Power distribution system multi-stage dynamic recovery method considering cold load starting characteristics

Abstract

The invention discloses a power distribution system multi-stage dynamic recovery method considering cold load starting characteristics. The power distribution system multi-stage dynamic recovery method comprises the steps of 1, obtaining an operation state when a power distribution system breaks down; 2, establishing operation models of the power distribution system and various installed controllable equipment; 3, constructing an optimal load recovery model in combination with operation models of various devices, obtaining a dynamic network reconstruction scheme and a flexible device action sequence of the power distribution system by solving the optimal load recovery model, and issuing the dynamic network reconstruction scheme and the flexible device action sequence to a dynamic recovery model; 4, establishing a dynamic recovery stage model by combining an operation model of the equipment and considering a load recovery model of a cold load start characteristic CLPU and voltage and frequency safety; and 5, solving the dynamic recovery model to obtain a complete power distribution system multi-stage dynamic recovery scheme. The method can improve the recovery level of the power distribution system and the utilization rate of renewable energy sources, thereby improving the adaptability of a recovery scheme, and enhancing the self-repairing capability of the power distribution system.

Description

Technical Field

[0001] The invention belongs to the field of power distribution system fault recovery, and in particular is a multi-stage dynamic recovery method for a power distribution system taking cold load startup characteristics into consideration. Background Art

[0002] In recent years, extreme disasters have occurred frequently, posing a great threat to the reliable power supply of power systems. Compared with the transmission system, the distribution system is directly connected to power users, and the reliable power supply of the distribution system has a greater impact on users. On the other hand, the large-scale access of distributed power sources and renewable energy sources (such as photovoltaics and wind turbines) has turned the traditional radial distribution system into a multi-power system. With the development of power electronics technology, the distribution system has been further promoted to evolve towards a flexible closed-loop structure. When a failure occurs in the distribution system, the complex structure and control method make fault recovery more difficult.

[0003] How to face the above challenges and make good use of the multi-source controllable resources (such as distributed resources and power electronic equipment) of the distribution system to improve the resilience of the distribution system under extreme events is an urgent problem to be solved. Soft switches have precise power flow control, fast power transfer, and active voltage / frequency support, and have broad application prospects in the field of distribution system fault recovery. The reconfigurable soft switch is a further improved design based on the multi-terminal soft switch. Its design goal is to improve the utilization efficiency of the voltage source converter without changing the total capacity of the multi-terminal soft switch. This makes the reconfigurable soft switch have better performance than the multi-terminal soft switch in scenarios with high power transmission requirements. However, in fact, the introduction of this type of new power electronic equipment in the distribution system fault recovery will affect the order of switch operations and the actions of multiple recovery resources. Therefore, it is urgent to design a collaborative optimization framework to seamlessly coordinate them.

[0004] On the other hand, most distribution system restoration studies only consider the constant load model, ignoring the cold load recovery phenomenon caused by the temperature control load recovery. In the distribution system, the proportion of temperature control loads represented by air conditioning has increased year by year, and the proportion of temperature control loads will increase further under extreme weather conditions. However, the temperature control load will experience a surge in load within a short period of time after commissioning. The load peak will affect the safety of the distribution system during the power supply restoration process. Ignoring this feature will cause the resulting restoration plan to deviate from practical applications. Summary of the invention

[0005] The present invention aims to solve the deficiencies of the above-mentioned prior art and proposes a multi-stage dynamic recovery method for a distribution system taking into account the cold load startup characteristics, in order to utilize the high-power transmission capability of reconfigurable soft switches, precise flow control, rapid power transfer, and active voltage / frequency support to effectively eliminate system voltage problems during recovery, integrate the controllable resources of the distribution system, and establish a multi-stage dynamic recovery method to improve adaptability to cold load recovery phenomena, thereby accelerating the recovery of power-off loads.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme:

[0007] The invention provides a method for multi-stage dynamic recovery of a power distribution system taking into account cold load startup characteristics, which comprises the following steps:

[0008] S1: Obtain the operating status of the power distribution system when it fails, including: the location and number of the faulty line of the power distribution system, and the location and number of the non-faulty line;

[0009] S2: When the distribution system fails, obtain the information of wind turbine Wind, photovoltaic equipment PV, reconfigurable soft switch R-SOP and the demand information of user load L in each period; thereby establishing the operation model of reconfigurable soft switch R-SOP, distribution system fault reconstruction model and distribution system fault power flow model;

[0010] S3: Modeling of various adjustable devices, including: operation model of on-load tap-changing transformer OLTC, operation model of distributed generation DG, operation model of switchable capacitor CBs, and operation model of energy storage system ESS:

[0011] S4: After building and solving the optimal load recovery phase model, the dynamic network reconstruction plan, energy storage capacity pre-allocation plan and discrete device action plan of the distribution system are obtained and sent to the dynamic recovery phase model;

[0012] S5: Considering the load recovery characteristics of cold load startup characteristics, the frequency safety response constraints of the distribution system during the power failure load recovery process are established and the operation model of the adjustable equipment is updated to construct a dynamic recovery stage model:

[0013] S6: After establishing and solving the dynamic recovery stage model, the action sequence including the reconfigurable soft switch, distributed power supply and controllable resources in the dynamic recovery stage is obtained, thereby obtaining a complete fault dynamic recovery solution;

[0014] S7: The multi-stage dynamic recovery constraints in the optimal load recovery stage model and the dynamic recovery stage model are converted into mixed integer second-order cone programming constraints and then solved to obtain a dynamic recovery scheme for the distribution system including the reconfigurable soft switch action, the on-load tap-changing transformer action, the switchable capacitor action, the energy storage system action, the load switch action, the section switch disconnection, and the tie switch disconnection;

[0015] The multi-stage dynamic recovery method of a power distribution system considering cold load startup characteristics described in the present invention is also characterized in that S2 includes the following steps:

[0016] S2-1: Using equations (1) to (6), the operation model of the reconfigurable soft switch R-SOP is obtained:

[0017] (1)

[0018] (2)

[0019] (3)

[0020] (4)

[0021] (5)

[0022] (6)

[0023] In formula (1) to formula (6), represents the set of nodes connected to the voltage source converter VSC, represents the set of all voltage source converters; for Time Node Total losses of all connected VSCs, is the total DC side active power of all VSCs, is the total AC side active power of all VSCs; for Time Node The apparent power transmitted by the reconfigurable soft-switching R-SOP, for Time Node Reactive power transmitted via reconfigurable soft switch R-SOP; for Time Node Voltage Source Converter The connectivity status; is the operating loss coefficient of each voltage source converter; Voltage Source Converter Capacity;

[0024] S2-2: Use equations (7) to (12) to establish a distribution system fault reconstruction model containing a reconfigurable soft switch R-SOP:

[0025] (7)

[0026] (8)

[0027] (9)

[0028] (10)

[0029] (11)

[0030] (12)

[0031] In formula (7) to formula (12), is the set of distribution line nodes, is the node of the distribution line, ; Representation Node To Node The line, Represents all slave nodes To Node The line; Is a node With Node Between lines The connectivity status; is the number of nodes in the power distribution system; is the number of substations in the distribution system; is the number of ports of the reconfigurable soft switch R-SOP; Is a node The connected reconfigurable soft switch R-SOP port is turned on. When the R-SOP is used as a virtual power supply output, is equal to 1, otherwise it is equal to 0; It is a line The virtual trend, It is a line Virtual trends on yes Time Node The virtual power flow from the virtual power source; Is a node Virtual load demand; M is a constant;

[0032] S2-3: Using equations (13) to (22), the fault flow model of the distribution system with reconfigurable soft switch R-SOP is obtained:

[0033] (13)

[0034] (14)

[0035] (15)

[0036] (16)

[0037] (17)

[0038] (18)

[0039] (19)

[0040] (20)

[0041] (twenty one)

[0042] (twenty two)

[0043] In formula (13) to formula (22), and They are Timeline Active power and reactive power transmitted on the and Line resistance and reactance; for Timeline The current transmitted on and They are Time Node The injected active power and reactive power; and They are Timeline Active power and reactive power transmitted on the , They are Time Node , The voltage at and Line resistance and reactance; for Timeline The current transmitted on for Time Node With Node Between lines The connectivity status; and They are Time Node The reduction in active power output of photovoltaic equipment PV and the reduction in photovoltaic active power; and They are Time Node The reduction in active power output of the wind turbine Wind and the reduction in active power of the wind turbine; , They are Time Node Active and reactive output of distributed power source DG; , for Time Node Active and reactive output of R-SOP; and They are Time Node The reactive power output of photovoltaic equipment PV and the reactive power output of wind turbine Wind; and They are Time Node The active power and reactive power consumed by the load L; yes Time Node Reactive power output of switchable capacitors CBs; for Time Node The load switch state that controls the load disconnection; and are the upper and lower limits of the node voltage of the distribution system respectively; It is the upper limit of the line current in the power distribution system.

[0044] Furthermore, S3 includes the following steps:

[0045] S3-1: Use equations (23) and (24) to establish the operation model of the on-load tap-changing transformer OLTC:

[0046] (twenty three)

[0047] (twenty four)

[0048] In formula (23)-formula (24), for Time Node The voltage at the auxiliary node aux is equal to the voltage on the secondary side of the on-load tap-changing transformer; for Timeline The tap position of the on-load tap changing transformer OLTC; for Timeline The voltage regulation rate of the on-load tap-changing transformer OLTC; For line Initial voltage regulation rate of the on-load tap-changing transformer OLTC; For line The change in voltage regulation rate at adjacent tap positions; is a set of integer variables;

[0049] S3-2: Use equations (25) to (28) to construct the operation model of distributed power generation DG and the operation model of switchable capacitors CBs:

[0050] (25)

[0051] (26)

[0052] (27)

[0053] (28)

[0054] In formula (25) to formula (28), , They are Time Node Active and reactive output of distributed power source DG; for Time Node The distributed power source DG is in the on state; , They are the maximum active output and the maximum reactive output of the distributed generation DG respectively; yes Time Node Reactive power output of switchable capacitors CBs; is the capacity of each group of switchable capacitors CBs, yes Time Node The number of groups of switchable capacitors CBs put into operation, The maximum number of groups of switchable capacitors CBs allowed to be put into operation;

[0055] S3-3: Use equations (29) to (33) to construct the operation model of the energy storage system ESS:

[0056] (29)

[0057] (30)

[0058] (31)

[0059] (32)

[0060] (33)

[0061] In formula (29) to formula (33), , are the minimum and maximum charging power of the energy storage system ESS, , are the minimum and maximum values ​​of the discharge power of the energy storage system ESS respectively; for The auxiliary variables set at all times, when the energy storage system ESS is charging, =1, when ESS is discharged, let =0; , They are respectively the energy storage system ESS Charging power and discharging power at the moment; , , are the minimum, maximum and initial values ​​of the energy storage charge state respectively; , are the total capacity of the energy storage system ESS and The capacity of the energy storage system ESS at any given moment; for The capacity of the energy storage system ESS at any given moment; , They are the charging efficiency and discharging efficiency of the energy storage system ESS respectively; is the time step.

[0062] Furthermore, S4 uses formula (34) to establish the objective function of the optimal load recovery stage model: , and take equations (7)-(12), (13)-(22), (1), (2), (4)-(6), (33), (23)-(32) as constraints of the optimal load recovery stage model;

[0063] (34)

[0064] In formula (34), , , They are the distribution system operation loss, load reduction and new energy output reduction. is the weight coefficient of the distribution system operation loss, is the weight coefficient of load reduction, is the weight coefficient of the reduction in renewable energy output; The failure time.

[0065] Further, S5 includes the following steps:

[0066] S5-1: Use equations (35) to (38) to establish a load recovery model that considers the cold load startup characteristics:

[0067] (35)

[0068] (36)

[0069] (37)

[0070] (38)

[0071] In formula (35) to formula (38), Indicates the cold start load recovery stage, , They are stage, Phase Node The multiple of the load demand when the cold start load demand is relatively stable; yes Phase Node The change in the cold start load demand ratio; , Node The ratio of the peak value of cold start load demand and the stable value of demand to the normal value; Is a node Attenuation factor of cold starting load; Is a node At the cold start load stage, the load starts to diversify from the state to Duration of the phase; Is a node The duration of the load from recovery to the state of diversification; is a piecewise function;

[0072] S5-2: Based on the proportion of cold load and normal load in the distribution system, the unrecovered nodes with cold start load are modeled using equations (39) to (41):

[0073] (39)

[0074] (40) (41)

[0075] In formula (39) to formula (41), , Respectively Time Node The active power demand and reactive power demand of the actual load L; and They are Time Node The active power and reactive power consumed by the load L; for Time Node The load switch state that controls the load disconnection; for Time Node The load switch state that controls the load disconnection; Is the cooling load at the node The proportion of

[0076] S5-3: Use equation (42) to construct the frequency response rate constraint:

[0077] (42)

[0078] In formula (42), It is the collection of load nodes in the distribution system that are not powered by the main grid. is the coefficient, is the maximum value of the active power transmitted by the reconfigurable soft switch R-SOP. is the maximum active power output of the distributed generation DG;

[0079] S5-4: In the dynamic recovery stage at a small time scale, the power ramp constraints of the energy storage system and distributed generation are constructed using equations (43) to (46):

[0080] (43)

[0081] (44)

[0082] (45)

[0083] (46)

[0084] In formula (43) to formula (46), , They are respectively the energy storage system ESS Charging power and discharging power at the moment; , They are respectively the energy storage system ESS Charging power and discharging power at the moment; is the limit value of energy storage power change; yes Time Node Active power output of distributed generation DG; yes Time Node Active power output of distributed generation DG; For Node The allowable value of the climbing power of the distributed power source DG, Solve for the time scale for the dynamic recovery phase.

[0085] Furthermore, in S6, equation (47) is used to establish the objective function of the dynamic recovery stage model, and equations (13)-(22), (42), (1), (2), (4)-(6), (33), (23)-(32), (43)-(46), (35)-(41) are used as constraints of the dynamic recovery stage model:

[0086] (47)

[0087] In formula (47), It is the weight coefficient of the energy deviation value of the two-stage energy storage.

[0088] Furthermore, S7 transforms the multi-stage dynamic recovery constraints into mixed integer second-order cone programming constraints in the following steps:

[0089] S7-1: Convert two nonlinear variables and Replaced by two linear variables and The big-M method is used to relax the fault power flow constraints (13)-(15) and (21)-(22) to transform them into linear constraints, and then the constraints that meet the fault scenario are constructed using (48)-(52):

[0090] (48)

[0091] (49)

[0092] (50)

[0093] (51)

[0094] (52)

[0095] In formula (48) to formula (52), and The square of the voltage at node i at time t and the square of the branch current between node i and node j at time t respectively;

[0096] S7-2: Use equation (53) to transform equation (16) into a second-order cone constraint:

[0097] (53)

[0098] S7-3: Modify equation (23) to construct the power flow constraint using equation (54);

[0099] (54)

[0100] In formula (54), Expressed as Time Node The square of the auxiliary node aux voltage at;

[0101] S7-4: Introduce auxiliary 0 / 1 variables and auxiliary voltage variables, and use equations (55) and (56) to obtain The transformed constraints are:

[0102] (55)

[0103] (56)

[0104] In formula (55), For line The maximum value at the OLTC tap position;

[0105] S7-5: Use equations (58) and (59) to construct linear constraints:

[0106] (58)

[0107] (59)

[0108] In formula (58) and formula (59), for Time corresponding to the node To Node On the line The tap position The closed state of If it is turned on, is 0, for Time corresponding to the node To Node Tap position on line The square of the upper voltage;

[0109] S7-6: Use equation (60) to transform the capacity constraint of R-SOP into a rotating cone constraint:

[0110] (60).

[0111] An electronic device of the present invention includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the multi-stage dynamic recovery method of the power distribution system, and the processor is configured to execute the program stored in the memory.

[0112] The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program executes the steps of the multi-stage dynamic recovery method of the power distribution system when the computer program is executed by a processor.

[0113] Compared with the prior art, the present invention has the following beneficial effects:

[0114] 1. The present invention proposes a multi-stage dynamic recovery model that adapts to the cold load recovery phenomenon caused by the temperature-controlled load recovery, taking into account the actual cold load recovery phenomenon caused by the temperature-controlled load recovery, thereby effectively meeting the actual load operation requirements and realizing reliable and adaptive recovery of the distribution system.

[0115] 2. The present invention integrates the scheduling of controllable devices and the dynamic reconstruction of the network, and establishes a multi-device collaborative optimization framework. The present invention enables various types of flexible devices to coordinate on different time scales according to their own characteristics and operating constraints. The recovery method integrates the reconstruction of the asymmetric voltage source converter and the sequential execution of the response to the cold load recovery phenomenon, thereby improving the adaptability of the recovery scheme, enhancing the self-repair capability of the distribution system, improving the power grid flow distribution, and improving the level of distribution service recovery.

[0116] 3. The present invention takes into account the voltage and frequency safety of the power distribution system during the fault recovery process, can ensure system safety during the recovery process, perform fault recovery in steps, and has better applicability. BRIEF DESCRIPTION OF THE DRAWINGS

[0117] Figure 1 This is a typical cooling load characteristic recovery curve;

[0118] Figure 2 It is the modified IEEE 33-node fault system topology diagram;

[0119] Figure 3 It is the action plan diagram of controllable resources of the distribution system;

[0120] Figure 4 is the total capacity of the voltage source converters connected to each feeder during the fault recovery process of the reconfigurable soft switch;

[0121] Figure 5 It is a schematic diagram of the load recovery percentage in each fault period under Case 1 and Case 2;

[0122] Figure 6 It is a schematic diagram of load recovery under Case 1 and Case 2;

[0123] Figure 7 It is a flow chart of the invention for implementing the method of multi-stage dynamic recovery model. DETAILED DESCRIPTION

[0124] In this embodiment, a multi-stage dynamic recovery method for a distribution system considering cold load recovery with the assistance of a reconfigurable soft switch is proposed. Taking into account the dynamic recovery characteristics of cold start load recovery, and combining the reconfigurable intelligent soft switch reconstruction, distribution network topology changes and flexible resource synergy, a multi-stage dynamic recovery model is constructed: the optimal load recovery stage and the dynamic recovery stage, which respond to cold start load demands on a faster time scale and perform fault recovery in a reliable and adaptive manner to improve the recovery level of the distribution system and the utilization rate of renewable energy, thereby improving the adaptability of the recovery plan and enhancing the self-repair capability of the distribution system. Specifically, if Figure 7 As shown, the method comprises the following steps:

[0125] S1: Obtain the operating status of the power distribution system when it fails, including: the location and number of the faulty line of the power distribution system, the location and number of the non-faulty line; to calculate the power-off area after the power distribution system fault is isolated;

[0126] S2: When a distribution system fails, obtain the information of wind turbine Wind, photovoltaic equipment PV, reconfigurable soft switch R-SOP and the demand information of user load L in each time period; thereby establishing the operation model of the reconfigurable soft switch R-SOP, the distribution system fault reconstruction model, and the distribution system fault power flow model.

[0127] S2-1: Using equations (1) to (6), the operation model of the reconfigurable soft switch R-SOP is obtained:

[0128] (1)

[0129] (2)

[0130] (3)

[0131] (4)

[0132] (5)

[0133] (6)

[0134] In formula (1) to formula (6), represents the set of nodes connected to the voltage source converter VSC, represents the set of all voltage source converters; for Time Node Total losses of all connected VSCs, is the total DC side active power of all VSCs, is the total AC side active power of all VSCs; for Time Node The apparent power transmitted by the reconfigurable soft-switching R-SOP, for Time Node Reactive power transmitted via reconfigurable soft switch R-SOP; for Time Node Voltage Source Converter The connectivity status; is the operating loss coefficient of each voltage source converter; Voltage Source Converter capacity.

[0135] S2-2: The virtual power flow method is used to establish a distribution system fault reconstruction model. The distribution system fault reconstruction model containing the reconfigurable soft switch R-SOP is established using equations (7) to (12) to obtain the distribution system fault reconstruction plan for the recovery process:

[0136] (7)

[0137] (8)

[0138] (9)

[0139] (10)

[0140] (11)

[0141] (12)

[0142] In formula (7) to formula (12), is the set of distribution line nodes, is the node of the distribution line, ; Representation Node To Node The line, Represents all slave nodes To Node The line; Is a node With Node Between lines The connectivity status; is the number of nodes in the power distribution system; is the number of substations in the distribution system; is the number of ports of the reconfigurable soft switch R-SOP; Is a node The connected reconfigurable soft switch R-SOP port is turned on. When the R-SOP is used as a virtual power supply output, is equal to 1, otherwise it is equal to 0; It is a line The virtual trend, It is a line Virtual trends on yes Time Node The virtual power flow from the virtual power source; Is a node virtual load demand; M is a constant.

[0143] S2-3: Using equations (13) to (22), the fault flow model of the distribution system with reconfigurable soft switch R-SOP is obtained:

[0144] (13)

[0145] (14)

[0146] (15)

[0147] (16)

[0148] (17)

[0149] (18)

[0150] (19)

[0151] (20)

[0152] (twenty one)

[0153] (twenty two)

[0154] In formula (13) to formula (22), and They are Timeline Active power and reactive power transmitted on the and Line resistance and reactance; for Timeline The current transmitted on and They are Time Node The injected active power and reactive power; and They are Timeline Active power and reactive power transmitted on the , They are Time Node , The voltage at and Line resistance and reactance; for Timeline The current transmitted on for Time Node With Node Between lines The connectivity status; and They are Time Node The reduction in active power output of photovoltaic equipment PV and the reduction in photovoltaic active power; and They are Time Node The reduction in active power output of the wind turbine Wind and the reduction in active power of the wind turbine; , They are Time Node Active and reactive output of distributed power source DG; , for Time Node Active and reactive output of R-SOP; and They are Time Node The reactive power output of photovoltaic equipment PV and the reactive power output of wind turbine Wind; and They are Time Node The active power and reactive power consumed by the load L; yes Time Node Reactive power output of switchable capacitors CBs; for Time Node The load switch state that controls the load disconnection; and are the upper and lower limits of the node voltage of the distribution system respectively; It is the upper limit of the line current in the power distribution system.

[0155] S3: Modeling of various adjustable devices, including: operation model of on-load tap-changing transformer OLTC, operation model of distributed generation DG, operation model of switchable capacitor CBs, and operation model of energy storage system ESS:

[0156] S3-1: An auxiliary node aux is introduced on the secondary side of the on-load tap-changing transformer (OLTC). After ignoring the loss of OLTC itself, the operation model of the on-load tap-changing transformer OLTC is established using equations (23)-(24):

[0157] (twenty three)

[0158] (twenty four)

[0159] In formula (23)-formula (24), for Time Node The voltage at the auxiliary node aux is equal to the voltage on the secondary side of the on-load tap-changing transformer; for Timeline The tap position of the on-load tap changing transformer OLTC; for Timeline The voltage regulation rate of the on-load tap-changing transformer OLTC; For line Initial voltage regulation rate of the on-load tap-changing transformer OLTC; For line The change in voltage regulation rate at adjacent tap positions; is a set of integer variables.

[0160] S3-2: Use equations (25) to (28) to construct the operation model of distributed power generation DG and the operation model of switchable capacitors CBs:

[0161] (25)

[0162] (26)

[0163] (27)

[0164] (28)

[0165] In formula (25) to formula (28), , They are Time Node Active and reactive output of distributed power source DG; for Time Node The distributed power source DG is in the on state; , They are the maximum active output and the maximum reactive output of the distributed generation DG respectively; yes Time Node Reactive power output of switchable capacitors CBs; is the capacity of each group of switchable capacitors CBs, yes Time Node The number of groups of switchable capacitors CBs put into operation, It is the maximum number of groups of switchable capacitors CBs allowed to be put into operation.

[0166] S3-3: Using equations (29) to (33) to construct the operation model of the energy storage system ESS, installing the energy storage system on the DC side of the R-SOP can enable the R-SOP to have energy storage capabilities and accurately control the power flow in both space and time dimensions. Modify equation (3) to equation (33):

[0167] (29)

[0168] (30)

[0169] (31)

[0170] (32)

[0171] (33)

[0172] In formula (29) to formula (33), , are the minimum and maximum charging power of the energy storage system ESS, , are the minimum and maximum values ​​of the discharge power of the energy storage system ESS respectively; for The auxiliary variables set at all times, when the energy storage system ESS is charging, =1, when ESS is discharged, let =0; , They are respectively the energy storage system ESS Charging power and discharging power at the moment; , , are the minimum, maximum and initial values ​​of the energy storage charge state respectively; , are the total capacity of the energy storage system ESS and The capacity of the energy storage system ESS at any given moment; for The capacity of the energy storage system ESS at any given moment; , They are the charging efficiency and discharging efficiency of the energy storage system ESS respectively; is the time step.

[0173] S4: After building and solving the optimal load recovery phase model, the dynamic network reconstruction plan, energy storage capacity pre-allocation plan and discrete device action plan of the distribution system are obtained and sent to the dynamic recovery phase model;

[0174] S4-1: Using equation (34) to establish the objective function of the optimal load recovery stage model :

[0175] (34)

[0176] In formula (34), , , They are the distribution system operation loss, load reduction and new energy output reduction. is the weight coefficient of the distribution system operation loss, is the weight coefficient of load reduction, is the weight coefficient of the reduction in renewable energy output; is the failure time;

[0177] S4-2: Use equations (7)-(12), (13)-(22), (1), (2), (4)-(6), (33), and (23)-(32) as constraints for the optimal load recovery stage model.

[0178] S5: Considering the load recovery characteristics of cold load startup characteristics, the frequency safety response constraints of the distribution system during the power failure load recovery process are established and the operation model of the adjustable equipment is updated to construct a dynamic recovery stage model:

[0179] S5-1: Adoption Figure 1 The exponential cold start load model shown is used to simulate the cold start load demand in the actual distribution system, and the load recovery model considering the cold load start characteristics is established using equations (35) to (38):

[0180] (35)

[0181] (36)

[0182] (37)

[0183] (38)

[0184] In formula (35) to formula (38), Indicates the cold start load recovery stage, , They are stage, Phase Node The multiple of the load demand when the cold start load demand is relatively stable; yes Phase Node The change in the cold start load demand ratio; , Node The ratio of the peak value of cold start load demand and the stable value of demand to the normal value; Is a node Attenuation factor of cold starting load; Is a node At the cold start load stage, the load starts to diversify from the state to Duration of the phase; Is a node The duration of the load from recovery to the state of diversification; is a piecewise function.

[0185] S5-2: Based on the proportion of cold load and normal load in the distribution system, the load modeling in equations (17) and (18) needs to be modified in the dynamic recovery model. The fully recovered cold start load can be treated as a normal load. When it is not fully recovered, the cold start load should not be reduced. The unrecovered nodes containing cold start loads are modeled using equations (39) to (41):

[0186] (39)

[0187] (40) (41)

[0188] In formula (39) to formula (41), , Respectively Time Node The active power demand and reactive power demand of the actual load L; and They are Time Node The active power and reactive power consumed by the load L; for Time Node The load switch state that controls the load disconnection; for Time Node The load switch state that controls the load disconnection; Is the cooling load at the node % of.

[0189] S5-3: In order to reduce the difficulty of solving the model of the frequency-constrained recovery scheme, a simplified frequency response rate constraint is used to approximate the frequency response, and the frequency response rate constraint is constructed using equation (42):

[0190] (42)

[0191] In formula (42), It is the collection of load nodes in the distribution system that are not powered by the main grid. is the coefficient, is the maximum value of the active power transmitted by the reconfigurable soft switch R-SOP. is the maximum active power output of the distributed generation DG;

[0192] S5-4: In the dynamic recovery stage at a small time scale, the power ramp constraints of the energy storage system and distributed generation are constructed using equations (43) to (46):

[0193] (43)

[0194] (44)

[0195] (45)

[0196] (46)

[0197] In formula (43) to formula (46), , They are respectively the energy storage system ESS Charging power and discharging power at the moment; , They are respectively the energy storage system ESS Charging power and discharging power at the moment; is the limit value of energy storage power change; yes Time Node Active power output of distributed generation DG; yes Time Node Active power output of distributed generation DG; For Node The allowable value of the climbing power of the distributed power source DG, Solve for the time scale for the dynamic recovery phase.

[0198] S6: After establishing and solving the dynamic recovery stage model, the action sequence including the reconfigurable soft switch, distributed power supply and controllable resources in the dynamic recovery stage is obtained, thereby obtaining a complete fault dynamic recovery solution;

[0199] S6-1: Use equation (47) to establish the objective function of the dynamic recovery stage model:

[0200] (47)

[0201] In formula (47), is the weight coefficient of the energy deviation value of the two-stage energy storage;

[0202] S6-2: Use equations (13)-(22), (42), (1), (2), (4)-(6), (33), (23)-(32), (43)-(46), and (35)-(41) as constraints for the dynamic recovery phase model.

[0203] S7: The multi-stage dynamic recovery constraints in the optimal load recovery stage model and the dynamic recovery stage model are converted into mixed integer second-order cone programming constraints and then solved to obtain a dynamic recovery scheme for the distribution system including the reconfigurable soft switch action, the on-load tap-changing transformer action, the switchable capacitor action, the energy storage system action, the load switch action, the section switch disconnection, and the tie switch disconnection;

[0204] S7-1: Convert two nonlinear variables and Replaced by two linear variables and The big-M method is used to relax the fault power flow constraints (13)-(15) and (21)-(22) to transform them into linear constraints, and then the constraints that meet the fault scenario are constructed using (48)-(52):

[0205] (48)

[0206] (49)

[0207] (50)

[0208] (51)

[0209] (52)

[0210] In formula (48) to formula (52), and The square of the voltage at node i at time t and the square of the branch current between node i and node j at time t respectively.

[0211] S7-2: Use equation (53) to transform equation (16) into a second-order cone constraint:

[0212] (53)

[0213] S7-3: Modify equation (23) to construct the power flow constraint using equation (54);

[0214] (54)

[0215] In formula (54), Expressed as Time Node The square of the auxiliary node aux voltage at;

[0216] S7-4: Introduce auxiliary 0 / 1 variables and auxiliary voltage variables, and use equations (55) and (56) to obtain The transformed constraints are:

[0217] (55)

[0218] (56)

[0219] In formula (55), For line The maximum value at the OLTC tap position;

[0220] S7-5: Use equations (58) and (59) to construct linear constraints:

[0221] (58)

[0222] (59)

[0223] In formula (58) and formula (59), for Time corresponding to the node To Node On the line The tap position The closed state of If it is turned on, is 0, for Time corresponding to the node To Node Tap position on line The square of the voltage on the

[0224] S7-6: Use equation (60) to transform the capacity constraint of R-SOP into a rotating cone constraint:

[0225] (60)

[0226] The multi-stage dynamic recovery constraints are converted into a mixed integer second-order cone programming problem using the above method. The optimal load recovery stage and the dynamic recovery stage are solved with the help of a solver, and a dynamic recovery plan for the distribution system including reconfigurable soft switch action, on-load tap-changing transformer action, switchable capacitor action, energy storage system action, load switch action, section switch disconnection, and tie switch disconnection can be obtained.

[0227] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0228] In this embodiment, a computer-readable storage medium stores a computer program on the computer-readable storage medium, and the computer program executes the steps of the above method when executed by a processor.

[0229] In order to enable those skilled in the art to better understand the present invention, the example analysis includes the following components:

[0230] 1. Case description and simulation results analysis

[0231] In order to verify its effectiveness, the present invention uses Figure 2 The modified IEEE 33-node fault system topology shown in the figure is used for example analysis. The system arranges photovoltaic and wind turbine systems at nodes 16 and 31, with a capacity of 2MVA each and a power factor of 0.85. Distributed power sources and switchable capacitors are set at nodes 27 and 28 respectively. The maximum active output of distributed power sources is 0.35MW and the maximum reactive output is 0.35MVar. The switchable capacitor can put into the capacitor bank capacity of 0.15MVar each time, and a maximum of 4 groups of capacitors can be put into use. The cold start load is set to decay to a stable value after 10 minutes. The total capacity of the 4-port reconfigurable soft switch is 4MVA. The capacity allocation of each voltage source converter port adopts the "golden ratio size" allocation method, which is 0.5, 0.309, 0.118, and 0.073 times of the total capacity respectively. The nodes available for feeder selection switches are set at nodes 12, 18, 22, and 33. M is set to 1000. The voltage amplitude is [0.95,1.05] pu. The time scales of the two stages are set to 30 minutes and 2 minutes respectively. Example setting Due to extreme weather, permanent faults occurred in lines 2 and 25 of the test system at 8 am. After the fault was isolated, the distribution system began to perform self-healing recovery. The fault repair time of the test system is set to 4 hours.

[0232] In order to verify the effectiveness of the multi-stage dynamic recovery scheme described in the invention, two schemes, Case 1 and Case 2, are set up in the example part for comparison.

[0233] Case 1: Fault recovery method for a distribution system with a 4-port reconfigurable soft switch. This method ignores the impact of cold start load characteristics and therefore adopts a long-term recovery scheme.

[0234] Case 2: Multi-stage dynamic recovery scheme with 4-port reconfigurable soft switch assisted dynamic repair.

[0235] The simulation program was implemented in Matlab environment in a computer with 64-bit Windows, Intel(R) CoreTM i5 CPU@3.5GHz, and 8GB memory. The YALMIP toolbox and Gurobi solver were used for solving.

[0236] exist Figure 2 In the IEEE 33-node test system shown in FIG. 1 , the above two schemes are executed simultaneously. The specific statistical data under the two schemes are shown in Table 1.

[0237] Table 1

[0238] Load loss / MWh Average voltage deviation / % System loss / MWh Reduction in renewable energy / MWh Case 1 3.2737 3.6574 1.2433 2.2901 Case 2 0.1464 3.6640 0.3953 0

[0239] As can be seen from Table 1, both cases can guarantee good voltage performance during the fault recovery process, and the average system voltage deviation is less than 5%, which shows that the reconfigurable soft switch has good voltage regulation capability. Case 2 adopts a multi-stage dynamic recovery scheme, which is superior to Case 1 in improving load recovery, reducing system losses and promoting new energy consumption. Case 2 achieves flexible allocation of voltage source converter capacity by virtue of the reconfigurable design of the reconfigurable soft switch, which improves the transmission power of the reconfigurable soft switch. The load loss during the entire fault recovery process is only 0.1464MWh, which is much smaller than the other case, and can achieve full consumption of new energy. It can be seen that the recovery method of the present invention utilizes the synergistic benefits of the control scheduling of the reconfigurable soft switch and the change of network topology, and adapts to the multi-stage dynamic recovery model of the cold start load characteristics, thereby effectively meeting the actual load operation behavior and realizing reliable and adaptive recovery of the distribution system.

[0240] like Figure 3 As shown in Figure 1, various flexible devices coordinate on different time scales according to their own characteristics and operating constraints, thereby improving the power flow distribution of the power grid and improving the level of power distribution service recovery. Figure 4 It can be seen that the total capacity of the voltage source converters connected to each feeder during the fault recovery process of the reconfigurable soft switch. It can be seen that the reconfigurable soft switch can dynamically allocate the total capacity of the voltage source converters connected to each feeder according to the actual power transmission requirements of the feeder during operation.

[0241] Depend on Figure 5 It can be seen that both cases set at the initial stage of fault recovery cannot reach a high recovery level immediately. On the one hand, because the power-off area mainly relies on the reconfigurable soft switch or the reconfigurable soft switch to transfer power from the main grid, the frequency stability of the power-off area needs to be considered during the recovery process, so the load that can be restored in a single step is limited. On the other hand, due to the starting characteristics of the cold load, the power demand of the cold start load is much greater than the load demand during stability at the initial stage of recovery. The reconfigurable soft switch cannot exceed the capacity limit of the voltage source converter device, so it cannot support more load recovery. Among the two cases, Case 2 has the best recovery effect. Case 1 does not consider the cold start load characteristics of the load and adopts a longer time scale to formulate a recovery plan, resulting in a recovery effect far inferior to that of Case 2.

[0242] Depend on Figure 6It can be seen that, unlike Case 1, Case 2 adopts a sequential recovery method after considering the cold start load characteristics and the frequency response constraints during the recovery process, and performs fault recovery in steps, which has better applicability. In the test scenario, the load recovery amount of Case 2 in each period is higher than that of Case 1. Case 1 does not consider the cold start load characteristics and adopts a long-term recovery plan. The load switch is actuated every 30 minutes. When the recovery is actually performed, the load switch cannot be closed according to the expected recovery plan, and the system cannot achieve the expected recovery goal.

[0243] In this specification, the schematic description of the present invention does not necessarily refer to the same embodiment or example, and those skilled in the art may combine and combine different embodiments or examples described in this specification. In addition, the contents of the embodiments of this specification are merely an enumeration of the implementation forms of the inventive concept, and the protection scope of the present invention should not be considered as limited to the specific forms described in the implementation cases. The protection scope of the present invention also includes equivalent technical means that those skilled in the art can think of based on the inventive concept.

Claims

1. A multi-stage dynamic recovery method for a power distribution system considering cold load startup characteristics, characterized in that: The following steps are involved: S1: Obtain the operating status of the power distribution system when it fails, including: the location and number of the faulty line of the power distribution system, and the location and number of the non-faulty line; S2: When the distribution system fails, obtain the information of wind turbine Wind, photovoltaic equipment PV, reconfigurable soft switch R-SOP and the demand information of user load L in each period; thereby establishing the operation model of reconfigurable soft switch R-SOP, distribution system fault reconstruction model and distribution system fault power flow model; S3: Modeling of various adjustable devices, including: operation model of on-load tap-changing transformer OLTC, operation model of distributed generation DG, operation model of switchable capacitor CBs, and operation model of energy storage system ESS: S4: After building and solving the optimal load recovery phase model, the dynamic network reconstruction plan, energy storage capacity pre-allocation plan and discrete device action plan of the distribution system are obtained and sent to the dynamic recovery phase model; S5: Considering the load recovery characteristics of cold load startup characteristics, the frequency safety response constraints of the distribution system during the power failure load recovery process are established and the operation model of the adjustable equipment is updated to construct a dynamic recovery stage model: S6: After establishing and solving the dynamic recovery stage model, the action sequence including the reconfigurable soft switch, distributed power supply and controllable resources in the dynamic recovery stage is obtained, thereby obtaining a complete fault dynamic recovery solution; S7: The multi-stage dynamic recovery constraints in the optimal load recovery stage model and the dynamic recovery stage model are converted into mixed integer second-order cone programming constraints and then solved to obtain a dynamic recovery plan for the distribution system including reconfigurable soft switch action, on-load tap-changing transformer action, switchable capacitor action, energy storage system action, load switch action, section switch disconnection, and tie switch disconnection.

2. A multi-stage dynamic recovery method for a power distribution system considering cold load startup characteristics according to claim 1, characterized in that: S2 includes the following steps: S2-1: Using equations (1) to (6), the operation model of the reconfigurable soft switch R-SOP is obtained: (1) (2) (3) (4) (5) (6) In formula (1) to formula (6), represents the set of nodes connected to the voltage source converter VSC, represents the set of all voltage source converters; for Time Node Total losses of all connected VSCs, is the total DC side active power of all VSCs, is the total AC side active power of all VSCs; for Time Node The apparent power transmitted by the reconfigurable soft-switching R-SOP, for Time Node Reactive power transmitted via reconfigurable soft switch R-SOP; for Time Node Voltage Source Converter The connectivity status; is the operating loss coefficient of each voltage source converter; Voltage Source Converter Capacity; S2-2: Use equations (7) to (12) to establish a distribution system fault reconstruction model containing a reconfigurable soft switch R-SOP: (7) (8) (9) (10) (11) (12) In formula (7) to formula (12), is the set of distribution line nodes, is the node of the distribution line, ; Representation Node To Node The line, Represents all slave nodes To Node The line; Is a node With Node Between lines The connectivity status; is the number of nodes in the power distribution system; is the number of substations in the distribution system; is the number of ports of the reconfigurable soft switch R-SOP; Is a node The connected reconfigurable soft switch R-SOP port is turned on. When the R-SOP is used as a virtual power supply output, is equal to 1, otherwise it is equal to 0; It is a line The virtual trend, It is a line Virtual trends on yes Time Node The virtual power flow from the virtual power source; Is a node Virtual load demand; M is a constant; S2-3: Using equations (13) to (22), the fault flow model of the distribution system with reconfigurable soft switch R-SOP is obtained: (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) In formula (13) to formula (22), and They are Timeline Active power and reactive power transmitted on the and Line resistance and reactance; for Timeline The current transmitted on and They are Time Node The injected active power and reactive power; and They are Timeline Active power and reactive power transmitted on the , They are Time Node , The voltage at and Line resistance and reactance; for Timeline The current transmitted on for Time Node With Node Between lines The connectivity status; and They are Time Node The reduction in active power output of photovoltaic equipment PV and the reduction in photovoltaic active power; and They are Time Node The reduction in active power output of the wind turbine Wind and the reduction in active power of the wind turbine; , They are Time Node Active and reactive output of distributed power source DG; , for Time Node Active and reactive output of R-SOP; and They are Time Node The reactive power output of photovoltaic equipment PV and the reactive power output of wind turbine Wind; and They are Time Node The active power and reactive power consumed by the load L; yes Time Node Reactive power output of switchable capacitors CBs; for Time Node The load switch state that controls the load disconnection; and are the upper and lower limits of the node voltage of the distribution system respectively; It is the upper limit of the line current in the power distribution system.

3. A multi-stage dynamic recovery method for a power distribution system considering cold load startup characteristics according to claim 2, characterized in that: S3 includes the following steps: S3-1: Use equations (23) and (24) to establish the operation model of the on-load tap-changing transformer OLTC: (23) (24) In formula (23)-formula (24), for Time Node The voltage at the auxiliary node aux is equal to the voltage on the secondary side of the on-load tap-changing transformer; for Timeline The tap position of the on-load tap changing transformer OLTC; for Timeline The voltage regulation rate of the on-load tap-changing transformer OLTC; For line Initial voltage regulation rate of the on-load tap-changing transformer OLTC; For line The change in voltage regulation rate at adjacent tap positions; is a set of integer variables; S3-2: Use equations (25) to (28) to construct the operation model of distributed power generation DG and the operation model of switchable capacitors CBs: (25) (26) (27) (28) In formula (25) to formula (28), , They are Time Node Active and reactive output of distributed power source DG; for Time Node The distributed power source DG is in the on state; , They are the maximum active output and the maximum reactive output of the distributed generation DG respectively; yes Time Node Reactive power output of switchable capacitors CBs; is the capacity of each group of switchable capacitors CBs, yes Time Node The number of groups of switchable capacitors CBs put into operation, The maximum number of groups of switchable capacitors CBs allowed to be put into operation; S3-3: Use equations (29) to (33) to construct the operation model of the energy storage system ESS: (29) (30) (31) (32) (33) In formula (29) to formula (33), , are the minimum and maximum charging power of the energy storage system ESS, , are the minimum and maximum values ​​of the discharge power of the energy storage system ESS respectively; for The auxiliary variables set at all times, when the energy storage system ESS is charging, =1, when ESS is discharged, let =0; , They are respectively the energy storage system ESS Charging power and discharging power at the moment; , , are the minimum, maximum and initial values ​​of the energy storage charge state respectively; , are the total capacity of the energy storage system ESS and The capacity of the energy storage system ESS at any given moment; for The capacity of the energy storage system ESS at any given moment; , They are the charging efficiency and discharging efficiency of the energy storage system ESS respectively; is the time step.

4. A method for multi-stage dynamic recovery of a power distribution system considering cold load startup characteristics according to claim 3, characterized in that: S4 is the objective function of the optimal load recovery stage model established using formula (34): , and take equations (7)-(12), (13)-(22), (1), (2), (4)-(6), (33), (23)-(32) as constraints of the optimal load recovery stage model; (34) In formula (34), , , They are the distribution system operation loss, load reduction and new energy output reduction. is the weight coefficient of the distribution system operation loss, is the weight coefficient of load reduction, is the weight coefficient of the reduction in renewable energy output; The failure time.

5. A method for multi-stage dynamic recovery of a power distribution system considering cold load startup characteristics according to claim 4, characterized in that: S5 includes the following steps: S5-1: Use equations (35) to (38) to establish a load recovery model that considers the cold load startup characteristics: (35) (36) (37) (38) In formula (35) to formula (38), Indicates the cold start load recovery stage, , They are stage, Phase Node The multiple of the load demand when the cold start load demand is relatively stable; yes Phase Node The change in the cold start load demand ratio; , Node The ratio of the peak value of cold start load demand and the stable value of demand to the normal value; Is a node Attenuation factor of cold starting load; Is a node At the cold start load stage, the load starts to diversify from the state to Duration of the phase; Is a node The duration of the load from recovery to the state of diversification; is a piecewise function; S5-2: Based on the proportion of cold load and normal load in the distribution system, the unrecovered nodes with cold start load are modeled using equations (39) to (41): (39) (40) (41) In formula (39) to formula (41), , Respectively Time Node The active power demand and reactive power demand of the actual load L; and They are Time Node The active power and reactive power consumed by the load L; for Time Node The load switch state that controls the load disconnection; for Time Node The load switch state that controls the load disconnection; Is the cooling load at the node The proportion of S5-3: Use equation (42) to construct the frequency response rate constraint: (42) In formula (42), It is the collection of load nodes in the distribution system that are not powered by the main grid. is the coefficient, is the maximum value of the active power transmitted by the reconfigurable soft switch R-SOP. is the maximum active power output of the distributed generation DG; S5-4: In the dynamic recovery stage at a small time scale, the power ramp constraints of the energy storage system and distributed generation are constructed using equations (43) to (46): (43) (44) (45) (46) In formula (43) to formula (46), , They are respectively the energy storage system ESS Charging power and discharging power at the moment; , They are respectively the energy storage system ESS Charging power and discharging power at the moment; is the limit value of energy storage power change; yes Time Node Active power output of distributed generation DG; yes Time Node Active power output of distributed generation DG; For Node The allowable value of the climbing power of the distributed power source DG, Solve for the time scale for the dynamic recovery phase.

6. A method for multi-stage dynamic recovery of a power distribution system considering cold load startup characteristics according to claim 5, characterized in that: In S6, equation (47) is used to establish the objective function of the dynamic recovery stage model, and equations (13)-(22), (42), (1), (2), (4)-(6), (33), (23)-(32), (43)-(46), (35)-(41) are used as constraints of the dynamic recovery stage model: (47) In formula (47), It is the weight coefficient of the energy deviation value of the two-stage energy storage.

7. A method for multi-stage dynamic recovery of a power distribution system considering cold load startup characteristics according to claim 6, characterized in that: S7 converts the multi-stage dynamic recovery constraints into mixed integer second-order cone programming constraints in the following steps: S7-1: Convert two nonlinear variables and Replaced by two linear variables and The big-M method is used to relax the fault power flow constraints (13)-(15) and (21)-(22) to transform them into linear constraints, and then the constraints that meet the fault scenario are constructed using (48)-(52): (48) (49) (50) (51) (52) In formula (48) to formula (52), and The square of the voltage at node i at time t and the square of the branch current between node i and node j at time t respectively; S7-2: Use equation (53) to transform equation (16) into a second-order cone constraint: (53) S7-3: Modify equation (23) to construct the power flow constraint using equation (54); (54) In formula (54), Expressed as Time Node The square of the auxiliary node aux voltage at; S7-4: Introduce auxiliary 0 / 1 variables and auxiliary voltage variables, and use equations (55) and (56) to obtain The transformed constraints are: (55) (56) In formula (55), For line The maximum value at the OLTC tap position; S7-5: Use equations (58) and (59) to construct linear constraints: (58) (59) In formula (58) and formula (59), for Time corresponding to the node To Node On the line The tap position The closed state of If it is turned on, is 0, for Time corresponding to the node To Node Tap position on line The square of the upper voltage; S7-6: Use equation (60) to transform the capacity constraint of R-SOP into a rotating cone constraint: (60)。 8. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the multi-stage dynamic recovery method for the power distribution system as described in any one of claims 1-7, and the processor is configured to execute the program stored in the memory.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the multi-stage dynamic restoration method for a power distribution system according to any one of claims 1 to 7 are executed.

Images