Multi-resource, multi-stage operation method for power distribution networks to cope with natural disasters and man-made attacks

Abstract

This invention discloses a multi-resource, multi-stage operation method for power distribution networks to cope with natural disasters and human attacks. The method assumes a high-risk, low-probability event of attackers launching malicious attacks after a natural disaster. It considers multiple resource types, including gas turbines, SESS (Self-Enhanced Energy Storage), MESS (Self-Enhanced Energy Storage), RC (Resistant Power Storage), and NR (Normative Energy Storage), and their coupling relationships. Based on the time difference between the natural disaster and the human attack, the operation process of the power distribution network is divided into five stages. In the context of power grid-transportation network coupling, the impact of the dynamic changes in schedulable resources and traffic flow at each stage on the movement of mobile energy storage and the repair process by emergency repair personnel is considered to establish an optimized scheduling model for each stage of the power distribution network. The model is then transformed into a five-level mixed-integer linear programming problem that can be solved using mature commercial optimization software, based on time-series relationships. This method can ensure power supply to the load and reduce system operating costs under extreme scenarios through flexible coordination among multiple resource types, effectively improving the resilience of the power distribution network.

Description

Technical Field

[0001] This invention belongs to the field of power distribution network operation and scheduling, and more specifically, relates to a multi-resource, multi-stage operation method for power distribution networks in response to natural disasters and human attacks. Background Technology

[0002] Because power distribution networks are mostly exposed to the elements and rely on widely deployed communication equipment for information exchange, they are highly vulnerable to attacks. In recent years, ecological damage and global warming have become increasingly severe, with frequent natural disasters such as typhoons and torrential rains. Furthermore, the continuous development of network and computer technologies has made malicious attacks more sophisticated and difficult to defend against. All of these factors pose a significant threat to the safe operation of power distribution networks. Therefore, improving their ability to cope with high-risk, low-probability events such as natural disasters and human attacks, and mitigating economic losses caused by power outages, is of paramount importance. Against this backdrop, the concept of resilience has emerged and gradually become a focus of academic research. It refers to the ability of a power grid to adapt to changing conditions, take defensive measures to reduce fault losses, and quickly restore normal operation when faced with high-risk, low-probability events.

[0003] Currently, scholars both domestically and internationally have proposed many effective measures to enhance the resilience of power distribution networks, such as pre-disaster deployment of emergency resources, reinforcement of distribution lines, load transfer through network reconfiguration (NR), microgrid formation using distributed generators or stationary energy storage systems (SESS), dynamic islanding using mobile energy storage systems (MESS), and dispatching repair crews (RC) to repair damaged equipment. However, most of these measures only consider a few of the resources, including distributed generators, SESS, MESS, RC, and NR, and rarely delve into their multi-stage coupling relationships and mutual influence analysis, especially the coordination between MESS, RC, and NR. Furthermore, most existing literature uses stochastic optimization methods to generate disaster scenarios and solve models, which, while simplifying the calculation process to some extent, cannot cover the most severe damage scenarios. Some literature has adopted the idea of ​​robust optimization, but it only considers a single instance of power grid disruption. While this can effectively cope with natural disasters such as typhoons and rainstorms, existing research cannot provide effective countermeasures if further damage occurs during the post-disaster recovery process, such as malicious human attacks. The literature review results indicate that there is currently a gap in research on simultaneously considering natural disasters and human attacks, and making full use of flexible resources for robust scheduling to improve the resilience of distribution networks. Summary of the Invention

[0004] In response to the above-mentioned deficiencies or improvement needs of existing technologies, this invention provides a multi-resource, multi-stage operation method for power distribution networks to cope with natural disasters and human attacks. In extreme scenarios, through flexible cooperation among multiple types of resources, it can ensure the power supply to the load as much as possible and reduce the operating costs of the system, effectively improving the resilience of the power distribution network.

[0005] To achieve the above objectives, according to a first aspect of the present invention, a multi-resource, multi-stage operation method for a power distribution network in response to natural disasters and human attacks is provided, comprising:

[0006] S1. Using gas turbines, SESS, and MESS as dispatchable resources, and aiming to minimize the first total operating cost of the distribution network, a dispatching model is established for the normal operation phase.

[0007] S2 establishes a disaster disturbance stage damage model with the goal of maximizing the load loss during the period from the disaster to the restoration of normal operation of the distribution network caused by natural disasters; and establishes a post-disaster degradation stage scheduling model with gas turbines, SESS, MESS, RC, and NR as dispatchable resources and the goal of minimizing the second total operating cost of the distribution network.

[0008] S3 sets the goal of maximizing the load loss during the period from the attack to the restoration of normal operation of the distribution network under human attack, and establishes a destruction model for the human attack phase; with gas turbine, SESS, RC, NR, and unattacked MESS as dispatchable resources, and with the goal of minimizing the second total operating cost of the distribution network, establishes a scheduling model for the defense and recovery phase.

[0009] S4. Based on the temporal relationships between the models established in S1-S3, the problem is transformed into a five-level mixed integer linear programming problem and solved to obtain the multi-stage operation scheme of the power distribution network.

[0010] Wherein, the natural disasters only damage the power distribution lines; the man-made attacks only damage the MESS; the first total operating cost includes the cost of purchasing electricity from the upper-level power grid, the cost of generating electricity from the gas turbine, the operating cost of the MESS and transportation costs, the operating cost of the SESS and network loss costs; the second total operating cost includes the first total operating cost, the cost of load shedding penalty and the operating cost of the tie lines.

[0011] According to a second aspect of the present invention, a multi-resource, multi-stage operation system for a power distribution network in response to natural disasters and man-made attacks is provided, comprising: a computer-readable storage medium and a processor;

[0012] The computer-readable storage medium is used to store executable instructions;

[0013] The processor is configured to read executable instructions stored in the computer-readable storage medium and execute the method as described in the first aspect.

[0014] According to a third aspect of the invention, a computer-readable storage medium is provided, characterized in that the computer-readable storage medium stores computer instructions for causing a processor to perform the method as described in the first aspect.

[0015] In summary, compared with the prior art, the above-described technical solutions conceived by this invention can achieve the following beneficial effects:

[0016] This invention provides a method that addresses the high-risk, low-probability event of attackers launching malicious attacks after a natural disaster. It considers multiple resource types, including gas turbines, SESS (Search Engine Safe), MESS (Search Engine Safe), RC (Resistant Power Supply), and NR (Resistant Grid), and their coupling relationships. This method enables multi-resource, multi-stage operation scheduling of the distribution network to cope with both natural disasters and man-made attacks, allowing operators to switch operations smoothly and defend effectively in extreme scenarios. First, based on the time difference between natural disasters and man-made attacks, the operation of the distribution network is divided into five stages: normal operation, disaster disturbance, post-disaster degradation, man-made attack, and defense and recovery. Second, within the context of power grid-transportation network coupling, and considering the schedulable resources at each stage, as well as the impact of dynamic traffic flow characteristics on the movement of mobile energy storage and the repair process by emergency personnel, an optimized scheduling model for each stage of the distribution network is established. Finally, the model is transformed into a five-level mixed-integer linear programming problem solvable using mature commercial optimization software, based on time-series relationships. This method can, through flexible coordination among multiple resource types in extreme scenarios, maximize the power supply to the load and reduce system operating costs, effectively improving the resilience of the distribution network. Attached Figure Description

[0017] Figure 1 This is a time-series relationship diagram between the various models provided in this invention.

[0018] Figure 2 This is a topology diagram of the improved IEEE 33-node system in Embodiment 1 of the present invention.

[0019] Figure 3 This is a schematic diagram of the load demand power curve of the distribution network in Embodiment 1 of the present invention.

[0020] Figure 4 This is a schematic diagram of the transportation network corresponding to the improved IEEE 33-node system in Embodiment 1 of the present invention.

[0021] Figure 5 This is a schematic diagram of the dispatching results of the distribution network during the normal operation phase in Embodiment 1 of the present invention.

[0022] Figure 6This is a schematic diagram of the SOC change curve of the energy storage device during normal operation in Embodiment 1 of the present invention.

[0023] Figure 7 In the diagrams (a)-(f), the topology changes of the distribution network in scenario 1 of embodiment 1 of the present invention are shown in the diagrams for the time periods 35, 36-38, 39-41, 42, 43-44, and 45-48, respectively.

[0024] Figure 8 This is a schematic diagram of the dispatching results of the power distribution network under scenario 1 in embodiment 1 of the present invention.

[0025] Figure 9 This is a schematic diagram of the SOC change curve of the energy storage device in scenario 1 of embodiment 1 of the present invention.

[0026] Figure 10 In the diagrams (a)-(h), the topology changes of the distribution network in scenario 2 of embodiment 1 of the present invention are shown in the time periods of 35, 36, 37, 38, 39, 40-41, 42-44, and 45-48, respectively.

[0027] Figure 11 In the diagrams (a)-(f), the topology changes of the distribution network under scenario 1-after in Embodiment 1 of the present invention are shown in the time periods of 37-38, 39-41, 42, 43-44, 45, and 46-48, respectively.

[0028] Figure 12 This is a schematic diagram of the SOC change curve of the energy storage device in scenario 1-after in Embodiment 1 of the present invention.

[0029] Figure 13 In the diagrams (a)-(g), the topology changes of the distribution network under scenario 2-after in Embodiment 1 of the present invention are shown in the time periods of 37, 38, 39, 40, 41, 42-44, and 45-48, respectively. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0031] This invention provides a method for multi-resource, multi-stage operation of a power distribution network in response to natural disasters and human attacks, including:

[0032] S1. Using gas turbines, SESS, and MESS as dispatchable resources, and aiming to minimize the first total operating cost of the distribution network, a dispatching model is established for the normal operation phase.

[0033] S2 establishes a disaster disturbance stage damage model with the goal of maximizing the load loss during the period from the disaster to the restoration of normal operation of the distribution network caused by natural disasters; and establishes a post-disaster degradation stage scheduling model with gas turbines, SESS, MESS, RC, and NR as dispatchable resources and the goal of minimizing the second total operating cost of the distribution network.

[0034] S3 sets the goal of maximizing the load loss during the period from the attack to the restoration of normal operation of the distribution network under human attack, and establishes a destruction model for the human attack phase; with gas turbine, SESS, RC, NR, and unattacked MESS as dispatchable resources, and with the goal of minimizing the second total operating cost of the distribution network, establishes a scheduling model for the defense and recovery phase.

[0035] S4. Based on the temporal relationships between the models established in S1-S3, the problem is transformed into a five-level mixed integer linear programming problem and solved to obtain the multi-stage operation scheme of the power distribution network.

[0036] Wherein, the natural disasters only damage the power distribution lines; the man-made attacks only damage the MESS; the first total operating cost includes the cost of purchasing electricity from the upper-level power grid, the cost of generating electricity from the gas turbine, the operating cost of the MESS and transportation costs, the operating cost of the SESS and network loss costs; the second total operating cost includes the first total operating cost, the cost of load shedding penalty and the operating cost of the tie lines.

[0037] Specifically, such as Figure 1 As shown, based on the temporal relationship between natural disasters and man-made attacks on the power distribution network, its operation process is divided into five stages: normal operation stage, disaster disturbance stage, post-disaster degradation stage, man-made attack stage, and defense and recovery stage. Among them, the dispatchable resources in the normal operation stage include gas turbines, SESS, and MESS; the dispatchable resources in the post-disaster degradation stage include gas turbines, SESS, MESS, RC, and NR; and the dispatchable resources in the defense and recovery stage include gas turbines, SESS, RC, NR, and unattacked MESS.

[0038] Then, combining the ideas of schedulable resources and robust optimization at each stage, models for five stages are established respectively: normal operation stage scheduling model, disaster disturbance stage damage model, post-disaster degradation stage scheduling model, human attack stage damage model, and defense and recovery stage scheduling model.

[0039] Finally, based on the temporal relationship of the five-stage model, it is transformed into a five-level mixed integer linear programming problem, and the exhaustive method and mature commercial optimization software are used to calculate and solve it, so as to obtain a multi-stage operation scheme for the distribution network that comprehensively considers multiple types of resources.

[0040] The objective function of the normal operation phase scheduling model is to minimize the overall operating cost of the distribution network, including the cost of purchasing electricity from the upstream grid, the cost of gas turbine power generation, the operating cost of MESS and transportation costs, the operating cost of SESS and network loss costs.

[0041]

[0042] In the above formula, t0 and t end These represent the start and end times of a day, t. a1 The period during which natural disasters occur, t r The period during which the distribution network completes fault repairs and restores normal operation; λ t grid Let P be the unit electricity purchase cost of the distribution network from the upper-level power grid during time period t. t grid The active power provided by the upstream power grid during time period t; I, M, S, and L G These are collections of gas turbines, MESS, SESS, and conventional lines in the power distribution network; α i Let be the fuel cost coefficient for the i-th gas turbine. w represents the active power output of the i-th gas turbine during time period t; m and w s These are the operating cost coefficients for MESS and SESS, respectively. and These represent the charging and discharging power of MESS m during time period t. and denoted as SESS charge / discharge power during time period t; γ is the mobility cost coefficient of MESS. β is the distance traveled by MESS m (from node n to node p) during time period t; β is the network loss cost coefficient, Pf l,t The active power on line l during time period t is the active power.

[0043] The constraints of the objective function include: prohibition of power backfeeding, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, initial energy constraints for energy storage, first constraint for MESS access status, road congestion constraints, and MESS travel distance and time constraints.

[0044] Power backfeed constraint prohibited:

[0045] P t grid ≥0 (2)

[0046] Node power balance constraints:

[0047]

[0048] In the above formula, O(l) and D(l) are the start and end node numbers of line l, respectively; PL n,t and QL n,t These represent the active and reactive loads at node n during time period t; The reactive power provided by the upstream power grid during time period t; Qf l,t The reactive power on line l during time period t; and These are the sets of MESS, SESS, and gas turbines connected to node n, respectively; Q m,t Q s,t and These represent the reactive power provided by MESS m, SESS s, and gas turbine i during time period t, respectively.

[0049] Line capacity constraints:

[0050]

[0051] In the above formula, and These represent the upper limits of active and reactive power for line l, respectively.

[0052] Current constraints:

[0053]

[0054] In the above formula, G l and B l These are the conductance and susceptance of line l, respectively; V n,t and θ n,t Let I(n) and E(n) be the voltage amplitude and phase angle of node n during time period t, respectively; I(n) and E(n) are the sets of input and output lines of node n, respectively.

[0055] Voltage constraint:

[0056]

[0057] In the above formula, V0 is the substation bus voltage; and These are the upper and lower voltage limits at node n, respectively. This represents the upper limit of the phase angle difference of node n.

[0058] Gas turbine operating constraints:

[0059]

[0060] In the above formula, P i g,max and These represent the upper limits of active and reactive power output of gas turbine i, respectively. This is the active power ramp-up limit for gas turbine i.

[0061] MESS operational constraints:

[0062]

[0063] In the above formula, and These are the charging and discharging status flag variables of MESS m during time period t; and These represent the upper limits of active and reactive power output of MESS m, respectively; E m,t Let ηm be the energy stored in MESS m during time period t, ηm be the charge / discharge efficiency of MESS, and Δt be the duration of the unit time period; SOC max and SOC min These are the upper and lower limits of the State of Charge (SOC) of the energy storage device. This is the rated capacity of MESS m.

[0064] SESS operational constraints:

[0065]

[0066] In the above formula, and These are the charging and discharging state flag variables of SESS s during time period t; P s max and These are the upper limits of active and reactive power output of SESS s, respectively; E s,t Let ηs be the energy stored in the SESS during time period t, and ηs be the charging and discharging efficiency of the SESS. This refers to the rated capacity of SESS s.

[0067] Initial energy constraints for energy storage:

[0068]

[0069] In the above formula, E ini This represents the initial energy value of the energy storage device.

[0070] MESS access status first constraint:

[0071]

[0072] In the above formula, N CM For the set of accessible nodes in MESS; u m,n,t v is a connection status flag variable between MESS m and accessible node n during time period t. m,t Let m be the movement status flag variable for MESS during time period t; Let t be the time required for MESS m to travel from node n to node p during time period t. The constraints, from top to bottom, are as follows: ① The connection and movement states of a MESS are mutually exclusive, and a MESS can only connect to at most one accessible node during the same time period; ② A MESS can only charge and discharge during the connection state; ③ Each accessible node can only connect to at most one MESS during each time period; ④ When MESS m connects to node n during time period t, it needs to travel at least... It can only be connected to node p after a certain time.

[0073] Road congestion constraints:

[0074]

[0075]

[0076] In the above formula, D n,p,t χ represents the travel distance between node n and node p during time period t, considering congestion. n,p,t Let be the congestion coefficient of the road between node n and node p during time period t. ρ represents the original distance of the road between node n and node p. n,p,t Let κ represent the traffic flow on the road between node n and node p during time period t. n,p This represents the traffic congestion limit for the road between node n and node p.

[0077] MESS driving distance and time constraints:

[0078]

[0079] In the above formula, Vel m Let m be the speed of MESS.

[0080] The objective function of the disaster disturbance stage damage model is to maximize the load loss of the distribution network during the period from the disaster to the restoration of normal operation.

[0081]

[0082] In the above formula, N is the set of nodes in the distribution network; The active load removed from node n during time period t.

[0083] The objective function is constrained by a disaster damage budget constraint:

[0084]

[0085] In the above formula, a l B is the variable indicating the damaged state of line l. disaster This represents the upper limit for the number of power lines damaged by disasters. To simplify the modeling process, it is assumed here that natural disasters only damage power distribution lines. This is because, for the power distribution network, natural disasters may cause problems such as broken conductors, overturned poles, and equipment damage. However, generally speaking, equipment such as gas turbine units and energy storage devices (including MESS located in charging and discharging stations) are mostly located inside factory buildings, and their defense mechanisms are more robust, so the probability of damage from disasters is relatively low. In contrast, power distribution lines and poles are directly exposed to the outdoors, are more vulnerable, and have poor disaster prevention capabilities, making them more likely to become disaster-bearing bodies. Although the specific fault manifestations after power distribution lines and poles are damaged are different, from a system perspective, both ultimately reflect as an interruption of power supply to the line, and therefore can be equivalent to damage to the power distribution line.

[0086] The objective function of the post-disaster degradation stage scheduling model is to minimize the overall operating cost of the distribution network. This includes not only the cost of purchasing electricity from the upstream grid, the cost of gas turbine power generation, the operating costs of the MESS (Medium-Stop Service) and transportation, the operating costs of the SESS (Short-Stop Service), and network loss costs, but also the cost of load shedding penalties and the operating costs of tie lines. It should be noted that the RC (Regulatory Control Unit) consists of the distribution network's own operation and maintenance personnel; therefore, their working costs are not considered in the objective function.

[0087]

[0088] In the above formula, L C δ represents the set of interconnecting lines in the distribution network; δ is the load loss penalty cost coefficient, which is related to the importance of the load; ζ is the operating cost coefficient of the interconnecting lines. and These are the forward and reverse power transmission status flag variables for line l during time period t.

[0089] The constraints of the objective function include power purchase constraints, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, MESS access status first constraint, road congestion constraints, MESS travel distance and time constraints, load shedding constraints, line power transmission status constraints, node status constraints, RC operating status constraints, and RC travel time constraints. Among these, the line capacity constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, MESS access status first constraint, road congestion constraints, and MESS travel distance and time constraints are the same as those in the normal operation phase scheduling model, as shown in equations (4), (6)-(9), and (11)-(14).

[0090] Power purchase constraints:

[0091]

[0092] In the above formula, This represents the maximum active power that the upstream power grid can provide after a natural disaster. This constraint indicates that the distribution network cannot feed power back to the upstream power grid, and its power purchasing capacity is limited after a natural disaster. This is because natural disasters may damage the upstream power grid, and to alleviate the power supply pressure on the upstream power grid, it is necessary to impose certain restrictions on the power purchasing capacity of the distribution network.

[0093] Node power balance constraints:

[0094]

[0095] In the above formula, This represents the reactive load removed from node n during time period t.

[0096] Current constraints:

[0097]

[0098] Loss of load constraint:

[0099]

[0100] Line power transmission state constraints:

[0101]

[0102] The constraints, from top to bottom, are as follows: ① If a conventional line is in normal operation, it can only transmit power in one direction, either forward or backward, while a damaged line must be taken out of operation; ② A tie line can only transmit power in either the forward or reverse direction after it is put into operation; ③ To ensure the radial topology of the distribution network, two or more power sources are not allowed to supply power to the loads on the same node at the same time.

[0103] Status constraints of the node to be inspected:

[0104]

[0105] In the above formula, N CR Let z be the set of nodes to be RC inspected; n,t Let be the maintenance status flag variable for node n in time period t; R is the RC set of the distribution network, c r,n,t Let be the operating status flag variable of RC r near node n during time period t. Here, this invention simplifies the maintenance process of power distribution lines, assuming that RC maintenance starts from the endpoints of the line, i.e., nodes. After reaching a certain node, it will begin maintenance from that node until the midpoint of the line. During this process, any problems discovered will be addressed immediately. Therefore, if RC visits nodes at both ends of a line, the line can return to normal operation. Based on this simplification, the constraints are expressed from top to bottom as follows: ① Only nodes connected to lines damaged by disasters constitute nodes to be maintained by RC; ② After a disaster, the status flag variables of all nodes to be maintained are all set to 1; ③ Only nodes that have been visited and maintained by RC can have their maintenance status variables set to 0; ④ If the maintenance status variables of nodes at both ends of a damaged line are both 0 (i.e., both have undergone RC maintenance), the line will return to normal operation.

[0106] RC operating state constraints:

[0107]

[0108] In the above formula, b r,t Let be the movement status flag variable of RC r during time period t; The maintenance time required near node n; Let t be the time required for RC r to travel from node n to node p during time period t. The constraints, from top to bottom, are as follows: ① The maintenance and movement states of an RC are mutually exclusive, and an RC can only handle one node to be maintained at any given time period; ② At most one RC can be at a node to be maintained at any given time period, and multiple RCs are not allowed to perform repetitive work; ③ Once RC r starts handling node n to be maintained during time period t, it must continue working until all problems at that node are resolved; ④ If RC r completes handling node n to be maintained during time period t, it must travel at least [time period missing]. The node p to be inspected can only be processed after a certain period of time.

[0109] RC travel time constraint:

[0110]

[0111] In the above formula, Vel r Let r be the moving speed of RC.

[0112] The objective function of the deliberate attack phase disruption model is to maximize the load loss of the distribution network during the period from the time it is attacked to the time it returns to normal operation.

[0113]

[0114] In the above formula, t a2 This refers to the time period during which a human-made attack occurred.

[0115] The constraint condition of the objective function is the attack budget constraint:

[0116]

[0117] In the above formula, a m B is the damaged state indicator variable for MESS m. cyber This represents the upper limit of the number of MESS nodes that can be attacked by humans. Similarly, to simplify the modeling process, it is assumed here that human attacks will only damage MESS nodes. This is because MESS nodes rely on wireless communication technology to communicate with operators, and their transmission medium is completely open, lacking physical protection. Therefore, compared to transmission lines, gas turbines, and SESS devices, MESS nodes are more vulnerable to attack and more difficult to defend against. Furthermore, since the number of accessible MESS nodes is limited, regardless of their movement across the network, they will ultimately interact with the power grid at accessible nodes, providing attackers with certain advantages. Attackers can employ a "wait-and-see" approach, deploying attack plans in advance near accessible MESS nodes and launching a rapid attack once a MESS node enters a controllable range.

[0118] The objective function of the defense and recovery phase scheduling model is also to minimize the overall operating cost of the distribution network, including the cost of purchasing electricity from the upper-level grid, the cost of gas turbine power generation, the operating cost of MESS and transportation costs, the operating cost of SESS, network loss costs, the cost of load shedding penalties, and the operating cost of tie lines. The operating cost of RC is not considered.

[0119]

[0120] The constraints of the objective function include power purchase constraints, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, MESS access status second constraints, road congestion constraints, MESS travel distance and time constraints, load shedding constraints, line power transmission status constraints, node status constraints, RC operating status constraints, and RC travel time constraints. Among them, the power purchase constraints, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, road congestion constraints, MESS travel distance and time constraints, load shedding constraints, line power transmission status constraints, node status constraints, RC operating status constraints, and RC travel time constraints are the same as those in the post-disaster degradation stage scheduling model, as shown in equations (18), (19), (4), (20), (6)-(9), (12)-(14), and (21)-(25).

[0121] MESS Access Status Second Constraint:

[0122]

[0123] The constraints, from top to bottom, are as follows: ① The connection and movement states of a MESS are mutually exclusive, and a MESS can only connect to at most one accessible node at any given time. If a MESS is attacked, it will exit operation; ② A MESS can only charge and discharge when it is connected; ③ Each accessible node can only connect to at most one MESS in each time period; ④ If the m-th MESS is connected to node n in time period t and is attacked, then other MESSes will not be allowed to go to node n for charging and discharging; ⑤ When MESS m is connected to node n in time period t, it needs to travel at least... It can only be connected to node p after a certain time.

[0124] It is important to note that once distribution network operators detect a malicious attack on a MESS (Mechanical Service Shield), they will immediately dispatch professionals to repair it. Considering the complexity of defending against and eliminating malicious attacks, this invention assumes that the attacked MESS will not return to normal operation until all damaged distribution lines have been repaired. Therefore, when the RC (Remote Control Center) completes the repair of all damaged lines, the attacked MESS will also return to normal, and the distribution network will switch back to normal operation.

[0125] The principle of this invention is explained as follows:

[0126] This invention provides a multi-resource, multi-stage operation method for power distribution networks in response to natural disasters and man-made attacks. It considers multiple resource types, including gas turbines, SESS (Self-Enhancing Energy System), MESS (Mechanical Energy System), RC (Resistant Controller), and NR (Normative Controller), and their coupling relationships, forming a power distribution network scheduling strategy aimed at improving resilience. This allows operators to switch smoothly and defend effectively in extreme scenarios. First, based on the time difference between natural disasters and man-made attacks, the operation process of the power distribution network is divided into five stages: normal operation stage, disaster disturbance stage, post-disaster degradation stage, man-made attack stage, and defense and recovery stage. Second, in the context of power grid-transportation network coupling, considering the schedulable resources at each stage and taking into account the impact of dynamic traffic flow characteristics on the MESS movement process and RC repair process, models for each stage of the power distribution network are established: a scheduling model for the normal operation stage, a disruption model for the disaster disturbance stage, a scheduling model for the post-disaster degradation stage, a disruption model for the man-made attack stage, and a scheduling model for the defense and recovery stage. Finally, according to the temporal relationship, the model is transformed into a five-level mixed-integer linear programming problem solvable using mature commercial optimization software. This design can ensure the power supply to the load and reduce the operating cost of the system by flexibly coordinating multiple types of resources in extreme scenarios, effectively improving the resilience of the distribution network.

[0127] The normal operation scheduling model aims to minimize the overall operating cost of the distribution network, including the cost of purchasing electricity from the upstream grid, the cost of gas turbine power generation, the operating costs of the MESS (Mechanical Energy Storage System) and transportation costs, the operating costs of the SESS (Short-Terminal Energy Storage System), and network loss costs. Constraints include prohibition of backfeeding, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, initial energy constraints for energy storage, first constraint on MESS access status, road congestion constraints, and MESS travel distance and time constraints.

[0128] Power backfeeding prohibition constraint: This constraint is used to ensure the local use of power in the distribution network and to prevent power from being fed back to the transmission network.

[0129] Node power balance constraint: This constraint is used to ensure the real-time balance of power generation and consumption in the distribution network.

[0130] Line capacity constraint: This constraint is used to ensure that the distribution network lines do not become overloaded.

[0131] Power flow constraint: This constraint is used to ensure that the distribution network satisfies the basic power flow equations.

[0132] Voltage constraint: This constraint is used to ensure that the voltage at each node of the distribution network is within a reasonable range.

[0133] Gas turbine operating constraints: These constraints are used to ensure that the output of the gas turbine is within the technical limits.

[0134] MESS operating constraints: These constraints are used to ensure that the output of MESS is within the technical limits.

[0135] SESS operating constraints: These constraints are used to ensure that the output of the SESS is within the technical limits.

[0136] Initial energy constraint for energy storage: This constraint is used to set the initial energy level of the energy storage.

[0137] MESS Access Status First Constraint: This constraint is used to ensure that the access and mobility of MESS meet the basic requirements of the power distribution network and transportation network.

[0138] Road congestion constraint: This constraint is used to reflect the impact of dynamic changes in traffic flow on the MESS movement process.

[0139] MESS travel distance and time constraints: This constraint is used to calculate the travel distance and travel time of MESS between two nodes.

[0140] Disaster disturbance phase damage model: The optimization objective is to maximize the load loss of the distribution network from the time of disaster to the time of normal operation, and the constraint is the disaster damage budget constraint.

[0141] Disaster damage budget constraint: This constraint is used to ensure that the number of power distribution lines damaged by natural disasters is within a credible probability range.

[0142] Post-disaster degradation stage scheduling model: The optimization objective is to minimize the overall operating cost of the distribution network. This includes the costs of purchasing electricity from the upstream grid, gas turbine power generation, MESS (Mechanical, Electrical, and Storage) operation, transportation costs, SESS (Support, and Service) operation, and network loss costs, as well as the cost of load shedding penalties and tie line operation. It is important to note that the RC (Responsible Controller) is composed of the distribution network's own operation and maintenance personnel, therefore their working costs are not considered. Constraints include: power purchase constraints, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, first constraint on MESS access status, road congestion constraints, MESS travel distance and time constraints, load shedding constraints, line power transmission status constraints, node status constraints, RC operating status constraints, and RC travel time constraints.

[0143] Power purchase constraint: This constraint is used to prevent the distribution network from feeding back power to the upper-level grid, while ensuring that its power purchase demand is within the power supply capacity of the upper-level grid.

[0144] Load shedding constraint: This constraint is used to ensure that the power factor of the distribution network remains unchanged.

[0145] Line power transmission status constraint: This constraint is used to ensure normal power transmission of the line while satisfying the radial topology of the distribution network.

[0146] Status constraint of nodes to be inspected: This constraint is used to reflect the status of damaged lines to be inspected, while ensuring that the lines that have been inspected are restored to normal operation.

[0147] RC operating state constraint: This constraint is used to ensure that the operation and movement of the RC meet the basic requirements of the power distribution network and transportation network.

[0148] RC travel time constraint: This constraint is used to calculate the travel time of the RC between two nodes.

[0149] Human-induced attack phase disruption model: The optimization objective is to maximize the load loss of the distribution network during the period from being attacked to returning to normal operation, and the constraint is the attack budget constraint.

[0150] Attack budget constraint: This constraint is used to ensure that the number of attacked MESS does not exceed the attack budget.

[0151] The defense and recovery phase scheduling model aims to minimize the overall operating cost of the distribution network, including the cost of purchasing electricity from the upstream grid, the cost of gas turbine power generation, the operating costs of the MESS (Mechanical Energy Separator) and transportation costs, the operating costs of the SESS (Supported Energy Separator), network loss costs, load shedding penalty costs, and the operating costs of tie lines. The operating costs of the RC (Resistant Controller) are not considered. Constraints include: power purchase constraints, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, a second constraint on MESS access status, road congestion constraints, MESS travel distance and time constraints, load shedding constraints, line power transmission status constraints, node status constraints, RC operating status constraints, and RC travel time constraints.

[0152] The second constraint of MESS access status: This constraint is used to ensure the safe operation of MESS that has not been attacked, to meet the basic requirements of the distribution network and transportation network, and at the same time to cause the attacked MESS to be taken out of operation.

[0153] The method provided by the present invention will be further illustrated below with a specific example.

[0154] See Figure 1 A multi-resource, multi-stage operation method for power distribution networks to cope with natural disasters and human attacks. Figure 2 The improved IEEE 33-node system shown is the implementation example. Its line numbers are shown in Table 1, and the parameters of the gas turbine and SESS are shown in Tables 2 and 3, respectively. The parameters of the four MESS units are identical, with active and reactive power output limits of 150kW and 120kVar, respectively, and a rated capacity of 500kWh. The active and reactive power limits of the lines are set to 5000kW and 2000kVar, respectively. The initial and upper / lower limits of the SOC of all energy storage devices are 0.5, 0.9, and 0.1, respectively, and the charge / discharge efficiency is set to 0.9. The upper and lower limits of the node voltage are set to 1.1 times and 0.9 times the per-unit value, respectively. The dispatch interval is set to 30 minutes, and the time-of-use pricing for the 48 time periods per day is shown in Table 4. The system load demand curve is shown below. Figure 3 As shown. The loss-of-load penalty cost for critical loads is ¥1000 / kW, while the loss-of-load penalty cost for ordinary loads is ¥20 / kW. The mobility cost coefficient for MESS is ¥0.6 / km, the operating cost coefficients for SESS and MESS are set at ¥0.04 / kWh and ¥0.06 / kWh respectively, the network loss cost coefficient is ¥0.005 / kWh, and the operating cost coefficient for tie lines is ¥5 / time. The power purchase limit from the distribution network to the upper-level grid after a natural disaster is set at 2500kW.

[0155] The corresponding transportation network of the system is as follows: Figure 4As shown, the distances marked are actual distances without considering congestion factors. During peak travel times, some roads will experience congestion. Table 5 lists the road sections at risk of congestion and their traffic flow at different times. The moving speed of both MESS and RC is 30 km / h. Considering the uncertainty of natural disasters, this embodiment sets 17:00-19:00 as the uncertainty interval based on weather forecast results, assuming that the disaster may occur at any time during this period (i.e., time periods 33, 34, 35, or 36). The budgets for natural disasters and man-made attacks are set to 3 and 1 respectively, the number of RCs is set to 1, and the initial positions of MESS and RCs are set at node 1. The line between nodes 1 and 2 serves as the substation outgoing line, laid in the form of cables, and is assumed to be unaffected by natural disasters. The repair time for damaged power distribution lines is set to 2 hours.

[0156] Table 1. Correspondence between line numbers and nodes

[0157] Table 2 Parameters of the gas turbine in the system

[0158]

[0159] Table 3. Parameters of SESS in the system

[0160]

[0161] Table 4 Time-of-use Electricity Prices

[0162]

[0163] Table 5 shows road sections at risk of congestion and their traffic volume at different times.

[0164]

[0165] Based on the above parameters, this embodiment proceeds according to the following steps:

[0166] Step 1: Based on the temporal relationship between natural disasters and human attacks on the power distribution network, its operation process is divided into five stages: normal operation stage, disaster disturbance stage, post-disaster degradation stage, human attack stage, and defense and recovery stage.

[0167] Step 2: Combining the schedulable resources and robust optimization ideas of each stage, establish models for five stages: normal operation stage scheduling model, disaster disturbance stage destruction model, post-disaster degradation stage scheduling model, human attack stage destruction model, and defense and recovery stage scheduling model; the objective function and constraints of each model are shown in equations (1)-(29).

[0168] Step 3: Based on the temporal relationship of the five-stage model, it is transformed into a five-level mixed integer linear programming problem. The exhaustive method and the MATLAB / YALMIP interface are used to call Gurobi 9.0.1 for calculation and solution, so as to obtain a multi-stage operation scheme for the distribution network that comprehensively considers multiple types of resources.

[0169] The dispatching results of the distribution network during normal operation are as follows: Figure 5 As shown in the figure, the distribution network relies on purchasing electricity from the upstream grid to meet most of its load demand, but it also starts gas turbine power generation. This is because, for most of the time (i.e., periods 15-48), the local marginal electricity price is significantly higher than the unit power generation cost of the gas turbines, so operators will start the gas turbines to improve overall operational economy. During periods 1-13, the local marginal electricity price is at a lower level, making it more economical to use electricity from the upstream grid; therefore, operators will shut down the gas turbines during this period. Period 14 is a transitional period between the two states. Due to ramp-up power limitations, each gas turbine operates at a lower technical output during this time so that it can directly switch to full power in the next period. Furthermore, it can be seen that although the output power of the energy storage device fluctuates, it will engage in "peak-valley arbitrage" based on the degree of change in the local marginal electricity price, i.e., charging when the electricity price is low (periods 1-14 and 23-32) and discharging when the electricity price is high (periods 15-22 and 33-40). During the period from 41 to 48, the energy storage device neither charges nor discharges, and does not participate in the optimized operation of the power system. To fully illustrate the above phenomenon, the SOC change curve of the energy storage device is plotted, as shown below. Figure 6 As shown in the figure, although the State of Charge (SOC) of each energy storage device varies across different time periods, it tends to peak around time periods 14 and 32, and trough around time periods 22 and 40. This is because electricity prices are lowest between time periods 1 and 14, prompting energy storage devices to charge during this period to prepare for the peak discharge period from time periods 15 to 22. Similarly, between time periods 23 and 32, devices also charge quickly to restore power supply capacity, preparing for the peak electricity consumption period from time periods 33 to 40. During time periods 41 to 48, the SOC of each energy storage device remains at its lower limit, as there is no further opportunity for "peak-valley arbitrage," and the energy storage devices tend to shut down.

[0170] During the disaster disturbance phase, to ensure the robustness of the model, it is necessary to generate the worst-case scenario (i.e., the scenario with the largest load loss) by combining the attack budget. The generated result is that lines 2, 18, and 20 simultaneously cease operation during time period 36. This is because during time periods 33-36, the load demand is highest in time period 36, and the SOC of each SESS and MESS is lowest. If lines 2 and 18 cease operation simultaneously at this time, the distribution network will be disconnected from the upstream grid, losing its main power source, resulting in the most severe disaster. In this context, only if any one of lines 19, 20, or 21 is simultaneously damaged will some nodes be disconnected from the main grid, forming small islands. When line 20 is damaged, the source-load difference in the small island is the largest, thus resulting in the largest load loss. For ease of description below, this scenario is defined as Scenario 1.

[0171] The distribution network topology change process, dispatch results, and SOC change curve of energy storage devices in scenario 1 of the post-disaster degradation stage are as follows: Figure 7 (a)-(f) Figure 8 and Figure 9 As shown in the figure, to enhance the resilience of the distribution network under fault scenarios and ensure economical operation as much as possible, all six gas turbines operated at maximum technical output, and all energy storage devices were also in a discharging state. During the period from the 36th to the 38th after the typhoon hit, lines 2, 18, and 20 were taken out of service due to faults, and the distribution network was disconnected from the upstream grid. Although the energy storage devices increased their output at this time, they were still unable to maintain the power supply of the entire network, resulting in a significant load loss. During the period from the 39th to the 41st, line 2 returned to normal, and the distribution network was able to re-establish contact with the upstream grid. At this time, tie line 33 was put into operation to restore the load located at nodes 21 and 22. However, since lines 18 and 20 were still in a fault state, and there was no independent power source at nodes 19 and 20, the distribution network still had a small amount of load loss. During periods 42-44, line 18 was also reactivated, ensuring connectivity with the upstream power grid at all nodes. However, the SOC of the energy storage devices had reached its lower limit and could no longer supply power. Due to power purchase constraints, the distribution network still needed to cut off some loads to maintain power balance. Considering the operating costs of the tie lines, to maximize system economy, operators ultimately chose to cut off some loads at nodes 21 and 22 during period 42 and take tie line 33 out of operation. During periods 43 and 44, the load demand of the distribution network decreased, making it capable of supplying power to the cut-off loads, so tie line 33 was reactivated. During periods 45-48, all lines returned to normal, and load demand further decreased, allowing all loads to be restored, and the distribution network switched back to normal operation.

[0172] Since Scenario 1 does not clearly demonstrate the coupling relationship between MESS, NR, and RC, this embodiment introduces Scenario 2: Lines 16, 21, and 31 simultaneously cease operation during the 36th time period to analyze the changes in distribution network topology after a natural disaster. The results are as follows: Figure 10 As shown in (a)-(h) of the figure, after the fault occurred, node 22 was disconnected, nodes 17 and 18 formed island 1, and nodes 32 and 33 formed island 2. In this context, tie line 35 was activated to restore the connectivity between node 22 and the main grid, and MESS2 started from node 31 and proceeded to node 32 to discharge. Because island 2 contains SESS2 assisting in power supply, the energy consumption rate of MESS2 is slower than that of MESS4. To achieve the best possible power balance between the two, during period 38, the operators activated tie line 36, with island 2 providing support for island 1. During period 39, line 16 returned to normal, island 1 ceased to exist, and tie line 36 was also taken out of operation. Island 2 continued to supply power to the local load under the combined action of SESS2 and MESS2. During periods 40 and 41, the power of both energy storage devices was exhausted, and the operators had to activate tie line 36 again to ensure the power supply of island 2. During the subsequent 42nd and 45th periods, lines 21 and 31 were restored to normal operation in turn, and connecting lines 35 and 36 were also gradually taken out of operation, and the distribution network was finally restored to normal operation.

[0173] The human-attack phase also employs robust optimization principles, combining budget to identify the attack scheme that maximizes the load loss in the distribution network, and determines the subsequent scheduling plan accordingly. It is assumed that a malicious attacker launches the attack immediately after the natural disaster (i.e., the 37th time period), causing a certain MESS to go offline, and its impact is not cleared until all lines damaged by the natural disaster return to normal operation. Based on this, the load loss and overall operating costs under various attack schemes for scenarios 1 and 2 are statistically analyzed, with results shown in Tables 6 and 7. As shown in the tables, the most severe attack scheme in scenario 1 is: MESS1 goes offline in the 37th time period, and node 29 no longer allows other normal MESS to connect. For ease of description, this attack scheme is defined as scenario 1-after. Considering that MESS2 is in a mobile state during the 37th time period in scenario 2, Table 7 does not calculate its attack scenarios. Therefore, the most severe attack scheme in scenario 2 is: MESS4 goes offline in the 37th time period, and node 18 no longer allows other MESS to connect. This attack scheme is defined as scenario 2-after.

[0174] Table 6 Comparison of attack schemes under Scenario 1

[0175] Table 7 Comparison of attack schemes in Scenario 2

[0176]

[0177] The topology changes of the distribution network and the SOC change curves of the energy storage device in the defense and recovery phase scenario 1-after are shown below. Figure 11 (a)-(f) and Figure 12 As shown in the figure, after MESS1 was attacked in period 37, the SOC decline rate of the remaining energy storage devices remained basically unchanged. This is mainly because the distribution network had already been disconnected from the main grid before this, and could not maintain the power supply for all loads. At this time, if the output of the remaining energy storage devices were increased rashly to make up for the power deficit caused by the shutdown of MESS1, although it could reduce the load loss of the distribution network in a short period of time, it would prematurely deplete the stored energy and enter an unavailable state, which would still cause large-scale load loss in subsequent periods, and might even cause greater economic losses due to the inability to meet the power supply needs of important loads. Therefore, the operators did not change the output plan of the remaining energy storage devices. In addition, in periods 37-44, except that MESS1 shut down and node 29 rejected the access of other MESS devices, the topology changes of the distribution network in scenario 1-after were consistent with scenario 1. In period 45 and thereafter, MESS1 resumed normal operation. It still had some remaining power, which would be discharged to node 32 to optimize the power flow distribution and reduce the overall operating cost of the distribution network.

[0178] Similarly, the topology change process of the distribution network in scenario 2-after is analyzed, such as... Figure 13As shown in (a)-(g) of the diagram, before the distribution network was attacked, lines 16, 21, and 31 were out of service, nodes 17 and 18 formed island 1, and nodes 32 and 33 formed island 2. During period 37, MESS4 went out of service and node 18 refused access from the remaining MESS nodes, causing island 1 to lose power. To minimize the load loss, operators had to activate tie line 36, allowing island 2 to support island 1. Simultaneously, MESS2 would discharge at the nearest node 32 to assist SESS2 in powering the load. Considering the high load demand of the two islands, if MESS2 continued to supply power, its SOC would quickly approach the lower limit. To avoid deteriorating its availability, during period 38, operators would dispatch MESS3 to take over the discharge. During period 39, line 16 returned to normal, and island 1 disappeared. At this point, only island 2 has a small power supply demand, which can be easily met by MESS3 while ensuring availability in subsequent periods. Therefore, the access location of MESS3 will not change. Furthermore, to minimize network losses, MESS2 will discharge at node 15 to optimize power flow distribution. During period 40, the SOC of both SESS2 and MESS3 is close to the lower limit, and operators will reactivate tie line 36 to restore connectivity between island 2 and the main network. In the subsequent period 41, MESS1 will go to node 32 to take over the load support task from MESS3, while MESS3 will discharge at node 12 to reduce network losses. Considering the large amount of energy stored in MESS1 and the cost of operating the tie line, operators will again deactivate tie line 36 to improve economic efficiency. During periods 42-44, the energy of all energy storage devices will be depleted, and operators will reactivate tie line 36 to ensure power supply to all loads. During the 45th period and thereafter, MESS4 resumes normal operation and will continue to discharge at node 18 to optimize power flow distribution.

[0179] The above analysis shows that after a high-risk, low-probability event, energy storage devices can be used as backup power to supply power to islanded loads. Compared to SESS (Self-Powered Energy Storage), MESS (Mechanical Energy Storage) is more flexible. It can discharge to the nearest node to be connected based on the island's emergency power support needs, ensuring the power supply to the load and minimizing the operating costs of tie lines caused by NR (Normally Resistant Line). For islands without gas turbines, SESS, or MESS nodes, NR can be used to restore connection with the main grid, thereby reducing load loss. The addition of RC (Resistant Grid Conversion) can quickly repair damaged lines, allowing the distribution network to gradually return to normal operation and preventing large-scale load loss after the energy stored in MESS and SESS is depleted. If a MESS is taken out of service due to a human attack, the remaining MESSs need to compensate for the power deficit as much as possible through cooperation with NR and RC, while ensuring their own subsequent power supply tasks are completed, thereby reducing the overall load loss and operating costs of the entire network. Through the cooperation of multiple types of resources, especially MESS, NR, and RC, the load loss can be effectively reduced, and the resilience of the distribution network under high-risk, low-probability events can be significantly improved.

[0180] In summary, the multi-resource, multi-stage operation method for power distribution networks proposed in this invention to cope with natural disasters and human attacks is effective and reasonable.

[0181] This invention provides a multi-resource, multi-stage operation system for power distribution networks to cope with natural disasters and man-made attacks, comprising: a computer-readable storage medium and a processor; the computer-readable storage medium is used to store executable instructions;

[0182] The processor is configured to read executable instructions stored in the computer-readable storage medium and execute the method as described in any of the above embodiments.

[0183] This invention provides a computer-readable storage medium storing computer instructions that cause a processor to perform the method described in any of the above embodiments.

[0184] Those skilled in the art will readily understand that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-resource, multi-stage operation method for power distribution networks in response to natural disasters and human-caused attacks, characterized in that, include: S1. Using gas turbines, SESS, and MESS as dispatchable resources, and aiming to minimize the first total operating cost of the distribution network, a dispatching model is established for the normal operation phase. S2, with the goal of maximizing the load loss of the distribution network during the period from the disaster to the restoration of normal operation, a disaster disturbance stage failure model is established; Using gas turbines, SESS, MESS, RC, and NR as dispatchable resources, and aiming to minimize the second total operating cost of the distribution network, a dispatch model for the post-disaster degradation stage is established. S3, with the goal of maximizing the load loss during the period from the time the distribution network is attacked to the time it returns to normal operation, establish a destruction model for the human attack phase; Using gas turbines, SESS, RC, NR, and unattacked MESS as schedulable resources, and aiming to minimize the second total operating cost of the distribution network, a defense and recovery phase scheduling model is established. S4. Based on the temporal relationships between the models established in S1-S3, the problem is transformed into a five-level mixed integer linear programming problem and solved to obtain the multi-stage operation scheme of the power distribution network. Wherein, the natural disasters only damage the power distribution lines; the man-made attacks only damage the MESS; the first total operating cost includes the cost of purchasing electricity from the upper-level power grid, the cost of generating electricity from the gas turbine, the operating cost of the MESS and transportation costs, the operating cost of the SESS and network loss costs; the second total operating cost includes the first total operating cost, the cost of load shedding penalty and the operating cost of the tie lines.

2. The method as described in claim 1, characterized in that, The objective function of the disaster disturbance stage damage model is: The constraints are: Among them, t a1 The period during which natural disasters occur, t r N represents the period during which the distribution network completes fault repair and restores normal operation; L represents the active load removed from node n during time period t; G A collection of regular routes; a l B is the variable indicating the damaged state of line l. disaster This is the upper limit for the number of lines damaged by disasters; The objective function for the destructive model during the artificial attack phase is: The constraints are: Among them, t a2 Let M be the time period during which the human attack occurred, and let M be the set of MESS values ​​for the distribution network. m B is a variable indicating the damaged state of MESSm. cyber This is the maximum number of MESS attacks initiated by human agents.

3. The method as described in claim 2, characterized in that, The objective function of the scheduling model during normal operation is: Where t0 and t end These are the start and end times of the day, respectively. Let P be the unit electricity purchase cost of the distribution network from the upper-level power grid during time period t. t grid α represents the active power supplied by the upstream power grid during time period t; I and S represent the sets of gas turbines and SESS in the distribution network, respectively; i Let be the fuel cost coefficient for the i-th gas turbine. w represents the active power output of the i-th gas turbine during time period t; m and w s These are the operating cost coefficients for MESS and SESS, respectively. and These represent the charging and discharging power of MESSm during time period t. and denoted as SESS charge / discharge power during time period t; γ is the mobility cost coefficient of MESS. Let t be the distance traveled by MESSm from node n to node p during time period t; β is the network loss cost coefficient, Pf l,t The active power on line l during time period t; The constraints include: prohibition of power backfeeding, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, initial energy constraints for energy storage, first constraint for MESS access status, road congestion constraints, and MESS travel distance and time constraints.

4. The method as described in claim 3, characterized in that, The objective function of the post-disaster degradation stage scheduling model is: Among them, L C δ represents the set of interconnecting lines in the distribution network; δ is the load loss penalty cost coefficient, which is related to the importance of the load; ζ is the operating cost coefficient of the interconnecting lines. and These are the forward and reverse power transmission status flag variables for line l during time period t; The constraints include: power purchase constraints, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, MESS access status first constraint, road congestion constraints, MESS travel distance and time constraints, load shedding constraints, line power transmission status constraints, node status constraints, RC operating status constraints, and RC travel time constraints.

5. The method as described in claim 3 or 4, characterized in that, The first constraint of the MESS access status is: Where, N CM For the set of accessible nodes in MESS; u m,n,t v is the connection status flag variable between MESSm and the accessible node n during time period t. m,t Let MESSm be the movement status flag variable for time period t; and These are the charging and discharging status flag variables of MESSm during time period t; Let MESSm be the time required for MESSm to travel from node n to node p during time interval t. The road congestion constraints are as follows: Among them, D n,p,t χ represents the travel distance between node n and node p during time period t, considering congestion. n,p,t Let be the congestion coefficient of the road between node n and node p in time period t. ρ represents the original distance of the road between node n and node p. n,p,t Let κ represent the traffic flow on the road between node n and node p during time period t. n,p Let n be the traffic congestion limit for the road between node n and node p; The MESS travel distance and time constraints are as follows: Among them, Vel m Let MESSm be the speed of travel, and Δt be the duration of a unit time period.

6. The method as described in claim 4, characterized in that, The objective function of the defense recovery phase scheduling model is: The constraints include: power purchase constraints, node power balance constraints, line capacity constraints, power flow constraints, voltage constraints, gas turbine operation constraints, MESS operation constraints, SESS operation constraints, MESS access status second constraint, road congestion constraints, MESS travel distance and time constraints, load shedding constraints, line power transmission status constraints, node status constraints, RC operating status constraints, and RC travel time constraints.

7. The method as described in claim 4 or 5, characterized in that, The RC operating state constraint is as follows: Where, N CR Let c be the set of nodes to be RC inspected. r,n,t b is a variable indicating the working state of RCr near node n during time period t; r,t Let R be the movement state flag variable of RCr during time period t; R is the set of RCs in the distribution network. The maintenance time required near node n; Let RCr be the time required for node n to travel to node p during time interval t. The RC travel time constraint is: Among them, Vel r RCr represents the movement speed.

8. A multi-resource, multi-stage operation system for power distribution networks to cope with natural disasters and man-made attacks, characterized in that, include: Computer-readable storage media and processors; The computer-readable storage medium is used to store executable instructions; The processor is configured to read executable instructions stored in the computer-readable storage medium and execute the method as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a processor to perform the method as described in any one of claims 1-7.

Images