Power distribution cyber-physical system post-disaster partition autonomous recovery method

Abstract

This invention discloses a post-disaster regional autonomous recovery method for a power distribution cyber-physical system. Specifically, the power distribution system is divided into multiple post-disaster autonomous recovery zones, with one and only one intelligent terminal in each zone. Each intelligent terminal identifies the power loss load within its zone and predicts the output power of distributed photovoltaic systems within that zone for a period after the fault. Based on the identified power loss load and the predicted output power of the distributed photovoltaic systems, the intelligent terminal sets constraints and an objective function, obtains candidate recovery schemes based on the constraints, and selects the final recovery scheme from among the candidate schemes based on the objective function. This invention can effectively support rapid post-disaster power restoration by considering power, communication, and computing resources when both power and communication networks are simultaneously affected by a disaster, thereby achieving optimal zone division and autonomous generation of power supply recovery schemes for each zone.

Description

Technical Field

[0001] This invention belongs to the field of power grid technology, and in particular relates to a method for autonomous regional recovery of a power distribution cyber-physical system after a disaster. Background Technology

[0002] With the increasing frequency of extreme natural disasters, the risk of damage to power distribution systems is constantly rising, severely threatening power supply reliability. Simultaneously, disasters often damage some power communication nodes and links, making it difficult to transmit real-time operational data in a timely manner, exacerbating the difficulty of post-disaster recovery. Against this backdrop, methods for rapid post-disaster recovery of power distribution cyber-physical systems have received increasing attention. In post-disaster scenarios, some distributed power sources can still operate normally. If the sensing and control functions of intelligent power distribution terminals can be rationally utilized and combined to achieve autonomous recovery in localized areas, it will help shorten the scope and duration of power outages and reduce power loss.

[0003] Current disaster recovery methods primarily employ a centralized control model, where the distribution network master station collects network-wide operational data, analyzes and calculates it to generate recovery plans, and then distributes these plans to each terminal for execution. However, in situations where disasters cause communication failures, the master station struggles to obtain complete measurement information, leading to difficulties in decision-making for recovery plans. Furthermore, existing regional recovery methods often divide regions based on power dimensions, neglecting communication link availability and the distribution of computing resources, resulting in some areas being unable to achieve independent recovery. Therefore, researching autonomous regional recovery methods for distribution cyber-physical systems after disasters is of great significance, especially under conditions of incomplete communication and dispersed computing power. Summary of the Invention

[0004] Purpose of the invention: In order to solve the problems existing in the prior art, the present invention provides a method for autonomous recovery of power distribution cyber-physical systems after disasters by partitioning.

[0005] Technical Solution: This invention discloses a method for autonomous regional recovery of a power distribution cyber-physical system after a disaster, specifically including the following steps:

[0006] The power distribution system is divided into multiple post-disaster autonomous recovery zones, and each post-disaster autonomous recovery zone has one and only one smart terminal.

[0007] Each smart terminal identifies the power loss load within its self-recovery area after a disaster and predicts the output power of distributed photovoltaic power within that area for a period of time after the fault.

[0008] The intelligent terminal sets constraints and objective functions based on the identified power loss load and the predicted output power of distributed photovoltaic power. Based on the constraints, candidate recovery schemes are obtained, and the final recovery scheme is selected from the candidate recovery schemes based on the objective function.

[0009] Furthermore, the power distribution system is divided into multiple self-recovery zones after a disaster, specifically:

[0010] Step A: Take each smart terminal and measurement device in the system as nodes, and the communication links between nodes as edges;

[0011] Step B: Based on the number of smart terminals n, divide the system into n post-disaster autonomous recovery zones, and use the smart terminals as temporary boundary nodes in the corresponding post-disaster autonomous recovery zones;

[0012] Step C: The boundary node sends an expansion message to neighboring nodes located outside the corresponding autonomous recovery zone via the adjacency communication link. If the neighboring node has not yet joined any autonomous recovery zone, the boundary node is added to the autonomous recovery zone and the boundary node is updated.

[0013] Step D: Repeat step C until every node has been added to the post-disaster autonomous recovery zone.

[0014] Furthermore, in step C, if a node has not joined any post-disaster autonomous recovery zone and receives expansion messages from two or more boundary nodes, then the cumulative communication latency in each expansion message is compared, and the node is joined to the post-disaster autonomous recovery zone corresponding to the minimum cumulative communication latency.

[0015] Furthermore, in step C, updating the boundary node specifically means: if all the neighboring communication links of a certain node in the post-disaster autonomous recovery area point to neighboring nodes that belong to this post-disaster autonomous recovery area, then the node is considered not a boundary node; otherwise, the node is considered a boundary node.

[0016] Furthermore, the intelligent terminal identifies the fault type of its post-disaster autonomous recovery area by analyzing measurement data within that area. Specifically, the intelligent terminal performs cross-validation on the measurement data within the post-disaster autonomous recovery area according to the following power flow equation, the expression of which is shown below:

[0017] ;

[0018] Where i and j represent two adjacent busbars, and V represents voltage. , , and The values ​​represent the resistance, reactance, active power, and reactive power of the feeder segment between the two busbars.

[0019] If the equation holds true, then the load with a power measurement value of zero connected to the bus is considered a power failure load, and the fault type is identified as a power outage; otherwise, the fault type is identified as a communication fault.

[0020] Furthermore, the output power of distributed photovoltaic power in the disaster-prone autonomous recovery area is predicted. Specifically, the intelligent terminal calculates the solar altitude angle based on the time data at the time of the fault and estimates the change of the solar altitude angle in the future period after the fault, thereby estimating the power prediction value of distributed photovoltaic power in the future period.

[0021] Furthermore, the projected power output of distributed photovoltaic systems in post-disaster self-recovery areas over the next period of time. The expression is as follows:

[0022] ;

[0023] Where t represents a future moment. The moment the failure occurred. For distributed photovoltaic in Power at any given moment; This represents the photovoltaic power generation efficiency of a distributed photovoltaic system at time t. The expression is as follows:

[0024]

[0025] in, and Let denot t represent the solar altitude angle and azimuth angle at time t, respectively. and These represent the tilt angle and azimuth angle of the photovoltaic panel, respectively.

[0026] Furthermore, the constraints include: upper and lower limits of bus voltage, upper limit of line power, and power flow constraints.

[0027] The power flow constraint is specifically as follows:

[0028]

[0029] Where t represents a future moment after the fault. and Indicates busbar The active and reactive power injected by the connected load or distributed photovoltaic system at time t; if the bus... If the load is connected, then , and This refers to the active and reactive power requirements of the load, and the reactive power requirement; if the bus... If it is a distributed photovoltaic system, then , Indicates busbar The predicted output power of the connected distributed photovoltaic system at time t. Indicates busbar The rated capacity of the inverter connected to the distributed photovoltaic system; busbar The voltage amplitude at time t, busbar and via feeder section and busbar The set of directly connected busbars; For the bus at time t and The phase angle difference between them busbar and The open / closed state of the switch on the feeder segment; if open, If closed, ; and For intermediate parameters, the expression is:

[0030] ;

[0031] Represents a set busbar busbar, and busbars and Conductivity and susceptance of the feeder segments;

[0032] The bus voltage upper and lower limit constraints and the line power upper limit constraints are as follows:

[0033] ;

[0034] in, and These are the lower and upper limits of the bus voltage, respectively. The apparent power limit for the feeder segment. and The expression is:

[0035] .

[0036] Furthermore, the objective function is constructed to minimize the power loss load:

[0037] ;

[0038] in, For the collection of loads in the post-disaster autonomous recovery area, For the first One load, In the recovery plan Power reduction for each load.

[0039] Furthermore, the final recovery plan is as follows:

[0040] The electrical feasibility of each restoration scheme is verified using a DC power flow calculation method, and restoration schemes that do not meet the constraints are filtered out. Then, a mixed-integer linear programming model is constructed based on the LinDistFlow approximation equation, and the optimal solution of this model is solved. A neighborhood search is performed on the optimal solution to obtain multiple candidate restoration schemes. The objective function value corresponding to each candidate restoration scheme is calculated to obtain the final restoration scheme. The neighborhood search of the optimal solution specifically involves changing the switch states in the optimal solution, and the total number of switches changed does not exceed a preset number.

[0041] Beneficial Effects: This invention addresses the problem that existing post-disaster recovery methods for power distribution systems rely excessively on the master station and are difficult to apply under communication failure conditions. It proposes a post-disaster regional autonomous recovery method for power distribution cyber-physical systems. This method can, when both power and communication networks are simultaneously affected by a disaster, take into account power, communication, and computing resources, achieving optimal regional division and autonomous generation of power supply recovery plans for each region, thereby effectively supporting rapid power supply restoration after a disaster. Attached Figure Description

[0042] Figure 1 This is a flowchart of the post-disaster regional autonomous recovery method for the power distribution information physical system according to an embodiment of the present invention;

[0043] Figure 2 This is a diagram of the power distribution information physical system structure used to illustrate the post-disaster regional autonomous recovery method, as described in an embodiment of the present invention.

[0044] Figure 3 This is a communication link diagram between various nodes of the communication network in this embodiment of the invention;

[0045] Figure 4 This is a diagram illustrating the post-disaster autonomous recovery area division scheme as described in an embodiment of the present invention;

[0046] Figure 5 This refers to the optimal recovery plan formulated by each smart terminal for its respective region as described in the embodiments of the present invention.

[0047] Explanation of the reference numerals in the attached diagram: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, and 11 are all communication nodes. Detailed Implementation

[0048] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0049] like Figure 1As shown, the post-disaster regional autonomous recovery method for a power distribution information physical system provided by the present invention includes the following steps:

[0050] S1: Based on the location distribution of smart terminals in the power distribution information physical system and the status of each communication link after the disaster, the power distribution system is divided into multiple post-disaster autonomous recovery areas, with one and only one smart terminal in each area.

[0051] Based on the location distribution of intelligent terminals in the power distribution cyber-physical system and the status of various communication links after a disaster, the power distribution system is divided into multiple post-disaster autonomous recovery zones, specifically including:

[0052] S1.1: The various intelligent terminals and measurement devices in the system are used as nodes of the graph, and the communication links between the nodes are used as edges of the graph.

[0053] S1.2: Perform region division initialization, establish autonomous recovery regions with the same number of smart terminals in the system. Each region contains one and only one smart terminal node. The smart terminal node in each region is the only core node of that region and also temporarily serves as the boundary node of that region.

[0054] S1.3: Each region boundary node sends an expansion message to its neighboring nodes located outside the region through the adjacency communication link. If the neighboring node has not yet joined any region, then the neighboring node is included in this region.

[0055] S1.4: Update the boundary nodes of the region;

[0056] S1.5: Repeat S1.3 and S1.4 until all external neighbor nodes of all boundary nodes in each region have joined other regions, forming the final partitioning scheme of autonomous recovery regions.

[0057] Step S1.3 specifically includes:

[0058] The expansion message sent by the regional boundary node to the neighboring node outside the region through the adjacency communication link consists of the number of the regional core intelligent terminal and the cumulative communication delay from the regional core intelligent terminal node to the current node. If a node receives expansion messages from multiple regions at the same time, it compares the cumulative communication delay in each message and joins the region with the shortest cumulative communication delay.

[0059] Step S1.4 specifically includes:

[0060] Each node within the region checks whether all neighboring nodes pointed to by its adjacent communication links belong to this region; if so, the node is not a region boundary node; otherwise, the node is a region boundary node.

[0061] S2: Each smart terminal identifies the power outage load within its area and predicts the output power of distributed photovoltaic systems within that area; specifically:

[0062] S2.1: Each intelligent terminal identifies all power-loss loads within its area by analyzing measurement data. Each intelligent terminal cross-validates the measurement data within its area according to the power flow equations. For nodes lacking measurement data or with zero measurement data, cross-validation distinguishes between power outages and communication failures. The power flow equations upon which the cross-validation of measurement data is based are:

[0063]

[0064] In the formula, V represents voltage, and i and j represent two adjacent busbars. , , and The values ​​represent the resistance, reactance, active power, and reactive power of the feeder segment between the two busbars.

[0065] If the equation holds true, then the load with a power measurement value of zero connected to the bus is considered a power failure load, and the fault type is identified as a power outage; otherwise, the fault type is identified as a communication fault.

[0066] S2.2: Each smart terminal uses an edge-side model to predict the output power of each distributed photovoltaic system in its area within 2 hours after the disaster.

[0067] Each smart terminal generates minute-level power predictions for the distributed photovoltaic (PV) systems within its designated area for the next two hours. Based on the current date and time, the smart terminal calculates the solar altitude angle and estimates its change over the next two hours. Therefore, based on the current power output of each distributed PV system, it predicts the power change over the next two hours, and the predicted PV power at time t is calculated. for:

[0068] ;

[0069] In the formula, and Let be the solar altitude angle and azimuth angle at time t. and For the tilt angle and azimuth angle of the photovoltaic panel, At the time of failure, This represents the photovoltaic power generation efficiency of distributed photovoltaic systems at time t.

[0070] S3: Each smart terminal formulates the optimal recovery plan for its area based on the identification and prediction results of the power outage load and distributed photovoltaic output in its area.

[0071] S3.1: Based on the upper and lower limits of bus voltage and line capacity, and combined with the power loss load and distributed photovoltaic power, construct a set of constraints for the optimization problem of restoration scheme in this region;

[0072] The power flow equation constraints are:

[0073] st

[0074] In the formula, n represents the number of busbars within the region and on the region boundary. and Let t represent the active and reactive power injected by the load or distributed photovoltaic system connected to bus i at time t. If bus i is a load, then... , and These represent the active power demand and reactive power demand of the load, respectively. If bus i is connected to distributed photovoltaic power, then... , Let be the predicted output power of the distributed photovoltaic system connected to bus i at time t. The rated capacity of the inverter for the connected distributed photovoltaic system. Let be the voltage amplitude of bus i at time t. Let i be the busbar i and the set of buses directly connected to busbar i via feeder segments. Let be the phase angle difference between bus i and j at time t. and The conductance and susceptance of the feeder segment between busbar i and busbar j. This represents the open / closed state of the switch on the feeder segment between bus i and bus j, with 0 for open and 1 for closed.

[0075] The upper and lower limits of bus voltage and the upper limit of line power are as follows:

[0076] st

[0077] In the formula, and These are the lower and upper limits of the bus voltage. This represents the upper limit of the apparent power of the feeder segment.

[0078] S3.2: With the objective of minimizing the power outage load, solve the optimization problem of restoration schemes in this region to obtain the optimal restoration scheme. Specifically: First, use the DC power flow calculation method to verify the electrical feasibility of each restoration scheme, filtering out schemes that do not meet the bus voltage and line power constraints, thus compressing the search space; then, construct a mixed-integer linear programming model based on the LinDistFlow approximation equation, solve the optimal solution of this optimization model, and perform a neighborhood search, making no more than three changes to the optimal solution's switching states to obtain multiple candidate restoration schemes; finally, use precise AC power flow equations to calculate the objective function value corresponding to each candidate restoration scheme, accurately selecting the optimal restoration scheme.

[0079] The objective function is:

[0080]

[0081] in, For the collection of loads in the post-disaster autonomous recovery area, For the first One load, In the recovery plan Power reduction for each load.

[0082] One embodiment of the present invention, Figure 2 The power distribution cyber-physical system shown comprises three intelligent terminals, located at the substation, switch SW2, and distributed photovoltaic PV4, respectively, serving as communication nodes 1, 3, and 11. The communication links between these nodes are as follows: Figure 3 As shown in the diagram. First, region partitioning and initialization are performed, establishing three regions: Region 1 contains communication node 1, and its boundary node is communication node 1; Region 2 contains communication node 3, and its boundary node is communication node 3; Region 3 contains communication node 11, and its boundary node is communication node 11. Then, the boundary nodes of each region send expansion messages to their external neighbor nodes along the communication link. Specifically, the boundary node of Region 1 sends an expansion message to communication node 2, the boundary node of Region 2 sends expansion messages to communication nodes 2, 3, 4, 5, and 9, and the boundary node of Region 3 sends expansion messages to communication nodes 6, 7, 8, and 10. Among these, communication node 2 receives expansion messages from both Region 1 and Region 2. The cumulative latency of the message from Region 1 is 2, and the cumulative latency of the message from Region 2 is 8. Therefore, communication node 2 chooses to join Region 1, which has the shorter cumulative latency.

[0083] Following this pattern, each region expands along the communication link, with the expansion speed depending on the latency of the communication link. The final region division scheme is as follows: Figure 4 As shown, the order in which nodes are added in each region is as follows:

[0084] Area 1: Communication Node 1 (Substation), Communication Node 2 (Normal Closed Switch SW1);

[0085] Area 2: Communication Node 3 (normally closed switch SW2), Communication Node 4 (distributed photovoltaic PV1), Communication Node 5 (normally open switch SW5), Communication Node 9 (distributed photovoltaic PV3);

[0086] Area 3: Communication Node 11 (Distributed Photovoltaic PV4), Communication Node 7 (Distributed Photovoltaic PV2), Communication Node 8 (Normally Open Switch SW6), Communication Node 6 (Normally Closed Switch SW3), Communication Node 10 (Normally Closed Switch SW4).

[0087] The second step is for each smart terminal to identify the power outage loads in its area and predict the output power of the distributed photovoltaic system in that area.

[0088] like Figure 4 and 5 As shown, in the obtained area division scheme, the measuring device in area 1 is communication node 2, located on switch SW1 of the feeder segment between loads L2 and L3, which can measure the voltage of loads L2 and L3. The measuring devices in area 2 include communication nodes 3, 5, 4, and 9, which are respectively located on switch SW2 of the feeder segment between loads L3 and L4, switch SW5 of the feeder segment between loads L5 and L9, and distributed photovoltaics PV1 and PV3, which can measure the voltage of loads L3, L4, L5 and L9, and distributed photovoltaics PV1 and PV3. The measuring devices in Zone 3 include communication nodes 6, 10, 8, 7, and 11, located on switch SW3 of the feeder segment between loads L5 and L6, switch SW4 of the feeder segment between loads L9 and L10, switch SW6 of the feeder segment between loads L7 and L11, and distributed photovoltaic (PV) systems PV2 and PV4, respectively. These devices can measure the voltage of loads L5, L6, L7, L9, L10, L11, and distributed photovoltaic (PV) systems PV2 and PV4. Zone 1 does not include distributed photovoltaic systems; Zone 2 includes distributed photovoltaic (PV) systems PV1 and PV3; and Zone 3 includes distributed photovoltaic (PV) systems PV2 and PV4.

[0089] First, each smart terminal identifies all power-loss loads within its area by analyzing measurement data. Taking area 3 as an example, it can calculate the active power flowing through the feeder segment between loads L5 and L6 based on the power measurement values ​​of each load and distributed photovoltaic system. and reactive power :

[0090] ;

[0091] In the formula, and These are the measured values ​​of active and reactive power flowing from the distribution network into the load or distributed photovoltaic system.

[0092] Based on this, and combining the voltage measurement values ​​from the measuring device on switch SW3 located in the feeder section between loads L5 and L6, verify whether the following equation holds true:

[0093] ;

[0094] In the formula, and Let L1 be the resistance and reactance of the feeder segment between loads L5 and L6. If the equation holds true, it indicates that the power measurement values ​​of the load and the distributed photovoltaic system are accurate, and the load with a power measurement value of zero is a power-off load. If the equation does not hold true, it indicates that the power measurement values ​​of the load and the distributed photovoltaic system are inaccurate, and the reason for the power measurement value of zero is a communication failure.

[0095] Then, each smart terminal uses an edge-side model to predict the output power of each distributed photovoltaic (PV) system within its area in the two hours following the disaster. Communication node 3 in area 2 predicts the output power of distributed PV systems PV1 and PV3 in the next two hours, and communication node 11 in area 3 predicts the output power of distributed PV systems PV2 and PV4 in the next two hours. The prediction formula is as follows:

[0096] ;

[0097] ;

[0098] ;

[0099] Where t represents a future moment. The moment the failure occurred. For distributed photovoltaic in Power at any given moment; This represents the photovoltaic power generation efficiency of a distributed photovoltaic system at time t. and Let denot t represent the solar altitude angle and azimuth angle at time t, respectively. and These represent the tilt angle and azimuth angle of the distributed photovoltaic panel, respectively.

[0100] The third step involves each smart terminal identifying and predicting the power outage load and distributed photovoltaic output in its area, and then formulating the optimal recovery plan for that area.

[0101] Taking Region 3 as an example, firstly, based on the upper and lower limits of bus voltage and line capacity, and combined with the power loss load and distributed photovoltaic power, a set of constraints is constructed for the optimization problem of the recovery scheme in this region, including power flow equation constraints, upper and lower limit constraints of bus voltage, and upper limit constraints of line power. The power flow equation constraints are:

[0102] st ;

[0103] In the formula, n represents the number of busbars within the region and on the region boundary. and Let t represent the active and reactive power injected by the load or distributed photovoltaic system connected to bus i at time t. If bus i is a load, then... , and These represent the active power demand and reactive power demand of the load, respectively. If bus i is connected to distributed photovoltaic power, then... , Let be the predicted output power of the distributed photovoltaic system connected to bus i at time t. The rated capacity of the inverter for the connected distributed photovoltaic system. Let be the voltage amplitude of bus i at time t. Let i be the busbar i and the set of buses directly connected to busbar i via feeder segments. Let be the phase angle difference between bus i and j at time t. and The conductance and susceptance of the feeder segment between busbar i and busbar j. This represents the open / closed state of the switch on the feeder segment between bus i and bus j, with 0 for open and 1 for closed.

[0104] st ;

[0105] Where i and j represent the loads or distributed photovoltaic buses mounted in the third region;

[0106] , , , , , .

[0107] The upper and lower limits of bus voltage and the upper limit of line power are as follows:

[0108] st ;

[0109] Then, construct the objective function that minimizes the power outage load:

[0110] .

[0111] Taking region 3 as an example, the optimization problem-solving process is as follows:

[0112] First, the electrical feasibility of each restoration scheme is verified using DC power flow calculation, filtering out schemes that do not meet the bus voltage and line power constraints, thus reducing the search space. For region 3, there are four restoration schemes: "SW3 closed, SW4 closed, SW6 closed", "SW3 closed, SW4 open, SW6 closed", "SW3 closed, SW4 closed, SW6 open", and "SW3 closed, SW4 open, SW6 open". The electrical feasibility of each scheme is quickly verified using DC power flow calculation, filtering out schemes "SW3 closed, SW4 closed, SW6 closed" and "SW3 closed, SW4 closed, SW6 open" that do not meet the line power constraints, retaining the two restoration schemes that meet the electrical feasibility: "SW3 closed, SW4 open, SW6 closed" and "SW3 closed, SW4 open, SW6 open".

[0113] Then, a mixed-integer linear programming model is constructed based on the LinDistFlow approximation equation. The optimal solution of this optimization model is solved, and a neighborhood search is performed. The optimal solution is then subjected to no more than three switch state changes to obtain a set of candidate recovery schemes. In this embodiment, the candidate recovery schemes are "SW3 closed, SW4 open, SW6 closed" and "SW3 closed, SW4 open, SW6 open".

[0114] Finally, using precise AC power flow equations, the objective function values ​​corresponding to each candidate recovery scheme are calculated, and the optimal recovery scheme is accurately selected. In this embodiment, the objective function of the recovery scheme "SW3 closed, SW4 open, SW6 closed" is the smallest, therefore this recovery scheme is the optimal scheme. Figure 5 As shown.

[0115] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the present invention will not describe the various possible combinations separately.

Claims

1. A method for autonomous regional recovery of a power distribution cyber-physical system after a disaster, characterized in that, Specifically, the steps include the following: The power distribution system is divided into multiple post-disaster autonomous recovery zones, and each post-disaster autonomous recovery zone has one and only one smart terminal. Each smart terminal identifies the power loss load within its self-recovery area after a disaster and predicts the output power of distributed photovoltaic power within that area for a period of time after the fault. The intelligent terminal sets constraints and objective functions based on the identified power loss load and the predicted output power of distributed photovoltaic power. Based on the constraints, candidate recovery schemes are obtained, and the final recovery scheme is selected from the candidate recovery schemes based on the objective function. Power forecast of distributed photovoltaic in post-disaster self-recovery areas in the near future The expression is as follows: ； Where t represents a future moment. The moment the failure occurred. For distributed photovoltaic in Power at any given moment; This represents the photovoltaic power generation efficiency of a distributed photovoltaic system at time t. The expression is as follows: ； in, and Let denot t represent the solar altitude angle and azimuth angle at time t, respectively. and These represent the tilt angle and azimuth angle of the photovoltaic panel, respectively. The constraints include: upper and lower limits of bus voltage, upper limit of line power, and power flow constraints. The power flow constraint is specifically as follows: ； Where t represents a future moment after the fault. and Indicates busbar The active and reactive power injected by the connected load or distributed photovoltaic system at time t; if the bus... If the load is connected, then , and This refers to the active and reactive power requirements of the load, and the reactive power requirement; if the bus... If it is a distributed photovoltaic system, then , Indicates busbar The predicted output power of the connected distributed photovoltaic system at time t. Indicates busbar The rated capacity of the inverter connected to the distributed photovoltaic system; busbar The voltage amplitude at time t, busbar and via feeder section and busbar The set of directly connected busbars; For the bus at time t and The phase angle difference between them busbar and The open / closed state of the switch on the feeder segment; if open, If closed, ; and For intermediate parameters, the expression is: ； Represents a set busbar busbar, and busbars and Conductivity and susceptance of the feeder segments; The bus voltage upper and lower limit constraints and the line power upper limit constraints are as follows: ； in, and These are the lower and upper limits of the bus voltage, respectively. The apparent power limit for the feeder segment. and The expression is: ； Construct the objective function to minimize the amount of power loss: ； in, For the collection of loads in the post-disaster autonomous recovery area, For the first One load, In the recovery plan Power reduction for each load.

2. The method for autonomous regional recovery of a power distribution information physical system after a disaster, as described in claim 1, is characterized in that... The power distribution system is divided into multiple self-recovery zones after a disaster, specifically: Step A: Take each smart terminal and measurement device in the system as nodes, and the communication links between nodes as edges; Step B: Based on the number of smart terminals n, divide the system into n post-disaster autonomous recovery zones, and use the smart terminals as temporary boundary nodes in the corresponding post-disaster autonomous recovery zones; Step C: The boundary node sends an expansion message to neighboring nodes located outside the corresponding autonomous recovery zone via the adjacency communication link. If the neighboring node has not yet joined any autonomous recovery zone, the boundary node is added to the autonomous recovery zone and the boundary node is updated. Step D: Repeat step C until every node has been added to the post-disaster autonomous recovery zone.

3. The method for autonomous regional recovery of a power distribution information physical system after a disaster, as described in claim 2, is characterized in that... In step C, if a node has not joined any post-disaster autonomous recovery zone and receives expansion messages from two or more boundary nodes, then the cumulative communication latency in each expansion message is compared, and the node is joined to the post-disaster autonomous recovery zone corresponding to the minimum cumulative communication latency.

4. A method for autonomous regional recovery of a power distribution information physical system after a disaster, as described in claim 2, is characterized in that... In step C, updating the boundary node specifically means: if all the neighboring communication links of a node in the post-disaster autonomous recovery area point to neighboring nodes that belong to this post-disaster autonomous recovery area, then the node is considered not a boundary node; otherwise, the node is considered a boundary node.

5. A method for autonomous regional recovery of a power distribution information physical system after a disaster, as described in claim 1, is characterized in that... The intelligent terminal identifies the fault type of its post-disaster autonomous recovery area by analyzing measurement data within that area. Specifically, the intelligent terminal performs cross-validation on the measurement data within the post-disaster autonomous recovery area according to the following power flow equation, the expression of which is shown below: ； in, j represents two adjacent busbars, and V represents voltage. , , and These represent the resistance, reactance, active power, and reactive power of the feeder segment between the two busbars, respectively. If the equation holds true, then the load with a power measurement value of zero connected to the bus is considered a power failure load, and the fault type is identified as a power outage; otherwise, the fault type is identified as a communication fault.

6. The method for autonomous regional recovery of a power distribution information physical system after a disaster, as described in claim 1, is characterized in that... The output power of distributed photovoltaic power in the disaster recovery area is predicted by the following steps: the smart terminal calculates the solar altitude angle based on the time data at the time of the fault, and estimates the change of the solar altitude angle in the future period after the fault, thereby estimating the power prediction value of distributed photovoltaic power in the future period.

7. A method for autonomous regional recovery of a power distribution information physical system after a disaster, as described in claim 1, is characterized in that... The final recovery plan is as follows: The electrical feasibility of each restoration scheme is verified using a DC power flow calculation method, and restoration schemes that do not meet the constraints are filtered out. Then, a mixed-integer linear programming model is constructed based on the LinDistFlow approximation equation, and the optimal solution of this model is solved. A neighborhood search is performed on the optimal solution to obtain multiple candidate restoration schemes. The objective function value corresponding to each candidate restoration scheme is calculated to obtain the final restoration scheme. The neighborhood search of the optimal solution specifically involves changing the switch states in the optimal solution, and the total number of switches changed does not exceed a preset number.

Images