected to the AC side of the VSC branch; VSC,D It is the node set connected to the DC side of the VSC branch; is the square of the potential in the VSC converter station of branch ij; and They are the upper and lower limits of reactive power compensation of the VSC converter station of branch ij respectively; is the per unit value of the square of the potential in the VSC converter station of branch ij; is the per-unit value of the square of the voltage at node j; constraints (17) to (18) are the DISTFLOW branch constraints of the VSC converter station equivalent circuit; constraint (19) is the second-order cone voltage constraint of the VSC converter station; constraint (20) is the reactive power compensation constraint of the VSC converter station.

7. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 1, characterized in that: The mobile emergency power supply dispatch constraints: Among them, Ω MEG MEG set; B M is the set of candidate nodes that can be connected to MEG; m k,i,t Indicates whether the mobile emergency power supply k is connected to node i. If so, then m k,i,t =1, otherwise m k,i,t =0; Constraint (21) indicates that each MEG can only be deployed to one location; Constraint (22) indicates that the MEG can only be deployed to a connectable candidate node; Constraint (23) indicates that the MEG will not be transferred after deployment.

8. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 1, characterized in that: The power output constraints are: Among them, g k,i Indicates whether substation k is located at node i. If so, g k,i =1, otherwise g k,i =0;d k,i Indicates whether distributed power source k is connected to node i. If yes, then d k,i =1, otherwise d k,i =0; represents the output upper limit of substation k; represents the output upper limit of distributed generation k; represents the output upper limit of the kth MEG; constraint (24) is the substation output constraint; constraint (25) is the distributed power output constraint; constraint (26) is the mobile emergency power output constraint.

9. The method for fault recovery of an AC / DC hybrid distribution network under extreme disasters according to claim 3, characterized in that: The safety constraints: in, and are the upper and lower limits of the voltage at node j respectively; is the upper limit of the current of branch ij; is the power upper limit of branch ij; constraint (27) is the node voltage range constraint; constraint (28) is the branch current capacity constraint; constraint (29) is the branch flow capacity constraint.

Images