Distributed self-healing method and device for power distribution network, electronic equipment and storage medium

Abstract

The invention discloses a distributed self-healing method and device for a power distribution network, electronic equipment and a storage medium. The method comprises the following steps: acquiring a historical load of a node of the power distribution network, a hardware parameter of an energy storage device, a network parameter of the power distribution network, a hardware parameter of a distributed photovoltaic device, a historical output and a hardware parameter of a reactive power compensation device; partitioning a power distribution network region; for each power distribution network region, constructing a distributed self-healing model according to the network parameters of the power distribution network; according to the historical load of the node, the network parameter of the power distribution network, the hardware parameter of the energy storage device, the hardware parameter of the distributed photovoltaic device and the hardware parameter of the reactive power compensation device, constructing a constraint condition of a distributed self-healing model; solving node recovery power, node recovery state and line current of the distributed self-healing models when all the distributed self-healing models are converged under constraint conditions; adjusting the power distribution network according to the node recovery power, the node recovery state and the line current; according to the invention, the reliability of the power distribution network can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems, and in particular to a distributed self-healing method, device, electronic equipment and storage medium for a distribution network. Background Art

[0002] In today's society, electricity is like the pulse of life, driving every corner of modern life. As the key link in the power system that connects the transmission network and users, the stable operation of the distribution network is crucial. However, various unforeseen factors, such as changes in the natural environment, aging of equipment and human operational errors, may cause distribution network failures. Once a failure occurs, it will not only affect people's daily life and industrial production, but may also cause huge economic losses.

[0003] At present, the self-healing method of distribution network is mostly centralized self-healing recovery, which monitors and analyzes the distribution network in real time through a centralized control center. Once a fault is detected, the control center will automatically take measures to isolate the fault and restore power supply. This method utilizes the regulation capabilities of distributed photovoltaic, energy storage devices and reactive power compensation devices to improve the safety of distribution network operation. However, the dependence of the centralized control center makes it face many challenges. Once the control center fails, the self-healing function of the entire distribution network may be paralyzed and the fault cannot be handled in time, thereby reducing the reliability of the distribution network. In addition, centralized control also needs to process a large amount of data and perform complex calculations, which may lead to longer decision-making time and affect the timeliness of fault handling. Summary of the invention

[0004] The embodiments of the present invention provide a distributed self-healing method, device, electronic device and storage medium for a distribution network. By implementing the present invention, when a fault is detected in the distribution network, a distributed self-healing strategy can be used to isolate the fault area and restore power supply as much as possible, thereby effectively dispersing processing tasks, reducing the burden on the control center, and improving the reliability of the distribution network and the timeliness of fault processing.

[0005] An embodiment of the present invention provides a distributed self-healing method for a power distribution network, comprising:

[0006] Obtain the historical load of the distribution network nodes, the hardware parameters of the energy storage device, the network parameters of the distribution network, the hardware parameters of the distributed photovoltaic system, the historical output of the distributed photovoltaic system and the hardware parameters of the reactive power compensation device.

[0007] The distribution network area is divided into zones to obtain several distribution network areas.

[0008] For each distribution network area, a corresponding distributed self-healing model is constructed according to the network parameters of the distribution network;

[0009] According to the historical load of the node, the network parameters of the distribution network, the hardware parameters of the energy storage device, the hardware parameters of the distributed photovoltaic system and the hardware parameters of the reactive power compensation device, the constraints of the distributed self-healing model are constructed.

[0010] Under the constraints, when all distributed self-healing models are converged, the node recovery power, node recovery state and line current corresponding to each distributed self-healing model are obtained.

[0011] The distribution network is adjusted according to the node recovery power, node recovery status and line current.

[0012] Furthermore, the corresponding distributed self-healing model is constructed according to the network parameters of the distribution network, including: according to the network parameters of the distribution network, using load recovery amount and network loss as evaluation indicators, constructing the following distributed self-healing model:

[0013]

[0014] Where φ is the target of the self-healing model; T is the total self-healing time of the distribution network; Π L is the collection of distribution network lines; I ij，t is the current of line ij during the self-healing period t; D is the node set of distribution network load; i is the load level of node i; a i，t is the recovery state of the load at node i during the self-healing period t; is the power restored by the load at node i during the self-healing period t; r ij is the resistance of the distribution network line ij.

[0015] Furthermore, when all distributed self-healing models are solved to converge under the constraints, the node recovery power, node recovery state and line current corresponding to each self-healing model include:

[0016] Optimization variable solving step: for each distributed self-healing model, under the constraints of the constraints, according to the current global variables and dual variables of each distributed self-healing model, the distributed self-healing model is solved to obtain the corresponding first optimization variable; wherein the first optimization variable includes node recovery power, node recovery status and line current.

[0017] Execution times judgment step: judge whether the execution times of the optimization variable solution step is less than the preset loop execution times. If so, jump to the model convergence judgment step; if not, obtain all global variables and dual variables generated after the first M executions of the model convergence judgment step, and calculate and generate the updated global variables and updated dual variables of each distributed self-healing model based on the neural network according to the output data of the distributed photovoltaic historical period, the historical load of the distribution network node, the obtained global variables and dual variables, and then jump to the optimization variable solution step.

[0018] Model convergence judgment step: for each distributed self-healing model, according to the corresponding current global variables and dual variables, based on the iterative update formula, generate updated global variables and updated dual variables; according to the global variables before update, the updated global variables and the updated dual variables, calculate the corresponding model residuals, compare the corresponding model residuals with the preset convergence accuracy, and judge whether the corresponding distributed self-healing model converges. If all distributed self-healing models converge, decompose the latest first optimization variable of each distributed self-healing model to obtain the node recovery power, node recovery status and line current; if any distributed self-healing model does not converge, jump to the optimization variable solution step.

[0019] Furthermore, the constraints of the distributed self-healing model include: distributed photovoltaic operation constraints, distribution network load constraints, energy storage charging and discharging constraints, energy storage charge constraints, reactive compensation device constraints, distribution network flow constraints and adjacent area coupling variable constraints.

[0020] The distributed photovoltaic operation constraints are specifically:

[0021]

[0022] in, is the active power transmission amount of distributed photovoltaic at node i to the distribution network at the self-healing period t; is the reactive power transmission of distributed photovoltaic power generation at node i to the distribution network during the self-healing period t; is the upper limit of the active power that can be transmitted by the distributed photovoltaic at node i during the self-healing period t; is the power generation capacity of distributed photovoltaic at the node during the self-healing period t.

[0023] The distribution network load constraints are specifically: Among them, κ i，t is the recovery factor of the load at node i during the self-healing period t; is the active power demand of the load at node i during the self-healing period t.

[0024] The energy storage charging and discharging constraints are specifically:

[0025]

[0026] in, is the charging state of the energy storage device at node i during the self-healing period t; is the discharge state of the energy storage device at node i during the self-healing period; is the charging power of the energy storage device at node i to the distribution network during the self-healing period t; is the discharge power of the energy storage device at node i to the distribution network during the self-healing period t; is the lower limit of the charging power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the charging power of the energy storage device by the distribution network at node i during the self-healing period; is the lower limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t.

[0027] The energy storage charge constraint is specifically:

[0028]

[0029] Among them, S i，t is the charge of the energy storage device at node i during the self-healing period t; S i，t-1 is the charge of the energy storage device at node i during the self-healing period t-1; is the charging efficiency of the energy storage device at node i during the self-healing period t; is the discharge efficiency of the energy storage device at node i during the self-healing period t; S i is the lower limit of the charge of the energy storage device at node i during the self-healing period t; is the upper limit of the charge of the energy storage device at node i during the self-healing period t.

[0030] The reactive power compensation device constraints are specifically: in, is the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the lower limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the upper limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t.

[0031] The power flow constraints of the distribution network are specifically:

[0032]

[0033] Among them, P i，t is the active node injection amount at node i during the self-healing period t; Q i，t is the node injection amount of reactive power at node i during the self-healing period t; L is the collection of distribution network lines; P ij，t is the active power at the head end of line ij during the self-healing period t; jk，t is the active power at the head end of line jk during the self-healing period t; Q ij，t is the reactive power at the head end of line ij during the self-healing period t; Q jk，t is the reactive power at the head end of line jk during the self-healing period t; ij，t is the current of line ij during the self-healing period t; r ij is the resistance of the distribution network line ij; x ij is the reactance of the distribution network line ij; a i，t is the recovery state of the load at node i during the self-healing period t; U i，t is the voltage at node i during the self-healing period t; η ij，t is the recovery state of line ij during the self-healing period t; J is a constant; is an auxiliary variable; U ij，t is the voltage of line ij during the self-healing period t; U i，t is the lower limit of the voltage at node i during the self-healing period; is the upper limit of the voltage at node i during the self-healing period t; is the upper limit of the current that line ij can withstand; P ij，t is the lower limit of the active power at the head end of line ij during the self-healing period t; is the upper limit of the active power at the head end of line ij during the self-healing period t; Q ij，t is the lower limit of reactive power at the head end of line ij during the self-healing period t; is the upper limit of reactive power at the head end of line ij during self-healing period t.

[0034] The adjacent area coupling variable constraint is specifically: for adjacent distribution network areas, the coupling variables exchanged between them through the same coupling branch should be the same.

[0035] Based on the above method embodiment, the present invention provides a corresponding device embodiment.

[0036] An embodiment of the present invention provides a distributed self-healing device for a power distribution network, comprising: a data acquisition module, a region division module, a model construction module, a constraint condition construction module, a model solving module and a power distribution network adjustment module.

[0037] The data acquisition module is used to obtain the historical load of the distribution network node, the hardware parameters of the energy storage device, the network parameters of the distribution network, the hardware parameters of the distributed photovoltaic, the historical output of the distributed photovoltaic and the hardware parameters of the reactive power compensation device.

[0038] The area division module is used to divide the distribution network area into zones to obtain a number of distribution network areas;

[0039] The model building module is used to build a corresponding distributed self-healing model for each distribution network area according to the network parameters of the distribution network.

[0040] The constraint condition building module is used to build the constraint conditions of the distributed self-healing model according to the historical load of the node, the network parameters of the distribution network, the hardware parameters of the energy storage device, the hardware parameters of the distributed photovoltaic system and the hardware parameters of the reactive power compensation device.

[0041] The model solving module is used to solve the node recovery power, node recovery state and line current corresponding to each distributed self-healing model when all distributed self-healing models converge under the constraint conditions;

[0042] The distribution network adjustment module is used to adjust the distribution network according to the node recovery power, the node recovery state and the line current.

[0043] Furthermore, the model building module builds a corresponding distributed self-healing model according to the network parameters of the distribution network, including: according to the network parameters of the distribution network, taking the load recovery amount and network loss as evaluation indicators, building the following distributed self-healing model:

[0044]

[0045] Where φ is the target of the self-healing model; T is the total self-healing time of the distribution network; Π L is the collection of distribution network lines; I ij，t is the current of line ij during the self-healing period t; D is the node set of distribution network load; i is the load level of node i; a i，t is the recovery state of the load at node i during the self-healing period t; is the power restored by the load at node i during the self-healing period t; r ij is the resistance of the distribution network line ij.

[0046] Furthermore, the model solving module is used to solve the node recovery power, node recovery state and line current corresponding to each distributed self-healing model when all distributed self-healing models converge under the constraint conditions, including:

[0047] Optimization variable solving step: for each distributed self-healing model, under the constraints of the constraints, according to the current global variables and dual variables of each distributed self-healing model, the distributed self-healing model is solved to obtain the corresponding first optimization variable; wherein the first optimization variable includes node recovery power, node recovery status and line current.

[0048] Execution times judgment step: judge whether the execution times of the optimization variable solution step is less than the preset loop execution times. If so, jump to the model convergence judgment step; if not, obtain all global variables and dual variables generated after the first M executions of the model convergence judgment step, and calculate and generate the updated global variables and updated dual variables of each distributed self-healing model based on the neural network according to the output data of the distributed photovoltaic historical period, the historical load of the distribution network node, the obtained global variables and dual variables, and then jump to the optimization variable solution step.

[0049] Model convergence judgment step: for each distributed self-healing model, according to the corresponding current global variables and dual variables, based on the iterative update formula, generate updated global variables and updated dual variables; according to the global variables before update, the updated global variables and the updated dual variables, calculate the corresponding model residuals, compare the corresponding model residuals with the preset convergence accuracy, and judge whether the corresponding distributed self-healing model converges. If all distributed self-healing models converge, decompose the latest first optimization variable of each distributed self-healing model to obtain the node recovery power, node recovery status and line current; if any distributed self-healing model does not converge, jump to the optimization variable solution step.

[0050] Furthermore, the constraint condition construction module includes: a distributed photovoltaic operation constraint construction unit, a distribution network load constraint construction unit, an energy storage charging and discharging constraint construction unit, an energy storage charge constraint construction unit, a reactive compensation device constraint construction unit, a distribution network flow constraint construction unit and an adjacent area coupling variable constraint construction unit.

[0051] The distributed photovoltaic operation constraint construction unit constructs the following distributed photovoltaic operation constraints:

[0052]

[0053] in, is the active power transmission amount of distributed photovoltaic at node i to the distribution network at the self-healing period t; is the reactive power transmission of distributed photovoltaic power generation at node i to the distribution network during the self-healing period t; is the upper limit of the active power that can be transmitted by the distributed photovoltaic at node i during the self-healing period t; is the power generation capacity of distributed photovoltaic at the node during the self-healing period t.

[0054] The distribution network load constraint construction unit constructs the following distribution network load constraints: Among them, κ i，t is the recovery factor of the load at node i during the self-healing period t; is the active power demand of the node load during the self-healing period t.

[0055] The energy storage charge and discharge constraint construction unit constructs the following energy storage charge and discharge constraints:

[0056]

[0057] in, is the charging state of the energy storage device at node i during the self-healing period t; is the discharge state of the energy storage device at node i during the self-healing period; is the charging power of the energy storage device at node i to the distribution network during the self-healing period t; is the discharge power of the energy storage device at node i to the distribution network during the self-healing period t; is the lower limit of the charging power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the charging power of the energy storage device by the distribution network at node i during the self-healing period; is the lower limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t.

[0058] The energy storage charge constraint construction unit constructs the following energy storage charge constraint:

[0059]

[0060] Among them, S i，t is the charge of the energy storage device at node i during the self-healing period t; S i，t-1 is the charge of the energy storage device at node i during the self-healing period t-1; is the charging efficiency of the energy storage device at node i during the self-healing period t; is the discharge efficiency of the energy storage device at node i during the self-healing period t; S i is the lower limit of the charge of the energy storage device at node i during the self-healing period t; is the upper limit of the charge of the energy storage device at node i during the self-healing period t.

[0061] The reactive power compensation device constraint construction unit constructs the following reactive power compensation device constraints: in, is the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the lower limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the upper limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t.

[0062] The power flow constraint construction unit of the distribution network constructs the following power flow constraint of the distribution network:

[0063]

[0064]

[0065] Among them, P i，t is the active node injection amount at node i during the self-healing period t; Q i，t is the node injection amount of reactive power at node i during the self-healing period t; π L is the collection of distribution network lines; P ij，t is the active power at the head end of line ij during the self-healing period t; jk，t is the active power at the head end of line jk during the self-healing period t; Q ij，t is the reactive power at the head end of line ij during the self-healing period t; Q jk，t is the reactive power at the head end of line jk during the self-healing period t; ij，t is the current of line ij during the self-healing period t; r ij is the resistance of the distribution network line ij; x ij is the reactance of the distribution network line ij; a i，t is the recovery state of the load at node i during the self-healing period t; U i，t is the voltage at node i during the self-healing period t; η ij，t is the recovery state of line ij during the self-healing period t; J is a constant; is an auxiliary variable; U ij，t is the voltage of line ij during the self-healing period t; U i，t is the lower limit of the voltage at node i during the self-healing period; is the upper limit of the voltage at node i during the self-healing period t; is the upper limit of the current that line ij can withstand; P ij，t is the lower limit of the active power at the head end of line ij during the self-healing period t; is the upper limit of the active power at the head end of line ij during the self-healing period t; Q ij，t is the lower limit of reactive power at the head end of line ij during the self-healing period t; is the upper limit of reactive power at the head end of line ij during self-healing period t.

[0066] The adjacent area coupling variable constraint construction unit constructs the adjacent area coupling variable constraint: for adjacent distribution network areas, the coupling variables exchanged between them through the same coupling branch should be the same.

[0067] Based on the above method embodiment, the present invention provides a corresponding electronic device embodiment.

[0068] An embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the distributed self-healing method for the distribution network described in any one of the above method embodiments can be implemented.

[0069] Based on the above method item embodiments, the present invention provides a corresponding storage medium item embodiment.

[0070] An embodiment of the present invention provides a storage medium having a computer program stored thereon. When the computer program is executed by a processor, the distributed self-healing method for a distribution network described in any one of the above method embodiments can be implemented.

[0071] Compared with the prior art, the present invention has the following beneficial effects:

[0072] The embodiment of the present invention provides a distributed self-healing method, device, electronic device and storage medium for a distribution network. The method partitions the distribution network area according to a preset partitioning algorithm; and constructs a distributed self-healing model for each distribution network area; for each distributed self-healing model, when all distributed self-healing models converge, the node recovery power, node recovery state and line current corresponding to each distributed self-healing model are solved under the constraints based on the constraint conditions. The distribution network is adjusted according to the node recovery power, node recovery state and line current.

[0073] The present invention divides the distribution network into regions and constructs a distributed self-healing model for each region. Each distributed self-healing model is solved to achieve distributed self-healing of the distribution network, thereby transforming centralized control into distributed control, effectively dispersing processing tasks, reducing the burden on the control center, and greatly improving the reliability of the distribution network and the timeliness of fault processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] Figure 1 It is a flow chart of a distributed self-healing method for a distribution network provided by an embodiment of the present invention.

[0075] Figure 2 It is a schematic diagram of the training structure of the long short-term memory neural network model provided by one embodiment of the present invention.

[0076] Figure 3 It is a distribution network topology diagram for simulation verification of a distributed self-healing model example system provided by an embodiment of the present invention.

[0077] Figure 4 It is a structural schematic diagram of a distributed self-healing device for a power distribution network provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0078] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0079] like Figure 1 As shown, an embodiment of the present invention provides a distributed self-healing method for a power distribution network, which at least includes the following steps:

[0080] Step S1, obtaining the historical load of the distribution network node, the hardware parameters of the energy storage device, the network parameters of the distribution network, the hardware parameters of the distributed photovoltaic, the historical output of the distributed photovoltaic and the hardware parameters of the reactive power compensation device.

[0081] Specifically, in the present invention, the data obtained from the historical load of the distribution network node includes: the active power demand of the load at the node i during the self-healing period

[0082] The data obtained from the hardware parameters of the energy storage device include: the lower limit of the charging power of the distribution network to the energy storage device at node i during the self-healing period t The upper limit of the charging power of the energy storage device by the distribution network at node i during the self-healing period t The lower limit of the power discharge of the distribution network to the energy storage device at node i during the self-healing period t The upper limit of the power discharge of the energy storage device by the distribution network at node i during the self-healing period t Charging efficiency of the energy storage device at node i during self-healing period t Discharge efficiency of the energy storage device at node i during self-healing period t The lower limit of the charge of the energy storage device at node i during the self-healing period t S i ; The upper limit of the charge of the energy storage device at node i during the self-healing period t

[0083] The data obtained from the network parameters of the distribution network include: the set of distribution network lines Π L ; Distribution network load node set Π D; Load level z of node i i ; Resistance r of distribution network line ij ij ; Reactance I of distribution network line ij ij ; The lower limit U of the voltage at node i during the self-healing period t i，t ; The upper limit of the voltage at node i during the self-healing period t The upper limit of the current that line ij can withstand The lower limit of the active power at the head end of line ij during the self-healing period t P ij，t ; The upper limit of the active power at the head end of line ij during the self-healing period t The lower limit of reactive power at the head end of line ij during the self-healing period t Q ij，t ; The upper limit of reactive power at the head end of line ij during self-healing period t

[0084] The data obtained from the hardware parameters of distributed photovoltaics include: the upper limit of the active power that can be transmitted by distributed photovoltaics at node i during the self-healing period t The power generation capacity of distributed photovoltaic at node i during the self-healing period t

[0085] The data obtained from the hardware parameters of the reactive power compensation device include: the lower limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t The upper limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t

[0086] The historical load of the above distribution network nodes, the hardware parameters of the energy storage device, the network parameters of the distribution network, the hardware parameters of the distributed photovoltaic, the hardware parameters of the reactive power compensation device, the recovery factor κ of the load at the node i at the self-healing period t i，t , constant J (in the present invention, J is 10 6 ), auxiliary variables All the known data are obtained from the power system.

[0087] The current I of the line ij during the self-healing period t mentioned in the invention is ij，t , the recovery state a of the load at node i during the self-healing period t i，t ; The power of the load restored at node i during the self-healing period t Active power transmission from distributed photovoltaic power station at node i to distribution network during self-healing period t Reactive power transmission of distributed photovoltaic power station at node i to distribution network during self-healing period t The charging state of the energy storage device at node i during the self-healing period t Discharge state of the energy storage device at node i during self-healing period t The charging power of the energy storage device to the distribution network at the node during the self-healing period t Discharge power of the energy storage device at node i to the distribution network during the self-healing period t The charge S of the energy storage device at node i during the self-healing period t i，t ; The charge S of the energy storage device at node i during the self-healing period t-1 i，t-1 ; The reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t The voltage U at node i during the self-healing period t i，t ; The recovery state η of line ij during the self-healing period t ij，t These data items are all unknown data items and are calculated in the subsequent process of solving the distributed self-healing model.

[0088] Step S2: partition the distribution network area to obtain a number of distribution network areas.

[0089] It should be noted that step S2 can partition the distribution network according to existing partitioning methods. For example, it can select a partitioning method based on electrical distance, a partitioning method based on the Newton-Raphson method, a partitioning method based on a comprehensive sensitivity matrix, and a regional attribution partitioning method.

[0090] According to the partitioning method, the distribution network area is partitioned to obtain several distribution network areas, and a distributed self-healing model will be constructed for each distribution network area later. Different distribution network areas are connected through coupling branches, so that different distribution network areas can determine and update related variables through coupling branches.

[0091] Step S3: For each distribution network area, a corresponding distributed self-healing model is constructed according to the network parameters of the distribution network.

[0092] In a preferred embodiment, the construction of a corresponding distributed self-healing model according to the network parameters of the distribution network includes: according to the network parameters of the distribution network, taking load recovery amount and network loss as evaluation indicators, constructing the following distributed self-healing model:

[0093]

[0094] Where φ is the target of the self-healing model; T is the total self-healing time of the distribution network; Π L is the collection of distribution network lines; I ij，t is the current of line ij during the self-healing period t; D is the node set of distribution network load; i is the load level of node i; ai，t is the recovery state of the load at node i during the self-healing period t; is the power restored by the load at node i during the self-healing period t; r ij is the resistance of the distribution network line ij.

[0095] By solving the current I of line ij in the self-healing period t in the distributed self-healing model ij，t , the recovery state of the load at node i during the self-healing period a i，t and the power restored by the load at node i during the self-healing period t A self-healing strategy for the distribution network can be formulated. Adjusting the distribution network according to the self-healing strategy of the distribution network can achieve isolation of the fault area and restore power supply as much as possible.

[0096] The alternating direction multiplier method is an effective algorithm for solving the distributed self-healing problem. The present invention solves the distributed self-healing model based on the alternating direction multiplier method. Specifically, when a distribution network is divided into two areas, the self-healing model is calculated and solved in the following way:

[0097]

[0098] Where x1 is the optimization variable in the first region during the self-healing process; x2 is the optimization variable in the second region during the self-healing process; f1(x1) is the distributed self-healing model in the first region; f2(x2) is the distributed self-healing model in the second region; is the penalty coefficient of the alternating direction multiplier method; is the coupling variable of the first regional coupling branch ij at the kth iteration; is the coupling variable of the first regional coupling branch ij at the k+1th iteration; is the coupling variable of the second regional coupling branch ij at the kth iteration; is the coupling variable of the second regional coupling branch ij at the k+1th iteration; is the global variable of the first regional coupling branch ij at the kth iteration; is the global variable of the second regional coupling branch ij at the kth iteration; is the dual variable of the first regional coupling branch ij at the kth iteration; is the dual variable of the second regional coupling branch ij at the kth iteration.

[0099] It should be noted here that the above formula is only described when a distribution network is divided into two areas. If the distribution network is divided into multiple areas, the information between other adjacent areas can be calculated by referring to the above formula.

[0100] Step S4: construct the constraints of the distributed self-healing model according to the historical load of the node, the network parameters of the distribution network, the hardware parameters of the energy storage device, the hardware parameters of the distributed photovoltaic system and the hardware parameters of the reactive power compensation device.

[0101] In a preferred embodiment, the constraints of the distributed self-healing model include: distributed photovoltaic operation constraints, distribution network load constraints, energy storage charging and discharging constraints, energy storage charge constraints, reactive compensation device constraints, distribution network power flow constraints and adjacent area coupling variable constraints;

[0102] The distributed photovoltaic operation constraints are specifically:

[0103]

[0104] in, is the active power transmission amount of distributed photovoltaic at node i to the distribution network at the self-healing period t; is the reactive power transmission of distributed photovoltaic power generation at node i to the distribution network during the self-healing period t; is the upper limit of the active power that can be transmitted by the distributed photovoltaic at node i during the self-healing period t; is the power generation capacity of distributed photovoltaic at the node during the self-healing period t;

[0105] The distribution network load constraints are specifically: Among them, κ i，t is the recovery factor of the load at node i during the self-healing period t; when κ i，t When κ is between 0 and 1, the load type at node i is controllable load. i，t When it is 1, the load type at node i is uncontrollable load; is the active power demand of the load at node i during the self-healing period t;

[0106] Active power demand of the load at node i during the self-healing period t It is obtained through the load of each node in the historical period. By setting the load recovery factor of different nodes, the load in the distribution network is divided into controllable load and uncontrollable load. When the load at the node is controllable, only the power required by the load can be provided for power supply; when it is an uncontrollable load, the full power required by the node load must be provided to restore power supply. The recovery factor is a preset value and can be flexibly adjusted according to actual conditions.

[0107] The load constraint of the distribution network is to constrain the active and reactive power in the power flow constraint to constrain the current I in the self-healing model. ij，t .

[0108] The energy storage charging and discharging constraints are specifically:

[0109]

[0110] in, is the charging state of the energy storage device at node i during the self-healing period t; is the discharge state of the energy storage device at node i during the self-healing period; is the charging power of the energy storage device at node i to the distribution network during the self-healing period t; is the discharge power of the energy storage device at node i to the distribution network during the self-healing period t; is the lower limit of the charging power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the charging power of the energy storage device by the distribution network at node i during the self-healing period; is the lower limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t.

[0111] Energy storage charging and discharging constraints are achieved by constraining the active power injected by nodes in the distribution network, thereby constraining the current I in the self-healing model. ij，t .

[0112] The energy storage charge constraint is specifically:

[0113]

[0114] Among them, S i，t is the charge of the energy storage device at node i during the self-healing period t; S i，t-1 is the charge of the energy storage device at node i during the self-healing period t-1; is the charging efficiency of the energy storage device at node i during the self-healing period t; is the discharge efficiency of the energy storage device at node i during the self-healing period t; S i is the lower limit of the charge of the energy storage device at node i during the self-healing period t; is the upper limit of the charge of the energy storage device at node i during the self-healing period t.

[0115] The energy storage charge constraint acts on the charging and discharging power of the energy storage by constraining the energy storage charge, and then acts on the node to inject active power, thereby constraining the line current I in the self-healing model. ij，t .

[0116] The reactive power compensation device constraints are specifically: in, is the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the lower limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the upper limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t.

[0117] The reactive power compensation device constraint is to constrain the reactive power compensation amount of the reactive power compensation device, and then constrain the reactive power injection power of the node, thereby constraining the current I in the self-healing model. ij，t .

[0118] The power flow constraints of the distribution network are specifically:

[0119]

[0120] Among them, P i，t is the active node injection amount at node i during the self-healing period t; Q i，t is the node injection amount of reactive power at node i during the self-healing period t; L is the collection of distribution network lines; P ij，t is the active power at the head end of line ij during the self-healing period t; jk，t is the active power at the head end of line jk during the self-healing period t; Q ij，t is the reactive power at the head end of line ij during the self-healing period t; Q jk，t is the reactive power at the head end of line jk during the self-healing period t; ij，t is the current of line ij during the self-healing period t; r ij is the resistance of the distribution network line ij; x ij is the reactance of the distribution network line ij; a i，t is the recovery state of the load at node i during the self-healing period t; U i，t is the voltage at node i during the self-healing period t; η ij，t is the recovery state of line ij during the self-healing period t; J is a constant; is an auxiliary variable; U ij，t is the voltage of line ij during the self-healing period t; U i，t is the lower limit of the voltage at node i during the self-healing period; is the upper limit of the voltage at node i during the self-healing period t; is the upper limit of the current that line ij can withstand; P ij，t is the lower limit of the active power at the head end of line ij during the self-healing period t; is the upper limit of the active power at the head end of line ij during the self-healing period t; Q ij，t is the lower limit of reactive power at the head end of line ij during the self-healing period t; is the upper limit of reactive power at the head end of line ij during self-healing period t.

[0121] The adjacent area coupling variable constraint is specifically: for adjacent distribution network areas, the coupling variables exchanged between them through the same coupling branch should be the same.

[0122] Step S5: Under the constraints, when all distributed self-healing models converge, the node recovery power, node recovery state, and line current corresponding to each distributed self-healing model are solved.

[0123] In order to solve the above problem, the node recovery power, node recovery state and line current corresponding to each self-healing model when all distributed self-healing models converge under the constraints are solved, including:

[0124] Optimization variable solving step: for each distributed self-healing model, under the constraints of the constraints, according to the current global variables and dual variables of each distributed self-healing model, the distributed self-healing model is solved to obtain the corresponding first optimization variable; wherein the first optimization variable includes node recovery power, node recovery status and line current.

[0125] Execution times judgment step: judge whether the execution times of the optimization variable solution step is less than the preset loop execution times. If so, jump to the model convergence judgment step; if not, obtain all global variables and dual variables generated after the previous M executions of the model convergence judgment step, and calculate and generate the updated global variables and updated dual variables of each distributed self-healing model based on the neural network according to the output data of the distributed photovoltaic historical period, the historical load of the distribution network node, the obtained global variables and dual variables, and then jump to the optimization variable solution step; wherein, the M times is the preset loop execution times. The preset loop execution times here is 5.

[0126] Model convergence judgment step: for each distributed self-healing model, according to the corresponding current global variables and dual variables, based on the iterative update formula, generate updated global variables and updated dual variables; according to the global variables before update, the updated global variables and the updated dual variables, calculate the corresponding model residuals, compare the corresponding model residuals with the preset convergence accuracy, and judge whether the corresponding distributed self-healing model converges. If all distributed self-healing models converge, decompose the latest first optimization variable of each distributed self-healing model to obtain the node recovery power, node recovery status and line current; if any distributed self-healing model does not converge, jump to the optimization variable solution step.

[0127] Specifically, when the distribution network is divided into two areas, for each distributed self-healing model, according to the corresponding current global variables and dual variables, based on the iterative update formula, the corresponding first global variable and first dual variable are generated, including:

[0128] For each distributed self-healing model, according to the corresponding current global variables and dual variables, the corresponding first coupling variables are generated through the distributed self-healing model calculation; according to the corresponding first coupling variables, the corresponding current global variables and dual variables, based on the iterative update formula, the corresponding first global variables and first dual variables are generated.

[0129] The iterative update formula is specifically:

[0130]

[0131] in, is the global variable of the first regional coupling branch ij at the k+1th iteration; is the global variable of the second regional coupling branch ij at the k+1th iteration; is the dual variable of the first regional coupling branch ij at the k+1th iteration; is the dual variable of the second regional coupling branch ij at the k+1th iteration.

[0132] Optionally, the calculating of the corresponding model residual according to the global variables before the update, the global variables after the update, and the dual variables after the update, and comparing the corresponding model residual with a preset convergence accuracy to determine whether the corresponding distributed self-healing model converges includes:

[0133] According to the global variables before update, the global variables after update and the dual variables after update, the residual of the model is calculated by the following formula:

[0134]

[0135] Among them, r k is the original residual at the kth iteration of the alternating direction multiplier method; s k Dual residual at the kth iteration of the alternating direction multiplier method.

[0136] According to the original residual and the dual residual, a larger value of the two is compared with a preset convergence precision. If the larger value of the two is less than or equal to the preset convergence precision, it is determined that the corresponding distributed self-healing model converges.

[0137] It should be noted here that different areas of the distribution network are connected through coupling branches. The dual variables and global variables are determined and updated through the relevant variables on the coupling branches of the two areas after solving their respective optimization models in the two adjacent areas. When the original residuals and dual residuals determined by the global variables and coupling variables of all coupled branches reach the convergence accuracy, the converged solution of the distributed computing of the entire distribution network is obtained at this time.

[0138] In actual operation, when the alternating direction multiplier method is used to solve the model, an initial global variable and a dual variable are required to start the algorithm. This initial value is obtained by setting the penalty coefficient to 0. When the algorithm is solved later, the penalty coefficient needs to be set manually. The present invention subsequently conducts comparative experiments on the results under different penalty coefficient settings.

[0139] like Figure 2 As shown, in a preferred embodiment, all global variables and dual variables generated after the previous M executions of the model convergence judgment step are obtained, and according to the output data of the distributed photovoltaic historical period, the historical load of the distribution network node, the obtained global variables and dual variables, based on the neural network, the updated global variables and updated dual variables of each distributed self-healing model are calculated and generated, including:

[0140] The output data of the distributed photovoltaic power generation in the historical period and the historical load of the distribution network node are divided into 24 groups of conditions according to 24 hours a day.

[0141] For each condition, a long short-term memory neural network model is trained.

[0142] It should be noted here that the long short-term memory neural network is a recursive neural network that can be used to process the regression problem of time series data. Usually, the long short-term memory neural network model is composed of multiple long short-term memory neural network units. Each unit consists of a unit, an input gate, an output gate and a forget gate. The unit is used to memorize values ​​within any time interval, and the three gates are designed to control the flow of information. In the present invention, we do not focus on the specific structure of the long short-term memory neural network unit, but only on its input and output. The structure used is as follows: Figure 2 As shown, H(m) and C(m) are the output vector (also known as the hidden state vector) and the unit state vector, respectively. H(M) is the final output for convergence value regression. The sequence length of the long short-term memory neural network model can be selected arbitrarily. In order to use less information to regress the convergence value and thus obtain better acceleration performance, the sequence length M is set to 5, that is, the parameters of the first five alternating direction multiplier method iterations are collected and then applied to regression, and the mean square error (MSE) is used as the loss function for model training.

[0143] Where M is the output data of distributed photovoltaic power generation in the historical period and the historical load of the distribution network node, which is collected by the alternating direction multiplier method for the training of the long short-term memory neural network. H M is the output value of the dual variable and global variable when training with the first M input data, X M is the Mth input global variable data value, λ M Enter the dual variable data values ​​for the Mth time.

[0144] It should be noted that when applying machine learning techniques, it is usually only necessary to normalize the input, not the output. However, the dual variables in the alternating direction multiplier method may change significantly during the iteration process. If the long short-term memory neural network model is directly trained using the unprocessed output, it will be difficult to converge. In this regard, the present invention performs a scaling process on the output data:

[0145]

[0146] In the formula, λ s is the dual variable of the scaled output in the LSTM neural network model; λ is the original output of the dual variable in the LSTM neural network model. v is the scaling factor, which is 1000 here.

[0147] When the trained model is used to predict the converged value, the predicted global variable X and the scaled dual variable λ are used in subsequent alternating direction multiplier iterations.

[0148] The long short-term memory neural network is used here because the dual variables and global variables have certain regularity in the iterative update process of distributed computing. Here, we can set the collection of the values ​​of the previous M times for training to predict the values ​​of global variables and dual variables in the subsequent algorithm iterations, thereby accelerating the convergence of the algorithm. The training is for all global variables and dual variables to predict the iterative values ​​of the subsequent updates and substitute them into the update of the alternating direction multiplier method. Figure 2 Provides a training structure for the long short-term memory neural network model.

[0149] In actual operation, the present invention builds a model based on the Matlab platform, and then solves the model through the Gurobi solver to output the optimal distribution network self-healing strategy.

[0150] Step S6: adjusting the distribution network according to the node recovery power, the node recovery status and the line current.

[0151] The distribution network self-healing strategy can be obtained based on the solved node recovery power, node recovery status and line current, and the distribution network can be adjusted according to the distribution network self-healing strategy.

[0152] In order to more clearly illustrate the technical solution and advantages of the present invention, the present invention performs an example simulation result analysis in a preferred embodiment. Figure 3 It is a distribution network topology diagram used for simulation verification of the distributed self-healing model example system, wherein the algorithms selected in the experiment are specifically: Algorithm 1 is a centralized distribution network self-healing optimization algorithm; Algorithm 2 is a distributed self-healing method for distribution network based on ADMM algorithm; Algorithm 3 is a distributed self-healing method for distribution network provided by the present invention.

[0153] Among them, the result data of the distributed self-healing algorithm under different initial penalty coefficients are as follows:

[0154]

[0155] It can be seen from the table that Algorithm 3 has the fastest convergence speed under different initial penalty parameters, takes the least time, and has higher convergence accuracy, which proves the effectiveness of the distributed self-healing method for distribution network provided by the present invention.

[0156] Based on the above method embodiment, the present invention provides a corresponding device embodiment.

[0157] like Figure 4 As shown, an embodiment of the present invention provides a distributed self-healing device for a distribution network, including: a data acquisition module 101, a region division module 102, a model construction module 103, a constraint condition construction module 104, a model solving module 105 and a distribution network adjustment module 106.

[0158] The data acquisition module 101 is used to obtain the historical load of the distribution network node, the hardware parameters of the energy storage device, the network parameters of the distribution network, the hardware parameters of the distributed photovoltaic, the historical output of the distributed photovoltaic and the hardware parameters of the reactive power compensation device.

[0159] The area division module 102 is used to divide the distribution network area into zones to obtain a plurality of distribution network areas.

[0160] The model building module 103 is used to build a corresponding distributed self-healing model for each distribution network area according to the network parameters of the distribution network.

[0161] The constraint condition building module 104 is used to build the constraint conditions of the distributed self-healing model according to the historical load of the node, the network parameters of the distribution network, the hardware parameters of the energy storage device, the hardware parameters of the distributed photovoltaic system and the hardware parameters of the reactive power compensation device.

[0162] The model solving module 105 is used to solve the node recovery power, node recovery state and line current corresponding to each distributed self-healing model when all distributed self-healing models converge under the constraint conditions.

[0163] The distribution network adjustment module 106 is used to adjust the distribution network according to the node recovery power, the node recovery status and the line current.

[0164] In an optional embodiment, the model building module 103 builds a corresponding distributed self-healing model according to the network parameters of the distribution network, including: according to the network parameters of the distribution network, using the load recovery amount and the network loss as evaluation indicators, building the following distributed self-healing model:

[0165]

[0166] Where φ is the target of the self-healing model; T is the total self-healing time of the distribution network; Π L is the collection of distribution network lines; I ij，t is the current of line ij during the self-healing period t; D is the node set of distribution network load; i is the load level of node i; a i，t is the recovery state of the load at node i during the self-healing period t; is the power restored by the load at node i during the self-healing period t; r ij is the resistance of the distribution network line ij.

[0167] In a preferred embodiment, the model solving module 105 is used to solve the node recovery power, node recovery state and line current corresponding to each distributed self-healing model when all distributed self-healing models converge under the constraint conditions, including:

[0168] Optimization variable solving step: for each distributed self-healing model, under the constraints of the constraints, according to the current global variables and dual variables of each distributed self-healing model, the distributed self-healing model is solved to obtain the corresponding first optimization variable; wherein the first optimization variable includes node recovery power, node recovery status and line current.

[0169] Execution times judgment step: judge whether the execution times of the optimization variable solution step is less than the preset loop execution times. If so, jump to the model convergence judgment step; if not, obtain all global variables and dual variables generated after the first M executions of the model convergence judgment step, and calculate and generate the updated global variables and updated dual variables of each distributed self-healing model based on the neural network according to the output data of the distributed photovoltaic historical period, the historical load of the distribution network node, the obtained global variables and dual variables, and then jump to the optimization variable solution step.

[0170] Model convergence judgment step: for each distributed self-healing model, according to the corresponding current global variables and dual variables, based on the iterative update formula, generate updated global variables and updated dual variables; according to the global variables before update, the updated global variables and the updated dual variables, calculate the corresponding model residuals, compare the corresponding model residuals with the preset convergence accuracy, and judge whether the corresponding distributed self-healing model converges. If all distributed self-healing models converge, decompose the latest first optimization variable of each distributed self-healing model to obtain the node recovery power, node recovery status and line current; if any distributed self-healing model does not converge, jump to the optimization variable solution step.

[0171] Optionally, the constraint condition construction module 104 includes: a distributed photovoltaic operation constraint construction unit 1041, a distribution network load constraint construction unit 1042, an energy storage charging and discharging constraint construction unit 1043, an energy storage charge constraint construction unit 1044, a reactive compensation device constraint construction unit 1045, a distribution network flow constraint construction unit 1046 and an adjacent area coupling variable constraint construction unit 1047.

[0172] The distributed photovoltaic operation constraint construction unit 1041 constructs the following distributed photovoltaic operation constraints:

[0173]

[0174] in, is the active power transmission amount of distributed photovoltaic at node i to the distribution network at the self-healing period t; is the reactive power transmission of distributed photovoltaic power generation at node i to the distribution network during the self-healing period t; is the upper limit of the active power that can be transmitted by the distributed photovoltaic at node i during the self-healing period t; is the power generation capacity of distributed photovoltaic at the node during the self-healing period t.

[0175] The distribution network load constraint constructing unit 1042 constructs the following distribution network load constraints: Among them, κ i，t is the recovery factor of the load at node i during the self-healing period t; is the active power demand of the load at node i during the self-healing period t.

[0176] The energy storage charge and discharge constraint constructing unit 1043 constructs the following energy storage charge and discharge constraints:

[0177]

[0178] in, is the charging state of the energy storage device at node i during the self-healing period t; is the discharge state of the energy storage device at node i during the self-healing period; is the charging power of the energy storage device at node i to the distribution network during the self-healing period t; is the discharge power of the energy storage device at node i to the distribution network during the self-healing period t; is the lower limit of the charging power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the charging power of the energy storage device by the distribution network at node i during the self-healing period; is the lower limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t.

[0179] The energy storage charge constraint constructing unit 1044 constructs the following energy storage charge constraint:

[0180]

[0181] Among them, S i，t is the charge of the energy storage device at node i during the self-healing period t; S i，t-1 is the charge of the energy storage device at node i during the self-healing period t-1; is the charging efficiency of the energy storage device at node i during the self-healing period t; is the discharge efficiency of the energy storage device at node i during the self-healing period t; S i is the lower limit of the charge of the energy storage device at node i during the self-healing period t; is the upper limit of the charge of the energy storage device at node i during the self-healing period t.

[0182] The reactive power compensation device constraint construction unit 1045 constructs the following reactive power compensation device constraints: in, is the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the lower limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the upper limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t.

[0183] The power flow constraint constructing unit 1046 of the distribution network constructs the following power flow constraint of the distribution network:

[0184]

[0185]

[0186] Among them, Pi，t is the active node injection amount at node i during the self-healing period t; Q i，t is the node injection amount of reactive power at node i during the self-healing period t; L is the collection of distribution network lines; P ij，t is the active power at the head end of line ij during the self-healing period t; jk，t is the active power at the head end of line jk during the self-healing period t; Q ij，t is the reactive power at the head end of line ij during the self-healing period t; Q jk，t is the reactive power at the head end of line jk during the self-healing period t; ij，t is the current of line ij during the self-healing period t; r ij is the resistance of the distribution network line ij; x ij is the reactance of the distribution network line ij; a i，t is the recovery state of the load at node i during the self-healing period t; U i，t is the voltage at node i during the self-healing period t; η ij，t is the recovery state of line ij during the self-healing period t; J is a constant; is an auxiliary variable; U ij，t is the voltage of line ij during the self-healing period t; U i，t is the lower limit of the voltage at node i during the self-healing period; is the upper limit of the voltage at node i during the self-healing period t; is the upper limit of the current that line ij can withstand; P ij，t is the lower limit of the active power at the head end of line ij during the self-healing period t; is the upper limit of the active power at the head end of line ij during the self-healing period t; Q ij，t is the lower limit of reactive power at the head end of line ij during the self-healing period t; is the upper limit of reactive power at the head end of line ij during self-healing period t.

[0187] The adjacent area coupling variable constraint construction unit 1047 constructs the adjacent area coupling variable constraint: for adjacent distribution network areas, the coupling variables exchanged between them through the same coupling branch should be the same.

[0188] It should be noted that the embodiments of the device described above correspond to the above embodiments of the present invention, and can implement any of the methods described above in the present invention. In addition, the embodiments of the above-mentioned device are merely schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which can be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art can understand and implement it without paying any creative work.

[0189] Based on the above method embodiment of the present invention, a corresponding electronic device embodiment is provided.

[0190] An embodiment of the present invention provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the distributed self-healing method for a distribution network described in any one of the present invention is implemented, or when the processor executes the computer program, the functions of the modules in the above-mentioned device embodiments are implemented.

[0191] Exemplarily, the computer program may be divided into one or more modules, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the terminal device.

[0192] The terminal device may be a computing device such as a desktop computer, a notebook, a palmtop computer, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0193] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the terminal device, and uses various interfaces and lines to connect various parts of the entire terminal device.

[0194] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the terminal device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application required for a function, etc.; the data storage area can store data created according to the use of the mobile phone, etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0195] Based on the above method embodiment, the present invention provides a storage medium embodiment;

[0196] Another embodiment of the present invention provides a storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the storage medium is located is controlled to execute any one of the above-mentioned distributed self-healing methods for distribution networks of the present invention.

[0197] The storage medium is a computer-readable storage medium, and the computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0198] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification and the features of different embodiments or examples, unless they are contradictory.

[0199] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A distributed self-healing method for a distribution network, characterized in that: include: Obtain the historical load of the distribution network nodes, the hardware parameters of the energy storage device, the network parameters of the distribution network, the hardware parameters of the distributed photovoltaic system, the historical output of the distributed photovoltaic system, and the hardware parameters of the reactive power compensation device; Divide the distribution network area into zones to obtain several distribution network areas; For each distribution network area, a corresponding distributed self-healing model is constructed according to the network parameters of the distribution network; According to the historical load of the node, the network parameters of the distribution network, the hardware parameters of the energy storage device, the hardware parameters of the distributed photovoltaic system and the hardware parameters of the reactive power compensation device, the constraints of the distributed self-healing model are constructed; Under the constraints, when all distributed self-healing models converge, the node recovery power, node recovery state and line current corresponding to each distributed self-healing model are solved; The distribution network is adjusted according to the node recovery power, node recovery status and line current.

2. The distributed self-healing method for power distribution network according to claim 1, characterized in that: The method of constructing a corresponding distributed self-healing model according to the network parameters of the distribution network includes: constructing the following distributed self-healing model according to the network parameters of the distribution network, taking load recovery amount and network loss as evaluation indicators: Where φ is the target of the self-healing model; T is the total self-healing time of the distribution network; Π L is the collection of distribution network lines; I ij， t is the current of line ij during the self-healing period t; D is the node set of distribution network load; i is the load level of node i; a i，t is the recovery state of the load at node i during the self-healing period t; is the power restored by the load at node i during the self-healing period t; r ij is the resistance of the distribution network line ij.

3. The distributed self-healing method for power distribution network according to claim 2, characterized in that: When all distributed self-healing models are solved to converge under the constraints, the node recovery power, node recovery state and line current corresponding to each self-healing model include: Optimization variable solving step: for each distributed self-healing model, under the constraints of the constraints, according to the current global variables and dual variables of each distributed self-healing model, solve the distributed self-healing model to obtain the corresponding first optimization variable; wherein the first optimization variable includes node recovery power, node recovery state and line current; Execution times judgment step: judge whether the execution times of the optimization variable solution step is less than the preset loop execution times. If so, jump to the model convergence judgment step; if not, obtain all global variables and dual variables generated after the previous M executions of the model convergence judgment step, and calculate and generate the updated global variables and updated dual variables of each distributed self-healing model based on the neural network according to the output data of the distributed photovoltaic historical period, the historical load of the distribution network node, the obtained global variables and dual variables, and then jump to the optimization variable solution step; Model convergence judgment step: for each distributed self-healing model, according to the corresponding current global variables and dual variables, based on the iterative update formula, generate updated global variables and updated dual variables; according to the global variables before update, the updated global variables and the updated dual variables, calculate the corresponding model residuals, compare the corresponding model residuals with the preset convergence accuracy, and judge whether the corresponding distributed self-healing model converges. If all distributed self-healing models converge, decompose the latest first optimization variable of each distributed self-healing model to obtain the node recovery power, node recovery status and line current; if any distributed self-healing model does not converge, jump to the optimization variable solution step.

4. The distributed self-healing method for power distribution network according to claim 3, characterized in that: The constraints of the distributed self-healing model include: distributed photovoltaic operation constraints, distribution network load constraints, energy storage charging and discharging constraints, energy storage charge constraints, reactive power compensation device constraints, distribution network power flow constraints and adjacent area coupling variable constraints; The distributed photovoltaic operation constraints are specifically: in, is the active power transmission amount of distributed photovoltaic at node i to the distribution network at the self-healing period t; is the reactive power transmission of distributed photovoltaic power generation at node i to the distribution network during the self-healing period t; is the upper limit of the active power that can be transmitted by the distributed photovoltaic at node i during the self-healing period t; is the power generation capacity of distributed photovoltaic at node i during the self-healing period t; The distribution network load constraints are specifically: Among them, κ i，t is the recovery factor of the load at node i during the self-healing period t; is the active power demand of the load at node i during the self-healing period t; The energy storage charging and discharging constraints are specifically: in, is the charging state of the energy storage device at node i during the self-healing period t; is the discharge state of the energy storage device at node i during the self-healing period t; is the charging power of the energy storage device at node i to the distribution network during the self-healing period t; is the discharge power of the energy storage device at node i to the distribution network during the self-healing period t; is the lower limit of the charging power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the charging power of the energy storage device by the distribution network at node i during the self-healing period t; is the lower limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; The energy storage charge constraint is specifically: Among them, S i，t is the charge of the energy storage device at node i during the self-healing period t; S i，t-1 is the charge of the energy storage device at node i during the self-healing period t-1; is the charging efficiency of the energy storage device at node i during the self-healing period t; is the discharge efficiency of the energy storage device at node i during the self-healing period t; S i is the lower limit of the charge of the energy storage device at node i during the self-healing period t; is the upper limit of the charge of the energy storage device at node i during the self-healing period t; The reactive power compensation device constraints are specifically: in, is the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the lower limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the upper limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; The power flow constraints of the distribution network are specifically: Among them, P i，t is the active node injection amount at node i during the self-healing period t; Q i，t is the node injection amount of reactive power at node i during the self-healing period t; L is the collection of distribution network lines; P ij，t is the active power at the head end of line ij during the self-healing period t; jk，t is the active power at the head end of line jk during the self-healing period t; Q ij，t is the reactive power at the head end of line ij during the self-healing period t; Q jk，t is the reactive power at the head end of line jk during the self-healing period t; ij，t is the current of line ij during the self-healing period t; r ij is the resistance of the distribution network line ij; x ij is the reactance of the distribution network line ij; a i，t is the recovery state of the load at node i during the self-healing period t; U i，t is the voltage at node i during the self-healing period t; η ij，t is the recovery state of line ij during the self-healing period t; J is a constant; is an auxiliary variable; U ij，t is the voltage of line ij during the self-healing period t; U i，t is the lower limit of the voltage at node i during the self-healing period t; is the upper limit of the voltage at node i during the self-healing period t; is the upper limit of the current that line ij can withstand; P ij，t is the lower limit of the active power at the head end of line ij during the self-healing period t; is the upper limit of the active power at the head end of line ij during the self-healing period t; Q ij，t is the lower limit of reactive power at the head end of line ij during the self-healing period t; is the upper limit of reactive power at the head end of line ij during self-healing period t; The adjacent area coupling variable constraint is specifically: for adjacent distribution network areas, the coupling variables exchanged between them through the same coupling branch should be the same.

5. A distributed self-healing device for a distribution network, characterized in that: include: Data acquisition module, area division module, model construction module, constraint condition construction module, model solving module and distribution network adjustment module; The data acquisition module is used to obtain the historical load of the distribution network node, the hardware parameters of the energy storage device, the network parameters of the distribution network, the hardware parameters of the distributed photovoltaic, the historical output of the distributed photovoltaic and the hardware parameters of the reactive power compensation device; The area division module is used to divide the distribution network area into zones to obtain a number of distribution network areas; The model building module is used to build a corresponding distributed self-healing model for each distribution network area according to the network parameters of the distribution network; The constraint condition building module is used to build the constraint conditions of the distributed self-healing model according to the historical load of the node, the network parameters of the distribution network, the hardware parameters of the energy storage device, the hardware parameters of the distributed photovoltaic system and the hardware parameters of the reactive power compensation device; The model solving module is used to solve the node recovery power, node recovery state and line current corresponding to each distributed self-healing model when all distributed self-healing models converge under the constraint conditions; The distribution network adjustment module is used to adjust the distribution network according to the node recovery power, the node recovery state and the line current.

6. The distributed self-healing device for power distribution network according to claim 5, characterized in that: The model building module builds the corresponding distributed self-healing model according to the network parameters of the distribution network, including: according to the network parameters of the distribution network, taking the load recovery amount and network loss as evaluation indicators, building the following distributed self-healing model: Where φ is the target of the self-healing model; T is the total self-healing time of the distribution network; Π L is the collection of distribution network lines; I ij，t is the current of line ij during the self-healing period t; D is the node set of distribution network load; i is the load level of node i; a i，t is the recovery state of the load at node i during the self-healing period t; is the power restored by the load at node i during the self-healing period t; r ij is the resistance of the distribution network line ij.

7. The distributed self-healing device for power distribution network according to claim 5, characterized in that: The model solving module is used to solve the node recovery power, node recovery state and line current corresponding to each distributed self-healing model when all distributed self-healing models converge under the constraint conditions, including: Optimization variable solving step: for each distributed self-healing model, under the constraints of the constraints, according to the current global variables and dual variables of each distributed self-healing model, solve the distributed self-healing model to obtain the corresponding first optimization variable; wherein the first optimization variable includes node recovery power, node recovery state and line current; Execution times judgment step: judge whether the execution times of the optimization variable solution step is less than the preset loop execution times. If so, jump to the model convergence judgment step; if not, obtain all global variables and dual variables generated after the previous M executions of the model convergence judgment step, and calculate and generate the updated global variables and updated dual variables of each distributed self-healing model based on the neural network according to the output data of the distributed photovoltaic historical period, the historical load of the distribution network node, the obtained global variables and dual variables, and then jump to the optimization variable solution step; Model convergence judgment step: for each distributed self-healing model, according to the corresponding current global variables and dual variables, based on the iterative update formula, generate updated global variables and updated dual variables; according to the global variables before update, the updated global variables and the updated dual variables, calculate the corresponding model residuals, compare the corresponding model residuals with the preset convergence accuracy, and judge whether the corresponding distributed self-healing model converges. If all distributed self-healing models converge, decompose the latest first optimization variable of each distributed self-healing model to obtain the node recovery power, node recovery status and line current; if any distributed self-healing model does not converge, jump to the optimization variable solution step.

8. The distributed self-healing device for power distribution network according to claim 5, characterized in that: The constraint condition building module includes: a distributed photovoltaic operation constraint building unit, a distribution network load constraint building unit, an energy storage charging and discharging constraint building unit, an energy storage charge constraint building unit, a reactive compensation device constraint building unit, a distribution network flow constraint building unit and an adjacent region coupling variable constraint building unit; The distributed photovoltaic operation constraint construction unit constructs the following distributed photovoltaic operation constraints: in, is the active power transmission amount of distributed photovoltaic at node i to the distribution network at the self-healing period t; is the reactive power transmission of distributed photovoltaic power generation at node i to the distribution network during the self-healing period t; is the upper limit of the active power that can be transmitted by the distributed photovoltaic at node i during the self-healing period t; is the power generation capacity of distributed photovoltaic at node i during the self-healing period t; The distribution network load constraint construction unit constructs the following distribution network load constraints: Among them, κ i，t is the recovery factor of the load at node i during the self-healing period t; is the active power demand of the load at node i during the self-healing period t; The energy storage charge and discharge constraint construction unit constructs the following energy storage charge and discharge constraints: in, is the charging state of the energy storage device at node i during the self-healing period t; is the discharge state of the energy storage device at node i during the self-healing period t; is the charging power of the energy storage device at node i to the distribution network during the self-healing period t; is the discharge power of the energy storage device at node i to the distribution network during the self-healing period t; is the lower limit of the charging power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the charging power of the energy storage device by the distribution network at node i during the self-healing period t; is the lower limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; is the upper limit of the discharge power of the distribution network to the energy storage device at node i during the self-healing period t; The energy storage charge constraint construction unit constructs the following energy storage charge constraint: Among them, S i，t is the charge of the energy storage device at node i during the self-healing period t; S i，t-1 is the charge of the energy storage device at node i during the self-healing period t-1; is the charging efficiency of the energy storage device at node i during the self-healing period t; is the discharge efficiency of the energy storage device at node i during the self-healing period t; S i is the lower limit of the charge of the energy storage device at node i during the self-healing period t; is the upper limit of the charge of the energy storage device at node i during the self-healing period t; The reactive power compensation device constraint construction unit constructs the following reactive power compensation device constraints: in, is the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the lower limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; is the upper limit of the reactive power transmission amount provided by the reactive power compensation device at node i to the distribution network during the self-healing period t; The power flow constraint construction unit of the distribution network constructs the following power flow constraint of the distribution network: Among them, P i，t is the active node injection amount at node i during the self-healing period t; Q i，t is the node injection amount of reactive power at node i during the self-healing period t; L is the collection of distribution network lines; P ij，t is the active power at the head end of line ij during the self-healing period t; jk，t is the active power at the head end of line jk during the self-healing period t; Q ij，t is the reactive power at the head end of line ij during the self-healing period t; Q jk，t is the reactive power at the head end of line jk during the self-healing period t; ij，t is the current of line ij during the self-healing period t; r ij is the resistance of the distribution network line ij; x ij is the reactance of the distribution network line ij; a i，t is the recovery state of the load at node i during the self-healing period t; U i，t is the voltage at node i during the self-healing period t; η ij，t is the recovery state of line ij during the self-healing period t; J is a constant; is an auxiliary variable; U ij，t is the voltage of line ij during the self-healing period t; U i，t is the lower limit of the voltage at node i during the self-healing period t; is the upper limit of the voltage at node i during the self-healing period t; is the upper limit of the current that line ij can withstand; P ij，t is the lower limit of the active power at the head end of line ij during the self-healing period t; is the upper limit of the active power at the head end of line ij during the self-healing period t; Q ij，t is the lower limit of reactive power at the head end of line ij during the self-healing period t; is the upper limit of reactive power at the head end of line ij during self-healing period t; The adjacent area coupling variable constraint construction unit constructs the adjacent area coupling variable constraint: for adjacent distribution network areas, the coupling variables exchanged between them through the same coupling branch should be the same.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the distributed self-healing method for the distribution network described in any one of claims 1 to 4 can be implemented.

10. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the distributed self-healing method for a distribution network described in any one of claims 1 to 4 can be implemented.

Images