Active power distribution network power supply recovery control method based on intelligent soft switch

Abstract

The invention discloses an active power distribution network power supply recovery control method based on an intelligent soft switch, and belongs to the field of active power distribution network power supply recovery. Wherein the key data comprises topological information, intelligent soft switching information and distributed power supply output information; according to the key data of the active power distribution network, constructing an active power distribution network power supply recovery model based on the intelligent soft switch; converting a non-convexity constraint in the power supply recovery model of the active power distribution network into a convexity constraint to obtain a mixed integer second-order cone programming model, and solving the mixed integer second-order cone programming model by adopting a Gurobi solver to obtain a fault recovery control strategy; and completing power supply recovery of the active power distribution network based on the fault recovery strategy. The problem that the fault recovery capability of the active power distribution network is insufficient in the prior art is solved.

Description

Technical Field

[0001] The present invention belongs to the field of power supply restoration of an active distribution network, and in particular relates to a power supply restoration control method of an active distribution network based on an intelligent soft switch. Background Art

[0002] With the rapid development of the economy, the distribution network has entered a new stage of development and has gradually formed a new form that includes various types of distributed power sources, electric vehicles and power electronic equipment. On the one hand, the structure of the distribution network has changed from the traditional single-source type to the multi-source type, thereby changing the system operation state. On the other hand, due to the increasing penetration rate of power electronic intelligent distribution devices in the distribution network, the power electronic level of the distribution network has been further improved. However, once a failure occurs in the distribution network, especially a large-scale power outage, it will have a great impact on the reliability of power supply. The output of distributed power sources in the active distribution network has the characteristics of randomness, volatility and intermittency. After being connected to the distribution network, it is very likely to cause adverse consequences such as secondary load shedding in the restored area during the power supply restoration process. With the development of power electronics technology, intelligent soft switches with flexible control and fast response capabilities will be used for fault recovery control of active distribution networks, which can more effectively realize power flow control and voltage support, and can replace traditional active distribution network tie switches to achieve operation optimization and power supply restoration of distribution networks coordinated with distributed power sources. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides an active distribution network power supply recovery control method based on intelligent soft switches to solve the problem of insufficient fault recovery capability of the active distribution network in the prior art.

[0004] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a power supply restoration control method of an active distribution network based on an intelligent soft switch, comprising the following steps:

[0005] S1. Acquire key data of the active distribution network; wherein the key data includes topology information, intelligent soft switch information and distributed power output information;

[0006] S2. According to the key data of the active distribution network, a power supply restoration model of the active distribution network based on intelligent soft switches is constructed;

[0007] S3, transforming the non-convex constraints in the active distribution network power supply restoration model into convex constraints, obtaining a mixed integer second-order cone programming model, and using the Gurobi solver to solve the mixed integer second-order cone programming model to obtain a fault recovery control strategy;

[0008] S4. Based on the fault recovery strategy, complete the power supply restoration of the active distribution network.

[0009] The beneficial effects of the present invention are as follows: the present invention applies intelligent soft switches to the power supply restoration of active distribution networks, based on the flexible control characteristics of intelligent soft switches, fully considers the synergistic effect of distributed power sources, and improves the power supply restoration capability and power supply reliability of active distribution networks. The method uses multiple types of distributed power sources to coordinate the power supply restoration of the distribution network, and can provide a longer power supply time for important loads during the power supply restoration period, reducing power outage losses; with the maximum load-weighted power supply time and the minimum total network loss as the goal, taking into account multiple power sources such as distributed synchronous generators, distributed energy storage and mobile emergency trams, it can achieve the optimal allocation of limited power generation resources in time and space dimensions, thereby improving the power supply restoration capability of the distribution network. This method has engineering and practical value for the fault recovery of distribution networks containing multiple types of distributed power sources.

[0010] Furthermore, the step S2 specifically constructs an active distribution network power supply restoration model with power flow constraints, operation constraints, intelligent soft switch operation constraints, limited energy constraints and distributed power supply output constraints as key constraints.

[0011] The beneficial effect of the above further scheme is: by fully considering the key constraints for achieving distribution network fault recovery, the solved fault recovery strategy can meet the various requirements of the distribution network, ensure that the fault recovery strategy makes the distribution network flow balanced and reliable, and can give full play to the fault recovery capabilities of intelligent soft switches and distributed power sources.

[0012] Furthermore, the expression of the power flow constraint is:

[0013]

[0014] Among them, S ki,t is the power flowing through line (k,i) in time period t; k:k→i is the downstream node of node i; V k,t is the voltage of node k in time period t; V i,t is the voltage of node i in time period t; is the conjugate complex number of the current flowing through line (k,i) in time period t; S i,t is the injected power of node i in time period t; S ij,t is the power flowing through line (i, j) in time period t; i, j and k are all nodes in the power supply island; N is the set of nodes in the power supply island; t is the time period; T all is the power supply recovery time range; j:i→j is the upstream node of node i; V j,t is the voltage of node j in time period t; β ij,t is the switching status of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; Z ij is the impedance value of line (i, j); Iij,t is the current flowing through line (i, j) in time period t; E is the set of lines in the power supply island; is the distributed power output power of node i in time period t; α i,t is the load status of node i in period t; is the load power demand of node i in time period t; P t Loss is the active power loss of the system network in time period t; is the active power output of the limited distributed generation at node i in time period t; is the load active power demand of node i in time period t; DG / PV is the set of all distributed power supply nodes in the power supply island except photovoltaic.

[0015] The beneficial effect of the above further solution is that the proposed fault recovery model can be solved normally and reliably through the power flow constraint processed by convex relaxation.

[0016] Furthermore, the operation constraints include node voltage constraints, line capacity constraints, other distributed power output constraints other than photovoltaics, and radial topology constraints:

[0017]

[0018] Among them, V i min is the lower limit of the operating voltage amplitude of node i; V i,t is the voltage of node i in time period t; V i max is the upper limit of the operating voltage amplitude of node i; i and j are nodes in the power supply island; N is the set of nodes in the power supply island; t is the time period; T all S is the power supply restoration time range; ij,t is the power flowing through line (i, j) in time period t; β ij,t is the switching status of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; is the apparent power upper limit of line (i, j); E is the set of lines in the power supply island; is the distributed power output power of node i in time period t; P i max is the upper limit of the active output of the distributed generation connected to node i; j is an imaginary unit; is the reactive power output upper limit of the distributed power source connected to node i; G is the set of distributed power source nodes of synchronous generators; P i ch,maxis the upper limit of the charging power of the energy storage or electric vehicle charging pile connected to node i; is the active power output of the limited distributed generation at node i in period t; P i dch,max is the upper limit of the discharge power of the energy storage or electric vehicle charging pile connected to node i; i min is the lower limit of the state of charge of the energy storage or mobile emergency power supply connected to node i; i,0 is the charge state of the energy storage or mobile emergency power supply connected to node i at the beginning of restoration; ρ i is the conversion coefficient of energy to charge state; T0 is the power supply recovery time range T all The span of several time periods divided; i max is the upper limit of the state of charge of the energy storage or mobile emergency power supply connected to node i; B is the set of distributed energy storage nodes; EV is the set of mobile emergency power supply nodes; n b is the total number of nodes in the power supply island; n s is the total number of root nodes in the power supply island; γ ij,t is a 0-1 variable, which takes the value 1 when j is the parent node of i, otherwise, it takes the value 0; ji,t is a 0-1 variable, which takes the value 1 when i is the parent node of j, otherwise, it takes the value 0; 1j,t An identifier indicating that the root node has no parent node.

[0019] The beneficial effects of the above further scheme are: the above operating constraints can ensure that the obtained fault recovery strategy can meet the voltage requirements of each node in the distribution network, ensure that the voltage does not exceed the limit, meet the capacity requirements of each line, prevent line overload, meet the output of distributed power sources, make full use of and improve the effect of distributed power sources on fault recovery, meet the radial topology constraints of the distribution network, avoid the formation of a ring network, and cause the distribution network to be unable to operate normally.

[0020] Furthermore, the expression of the intelligent soft switch operation constraint is:

[0021]

[0022] Among them, P i S is the active power injected into node i by the intelligent soft switch; is the active power injected into node j by the intelligent soft switch; ΔP i S is the active power loss of the converter with the intelligent soft switch connected to node i; is the active power loss of the converter with the intelligent soft switch connected to node j; is the loss coefficient of the converter connected to node i; is the loss coefficient of the converter connected to node j; is the reactive power injected into node i by the intelligent soft switch; is the reactive power injected into node j by the intelligent soft switch; is the capacity of the converter connected to node i; is the capacity of the converter connected to node j; is the voltage control value of the node connected to the intelligent soft switch on the fault side; U0 is the lower limit of the voltage amplitude of the node on the fault side.

[0023] The beneficial effects of the above further solution are: being able to ensure the normal operation of the intelligent soft switch and improving the effect of the intelligent soft switch during fault recovery.

[0024] Furthermore, the expression of the finite energy constraint is:

[0025]

[0026] Among them, T0 is the power supply recovery time range T all The span of several time periods divided; t is the time period; T all The time frame for power restoration; is the active power output of the limited distributed generation at node i in time period t; E i,0 is the remaining energy resource of node i at the beginning of the blackout; i is the node in the power supply island; DG / PV is the set of all distributed power supply nodes in the power supply island except photovoltaic.

[0027] The beneficial effect of the above further solution is: when a power supply island is formed due to fault recovery, the limited local power generation resources in the power supply island can be fully utilized.

[0028] Furthermore, the expression of the distributed power generation output constraint is:

[0029]

[0030] in, is the active power output of the limited distributed generation at node i in period t; e avg is the mean absolute percentage error; is the predicted output value of node i in time period t; i is the node in the power supply island; PV is the photovoltaic power node; t is the time period; T all This is the time range for power restoration.

[0031] The beneficial effect of the above further scheme is: through the prediction and constraint conditions of distributed power sources, the effect of distributed power sources in the recovery process can be fully utilized, and the impact of the uncertainty of distributed power source output caused by environmental factors on fault recovery can be reduced.

[0032] Furthermore, the objective function of the active distribution network power supply restoration model is constructed based on the load-weighted power supply time and the active power loss of the system network:

[0033] maxf=λ1f1-λ2f2

[0034]

[0035] Among them, f is the objective function of the active distribution network power supply restoration model; λ1 is the weight of the load weighted power supply time; f1 is the load weighted power supply time; λ2 is the weight of the system network active power loss; f2 is the system network active power loss; i is the node in the power supply island; L is the set of all load nodes in the power supply island; t is the time period; T all The power supply restoration time range; i is the important weight coefficient of node i; ω t is the important weight coefficient of time period t; T0 is the power supply recovery time range T all The span of several time periods divided; α i,t is the load state of node i in period t; P t Loss is the active power loss of the system network in time period t; S base is the base power of the system.

[0036] The beneficial effects of the above further scheme are: based on distributed power sources and intelligent soft switches, with maximizing the load-weighted power supply time as the main goal and minimizing the active power loss of the system network as the secondary goal, it can further improve the power supply recovery effect of the distribution network, and even in the event of a major power outage, by reducing non-essential loads, improve the power supply capacity of the recoverable part of the load at each node.

[0037] Furthermore, the step S3 is specifically as follows:

[0038] S301. Introduce non-negative real variables and use the big M method to linearize the voltage-current relationship in the power flow constraint to obtain the improved power flow constraint:

[0039]

[0040]

[0041] Among them, S ki,t is the power flowing through line (k,i) in time period t; k:k→i is the downstream node of node i; Z ki is the impedance value of line (k,i); l ki,t 、v i,t 、v j,t , l ij,t is a non-negative real variable; S i,tis the injected power of node i in time period t; i, j and k are all nodes in the power supply island; E is the set of lines in the power supply island; t is the time period; T all is the power supply recovery time range; N is the node set in the power supply island; j:i→j is the upstream node of node i; S ij,t is the apparent power flowing through line (i, j) in time period t; Re is the sign of the real part of the complex number; Z ij The conjugate complex number of ij is the impedance value of line (i, j); X ij,t is the reactance of line (i, j) in period t; β ij,t is the switching state of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; M is a positive real number; For S ij,t The conjugate complex number of ij,t is the power flowing through line (i, j) in time period t; V i,t is the voltage of node i in time period t; V i,t The conjugate complex number of ij,t is the current flowing through line (i, j) in time period t; For I ij,t The complex conjugate of ;

[0042] S302, the equality constraint in the improved power flow constraint Relax to inequality constraints and obtain the final power flow constraints; The expression after relaxation to inequality constraints is:

[0043]

[0044] S303: Nonlinear equality constraints in intelligent soft switch operation constraints Convex relaxation is performed to obtain the improved intelligent soft switching constraint; The expression after convex relaxation is:

[0045]

[0046] Among them, ΔP i S is the active power loss of the converter with the intelligent soft switch connected to node i; is the active power loss of the converter with the intelligent soft switch connected to node j; is the loss coefficient of the converter connected to node i; is the loss coefficient of the converter connected to node j; P i Sis the active power injected into node i by the intelligent soft switch; is the active power injected into node j by the intelligent soft switch; is the reactive power injected into node i by the intelligent soft switch; is the reactive power injected into node j by the intelligent soft switch; is the active power injected into node i by the intelligent soft switch in time period t; is the reactive power injected into node i by the intelligent soft switch in period t; is the active power injected into node j by the intelligent soft switch in period t; is the reactive power injected into node j by the intelligent soft switch in period t; is the active power loss of the converter with the intelligent soft switch connected to node i in time period t; is the active power loss of the converter with the intelligent soft switch connected to node j in time period t;

[0047] S304, converting the improved intelligent soft switch constraint into a standard second-order cone form to obtain the final intelligent soft switch constraint:

[0048]

[0049] in, is the capacity of the converter connected to node i; is the capacity of the converter connected to node j;

[0050] S305, replacing the power flow constraint and the intelligent soft switch operation constraint in the active distribution network power supply restoration model with the final power flow constraint and the final intelligent soft switch constraint, respectively, to obtain a mixed integer second-order cone programming model;

[0051] S306. Use the Gurobi solver to solve the mixed integer second-order cone programming model to obtain a fault recovery control strategy.

[0052] The beneficial effect of the above further scheme is: since some constraints of the above model are non-convex constraints, the relevant non-convex constraints are processed into convex constraints, and the non-convex model is converted into a convex optimization model, so that the proposed fault recovery model can solve the optimization result, make full use of the role of intelligent soft switches, and improve the fault recovery level of the distribution network containing distributed power sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The figure is a flow chart of the method of the present invention.

[0054] Figure 2 4 is a system equivalent topology diagram of an embodiment.

[0055] Figure 3 This is the system equivalent topology diagram of comparative example 1.

[0056] Figure 4 This is the system equivalent topology diagram of comparative example 2.

[0057] Figure 5 Schematic diagram of the conservative output principle of distributed power sources.

[0058] Figure 6 This is a schematic diagram of load recovery results in each period.

[0059] Figure 7 Schematic diagram of node voltage amplitude optimization results in different time periods under different scenarios.

[0060] Figure 8 Schematic diagram of intelligent soft switch port parameters in each time period of comparative example 2.

[0061] Fig. 9 Schematic diagram of intelligent soft switch port parameters in each time period of the embodiment.

[0062] Fig.10 This is a schematic diagram of power supply restoration results for different power outage durations in different scenarios. DETAILED DESCRIPTION

[0063] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0064] Example 1

[0065] like Figure 1 As shown, in one embodiment of the present invention, a method for controlling power supply restoration of an active distribution network based on an intelligent soft switch comprises the following steps:

[0066] S1. Acquire key data of the active distribution network; wherein the key data includes topology information, intelligent soft switch information and distributed power output information;

[0067] S2. According to the key data of the active distribution network, a power supply restoration model of the active distribution network based on intelligent soft switches is constructed;

[0068] S3, transforming the non-convex constraints in the active distribution network power supply restoration model into convex constraints, obtaining a mixed integer second-order cone programming model, and using the Gurobi solver to solve the mixed integer second-order cone programming model to obtain a fault recovery control strategy;

[0069] S4. Based on the fault recovery strategy, complete the power supply restoration of the active distribution network.

[0070] In this embodiment, key data of the active distribution network is obtained, and key data for the active distribution network to achieve rapid fault recovery is obtained through image model files, historical data of monitoring equipment, intelligent soft switch parameters, and distributed power output prediction parameters. The specific process is:

[0071] A. Based on the active distribution network model file, obtain the topological connection relationship, line impedance parameters and power flow conditions, node power demand and voltage range, system network loss, etc. of the active distribution network;

[0072] B. Based on the intelligent soft switch parameter file, obtain the control mode, operation status, power transmission capability, power loss parameters, capacity parameters, voltage range, etc. of the intelligent soft switch;

[0073] C. Based on the operating conditions of various distributed power sources / energy storage power sources, obtain the output conditions of various distributed power sources, the charge state of energy storage power sources, the energy conversion coefficient, etc.

[0074] The step S2 specifically constructs an active distribution network power supply restoration model with power flow constraints, operation constraints, intelligent soft switch operation constraints, limited energy constraints and distributed power output constraints as key constraints.

[0075] The expression of the power flow constraint is:

[0076]

[0077] Among them, S ki,t is the power flowing through line (k,i) in time period t; k:k→i is the downstream node of node i; V k,t is the voltage of node k in time period t; V i,t is the voltage of node i in time period t; is the conjugate complex number of the current flowing through line (k,i) in time period t; S i,t is the injected power of node i in time period t; S ij,t is the power flowing through line (i, j) in time period t; i, j and k are all nodes in the power supply island; N is the set of nodes in the power supply island; t is the time period; T all is the power supply recovery time range; j:i→j is the upstream node of node i; V j,t is the voltage of node j in time period t; β ij,t is the switching status of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; Z ij is the impedance value of line (i, j); I ij,tis the current flowing through line (i, j) in time period t; E is the set of lines in the power supply island; is the distributed power output power of node i in time period t; α i,t is the load status of node i in period t; is the load power demand of node i in time period t; P t Loss is the active power loss of the system network in time period t; is the active power output of the limited distributed generation at node i in time period t; is the load active power demand of node i in time period t; DG / PV is the set of all distributed power supply nodes in the power supply island except photovoltaic.

[0078] In this embodiment, the power supply island system needs to satisfy the following during the power supply restoration process: node KCL equation and power definition equation, VCR equation, node power balance equation, and system network active power loss equation.

[0079] The operation constraints include node voltage constraints, line capacity constraints, other distributed power generation output constraints other than photovoltaic power generation, and radial topology constraints:

[0080]

[0081] Among them, V i min is the lower limit of the operating voltage amplitude of node i; V i,t is the voltage of node i in time period t; V i max is the upper limit of the operating voltage amplitude of node i; i and j are nodes in the power supply island; N is the set of nodes in the power supply island; t is the time period; T all S is the power supply restoration time range; ij,t is the power flowing through line (i, j) in time period t; β ij,t is the switching status of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; is the apparent power upper limit of line (i, j); E is the set of lines in the power supply island; is the distributed power output power of node i in time period t; P i max is the upper limit of the active output of the distributed generation connected to node i; j is an imaginary unit; is the reactive power output upper limit of the distributed power source connected to node i; G is the set of distributed power source nodes of synchronous generators; P i ch,maxis the upper limit of the charging power of the energy storage or electric vehicle charging pile connected to node i; is the active power output of the limited distributed generation at node i in period t; P i dch,max is the upper limit of the discharge power of the energy storage or electric vehicle charging pile connected to node i; is the lower limit of the state of charge of the energy storage or mobile emergency power supply connected to node i; i,0 is the charge state of the energy storage or mobile emergency power supply connected to node i at the beginning of restoration; ρ i is the conversion coefficient of energy to charge state; T0 is the power supply recovery time range T all The span of several time periods divided; is the upper limit of the state of charge of the energy storage or mobile emergency power supply connected to node i; B is the set of distributed energy storage nodes; EV is the set of mobile emergency power supply nodes; n b is the total number of nodes in the power supply island; n s is the total number of root nodes in the power supply island; γ ij,t is a 0-1 variable, which takes the value 1 when j is the parent node of i, otherwise, it takes the value 0; ji,t is a 0-1 variable, which takes the value 1 when i is the parent node of j, otherwise, it takes the value 0; 1j,t An identifier indicating that the root node has no parent node.

[0082] In this embodiment, the power supply island system needs to operate within a safe range during the power supply restoration process, and its operation constraints include node voltage constraints, line capacity constraints, other distributed power output constraints, and radial topology constraints. The output constraints of distributed energy storage or mobile emergency power supply are shown in the following formula. In order to ensure the normal operation of distributed energy storage and mobile emergency power supply equipment, the state of charge (SOC) needs to set upper and lower limits. The distribution network is generally required to operate in a radial manner, that is, there is no ring operation topology in the network.

[0083] The expression of the intelligent soft switch operation constraint is:

[0084]

[0085] Among them, P i S is the active power injected into node i by the intelligent soft switch; is the active power injected into node j by the intelligent soft switch; ΔP i S is the active power loss of the converter with the intelligent soft switch connected to node i; is the active power loss of the converter with the intelligent soft switch connected to node j; is the loss coefficient of the converter connected to node i; is the loss coefficient of the converter connected to node j; is the reactive power injected into node i by the intelligent soft switch; is the reactive power injected into node j by the intelligent soft switch; is the capacity of the converter connected to node i; is the capacity of the converter connected to node j; is the voltage control value of the node connected to the intelligent soft switch on the fault side; U0 is the lower limit of the voltage amplitude of the node on the fault side.

[0086] The expression of the finite energy constraint is:

[0087]

[0088] Among them, T0 is the power supply recovery time range T all The span of several time periods divided; t is the time period; T all The time frame for power restoration; is the active power output of the limited distributed generation at node i in time period t; E i,0 is the remaining energy resource of node i at the beginning of the blackout; i is the node in the power supply island; DG / PV is the set of all distributed power supply nodes in the power supply island except photovoltaic.

[0089] In this embodiment, after a major power outage occurs, the local power generation resources in the power supply island are limited, so the limited energy constraint of each power source needs to be considered.

[0090] The expression of the distributed generation output constraint is:

[0091]

[0092] in, is the active power output of the limited distributed generation at node i in period t; e avg is the mean absolute percentage error; is the predicted output value of node i in time period t; i is the node in the power supply island; PV is the photovoltaic power node; t is the time period; T all This is the time range for power restoration.

[0093] In this embodiment, the distributed power generation adopts a conservative output strategy, that is, during the recovery process, the dispatched output value of the distributed power generation is always lower than the predicted value taking into account the average prediction error.

[0094] The objective function of the active distribution network power supply restoration model is constructed based on the load-weighted power supply time and the active power loss of the system network:

[0095] maxf=λ1f1-λ2f2

[0096]

[0097]

[0098] Among them, f is the objective function of the active distribution network power supply restoration model; λ1 is the weight of the load weighted power supply time; f1 is the load weighted power supply time; λ2 is the weight of the system network active power loss; f2 is the system network active power loss; i is the node in the power supply island; L is the set of all load nodes in the power supply island; t is the time period; T all The power supply restoration time range; i is the important weight coefficient of node i; ω t is the important weight coefficient of time period t; T0 is the power supply recovery time range T all The span of several time periods divided; α i,t is the load state of node i in period t; P t Loss is the active power loss of the system network in time period t; S base is the base power of the system.

[0099] In this embodiment, an active distribution network power supply recovery model based on intelligent soft switches is established, including objective functions and constraints. Among them, the objective function is based on the load-weighted power supply time and the active power loss of the system network, and the constraints include: flow constraints, operation constraints, intelligent soft switch operation constraints, limited energy constraints, and distributed power output constraints. The power supply recovery model is constructed by the above objective functions and constraints, making full use of the advantages of flexible flow control and rapid response of intelligent soft switches, and realizing the rapid fault recovery of active distribution networks coordinated by intelligent soft switches and distributed power sources.

[0100] In this embodiment, after the fault occurs and is removed, it is necessary to perform fault recovery on the non-faulty area, where the local power generation resources in the power supply island support the power supply, and the limited energy constraint needs to be considered; the distributed power source has output volatility, and in order to reduce the adverse effects of its volatility on power supply recovery, it is necessary to limit the dispatching output of the distributed power source during the power supply recovery process, and the output constraint of the distributed power source needs to be considered; in order to achieve normal operation of the system and be within a safe range, the flow constraint and operation constraint of the system need to be considered; in addition, in order to avoid frequent changes in the load state of the same node in the recovery strategy, the load state change constraint needs to be considered. Therefore, the constraints that need to be considered in the distribution network power supply recovery model with multiple types of distributed power sources include limited energy constraints, distributed power output constraints, flow constraints, operation constraints, and intelligent soft switch operation constraints.

[0101] The step S3 is specifically as follows:

[0102] S301. Introduce non-negative real variables and use the big M method to linearize the voltage-current relationship in the power flow constraint to obtain the improved power flow constraint:

[0103]

[0104] Among them, S ki,t is the power flowing through line (k,i) in time period t; k:k→i is the downstream node of node i; Z ki is the impedance value of line (k,i); l ki,t 、v i,t 、v j,t , l ij,t is a non-negative real variable; S i,t is the injected power of node i in time period t; i, j and k are all nodes in the power supply island; E is the set of lines in the power supply island; t is the time period; T all is the power supply recovery time range; N is the node set in the power supply island; j:i→j is the upstream node of node i; S ij,t is the apparent power flowing through line (i, j) in time period t; Re is the sign of the real part of the complex number; Z ij The conjugate complex number of ij is the impedance value of line (i, j); X ij,t is the reactance of line (i, j) in period t; β ij,t is the switching state of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; M is a positive real number; For S ij,t The conjugate complex number of ij,t is the power flowing through line (i, j) in time period t; V i,t is the voltage of node i in time period t; V i,t The conjugate complex number of ij,t is the current flowing through line (i, j) in time period t; For I ij,t The complex conjugate of

[0105] S302, the equality constraint in the improved power flow constraint Relax to inequality constraints and obtain the final power flow constraints; The expression after relaxation to inequality constraints is:

[0106]

[0107] S303: Nonlinear equality constraints in intelligent soft switch operation constraints Convex relaxation is performed to obtain the improved intelligent soft switching constraint; The expression after convex relaxation is:

[0108]

[0109] Among them, ΔP i S is the active power loss of the converter with the intelligent soft switch connected to node i; is the active power loss of the converter with the intelligent soft switch connected to node j; is the loss coefficient of the converter connected to node i; is the loss coefficient of the converter connected to node j; P i S is the active power injected into node i by the intelligent soft switch; is the active power injected into node j by the intelligent soft switch; is the reactive power injected into node i by the intelligent soft switch; is the reactive power injected into node j by the intelligent soft switch; is the active power injected into node i by the intelligent soft switch in time period t; is the reactive power injected into node i by the intelligent soft switch in period t; is the active power injected into node j by the intelligent soft switch in period t; is the reactive power injected into node j by the intelligent soft switch in period t; is the active power loss of the converter with the intelligent soft switch connected to node i in time period t; is the active power loss of the converter with the intelligent soft switch connected to node j in time period t;

[0110] S304, converting the improved intelligent soft switch constraint into a standard second-order cone form to obtain the final intelligent soft switch constraint:

[0111]

[0112] in, is the capacity of the converter connected to node i; is the capacity of the converter connected to node j;

[0113] S305, replacing the power flow constraint and the intelligent soft switch operation constraint in the active distribution network power supply restoration model with the final power flow constraint and the final intelligent soft switch constraint, respectively, to obtain a mixed integer second-order cone programming model;

[0114] S306. Use the Gurobi solver to solve the mixed integer second-order cone programming model to obtain a fault recovery control strategy.

[0115] In this embodiment, a mixed integer second-order cone programming model is used to improve the power supply restoration model, and a Gurobi solver is used to obtain a fault recovery control strategy.

[0116] Since in the second step, some constraints contain variable product terms, which are non-convex constraints and difficult to solve in the optimization problem, the present invention adopts linearization or convex relaxation methods to transform the non-convex constraints into convex constraints, thereby transforming the power supply restoration model into a mixed integer second-order cone programming model, which is easier to solve using a mature optimization solver.

[0117] Example 2

[0118] The present invention proposes a power supply restoration control method for an active distribution network based on an intelligent soft switch, and the specific implementation method thereof is as follows:

[0119] Step 1: Obtain key data of the active distribution network, and obtain key data for rapid fault recovery of the active distribution network through image model files, historical data of monitoring equipment, intelligent soft switch parameters, and distributed power output prediction parameters. The specific process is:

[0120] A. Obtain the key parameters of the active distribution network and the topological connection relationship of the active distribution network: Figure 2 The system equivalent topology diagram of the embodiment is shown. Figure 3 , Figure 4 They are the system equivalent topology diagrams of comparative examples 1 and 2 respectively; the line impedance parameters are shown in Table 1; and the node load parameters are shown in Table 2.

[0121] Table 1

[0122]

[0123]

[0124] Table 2

[0125]

[0126] B. Obtain key parameters of the intelligent soft switch. The control mode of the intelligent soft switch is shown in Table 3.

[0127] Table 3

[0128]

[0129] The active power constraint of the intelligent soft switch is:

[0130]

[0131] Where P i S , and They represent the active power and reactive power injected by SOP (intelligent soft switch) into node i and node j respectively; ΔP i S and They represent the active power loss of the converter when the SOP is connected to the node i and the node j respectively; and They represent the loss coefficients of the converters connected to nodes i and j respectively.

[0132] The capacity constraint of the intelligent soft switch is:

[0133]

[0134] In the formula, and Respectively represent the converter capacity connected to node i and node j.

[0135] The fault side voltage constraint of the intelligent soft switch is:

[0136]

[0137] In the formula, It indicates the voltage control value of the SOP connection node on the fault side; U0 indicates the lower limit of the voltage amplitude of the node on the fault side, which is generally 1.0.

[0138] C. Obtain the operating conditions of various distributed power sources. Since there is a certain prediction error in the output prediction of distributed power sources, the actual output value may be higher or lower than the predicted value. The conservative output strategy can reduce the impact of the uncertainty of the output of distributed power sources, such as Figure 5 The conservative output strategy is to set the upper limit of the dispatchable distributed power output to the lower limit of the average prediction error interval, that is, during the power supply restoration process, the output value of the distributed power is always lower than the predicted value taking into account the average prediction error, as shown in the following formula:

[0139]

[0140] Where P PV,t represents the dispatch value of photovoltaic output at time t; It represents the predicted value of photovoltaic output at time t; eavg represents the average absolute percentage error.

[0141] Step 2: Establish an active distribution network power supply recovery model based on intelligent soft switches, including objective functions and constraints. Among them, the objective function is based on load-weighted power supply time and system network active power loss, and the constraints include: flow constraints, operation constraints, intelligent soft switch operation constraints, limited energy constraints, and distributed power output constraints. The power supply recovery model is constructed through the above objective functions and constraints, making full use of the advantages of flexible flow control and rapid response of intelligent soft switches, and realizing the rapid fault recovery of active distribution networks coordinated by intelligent soft switches and distributed power sources.

[0142] A. The fault recovery objective function takes the maximum load-weighted power supply time as the main objective and the minimum system network active power loss as the secondary objective. After normalization, the following two sub-objective functions are obtained:

[0143]

[0144] Where: f1 and f2 correspond to the load weighted power supply time and the system network active loss respectively, L represents the set of all load nodes in the power supply island; w i represents the important weight coefficient of load i; w t Indicates the important weight coefficient of time period t. In order to provide "shutdown transition time" for important loads, a higher weight is given to the first time period after the power outage, and the remaining time periods are basic weights; a i,t represents the power supply recovery ratio of load i in time period t, which is a continuous variable in the interval [0,1]; P t Loss S represents the network active power loss of the system in time period t; base Represents the base power of the system; the comprehensive objective function is obtained by setting weights for the above two objective functions as shown in the following formula:

[0145] maxf=λ1f1-λ2f2

[0146] Where: l1 and l2 represent the target weight coefficients of load-weighted power supply time and system network loss, respectively. Since the maximum load-weighted power supply time is the primary goal and the minimum system network active loss is the secondary goal, l1 is set to 0.9 and l2 is set to 0.1. In order to further improve the power supply restoration effect, it is assumed that in the case of a large power outage, the dispatcher can communicate with the user to reduce unnecessary loads, that is, considering that the load of each node can partially restore power supply.

[0147] B. After the fault occurs and is removed, it is necessary to restore the fault in the non-faulty area. The local power generation resources in the power supply island support the power supply, and the limited energy constraint needs to be considered; the distributed power source has output volatility. In order to reduce the adverse impact of its volatility on power supply recovery, it is necessary to limit the dispatching output of the distributed power source during the power supply recovery process, and the output constraint of the distributed power source needs to be considered; in order to achieve normal operation of the system and be within a safe range, the flow constraint and operation constraint of the system need to be considered. Therefore, the constraints that need to be considered in the distribution network power supply recovery model for multiple types of distributed power sources include limited energy constraints, distributed power output constraints, flow constraints, and operation constraints, including:

[0148] (1) Limited energy constraint: After a blackout, the local power generation resources in the power supply island are limited. Therefore, the limited energy constraint of each power source needs to be considered, as shown in the following formula:

[0149]

[0150] Where: DG / PV represents the collection of all distributed power sources in the power supply island except photovoltaic power, including synchronous generators, distributed energy storage and mobile emergency power supplies. Among them, mobile emergency power supplies include mobile emergency energy storage vehicles and idle electric buses, which can be used as energy storage equipment to connect to the distribution network through charging piles to provide power support when a major power outage occurs; Indicates the active power value of power source i at time period t; E i,0 It represents the remaining energy resources of power source i at the beginning of the blackout. For the convenience of calculation, all different energy forms are converted into electrical energy for calculation.

[0151] (2) Distributed power generation output constraint: Distributed power generation adopts a conservative output strategy, that is, the dispatched output value of distributed power generation during the recovery process is always lower than the predicted value taking into account the average prediction error, as shown in the following formula:

[0152]

[0153] In the formula, represents the dispatch output value of the i-th photovoltaic power source in time period t; represents the predicted output value of the ith photovoltaic power source in time period t; e avg represents the mean absolute percentage error.

[0154] (3) Power flow constraints: During the power restoration process, the power supply island system needs to satisfy the following: node KCL equation and power definition equation, VCR equation, node power balance equation, and system network active power loss equation, as shown in the following equations:

[0155]

[0156] Where N and E represent the node set and line set in the power supply island, respectively; k:k→i represents the downstream node set of node i; b ij,t Indicates the switching status of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, then b ij,t =1; V i,t , S i,t , and They represent the voltage of node i, the injected power of node i, the output power of distributed generation at node i and the load power demand at node i in time period t respectively; S ij,t and I ij,t They represent the power and current flowing through line (i, j) in time period t respectively; Z ij Indicates the impedance value of line (i, j); P t Loss , and They respectively represent the active power loss of the system network, the active power output of the limited distributed generation at node i, and the active power demand of the load at node i in time period t.

[0157] (4) Operation constraints: The power supply island system needs to operate within a safe range during the power supply restoration process. Its operation constraints include node voltage constraints, line capacity constraints, other distributed power generation output constraints, and radial topology constraints.

[0158] 1) Node voltage constraints

[0159]

[0160] Where V i min and V i max They represent the lower limit and upper limit of the operating voltage amplitude of node i respectively.

[0161] 2) Line capacity constraints

[0162]

[0163] In the formula, is the apparent power upper limit of line (i, j); b ij,t The effect of the coefficient is that when line (i, j) is disconnected, its line power is equal to 0.

[0164] 3) Other distributed power output constraints

[0165] The output constraint of synchronous generator type distributed power source is as follows:

[0166]

[0167] The output constraint of distributed energy storage or mobile emergency power supply is shown in the following formula. In order to ensure the normal operation of distributed energy storage and mobile emergency power supply equipment, the upper and lower limits of their state of charge (SOC) need to be set.

[0168]

[0169] In the formula, G, B and EV represent the set of synchronous generator type distributed power sources, the set of distributed energy storage and the set of mobile emergency power sources respectively; P i max and They represent the upper limit of active power output and reactive power output of the distributed power source connected to node i respectively; P i ch,max and P i dch,max They represent the upper limit of charging power and the upper limit of discharging power of the energy storage or electric vehicle charging pile connected to node i respectively; and They represent the lower and upper limits of the state of charge of the energy storage or mobile emergency power supply connected to node i respectively; i,0 represents the charge state of the energy storage or mobile emergency power supply connected to node i at the beginning of restoration; ρ i The conversion factor that represents the conversion of energy to state of charge.

[0170] 4) Radial topology constraints

[0171] The distribution network is generally required to operate in a radial manner, that is, there is no ring operation topology in the network, as shown in the following formula:

[0172]

[0173] Where n b and n s They represent the total number of nodes and root nodes in the power supply island. Generally, only one root node is set in one power supply island. 1j,t Indicates that the root node does not have a parent node; g ij,t It is a 0-1 variable and takes the value 1 when j is the parent node of i.

[0174] (5) SOP operation constraints

[0175] Under power recovery conditions, SOP adopts V dc Q-Vf control mode, the operating constraints are as follows:

[0176]

[0177] In the formula, and They represent the active power and reactive power injected by SOP to node i and node j in period t respectively; and They represent the active power loss of the converter when SOP is connected to node i and node j during period t respectively; and Respectively represent the loss coefficient of the converter connected to node i and node j; It represents the voltage control value of the SOP connection node on the fault side during period t. U0 represents the lower limit of the node voltage on the fault side, which is generally 1.0pu.

[0178] Step 3: Use the mixed integer second-order cone programming model to improve the power supply restoration model and use the Gurobi solver to obtain the fault recovery control strategy.

[0179] Since in the second step, some constraints contain variable product terms, which are non-convex constraints and difficult to solve in the optimization problem, the present invention adopts linearization or convex relaxation methods to transform the non-convex constraints into convex constraints, thereby transforming the power supply restoration model into a mixed integer second-order cone programming model, which is easier to solve using a mature optimization solver.

[0180] In order to eliminate the quadratic term, a non-negative real variable v is introduced. i,t =V i,t V i,t * and l ij,t =I ij,t I ij,t * ; At the same time, the VCR equation in the power flow constraint is linearized using the "big M method" to improve the following power flow constraint:

[0181]

[0182] Where M represents a large positive real number. If line (i, j) is disconnected, that is, b ij,t = 0, for X ij,t No constraints.

[0183] There are still nonlinear equality constraints in the above formula, which are non-convex constraints and need to be further processed to relax them into inequality constraints. The second-order cone constraints are as follows:

[0184]

[0185] Convert the above formula into standard second-order cone form:

[0186]

[0187] At this point, the original power supply restoration model is converted into a mixed integer second-order cone programming model, which can be solved using a mature optimization solver. In this embodiment, a modified IEEE-33 node system is used for comparative testing. The embodiment and the comparative example are respectively as follows: Figure 2 , Figure 3 , Figure 4 The system parameters are shown in Tables 1, 2 and 3. Furthermore, the parameters of the distributed synchronous generator, distributed energy storage and mobile emergency power supply are shown in Tables 4 and 5.

[0188] Table 4 Distributed synchronous generator parameters

[0189]

[0190] Table 5 Distributed energy storage parameters and mobile emergency power supply parameters

[0191]

[0192] Based on the topology and model of the embodiment and the comparative example, the Yalmip toolbox was used to build the model and the solver Gurobi 10.0 was used to solve the problem. The upper limit of the relative optimal gap was set to 1×10 -5 , the CPU model is Intel Core I5, the main frequency is 2.30GHz, and the RAM is 16GB. The detailed comparison results are analyzed as follows:

[0193] In order to verify the effectiveness of the proposed power restoration strategy, the power restoration effects with and without intelligent soft switches and with different numbers of intelligent soft switches are compared. The following three scenarios are set:

[0194] Comparative Example 1 (such as Figure 3 As shown in the figure): a tie switch is used, the topology does not contain an intelligent soft switch, and a commonly used power supply recovery strategy for multiple types of distributed power sources is adopted.

[0195] Comparative Example 2 (such as Figure 4 As shown): the contact switches 11-21 are replaced with intelligent soft switches, whose capacity is 1MVA and loss coefficient is 0.02, and the power supply restoration strategy proposed by the present invention is adopted.

[0196] Example (such as Figure 2 As shown): the contact switches 11-21 and 17-32 are replaced with intelligent soft switches with a capacity of 1MVA and a loss coefficient of 0.02, and the power supply restoration strategy proposed in the present invention is adopted.

[0197] Assume that a power outage occurs at 12:00 noon on a certain day, and line 0-1 is disconnected and isolated. At this time, the entire distribution network area cannot be powered by the main grid. According to experience and actual information such as the scope of the accident, it is estimated that the large power grid will restore power to the distribution network area in about 3 hours. The power supply restoration results under different scenarios are shown in 0. The load recovery amount in each period is shown in Figure 6 As shown in the figure, the voltage amplitude of the system nodes in each period is as follows: Figure 7 As shown, the intelligent soft switch port parameters of each time period in Comparative Example 2 and the embodiment are respectively as follows: Figure 8 and Fig. 9 shown.

[0198] Table 6

[0199]

[0200]

[0201] Combining 0 and Figure 6 It can be seen that due to the scarcity of local power generation resources (the remaining energy at the end of recovery in the three scenarios is 70kW·h), the power supply recovery effect of Comparative Example 2 and the embodiment is still improved by 1.6% and 1.9% respectively compared with Comparative Example 1. Among them, in the first period of power supply recovery, the load recovery amount of Comparative Example 2 and the embodiment is 13.28% and 18.56% higher than that of Comparative Example 1 respectively. The above results show that the intelligent soft switch has an improvement effect on the power supply recovery effect in the power outage scenario.

[0202] Depend on Figure 7 It can be seen that compared with Comparative Example 1, the node voltage amplitude of the system in Comparative Example 2 is greatly improved. Figure 8 , Fig. 9 It can be seen that this is because the intelligent soft switch provides reactive power compensation for the power supply island and plays a role in voltage support. Since the embodiment is equipped with two intelligent soft switches, the node voltage amplitude of the system is further improved compared with comparative example 2. It can be seen that the intelligent soft switch can provide reactive power compensation for the power supply island after a power outage accident. In addition to restoring more important loads, it can also increase the voltage amplitude of the system, thereby reducing the system network loss under the restoration of the same amount of load.

[0203] In order to further test the applicability of the restoration strategy proposed in this invention under different power outage durations, based on the above comparison scenarios, the power outage durations were set to 1.5, 2.0, 2.5, 3.0, 3.5, and 4.0 hours, and the time of the power outage accident was set to 12:00 noon on a certain day. The power supply restoration results under different power outage durations in different scenarios are shown in Figure 2. Fig.10As shown, in each scenario, the first-level load is fully restored, so it is not shown. As can be seen from the figure, under the same power outage duration, the recovery effects of comparison examples 1, 2, and 3 all increase in sequence; and the shorter the power outage duration, that is, the more abundant the local power generation resources are relative to the load, the more obvious the power supply recovery effect of comparison examples 2 and 3 containing intelligent soft switches is compared with comparison example 1 based on traditional contact switches.

[0204] In summary, the power restoration strategy provided by the present invention has better restoration effect in different power outage duration scenarios compared to the restoration strategy based on the tie switch, and has good applicability.

Claims

1. A power supply recovery control method for an active distribution network based on an intelligent soft switch, characterized in that: The following steps are involved: S1. Acquire key data of the active distribution network; wherein the key data includes topology information, intelligent soft switch information and distributed power output information; S2. According to the key data of the active distribution network, a power supply restoration model of the active distribution network based on intelligent soft switches is constructed; S3, transforming the non-convex constraints in the active distribution network power supply restoration model into convex constraints, obtaining a mixed integer second-order cone programming model, and using the Gurobi solver to solve the mixed integer second-order cone programming model to obtain a fault recovery control strategy; S4. Based on the fault recovery strategy, complete the power supply restoration of the active distribution network.

2. The active distribution network power supply restoration control method based on intelligent soft switch according to claim 1 is characterized in that: The step S2 specifically constructs an active distribution network power supply restoration model with power flow constraints, operation constraints, intelligent soft switch operation constraints, limited energy constraints and distributed power output constraints as key constraints.

3. The active distribution network power supply recovery control method based on intelligent soft switch according to claim 2 is characterized in that: The expression of the power flow constraint is: Among them, S ki,t is the power flowing through line (k,i) in time period t; k:k→i is the downstream node of node i; V k,t is the voltage of node k in time period t; V i,t is the voltage of node i in time period t; is the conjugate complex number of the current flowing through line (k,i) in time period t; S i,t is the injected power of node i in time period t; S ij,t is the power flowing through line (i, j) in time period t; i, j and k are all nodes in the power supply island; N is the set of nodes in the power supply island; t is the time period; T all is the power supply recovery time range; j:i→j is the upstream node of node i; V j,t is the voltage of node j in time period t; β ij,t is the switching status of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; Z ij is the impedance value of line (i, j); I ij,t is the current flowing through line (i, j) in time period t; E is the set of lines in the power supply island; is the distributed power output power of node i in time period t; α i,t is the load status of node i in period t; is the load power demand of node i in time period t; P t Loss is the active power loss of the system network in time period t; is the active power output of the limited distributed generation at node i in time period t; is the load active power demand of node i in time period t; DG / PV is the set of all distributed power supply nodes in the power supply island except photovoltaic.

4. The active distribution network power supply restoration control method based on intelligent soft switch according to claim 2 is characterized in that: The operation constraints include node voltage constraints, line capacity constraints, other distributed power generation output constraints other than photovoltaic power generation, and radial topology constraints: Among them, V i min is the lower limit of the operating voltage amplitude of node i; V i,t is the voltage of node i in time period t; V i max is the upper limit of the operating voltage amplitude of node i; i and j are nodes in the power supply island; N is the set of nodes in the power supply island; t is the time period; T all S is the power supply restoration time range; ij,t is the power flowing through line (i, j) in time period t; β ij,t is the switching status of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; is the apparent power upper limit of line (i, j); E is the set of lines in the power supply island; is the output power of the distributed power source at node i in time period t; P i max is the upper limit of the active output of the distributed generation connected to node i; j is an imaginary unit; is the reactive power output upper limit of the distributed power source connected to node i; G is the set of distributed power source nodes of synchronous generators; P i ch,max is the upper limit of the charging power of the energy storage or electric vehicle charging pile connected to node i; is the active power output of the limited distributed generation at node i in period t; P i dch,max is the upper limit of the discharge power of the energy storage or electric vehicle charging pile connected to node i; is the lower limit of the state of charge of the energy storage or mobile emergency power supply connected to node i; i,0 is the charge state of the energy storage or mobile emergency power supply connected to node i at the beginning of restoration; ρ i is the conversion coefficient of energy to charge state; T0 is the power supply recovery time range T all The span of several time periods divided; is the upper limit of the state of charge of the energy storage or mobile emergency power supply connected to node i; B is the set of distributed energy storage nodes; EV is the set of mobile emergency power supply nodes; n b is the total number of nodes in the power supply island; n s is the total number of root nodes in the power supply island; γ ij,t is a 0-1 variable, which takes the value 1 when j is the parent node of i, otherwise, it takes the value 0; ji,t is a 0-1 variable, which takes the value 1 when i is the parent node of j, otherwise, it takes the value 0; 1j,t An identifier indicating that the root node has no parent node.

5. The active distribution network power supply restoration control method based on intelligent soft switch according to claim 2 is characterized in that: The expression of the intelligent soft switch operation constraint is: Among them, P i S is the active power injected into node i by the intelligent soft switch; is the active power injected into node j by the intelligent soft switch; ΔP i S is the active power loss of the converter with the intelligent soft switch connected to node i; is the active power loss of the converter with the intelligent soft switch connected to node j; is the loss coefficient of the converter connected to node i; is the loss coefficient of the converter connected to node j; is the reactive power injected into node i by the intelligent soft switch; is the reactive power injected into node j by the intelligent soft switch; is the capacity of the converter connected to node i; is the capacity of the converter connected to node j; is the voltage control value of the node connected to the intelligent soft switch on the fault side; U0 is the lower limit of the voltage amplitude of the node on the fault side.

6. The active distribution network power supply restoration control method based on intelligent soft switch according to claim 2 is characterized in that: The expression of the finite energy constraint is: Among them, T0 is the power supply recovery time range T all The span of several time periods divided; t is the time period; T all The time frame for power restoration; is the active power output of the limited distributed generation at node i in time period t; E i,0 is the remaining energy resource of node i at the beginning of the blackout; i is the node in the power supply island; DG / PV is the set of all distributed power supply nodes in the power supply island except photovoltaic.

7. The active distribution network power supply restoration control method based on intelligent soft switch according to claim 2 is characterized in that: The expression of the distributed generation output constraint is: in, is the active power output of the limited distributed generation at node i in period t; e avg is the mean absolute percentage error; is the predicted output value of node i in time period t; i is the node in the power supply island; PV is the photovoltaic power node; t is the time period; T all This is the time range for power restoration.

8. The active distribution network power supply restoration control method based on intelligent soft switch according to claim 1 is characterized in that: The objective function of the active distribution network power supply restoration model is constructed based on the load-weighted power supply time and the active power loss of the system network: maxf=λ1f1-λ2f2 Among them, f is the objective function of the active distribution network power supply restoration model; λ1 is the weight of the load weighted power supply time; f1 is the load weighted power supply time; λ2 is the weight of the system network active power loss; f2 is the system network active power loss; i is the node in the power supply island; L is the set of all load nodes in the power supply island; t is the time period; T all The power supply restoration time range; i is the important weight coefficient of node i; ω t is the important weight coefficient of time period t; T0 is the power supply recovery time range T all The span of several time periods divided; α i,t is the load state of node i in period t; P t Loss is the active power loss of the system network in time period t; S base is the base power of the system.

9. The active distribution network power supply restoration control method based on intelligent soft switch according to claim 1 is characterized in that: The step S3 is specifically as follows: S301. Introduce non-negative real variables and use the big M method to linearize the voltage-current relationship in the power flow constraint to obtain the improved power flow constraint: Among them, S ki,t is the power flowing through line (k,i) in time period t; k:k→i is the downstream node of node i; Z ki is the impedance value of line (k,i); l ki,t 、v i,t 、v j,t , l ij,t is a non-negative real variable; S i,t is the injected power of node i in time period t; i, j and k are all nodes in the power supply island; E is the set of lines in the power supply island; t is the time period; T all is the power supply recovery time range; N is the node set in the power supply island; j:i→j is the upstream node of node i; S ij,t is the apparent power flowing through line (i, j) in time period t; Re is the sign of the real part of the complex number; Z ij The conjugate complex number of ij is the impedance value of line (i, j); X ij,t is the reactance of line (i, j) in period t; β ij,t is the switching state of line (i, j) in time period t, which is a 0-1 variable. If the line is in operation, it is 1, otherwise, it is 0; M is a positive real number; For S ij,t The conjugate complex number of ij,t is the power flowing through line (i, j) in time period t; V i,t is the voltage of node i in time period t; V i,t The conjugate complex number of ij,t is the current flowing through line (i, j) in time period t; For I ij,t The complex conjugate of ; S302, the equality constraint in the improved power flow constraint Relax to inequality constraints and obtain the final power flow constraints; The expression after relaxation to inequality constraints is: S303: Nonlinear equality constraints in intelligent soft switch operation constraints Convex relaxation is performed to obtain the improved intelligent soft switching constraint; The expression after convex relaxation is: Among them, ΔP i S is the active power loss of the converter with the intelligent soft switch connected to node i; is the active power loss of the converter with the intelligent soft switch connected to node j; is the loss coefficient of the converter connected to node i; is the loss coefficient of the converter connected to node j; P i S is the active power injected into node i by the intelligent soft switch; is the active power injected into node j by the intelligent soft switch; is the reactive power injected into node i by the intelligent soft switch; is the reactive power injected into node j by the intelligent soft switch; is the active power injected into node i by the intelligent soft switch in time period t; is the reactive power injected into node i by the intelligent soft switch in period t; is the active power injected into node j by the intelligent soft switch in period t; is the reactive power injected into node j by the intelligent soft switch in period t; is the active power loss of the converter with the intelligent soft switch connected to node i in time period t; is the active power loss of the converter with the intelligent soft switch connected to node j in time period t; S304, converting the improved intelligent soft switch constraint into a standard second-order cone form to obtain the final intelligent soft switch constraint: in, is the capacity of the converter connected to node i; is the capacity of the converter connected to node j; S305, replacing the power flow constraint and the intelligent soft switch operation constraint in the active distribution network power supply restoration model with the final power flow constraint and the final intelligent soft switch constraint, respectively, to obtain a mixed integer second-order cone programming model; S306. Use the Gurobi solver to solve the mixed integer second-order cone programming model to obtain a fault recovery control strategy.

Images