Power-Transportation Coupled Charging Network Resilience Assessment Method and System

Abstract

The present invention provides a method and system for evaluating the resilience of a power-transportation coupled charging network, belonging to the technical field of power system analysis. In the method, by constructing a power-transportation coupled charging network, combining the Bezier curve to simulate the typhoon path and the Holland wind field model to calculate the wind speed, establishing a dynamic traffic flow distribution, an AC optimal power flow and a cascading fault model, using Monte Carlo simulation to generate a fault data set, and training a fault prediction model through a dynamic Bayesian network, after inputting the measured wind speed, a dynamic probability map is output, and a dynamic resilience index is calculated to evaluate the resilience of the power-transportation coupled charging network. The present invention considers the problems of insufficient fault propagation and dynamic evaluation between systems, and improves the accuracy of the resilience evaluation of the power-transportation coupled charging network under typhoon disasters.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system analysis, and particularly relates to a method and system for evaluating the resilience of a power-transportation coupled charging network. Background Art

[0002] Power and transportation are two major cores of the urban infrastructure system. Urban distribution network lines are at the end of the main power grid. The network scale is huge, there are numerous lines and a large number of them are installed outdoors, making them extremely vulnerable to the impact of typhoon disasters. After being affected, large-scale equipment failures and power outages may occur, directly affecting residents' production and life. With the development of electric vehicles in recent years, the traffic flow of electric vehicles in the urban transportation system has gradually increased, and electric vehicle charging stations have also developed rapidly, resulting in an increasingly close coupling relationship between the transportation system and the power system, making the power system more sensitive to the impact of extreme events. Therefore, improving the resilience of the power-transportation coupled charging network to extreme events is one of the keys to ensuring the safe and stable development of the city.

[0003] Currently, research on the resilience of the transportation system focuses on the road network in which various types of vehicles are involved. Taking electric vehicles as the main body, there is no comprehensive theoretical research on the impact of the redistribution behavior of traffic load after the failure of electric vehicle charging stations. The research on the resilience evaluation of the power-transportation coupled charging network only considers the connection between single systems and does not consider the propagation problem of failures between different systems. At the same time, the existing resilience evaluation of the power-transportation coupled system has static limitations and cannot reasonably evaluate the dynamic ability of the power-transportation coupled system to resist typhoon disasters, making the existing evaluation methods lack dynamic characteristics. Summary of the Invention

[0004] The present invention provides a method and system for evaluating the resilience of a power-transportation coupled charging network to solve the problems in the prior art that only the connection between single systems is considered, the propagation problem of failures between different systems is not considered, and the dynamic ability of the power-transportation coupled system to resist typhoon disasters cannot be reasonably evaluated.

[0005] In the first aspect, the present invention provides a method for evaluating the resilience of a power-transportation coupled charging network, including:

[0006] Construct a power-transportation coupled charging network according to the power system, transportation system, and electric vehicle charging stations in the target area;

[0007] According to the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial moment, use the Bessel curve to simulate the random path of the typhoon operation, and determine the wind field wind speed of each node on all transmission lines at different times according to the Holland wind field model to construct the initial fault model of the power system transmission line;

[0008] Construct a dynamic traffic flow allocation model based on cellular transmission according to the time-coupled flow propagation characteristics of traffic allocation and the redistribution principle of electric vehicles;

[0009] Construct an AC optimal power flow convergence model according to the bus voltage phase angle, voltage amplitude, generator active power, and generator reactive power of each transmission line in the power system;

[0010] Construct a cascading fault model according to the dynamic traffic flow allocation model, the AC optimal power flow convergence model, and the situation of charging stations out of service;

[0011] According to the initial fault model and the cascading fault model, for each node of each transmission line, determine the wind speed data set of each transmission line at different times based on the historical typhoon wind speed, and determine the fault state data set of each transmission line and each charging station according to the Monte Carlo simulation method;

[0012] Construct a fault prediction model based on a dynamic Bayesian network, and use the wind speed data set, fault state data set of each transmission line, and fault state data set of each charging station as inputs, and the conditional probability parameters of the dynamic Bayesian network model as outputs for training to obtain a trained fault prediction model;

[0013] Input the measured typhoon wind speed into the trained fault prediction model to obtain the dynamic probability map of the fault states of each transmission line and each charging station, determine the dynamic resilience index of the power-traffic coupled charging network, and evaluate the resilience of the power-traffic coupled charging network according to the dynamic resilience index.

[0014] Optionally, the constructing of the initial fault model of the transmission lines of the power system according to the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial moment, using a Bessel curve to simulate the random path of the typhoon operation, and determining the wind field wind speed affected by the typhoon for each node on all transmission lines at different times according to the Holland wind field model, includes:

[0015] Calculate the power outage probability of each transmission line at the target moment according to the following formula:

[0016] ;

[0017] ;

[0018] ;

[0019] ;

[0020] ;

[0021] where P L(H L (t)) is the power outage probability of the transmission line L at time t; H L (t) is the wind field wind speed of the transmission line L affected by the typhoon at time t; is the maximum wind resistance level of the transmission line L; exp(·) is the exponential function with the natural constant e as the base; V max is the maximum wind speed of the typhoon; R max is the maximum influence radius of the typhoon; B is the order of the Bessel curve; r(t) is the distance of the target transmission line from the typhoon center at time t; BH is the Holland coefficient; x(t) is the abscissa of the typhoon on the geographic coordinate map at time t; x b is the abscissa of the randomly generated intermediate control point; y(t) is the ordinate of the typhoon on the geographic coordinate map at time t; y b is the ordinate of the randomly generated intermediate control point; t b is the b-th power of t; b is the number of the path control points on the Bessel curve; represents the combination number operation;! represents the factorial operator.

[0022] Optionally, according to the initial fault model and the cascading fault model, for each node of each transmission line, the wind speed data set of each transmission line at different times is determined based on the historical typhoon wind speed, and the fault state data set of each transmission line and each charging station is determined according to the Monte Carlo simulation method, including:

[0023] Obtain the wind field wind speed of the target transmission line affected by the typhoon at the target time;

[0024] Determine the power outage probability of the target transmission line at the target time according to the wind field wind speed of the target transmission line affected by the typhoon at the target time;

[0025] Use the Monte Carlo simulation method to generate a random number of the power outage probability of the target transmission line at the target time;

[0026] In the case that the random number is greater than or equal to the power outage probability, it is determined that the target transmission line fails at the target time; in the case that the random number is less than the power outage probability, it is determined that the target transmission line does not fail at the target time; traverse all transmission lines to obtain the fault state data set of the transmission lines in a single typhoon accident;

[0027] In the case that all the transmission lines connected to the target charging station fail, it is determined that the target charging station fails; traverse all charging stations to obtain the fault state data set of the charging stations in a single typhoon accident;

[0028] In the case of at least one charging station failure, the cascading fault model is called. If the power system converges and the cascading fault model does not trip other transmission lines, go to the next moment and re-obtain the power-off probability of the target transmission line at the target moment. If the power system converges and the cascading fault model trips other transmission lines, update the transmission line fault status dataset and the charging station fault status dataset until the current moment is the preset total number of moments, stop updating, and obtain the final transmission line fault status dataset and the charging station fault status dataset.

[0029] Optionally, constructing a fault prediction model based on a dynamic Bayesian network and training it with the wind speed dataset, fault status dataset of each transmission line, and fault status dataset of each charging station as inputs and the conditional probability parameters of the dynamic Bayesian network model as outputs to obtain a trained fault prediction model, including:

[0030] Construct a fault prediction model G based on a dynamic Bayesian network B Expression:

[0031] ;

[0032] ;

[0033] ;

[0034] ;

[0035] ;

[0036] ;

[0037] ;

[0038] Where, N B is the set of power system nodes described by the Bayesian network; L B is the set of transmission lines of the power system described by the Bayesian network; θ B is the conditional probability distribution of the power system nodes described by the Bayesian network; t = 0, 1, 2,..., T; T is the preset total number of moments; S L is the fault status of transmission line L; S L = 1 indicates that transmission line L is normal; S L = 0 indicates that transmission line L is faulty; S c is the fault status of charging station c; S c = 1 indicates that charging station c is normal; S c = 0 indicates that charging station c is faulty; c = 1, 2,..., C'; C' is the total number of charging stations; H LIndicates the wind speed state of the transmission line L at the total number of preset time instants T; H L = 0 indicates the tropical depression wind speed; H L = 1 indicates the tropical storm and severe tropical storm wind speeds; H L = 2 indicates the typhoon, severe typhoon and super typhoon wind speeds; L = 1, 2, ..., L'; L' is the total number of transmission lines in the power system; is the wind speed state of the transmission line L at time t; is the fault state of the transmission line L at time t; is the fault state of the transmission line L at time t; P(·) is the conditional probability of the nodes in the power system described by the Bayesian network;

[0039] Construct the expression of the set D of the wind speed data set, fault state data set of each transmission line and fault state data set of each charging station:

[0040] ;

[0041] where, H L' is the wind speed data set of all transmission lines; S L' is the fault state data set of all transmission lines; S C' is the fault state data set of all charging stations; q is the number of typhoon accident simulations;

[0042] Using the set D as the input and the conditional probability parameters of the dynamic Bayesian network model as the output, the maximum a posteriori estimation method and the EM parameter learning method are used to train the fault prediction model G B to obtain the trained fault prediction model.

[0043] Optionally, input the measured typhoon wind speed into the trained fault prediction model to obtain the dynamic probability map of the fault states of each transmission line and each charging station, so as to determine the dynamic resilience index of the power-transportation coupled charging network, and evaluate the resilience of the power-transportation coupled charging network according to the dynamic resilience index, including:

[0044] Calculate the resilience index of the target charging station at the target time according to the following formula:

[0045] ;

[0046] where, R c (t) is the resilience index of the charging station c at time t; P c (S t = 0) represents the probability that the charging station c fails at time t; S t = 0 indicates that the state value at time t is 0; T is the total number of preset time instants;

[0047] Calculate the resilience index of the power-transportation coupled charging network at different times according to the following formula:

[0048] ;

[0049] where R NET (t) is the resilience index of the power-transportation coupled charging network at time t; w c is the weight of the resilience index of charging station c determined by the entropy weight method; C' is the total number of charging stations.

[0050] In a second aspect, the present invention provides a power-transportation coupled charging network resilience evaluation system, including:

[0051] A first construction module for constructing a power-transportation coupled charging network according to the power system, transportation system, and electric vehicle charging stations in the target area;

[0052] A second construction module for simulating the random path of typhoon operation using Bessel curves according to the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial time, and determining the wind field wind speed affected by the typhoon at each node on all transmission lines at different times according to the Holland wind field model, so as to construct an initial fault model of the power system transmission line;

[0053] A third construction module for constructing a dynamic traffic flow distribution model based on cellular transmission according to the time-coupled flow propagation characteristics of traffic distribution and the electric vehicle reallocation principle;

[0054] A fourth construction module for constructing an AC optimal power flow convergence model according to the bus voltage phase angle, voltage amplitude, generator active power, and generator reactive power of each transmission line of the power system;

[0055] A fifth construction module for constructing a cascading fault model according to the dynamic traffic flow distribution model, the AC optimal power flow convergence model, and the situation of charging stations stopping operation;

[0056] A first determination module for determining the wind speed data set of each transmission line at different times with the historical typhoon wind speed for each node of each transmission line according to the initial fault model and the cascading fault model, and determining the fault state data set of each transmission line and each charging station according to the Monte Carlo simulation method;

[0057] A sixth construction module for constructing a fault prediction model based on a dynamic Bayesian network, and training with the wind speed data set of each transmission line, the fault state data set, and the fault state data set of each charging station as inputs and the conditional probability parameters of the dynamic Bayesian network model as outputs to obtain a trained fault prediction model;

[0058] The second determination module is configured to input the measured typhoon wind speed into the trained fault prediction model to obtain the dynamic probability diagrams of the fault states of each transmission line and each charging station, determine the dynamic resilience index of the power-transportation coupled charging network, and evaluate the resilience of the power-transportation coupled charging network according to the dynamic resilience index.

[0059] Optionally, the second construction module includes:

[0060] The first calculation unit is configured to calculate the power-off probability of each transmission line at the target time according to the following formula:

[0061] ;

[0062] ;

[0063] ;

[0064] ;

[0065] [[ID=2D]] ;

[0066] where P L (H L (t)) is the power-off probability of transmission line L at time t; H L (t) is the wind field wind speed of transmission line L affected by the typhoon at time t; is the maximum wind resistance level of transmission line L; exp(·) is the exponential function with the natural constant e as the base; V max is the maximum wind speed of the typhoon; R max is the maximum influence radius of the typhoon; B is the order of the Bessel curve; r(t) is the distance of the target transmission line from the typhoon center at time t; BH is the Holland coefficient; x(t) is the abscissa of the typhoon on the geographic coordinate map at time t; x b is the randomly generated abscissa of the intermediate control point; y(t) is the ordinate of the typhoon on the geographic coordinate map at time t; y b is the randomly generated ordinate of the intermediate control point; t b is the b-th power of t; b is the number of the path control points on the Bessel curve; represents the combination number operation;! represents the factorial operator.

[0067] Optionally, the first determination module includes:

[0068] The acquisition unit is configured to acquire the wind field wind speed of the target transmission line affected by the typhoon at the target time;

[0069] The first determination unit is configured to determine the power outage probability of the target transmission line at the target moment according to the wind field wind speed of the target transmission line affected by the typhoon at the target moment;

[0070] The random number generation unit is configured to generate a random number of the power outage probability of the target transmission line at the target moment by using the Monte Carlo simulation method;

[0071] The second determination unit is configured to determine that the target transmission line fails at the target moment when the random number is greater than or equal to the power outage probability; and determine that the target transmission line does not fail at the target moment when the random number is less than the power outage probability; traverse all transmission lines to obtain a dataset of the fault status of the transmission lines in a single typhoon accident;

[0072] The third determination unit is configured to determine that the target charging station fails when all the transmission lines connected to the target charging station fail; traverse all charging stations to obtain a dataset of the fault status of the charging stations in a single typhoon accident;

[0073] The fourth determination unit is configured to, when at least one charging station fails, call a cascading fault model. If the power system converges and the cascading fault model does not disconnect other transmission lines, proceed to the next moment and re-obtain the power outage probability of the target transmission line at the target moment; if the power system converges and the cascading fault model disconnects other transmission lines, update the dataset of the fault status of the transmission lines and the dataset of the fault status of the charging stations until the current moment reaches the preset total number of moments, stop the update, and obtain the final dataset of the fault status of the transmission lines and the dataset of the fault status of the charging stations.

[0074] Optionally, the sixth construction module includes:

[0075] The first construction unit is configured to construct an expression of a fault prediction model G B based on a dynamic Bayesian network:

[0076] ;

[0077] ;

[0078] ;

[0079] ;

[0080] ;

[0081] ;

[0082] ;

[0083] where N BThe set of power system nodes described by the Bayesian network; L B The set of transmission lines in the power system described by the Bayesian network; θ B The conditional probability distribution of the power system nodes described by the Bayesian network; t = 0, 1, 2,..., T; T is the total number of preset time instants; S L The fault state of the transmission line L; S L = 1 indicates that the transmission line L is normal; S L = 0 indicates that the transmission line L is faulty; S c The fault state of the charging station c; S c = 1 indicates that the charging station c is normal; S c = 0 indicates that the charging station c is faulty; c = 1, 2,..., C'; C' is the total number of charging stations; H L Indicates the wind speed state of the transmission line L at the total number of preset time instants T; H L = 0 indicates tropical depression wind speed; H L = 1 indicates tropical storm and severe tropical storm wind speed; H L = 2 indicates typhoon, severe typhoon and super typhoon wind speed; L = 1, 2,..., L'; L' is the total number of transmission lines in the power system; The wind speed state of the transmission line L at time t; The fault state of the transmission line L at time t; The fault state of the transmission line L at time t; P(·) is the conditional probability of the power system nodes described by the Bayesian network;

[0084] The second construction unit is used to construct the expression of the set D of the wind speed data set, the fault state data set of each transmission line, and the fault state data set of each charging station:

[0085] ;

[0086] where H L' Is the wind speed data set of all transmission lines; S L' Is the fault state data set of all transmission lines; S C' Is the fault state data set of all charging stations; q is the number of typhoon accident simulations;

[0087] The model training unit is used to take the set D as the input and the conditional probability parameters of the dynamic Bayesian network model as the output, and use the maximum a posteriori estimation method and the EM parameter learning method to train the fault prediction model G B To obtain the trained fault prediction model.

[0088] Optionally, the second determination module includes:

[0089] A second calculation unit, configured to calculate a resilience index of a target charging station at a target time according to the following formula:

[0090] ;

[0091] wherein, R c (t) is the resilience index of charging station c at time t; P c (S t = 0) represents the probability of failure of charging station c at time t; S t = 0 indicates that the state value at time t is 0; T is the total number of preset times;

[0092] A third calculation unit, configured to calculate a resilience index of the power-transportation coupled charging network at different times according to the following formula:

[0093] ;

[0094] wherein, R NET (t) is the resilience index of the power-transportation coupled charging network at time t; w c is the weight of the resilience index of charging station c determined by the entropy weight method; C' is the total number of charging stations.

[0095] The present invention provides a method and a system for evaluating the resilience of a power-transportation coupled charging network. The method comprehensively considers the coupling relationship between the power network, the transportation network, and the charging network, and can integrate the interdependent systems into a whole for studying their characteristics; the cascade failure propagation mechanism between the coupled systems is complex. The present invention can accurately analyze the secondary impact of the transportation system on the power system through a cascade failure model considering the dynamic traffic flow redistribution, considering the constraints of the bus voltage and the safety constraints of the line power flow; based on the dynamic Bayesian network, the present invention studies the operation changes of the coupled system in the case of encountering disturbances or failures, and can accurately predict the possible failure situations at different times. BRIEF DESCRIPTION OF THE DRAWINGS

[0096] In order to more clearly illustrate the technical solutions of the present invention, the drawings required for the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0097] Figure 1 is a schematic flowchart of a method for evaluating the resilience of a power-transportation coupled charging network provided by an embodiment of the present invention;

[0098] Figure 2 is a schematic diagram of a fault simulation of a typhoon on a power distribution network provided by an embodiment of the present invention;

[0099] Figure 3 The figure of the prediction result of the dynamic power outage probability of the transmission line under the low - voltage wind speed level provided by the embodiment of the present invention;

[0100] Figure 4 The figure of the prediction result of the dynamic power outage probability of the transmission line under the storm wind speed level provided by the embodiment of the present invention;

[0101] Figure 5 The figure of the prediction result of the dynamic power outage probability of the transmission line under the severe typhoon wind speed level provided by the embodiment of the present invention;

[0102] Figure 6 The figure of the prediction result of the power outage probability of the charging station under the low - voltage wind speed level provided by the embodiment of the present invention;

[0103] Figure 7 The figure of the prediction result of the power outage probability of the charging station under the storm wind speed level provided by the embodiment of the present invention;

[0104] Figure 8 The figure of the prediction result of the power outage probability of the charging station under the severe typhoon wind speed level provided by the embodiment of the present invention;

[0105] Figure 9 The schematic diagram of the resilience evaluation calculation provided by the embodiment of the present invention;

[0106] Figure 10 The curve graph of the change of the resilience of the charging network provided by the embodiment of the present invention;

[0107] Figure 11 The schematic diagram of the structure of a power - traffic coupled charging network resilience evaluation system provided by the embodiment of the present invention. Specific embodiments

[0108] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0109] Embodiment 1

[0110] As Figure 1 shown, the embodiment of the present invention provides a method for evaluating the resilience of a power - traffic coupled charging network, including:

[0111] Step 101, construct a power - traffic coupled charging network according to the power system, traffic system and electric vehicle charging stations in the target area.

[0112] In this step, exemplarily, a power-transportation coupled charging network is constructed according to Equations (1)-(10).

[0113] G r (V r ,E r ,W r )(1)

[0114] G e (V e ,E e ,W e )(2)

[0115] G p (V p ,E0] p ,W p )(3)

[0116] G(V,E,W)(4)

[0117] (5)

[0118] (6)

[0119] (7)

[0120] (8)

[0121] (9)

[0122] (10)

[0123] In Equation (1), G r represents the transportation system (i.e., the transportation network); V r represents the set of nodes of the transportation system; E r represents the set of roads of the transportation system; W r represents the entropy matrix of the transportation system. Each road in the transportation system corresponds to an entropy weight in the entropy matrix, and each entropy weight is the time required for the corresponding road; in Equation (2), G e represents the charging station network; V e represents the set of charging stations; E e represents the set of roads of the charging station network; W e represents the entropy matrix of the charging station network, and the entropy weights in the entropy matrix are expressed as ; in Equation (3), G p represents the power system; V p represents the set of nodes of the power system; E p represents the set of lines of the power system; W prepresents the entropy matrix of the power system, and each entropy weight in the entropy matrix is expressed as a percentage of the remaining capacity of the power system; in Equation (4), G represents the power-transportation charging coupling network; V represents the set of nodes in the power-transportation coupling charging network, which consists of V e and V p ; E represents the set of lines in the power-transportation coupling network, which consists of E r , E e and E c , where E c is the line coupling link; W represents the entropy matrix of the power-transportation coupling network, which consists of W p , W e and W c , where W c is the entropy matrix of the coupling link; Equations (5) and (6) represent the road impedance function; t r represents the road equivalent impedance of road r; represents the free driving time of road r; c r is the traffic capacity of road r; f r is the traffic flow of road r; is the normalized representation of t r ; and represent the minimum and maximum values of t r respectively; in Equation (7), R mn represents the equivalent resistance between nodes m and n in the charging station network; is the Laplacian matrix of the power-transportation charging coupling network; , and are all elements in the matrix ; e m and e n represent the m-th and n-th standard vectors of the Laplacian matrix respectively; in Equation (8) represents the available capacity of line d in the power network; I d represents the load percentage of line d; Equation (9) represents the entropy matrix of the power-transportation coupling charging network; is the transpose matrix of the entropy matrix of the coupling link; in Equation (10) represents the remaining available traffic capacity of link c1 in the power-transportation coupling charging network; φ max is the maximum charging power of the charging station; φ k' is the actual charging power.

[0124] Step 102: According to the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial moment, use the Bessel curve to simulate the random path of the typhoon, and determine the wind field wind speed affected by the typhoon at each node on all transmission lines at different times according to the Holland wind field model, so as to construct the initial fault model of the transmission lines of the power system.

[0125] The impact of typhoon disasters on the power system is mainly reflected in the strong wind weather brought by the typhoon, which will cause direct damage to the transmission lines of the power system. Using the Bessel curve to simulate the random movement path of the typhoon and the Holland wind field model can simulate the wind speed change of the typhoon, and further calculate the impact of different typhoons on each line at different times.

[0126] For the transmission lines of the power system in the target area, it is necessary to strictly calculate the wind speed impact on the line area at different times according to the actual distance between each line and the nodes on each line.

[0127] The deviation of the typhoon's movement path in a city is not too large. Therefore, when using the Bessel curve mathematical model to simulate the random path of the typhoon in the urban distribution network, the order cannot be too high; most typhoons in China land in coastal areas. Therefore, the selection of the B coefficient in the Holland model used to calculate the wind speed impact on the line area at different times should conform to the actual situation in coastal areas.

[0128] Exemplarily, calculate the power-off probability of each transmission line at the target moment according to the following formula:

[0129] (11)

[0130] (12)

[0131] (13)

[0132] (14)

[0133] (15)

[0134] Among them, P L (H L (t)) is the power-off probability of transmission line L at time t; H L (t) is the wind field wind speed of transmission line L affected by the typhoon at time t; is the maximum wind resistance level of transmission line L; exp(·) is the exponential function with the natural constant e as the base; V max is the maximum wind speed of the typhoon; R maxis the maximum influence radius of the typhoon; B is the order of the Bessel curve; r(t) is the distance between the target transmission line and the typhoon center at time t; BH is the Holland coefficient, which is used to represent the shape of the typhoon wind field and indirectly characterize the typhoon influence range and intensity; x(t) is the abscissa of the typhoon on the geographic coordinate map at time t; x b is the abscissa of the randomly generated intermediate control point; y(t) is the ordinate of the typhoon on the geographic coordinate map at time t; y b is the ordinate of the randomly generated intermediate control point; t b is the b-th power of t; b is the number of the path control points on the Bessel curve; represents the combination number operation;! represents the factorial operator; in this embodiment, the order of the Bessel curve is selected as the fifth order, and the Holland coefficient is selected as 1.9, which characterizes the coastal typhoon wind field characteristics.

[0135] Step 103: Construct a dynamic traffic flow distribution model based on cell transmission according to the time-coupled flow propagation characteristics of traffic assignment and the electric vehicle redistribution principle.

[0136] In this step, when constructing a dynamic traffic flow distribution model based on cell transmission, the dynamic system optimal scheme should be used, that is, the user participants need to obey the dispatcher's command to minimize the total travel cost of all users, so as to determine the traffic flow characteristics after the charging station stops operating.

[0137] The construction formula of the cell transmission model is as shown in equations (16)-(26):

[0138] (16)

[0139] (17)

[0140] (18)

[0141] (19)

[0142] (20)

[0143] (21)

[0144] (22)

[0145] (23)

[0146] (24)

[0147] (25)

[0148] (26)

[0149] Equations (16) - (19) are the mathematical models of ordinary cells. Equation (16) represents the number of vehicles in cell i at time t + 1 equal to the number of vehicles in cell i at time t plus the number of incoming vehicles in cell i - 1 , and then minus the number of outgoing vehicles in cell i ; Equation (17) represents the inflow at cell i at time t depending on the smaller value between the sending volume of cell i at time t and the receiving volume of cell i + 1 ; Equation (18) represents the sending volume of cell i at time t depending on the smaller value between the number of vehicles in cell i at the current time and the outflow capacity ; Equation (19) represents the receiving volume of cell i + 1 at time t depending on the smaller value between the outflow capacity of cell i + 1 at the current time and the remaining carrying capacity ; w is the traffic flow congestion wave speed; v f is the free flow speed; is the vehicle occupancy rate of cell i + 1 at time t; is the number of vehicles in cell i + 1 at time t; Equations (20) - (22) are the mathematical models of the shunt cell, where α is the shunt coefficient, set externally; in Equation (20), represents the flow transmitted from cell i to cell j at time t; in Equation (21), represents the flow transmitted from cell i to cell k; Equation (22) represents the change in the calculation formula of the cell flow after the shunt coefficient; represents the number of outgoing vehicles in cell i; represents the sending volume of cell i; represents the receiving volume of cell j; represents the receiving volume of cell k; Equations (23) - (26) are the mathematical models of the confluence cell, where β is the confluence coefficient, which is the set value of the externally input parameter; Equations (23) - (24) represent and the number of outgoing vehicles needs to be less than the receiving volume of cell i; represents the flow transmitted from cell j to cell i at time t; represents the sending volume of cell j at time t; represents the flow transmitted from cell k to cell i at time t; represents the sending volume of cell k at time t; represents the acceptance amount of cell i at time t; Equation (25) represents that cell i is in an unobstructed state when the sum of the sending amounts of cells j and k is less than the acceptance amount of cell i; represents the number of vehicles in cell j at time t; represents the number of vehicles in cell k at time t; represents the inflow capacity of cell j at time t; represents the inflow capacity of cell k at time t; Equation (26) represents that cell i is in a congested state when the sum of the sending amounts of cells j and k is greater than the acceptance amount of cell i; represents the vehicle occupancy rate of cell i at time t.

[0150] The charging cell is defined as the section connected to the charging pile and is a virtual section. To prevent congestion in the charging cell, it is necessary to consider the queuing and non-queuing situations to describe the electric vehicles entering or leaving the charging cell. Assume that the required charging time of the electric vehicle in charging cell i is T c and the number of charging vehicles in the charging cell at time t is NC i (t) is the maximum number of charging vehicles in the charging cell at time t, and the charging cell constraints are as follows:

[0151] (27)

[0152] (28)

[0153] (29)

[0154] (30)

[0155] (31)

[0156] (32)

[0157] (33)

[0158] (34)

[0159] (35)

[0160] (36)

[0161] (37)

[0162] (38)

[0163] (39)

[0164] (40)

[0165] Among them, C C is the set of charging cells; R w is the set of all paths in OD pair (origin and destination, start - end point) w; T s is the charging start time; T f is the transition time from non - queuing state to queuing state; Equation (27) means that in the non - queuing case, the vehicle occupancy rate of charging cell i is not greater than the maximum number of charging vehicles in the charging cell at time t; Equation (28) means that in the non - queuing case, the vehicle flow rate flowing into the charging cell at time T s is equal to the vehicle flow rate flowing out of the charging cell at time T s +T c ; is the number of vehicles flowing from cell k to cell i through path r without charging state demand; is the number of vehicles flowing from cell k to cell i through path r with charging state demand; Equation (29) means that in the queuing case, the vehicle occupancy rate of charging cell i is not less than the maximum number of charging vehicles in the charging cell at time t; Equation (30) means that in the queuing case, the number of vehicles that can be charged at time T s is equal to the vehicle flow rate flowing out of the charging cell at time T s +T c ; is the number of vehicles flowing from cell k to cell i through path r without charging state demand; Equation (31) means that when transitioning from non - queuing to queuing, the vehicle occupancy rate of charging cell i is not greater than the maximum number of charging vehicles in the charging cell at time t; Equation (32) means that when transitioning from non - queuing to queuing, the vehicle occupancy rate of charging cell i is not less than the maximum number of charging vehicles in the charging cell at time t; Equation (33) means that the vehicle flow rate flowing out of the charging cell at time T s +T c is equal to the number of vehicles that can just be charged at the charging start time T s ; Equations (34) and (35) are the constraints on vehicle occupancy rate when transitioning from queuing to non - queuing; Equation (36) means that the vehicle flow rate flowing out of the charging cell at time T s +T c is equal to the number of vehicles that can just be charged at the charging start time T s ; Equation (37) means that the vehicles flowing into the charging cell have charging demands, while the vehicles flowing out of the charging cell have no charging demands; Equation (38) means that the vehicle flow of charging station i is conserved; Equation (39) means that the inflow and outflow should be less than the maximum passing capacity of charging cell i; Q i(t) is the acceptance capacity of charging cell i at time t; T is the total number of preset times; Equation (40) indicates that there is no vehicle with a charging demand in end cell i.

[0166] Expand the cell transmission model into a dynamic traffic assignment model, and make the merging and diverging coefficients not affect the optimality of the dynamic traffic flow. Introduce charging cells to capture the charging load of electric vehicles, and at the same time satisfy the accurate simulation of traffic flow at the path level. Construct the dynamic traffic assignment model according to Equations (41)-(55).

[0167] (41)

[0168] (42)

[0169] (43)

[0170] (44)

[0171] (45)

[0172] (46)

[0173] (47)

[0174] (48)

[0175] (49)

[0176] (50)

[0177] (51)

[0178] (52)

[0179] (53)

[0180] (54)

[0181] (55)

[0182] Among them, in Equation (41), f1 is the objective function of the dynamic traffic flow model, which is to minimize the total travel time of all vehicles, and x i,a (t) represents the vehicle occupancy rate of cell i at time t in state a; a represents whether there is a charging demand. a = 1 indicates that the vehicle has a charging demand, and a = 0 indicates that the vehicle has no charging demand; T is the total number of preset times; C is the total number of ordinary cells; C sis the total number of charging cells; Equation (42) is the demand constraint, indicating that the traffic demand on each path within the departure time is equal to the total OD demand; represents the demand for path r at time t in state a; W is the set of road OD pairs; T d is the maximum departure time; D w (t) is the flow demand constraint of the transportation network; Equations (43)-(55) are the constraints of ordinary cells, where Equations (43) and (44) represent the cell occupancy rate and the inflow and outflow capabilities at the aggregation level; is the number of vehicles of cell i on path r in state a at time t; y (i,j)a (t) represents the number of inflowing vehicles from cell i to cell j at time t in state a; is the number of inflowing vehicles from cell i to cell j on path r at time t in state a; Equations (45)-(46) respectively represent the flow balance of the starting cell C R at the departure time and the non-departure time; is the demand for cell i on path r at time t-1 in state a; Equation (47) is the flow conservation expression of ordinary cells; C\{C R ,C s} represents the set of all cells excluding the starting cell and the charging cells; Γ -1 (i) represents the set of the previous cells of cell i; represents the number of vehicles from cell k to cell i passing through path r at time t-1 in state a; Γ(i) represents the set of the next cells of cell i; represents the number of vehicles from cell i to cell j passing through path r at time t-1 in state a; Equation (48) indicates that the traffic flow from cell i to cell j does not exceed the number of vehicles of cell i at time t; Equations (49)-(50) indicate that the inflow and outflow of cell i at time t do not exceed the maximum passing capacity of the link; Q i (t) is the acceptance capacity of cell i at time t; Equation (51) indicates that when there is traffic congestion, the total inflow capacity of cell i does not exceed the remaining occupancy rate of the cell; represents the number of vehicles from cell k to cell i passing through path r at time t; δ i is the blocking density of cell i; N i (t) represents the maximum capacity of cell i; is the number of vehicles of cell i at time t; Equations (52)-(53) are the initial values of the cell occupancy rate and the link flow; Equations (54)-(55) are the non-negativity constraints of the cell occupancy rate and the link flow.

[0183] Step 104: Construct an AC optimal power flow convergence model according to the bus voltage phase angle, voltage amplitude, generator active power, and generator reactive power of each transmission line in the power system.

[0184] In this step, an AC optimal power flow convergence model is constructed, which is specifically a model that can be cyclically called by the initial fault of the power transmission line in the power system to shed the load of the target transmission line when the target transmission line is overloaded, that is, it is used to check whether the system power flow converges; if the system does not converge, the overloaded load needs to be trimmed continuously until the system power flow converges. The expression of the AC optimal power flow convergence model is:

[0185] (56)

[0186] Equation (56) represents the objective function for each generator, expressed as the cost function of active power and reactive power injection; is the active power of generator A Injection cost function; is the reactive power of generator A Injection cost function; n g is the total number of generators; θ is the voltage phase angle; V m is the voltage amplitude.

[0187] The active power balance constraint and reactive power balance constraint of the generator represented by Equation (56) are as shown in Equations (57)-(58):

[0188] P p (θ, V m , P g ) = P bus (θ, V m ) + P d - C g P g = 0(57)

[0189] Q p (θ, V m , Q g ) = Q bus (θ, V m ) + Q d - C g Q g = 0(58)

[0190] P p (θ, V m , P g ) is the active power balance constraint of the generator; P bus (θ, V m ) is the nodal active power; P d is the line active power; C g is an n b ×n g , matrix, n b is the total number of nodes. For Cg Internal element (M, A), if generator A is on bus M, then this element is 1, otherwise it is 0; in formula (58), Q p (θ,V m ,Q g ) is the generator reactive power balance constraint; Q bus (θ,V m ) is the node reactive power; Q d is the line reactive power.

[0191] The nonlinear functions of bus voltage phase angle and voltage amplitude are as follows:

[0192] h f (θ,V m )=|F f (θ,V m )|-F max ≤0(59)

[0193] h t (θ,V m )=|F t (θ,V m )|-F max ≤0(60)

[0194] (61)

[0195] (62)

[0196] (63)

[0197] (64)

[0198] Equation (59) is used for the starting end of each branch; Equation (60) is used for the ending end of each branch; h f (θ,V m ) is the nonlinear function of the branch flow at the starting end; F f (θ,V m ) is the flow vector of the starting branch; F max is the branch flow restriction vector; h t (θ,V m ) is the nonlinear function of the terminal branch flow; F t (θ,V m ) is the flow vector of the terminating branch; S f (θ,V m ) is the apparent power; P f (θ,V m ) is the real active power; I f (θ,V m) is the current; S f (V) is the true apparent power; V is the node voltage matrix; is the starting - end current vector; C f and C t are both n l × n b sparse connection matrices; n l is the total number of branches; is the starting - end admittance matrix; is the voltage vector; Y f and Y t are both n l × n b admittance matrices; S t (θ, V m ) is the terminating - end apparent power; P t (θ, V m ) is the terminating - end active power; I t (θ, V m ) is the terminating - end current; S t (V) is the terminating - end true apparent power; is the terminating - end current vector; is the terminating - end admittance matrix.

[0199] The constraint equations are specifically as shown in Equations (65) - (68):

[0200] (65)

[0201] (66)

[0202] (67)

[0203] (68)

[0204] Equation (65) is the reference - node voltage phase - angle constraint; θ i' is the voltage phase - angle of node i'; Γ ref represents the set of given bus - reference phase angles; Equation (66) is the node - voltage maximum and minimum limits; Equation (67) is the active - power injection limit; Equation (68) is the reactive - power injection limit; is the minimum voltage phase - angle; is the maximum voltage phase - angle; is the voltage amplitude; is the minimum voltage amplitude; is the maximum voltage amplitude; is the minimum injected active power; is the injected active power; is the maximum injected active power; is the minimum value of the injected reactive power; is the injected reactive power; is the maximum value of the injected reactive power.

[0205] Step 105: Construct a cascading fault model according to the dynamic traffic flow distribution model, the AC optimal power flow convergence model, and the situation of the charging station stopping operation.

[0206] The power system and the traffic system highly depend on and interact with each other over time and are coupled through electric vehicle charging stations. When the power system suffers from a typhoon disaster, an initial fault will occur in the transmission line. The disturbance of the power system may spread to the electric vehicle charging station, causing the charging station to be interrupted. The interruption or inaccessibility of the charging station may lead to the affected electric vehicles seeking charging services at other normally operating charging stations, which constitutes the redistribution of the dynamic traffic flow of electric vehicles. After the traffic load is redistributed and the power line is overloaded, load shedding must be performed. In this step, a cascading fault model is constructed to simulate the impact of the redistribution of the charging load of the traffic system on the power system after the charging station fails.

[0207] The cycle process of cascading faults is as follows:

[0208] At time t, perform AC optimal power flow convergence calculation on the power system in the target area. If the power system does not converge, identify the branch with the highest overload rate. With the midpoint of the branch with the highest overload rate as the center, reduce the load of all loads within a preset radius R by 5%. Then run the AC optimal power flow convergence model again to check whether its output result converges. If it converges, enter the next time t + 1; otherwise, continue to reduce the load by 5%. If this is repeated 20 times and the target power system still does not converge, cut off this branch and re-perform the convergence calculation, repeating this process until there are no more overloaded branches. If all branches connected to the charging station are cut off, calculate the transfer of the electric vehicle charging load after the charging station stops operating according to Equation (69):

[0209] (69)

[0210] The charging cell in the cellular model is the charging station on the road. Then, in Equation (71), TEL i→j represents the load transferred from charging station i to charging station j; K i represents the set of power source nodes adjacent to charging station i; R ij represents the effective resistance between charging station i and charging station j in the charging station network; DE i is the power load at charging station i; V e represents the set of charging stations.

[0211] Step 106: According to the initial fault model and the cascading fault model, for each node of each transmission line, determine the wind speed data sets of each transmission line at different times based on the historical typhoon wind speed, and determine the fault state data sets of each transmission line and each charging station according to the Monte Carlo simulation method.

[0212] Using random number generation, compare the fault probability of the distribution network line affected by the typhoon with the generated random number to determine the fault state of the distribution network line equipment. The Monte Carlo simulation method can effectively estimate the state and parameters of the system by generating a large number of random samples to approximate the state distribution of the system and gradually updating these samples to reflect the evolution of the system over time.

[0213] Exemplarily, this step includes:

[0214] Obtain the wind field wind speed of the target transmission line affected by the typhoon at the target time.

[0215] Determine the power outage probability of the target transmission line at the target time according to the wind field wind speed of the target transmission line affected by the typhoon at the target time.

[0216] Adopt the Monte Carlo simulation method to generate a random number of the power outage probability of the target transmission line at the target time.

[0217] In the case where the random number is greater than or equal to the power outage probability, it is determined that the target transmission line fails at the target time; in the case where the random number is less than the power outage probability, it is determined that the target transmission line does not fail at the target time; traverse all transmission lines to obtain the transmission line fault state data set in a single typhoon accident.

[0218] In the case where all the transmission lines connected to the target charging station fail, it is determined that the target charging station fails; traverse all charging stations to obtain the charging station fault state data set in a single typhoon accident.

[0219] In the case where at least one charging station fails, call the cascading fault model. If the power system converges and the cascading fault model does not trip other transmission lines, go to the next time and re-obtain the power outage probability of the target transmission line at the target time; if the power system converges and the cascading fault model trips other transmission lines, update the transmission line fault state data set and the charging station fault state data set until the current time is the preset total number of times (that is, the value of the current time is equal to the preset total number of times), stop updating, and obtain the final transmission line fault state data set and the charging station fault state data set.

[0220] The simulation results of a single accident are as Figure 2 shown, where red represents that the line fails and green represents that the line is operating normally:

[0221] 1) Typhoon wind field simulation: At time t, for each line node, according to the wind speed of historical typhoons, the wind field wind speed of each line is calculated by Equations (11)-(15) to obtain the wind field wind speed H L (t). For each wind speed H L (t), the empirical power outage probability P L (H L (t)) of each line can be calculated by Equation (11).

[0222] 2) Monte Carlo simulation: For each P L (H L (t)), a random number is generated and compared with it. If the random number is greater than P L (H L (t)), it is determined that the line L fails at time t; if the random number is less than P L (H L (t)), it is determined that the line L is normal at time t, and the transmission line fault status data S L' in a single typhoon accident is obtained. When all the transmission lines connected to the electric vehicle charging stations fail, the charging station is also damaged, and the electric vehicle charging station fault status data S C in this typhoon accident is obtained.

[0223] 3) When an electric vehicle charging station is damaged, the cascading fault model is called. If the system converges and no other lines are removed, the next step is entered; if the cascading fault model removes other lines, the transmission line fault status data S L' and the electric vehicle charging station fault status data S C are updated.

[0224] 4) Return to step 1) and enter time t + 1. When t is equal to the set time slice T, this simulation ends.

[0225] The simulation should be carried out from different directions with the same historical typhoon wind speed to eliminate the influence of the single wind direction traveling direction. For the typhoon wind speeds of different levels in the history of the target area, each typhoon is simulated 1000 times in each direction to obtain the data sets of all simulation simulations.

[0226] Step 107, construct a fault prediction model based on a dynamic Bayesian network, and use the wind speed data set, fault status data set of each transmission line, and fault status data set of each charging station as inputs, and the conditional probability parameters of the dynamic Bayesian network model as outputs for training to obtain a trained fault prediction model.

[0227] Exemplarily, construct the expression of the fault prediction model G B based on the dynamic Bayesian network:

[0228] (70)

[0229] (71)

[0230] (72)

[0231] (73)

[0232] (74)

[0233] (75)

[0234] (76)

[0235] wherein, N B is the set of power system nodes described by the Bayesian network; L B is the set of transmission lines of the power system described by the Bayesian network; θ B is the conditional probability distribution of the power system nodes described by the Bayesian network; t = 0, 1, 2,..., T; T is the total number of preset time instants; Equation (71) is the expansion of the network nodes, indicating that it includes three node variables: line wind speed, transmission line status, and electric vehicle charging station status; Equation (73) is the expansion form of the transmission line status variable, indicating that the status of each line in each time slice (the duration of each time slice is T) is one of two states, 0 or 1, representing fault and normal respectively. S L is the fault status of transmission line L; S L = 1 indicates that transmission line L is normal; S L = 0 indicates that transmission line L is faulty; Equation (74) is the expansion form of the electric vehicle charging station status variable, indicating that the status of each electric vehicle charging station in each time slice is one of two states, 0 or 1, representing fault and normal respectively; S c is the fault status of charging station c; S c = 1 indicates that charging station c is normal; S c = 0 indicates that charging station c is faulty; c = 1, 2,..., C'; C' is the total number of charging stations; Equation (72) is the expansion form of the line wind farm wind speed variable, indicating that the wind speed of each line in each time slice is one of three states, 0, 1, or 2; H L represents the wind speed status of transmission line L in a time slice T; H L = 0 indicates tropical depression wind speed; H L = 1 indicates tropical storm and severe tropical storm wind speed; H L= 2 represents the wind speeds of typhoons, severe typhoons, and super typhoons; the results of various typhoon types and their wind speeds are shown in Table 1; L = 1, 2,..., L'; L' is the total number of transmission lines in the power system; is the wind speed state of transmission line L at time t; is the fault state of transmission line L at time t; is the fault state of transmission line L at time t; P(·) is the conditional probability of the nodes in the power system described by the Bayesian network.

[0236] Table 1 Results of various typhoon types and their wind speeds

[0237]

[0238] Equation (75) is the directed line of the causal relationship of the dynamic Bayesian network, indicating that in each time slice, the line wind speed is the parent node of the transmission line state, and the transmission line state is the parent node of the electric vehicle charging station state; in different time slices, the line wind speed at the previous moment is the parent node of the line wind speed at the next moment, the transmission line state at the previous moment is the parent node of the transmission line state at the next moment, and the electric vehicle charging station state at the previous moment is the parent node of the electric vehicle charging station state at the next moment; Equation (76) is the conditional parameter of the dynamic Bayesian network, indicating the state probability of the parent node given the child node. In the constructed dynamic Bayesian network, the parameter nodes and the causal variable relationships are known, while the conditional probability parameters are unknown and need to take the simulation dataset D as the input and the conditional probability parameters of the dynamic Bayesian network as the output.

[0239] Construct the expression of the set D of the wind speed dataset, fault state dataset of each transmission line, and fault state dataset of each charging station:

[0240] (77)

[0241] where H L' is the wind speed dataset of all transmission lines; S L' is the fault state dataset of all transmission lines; S C' is the fault state dataset of all charging stations; q is the number of typhoon accident simulations.

[0242] Taking the set D as the input and the conditional probability parameters of the dynamic Bayesian network model as the output, use the maximum a posteriori estimation method and the EM parameter learning method to train the fault prediction model G B to obtain the trained fault prediction model.

[0243] Exemplarily, according to Equations (78) - (84), the maximum a posteriori estimation method and the EM parameter learning method are combined for model training:

[0244] (78)

[0245] (79)

[0246] (80)

[0247] (81)

[0248] (82)

[0249] (83)

[0250] (84)

[0251] Equation (78) is the basic formula for the conditional probability of the Bayesian network. P(θ|D) is the probability with parameter θ under the condition that the data is known; P(θ) is the prior distribution, representing the probability distribution of parameter θ before observing the data; P(D|θ) is the likelihood function, representing the probability of observing data D given parameter θ; P(D) is the evidence factor; Equation (79) represents the normalization of the posterior distribution of discrete parameters to ensure that its integral equals 1; Equation (80) is the maximum a posteriori estimation formula; represents the initial distribution of the conditional probability parameters of the network nodes after training; Equations (81)-(82) are the E-step of the EM parameter learning; Equation (81) represents the transition probability a oh for different time periods o, h, and g, the calculated value of the expected sufficient statistic N oh ; x t is the variable node at time t; Equation (82) represents the observation probability b h (g) for the same time slice, the calculated value of the expected sufficient statistic M hg ; Equations (83)-(84) are the M-step of the EM parameter learning, representing updating the transition probability and the observation probability of the conditional probability parameters according to the expected sufficient statistics calculated by Equations (81)-(82); N og is the updated expected sufficient statistic; T is the total number of preset time instants.

[0252] Step 108: Input the measured typhoon wind speed into the trained fault prediction model to obtain the dynamic probability diagrams of the fault states of each transmission line and each charging station, determine the dynamic resilience index of the power-transportation coupled charging network, and evaluate the resilience of the power-transportation coupled charging network according to the dynamic resilience index.

[0253] In this step, the calculation formula for the dynamic probability is as shown in Equations (85)-(86):

[0254] (85)

[0255] (86)

[0256] In the formula, is the joint probability distribution, representing the probability of all N variables X from time t = 0 to T; represents the set of parent nodes of the u-th variable X at time 0; represents the conditional probability distribution of the u-th variable X at time 0; represents the dynamic conditional probability distribution of the u-th variable at time t≥1; is the observed data at T time instants, is the probability of all variables X (from variable 1 to variable N) at T time instants under this observed data; is the observed probability after the states of all variables at time 0; is the observed probability after the states of all variables at time t. P(D 0:T ) is the posterior probability given the observed evidence from time 0 to time T; According to Equations (85)-(86), for the power-transportation coupled charging network constructed for the IEEE 14-node system, by inputting the wind speed at the initial stage of the actual typhoon approaching, the dynamic probability diagrams of the states of each transmission line and charging station are obtained, as shown in Figure 3 、 Figure 4 and Figure 5 shown, and Figure 6 、 Figure 7 and Figure 8 shown, where t0 to t5 are the time instants.

[0257] The dynamic resilience index of the charging network proposed in this embodiment is defined as the expected value of the remaining power supply of the charging station at the current time divided by the expected value of the power loss at the current time. As shown in Figure 9 shown, in mathematical form, it is the multiple of the area of the slanted solid line region divided by the area of the vertical dashed line, indicating the duration for which the expected value of the remaining power supply of the charging station can sustain the expected value of the power loss at the current time. Since the expected value of power is the product of the actual power and the probability, the actual probability value is cancelled out during the ratio calculation. Therefore, this embodiment calculates the resilience index of the target charging station at the target time according to the following formula:

[0258] (87)

[0259] where, R c (t) is the resilience index of charging station c at time t; P c (S t =0) represents the probability of charging station c failing at time t, and the dynamic probability diagrams of the resilience levels of different charging stations are calculated; S t=0 indicates that the state value at time t is 0; T is the total number of preset times.

[0260] After calculating the resilience of each charging station at each time, the weights of each charging station at different times are calculated according to the entropy weight method, and then the resilience of each charging station is weighted and summed to obtain the resilience of the entire charging network at different times. The resilience index of the power-traffic coupled charging network at different times is calculated according to the following formula:

[0261] (88)

[0262] Among them, R NET (t) is the resilience index of the power-traffic coupled charging network at time t; w c is the weight of the resilience index of charging station c determined by the entropy weight method; C' is the total number of charging stations; the change in the resilience level of the charging network is calculated as Figure 10 shown.

[0263] In summary, the power-traffic coupled charging network resilience evaluation method provided in this embodiment comprehensively considers the coupling relationship between the power network, the traffic network and the charging network, and can integrate the interdependent systems into a whole for studying its characteristics; this embodiment can accurately analyze the secondary impact of the traffic system on the power system by considering the cascading failure model considering the dynamic traffic flow redistribution, considering the constraints of the bus voltage and the safety constraints of the line power flow; this embodiment is based on the dynamic Bayesian network to study the operation changes of the coupled system in the event of disturbances or faults, and can accurately predict the possible fault situations at different times.

[0264] Embodiment 2

[0265] Based on the same inventive concept as Embodiment 1, this embodiment also provides a power-traffic coupled charging network resilience evaluation system. Since the principle of this system to solve problems is similar to the aforementioned power-traffic coupled charging network resilience evaluation method, the implementation of this system can refer to the implementation of the power-traffic coupled charging network resilience evaluation method.

[0266] As Figure 11 shown, the power-traffic coupled charging network resilience evaluation system includes:

[0267] The first construction module 10 is used to construct a power-traffic coupled charging network according to the power system, the traffic system and the charging stations of electric vehicles in the target area.

[0268] The second construction module 20 is used to simulate the random path of the typhoon operation by using the Bessel curve according to the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial moment, and determine the wind field speed affected by the typhoon at each node on all transmission lines at different times according to the Holland wind field model, so as to construct the initial fault model of the transmission lines of the power system.

[0269] The third construction module 30 is used to construct a dynamic traffic flow distribution model based on cellular transmission according to the time-coupled flow propagation characteristics of traffic assignment and the electric vehicle redistribution principle.

[0270] The fourth construction module 40 is used to construct an AC optimal power flow convergence model according to the bus voltage phase angle, voltage amplitude, generator active power and generator reactive power of each transmission line of the power system.

[0271] The fifth construction module 50 is used to construct a cascading fault model according to the dynamic traffic flow distribution model, the AC optimal power flow convergence model and the situation of the charging station stopping operation.

[0272] The first determination module 60 is used to determine the wind speed data set of each transmission line at different times with the historical typhoon wind speed for the nodes of each transmission line respectively according to the initial fault model and the cascading fault model, and determine the fault state data set of each transmission line and each charging station according to the Monte Carlo simulation method.

[0273] The sixth construction module 70 is used to construct a fault prediction model based on the dynamic Bayesian network, and take the wind speed data set, fault state data set of each transmission line and the fault state data set of each charging station as inputs, and take the conditional probability parameters of the dynamic Bayesian network model as outputs for training to obtain a trained fault prediction model.

[0274] The second determination module 80 is used to input the measured typhoon wind speed into the trained fault prediction model to obtain the dynamic probability diagram of the fault states of each transmission line and each charging station, so as to determine the dynamic resilience index of the power-traffic coupled charging network, and evaluate the resilience of the power-traffic coupled charging network according to the dynamic resilience index.

[0275] Exemplarily, the second construction module includes:

[0276] The first calculation unit is used to calculate the power-off probability of each transmission line at the target moment according to the following formula:

[0277] ;

[0278] ;

[0279] ;

[0280] ;

[0281] 。

[0282] Among them, P L (H L (t)) is the power outage probability of the transmission line L at time t; H L (t) is the wind field wind speed of the transmission line L affected by the typhoon at time t; is the maximum wind resistance level of the transmission line L; exp(·) is the exponential function with the natural constant e as the base; V max is the maximum wind speed of the typhoon; R max is the maximum influence radius of the typhoon; B is the order of the Bessel curve; r(t) is the distance between the target transmission line and the typhoon center at time t; BH is the Holland coefficient; x(t) is the abscissa of the typhoon on the geographical coordinate map at time t; x b is the abscissa of the randomly generated control point; y(t) is the ordinate of the typhoon on the geographical coordinate map at time t; b is the number of the path control point on the Bessel curve; y b is the ordinate of the randomly generated control point; t b is the b-th power of t; represents the combination number operation;! represents the factorial operator.

[0283] Exemplarily, the first determination module includes:

[0284] An acquisition unit, configured to acquire the wind field wind speed of the target transmission line affected by the typhoon at the target time.

[0285] A first determination unit, configured to determine the power outage probability of the target transmission line at the target time according to the wind field wind speed of the target transmission line affected by the typhoon at the target time.

[0286] A random number generation unit, configured to generate a random number of the power outage probability of the target transmission line at the target time by using the Monte Carlo simulation method.

[0287] A second determination unit, configured to determine that the target transmission line fails at the target time when the random number is greater than or equal to the power outage probability; and determine that the target transmission line does not fail at the target time when the random number is less than the power outage probability; traverse all transmission lines to obtain a dataset of the failure states of the transmission lines in a single typhoon accident.

[0288] A third determination unit, configured to determine that the target charging station fails when all the transmission lines connected to the target charging station fail; traverse all charging stations to obtain a dataset of the failure states of the charging stations in a single typhoon accident.

[0289] The fourth determination unit is configured to, in the case of at least one charging station failure, call a cascading fault model. If the power system converges and the cascading fault model does not disconnect other transmission lines, enter the next moment and re-obtain the power-off probability of the target transmission line at the target moment. If the power system converges and the cascading fault model disconnects other transmission lines, update the transmission line fault status data set and the charging station fault status data set until the current moment reaches the preset total number of moments, stop the update, and obtain the final transmission line fault status data set and the charging station fault status data set.

[0290] Exemplarily, the sixth construction module includes:

[0291] The first construction unit is configured to construct a fault prediction model G B of the expression:

[0292] ;

[0293] ;

[0294] ;

[0295] ;

[0296] ;

[0297] ;

[0298] .

[0299] Where N B is the set of power system nodes described by the Bayesian network; L B is the set of transmission lines of the power system described by the Bayesian network; θ B is the conditional probability distribution of the power system nodes described by the Bayesian network; t = 0, 1, 2,..., T; T is the preset total number of moments; S L is the fault status of the transmission line L; S L = 1 indicates that the transmission line L is normal; S L = 0 indicates that the transmission line L fails; S c is the fault status of the charging station c; S c = 1 indicates that the charging station c is normal; S c = 0 indicates that the charging station c fails; c = 1, 2,..., C; C is the total number of charging stations; H L represents the wind speed status of the transmission line L at the preset total number of moments T; H L = 0 indicates a tropical depression wind speed; H L= 1 represents the wind speeds of tropical storms and severe tropical storms; H L = 2 represents the wind speeds of typhoons, severe typhoons and super typhoons; L = 1, 2, ..., L'; L' is the total number of transmission lines in the power system; is the wind speed state of transmission line L at time t; is the fault state of transmission line L at time t; is the fault state of transmission line L at time t; P(·) is the conditional probability of the nodes in the power system described by the Bayesian network.

[0300] The second construction unit is used to construct the expression of the set D of the wind speed data set, the fault state data set of each transmission line, and the fault state data set of each charging station:

[0301] 。

[0302] Among them, H L' is the wind speed data set of all transmission lines; S L' is the fault state data set of all transmission lines; S C' is the fault state data set of all charging stations; q is the number of typhoon accident simulations.

[0303] The model training unit is used to take the set D as the input and the conditional probability parameters of the dynamic Bayesian network model as the output, and use the maximum a posteriori estimation method and the EM parameter learning method to train the fault prediction model G B to obtain the trained fault prediction model.

[0304] Exemplarily, the second determination module includes:

[0305] The second calculation unit is used to calculate the resilience index of the target charging station at the target time according to the following formula:

[0306] 。

[0307] Among them, R c (t) is the resilience index of charging station c at time t; P c (S t = 0) represents the probability that charging station c fails at time t; S t = 0 represents that the state value at time t is 0; T is the total number of preset times.

[0308] The third calculation unit is used to calculate the resilience index of the power-transportation coupled charging network at different times according to the following formula:

[0309] 。

[0310] Among them, R NET(t) is the resilience index of the power - transportation coupled charging network at time t; w c is the weight of the resilience index of charging station c determined by the entropy weight method; C' is the total number of charging stations.

[0311] For the more specific working processes of the above - mentioned various modules, reference can be made to the corresponding content disclosed in Embodiment 1, and details will not be elaborated here.

[0312] Embodiment 3

[0313] This embodiment provides a computer device, including a processor and a memory; wherein, when the processor executes the computer program stored in the memory, the steps of the power - transportation coupled charging network resilience evaluation method described in Embodiment 1 are implemented.

[0314] For the more specific process of the above - mentioned method, reference can be made to the corresponding content disclosed in Embodiment 1, and details will not be elaborated here.

[0315] Embodiment 4

[0316] This embodiment provides a computer - readable storage medium for storing a computer program; when the computer program is executed by a processor, the steps of the power - transportation coupled charging network resilience evaluation method described in Embodiment 1 are implemented.

[0317] For the more specific process of the above - mentioned method, reference can be made to the corresponding content disclosed in Embodiment 1, and details will not be elaborated here.

[0318] Embodiment 5

[0319] This embodiment provides a computer program product, including computer - executable instructions or a computer program. When the computer - executable instructions or the computer program are executed by a processor, the steps of the power - transportation coupled charging network resilience evaluation method described in Embodiment 1 are implemented.

[0320] For the more specific process of the above - mentioned method, reference can be made to the corresponding content disclosed in Embodiment 1, and details will not be elaborated here.

[0321] In this specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the embodiments can be referred to each other. For the systems, devices, storage media, and computer program products disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0322] Those skilled in the art can clearly understand that the technology in the embodiments of the present invention can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solutions in the embodiments of the present invention, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0323] In some embodiments, the computer-executable instructions can be in the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and can be deployed in any form, including being deployed as an independent program or being deployed as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0324] As an example, the computer-executable instructions may or may not correspond to files in a file system, and may be stored as part of a file that stores other programs or data. For example, they may be stored in one or more scripts in a Hyper Text Markup Language (HTML) document, stored in a single file dedicated to the program being discussed, or stored in multiple cooperating files (for example, files that store one or more modules, subroutines, or code portions).

[0325] As an example, the computer-executable instructions can be deployed to be executed on one electronic device, or on multiple electronic devices located at one location, or on multiple electronic devices distributed at multiple locations and interconnected by a communication network.

[0326] The present invention has been described in detail above in conjunction with specific embodiments and exemplary examples. However, these descriptions should not be construed as limiting the present invention. Those skilled in the art understand that without departing from the spirit and scope of the present invention, various equivalent replacements, modifications, or improvements can be made to the technical solutions of the present invention and their implementation manners, and these all fall within the scope of the present invention. The protection scope of the present invention is subject to the appended claims.

Claims

1. A method for evaluating the resilience of a power-transportation coupled charging network, characterized in that, include: Build an electricity-transportation coupled charging network based on the power system, transportation system, and electric vehicle charging stations in the target area; Based on the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial time, the Bezier curve is used to simulate the random path of the typhoon. The wind field speed of each node on all transmission lines affected by the typhoon at different times is determined based on the Holland wind field model to construct the initial fault model of the power system transmission line. According to the time-coupled flow propagation characteristics of traffic distribution and the electric vehicle redistribution principle, a dynamic traffic flow distribution model based on cellular transmission is constructed; An AC optimal power flow convergence model is constructed based on the bus voltage phase angle, voltage amplitude, generator active power, and generator reactive power of each transmission line in the power system; A cascading failure model is constructed based on the dynamic traffic flow distribution model, the AC optimal power flow convergence model, and the situation where the charging station stops operating; Based on the initial fault model and cascading fault model, the wind speed datasets of each transmission line at different times were determined using historical typhoon wind speeds for each node of each transmission line. The fault status datasets of each transmission line and each charging station were then determined using the Monte Carlo simulation method. A fault prediction model based on a dynamic Bayesian network was constructed. The wind speed dataset, fault status dataset of each transmission line, and fault status dataset of each charging station were used as input, and the conditional probability parameters of the dynamic Bayesian network model were used as output for training. A trained fault prediction model was obtained. By inputting the measured typhoon wind speed into the trained fault prediction model, a dynamic probability map of the fault status of each transmission line and each charging station is obtained to determine the dynamic resilience index of the power-transportation coupled charging network. The resilience of the power-transportation coupled charging network is then evaluated based on the dynamic resilience index. The method uses a Bezier curve to simulate the random path of a typhoon based on the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial moment, and determines the wind field wind speed of each node on all transmission lines affected by the typhoon at different times based on the Holland wind field model to construct an initial fault model of the power system transmission line, including: The power outage probability of each transmission line at the target time is calculated according to the following formula: ； ； ； ； ； Among them, P L (H L (t)) is the power outage probability of transmission line L at time t; H L (t) is the wind speed of the transmission line L in the wind field affected by the typhoon at time t; is the maximum wind resistance level of the transmission line L; exp(·) is the exponential function with the natural constant e as the base; V max is the maximum wind speed of the typhoon; R max is the maximum impact radius of the typhoon; B is the order of the Bezier curve; r(t) is the distance between the target transmission line and the typhoon center at time t; BH is the Holland coefficient; x(t) is the horizontal coordinate of the typhoon on the geographical coordinate map at time t; x b is the horizontal coordinate of the randomly generated intermediate control point; y(t) is the vertical coordinate of the typhoon on the geographic coordinate map at time t; b is the randomly generated ordinate of the middle control point; t b is t raised to the power of b; b is the number of the path control point on the Bezier curve; Represents a combination of number operations; ! represents the factorial operator.

2. The power-transportation coupled charging network resilience evaluation method according to claim 1, characterized in that The method of determining wind speed datasets of each transmission line at different times based on the initial fault model and the cascading fault model for each transmission line node using historical typhoon wind speeds, and determining fault status datasets of each transmission line and each charging station using the Monte Carlo simulation method includes: Obtain the wind speed of the target transmission line affected by the typhoon at the target time; Determine the power outage probability of the target transmission line at the target time according to the wind speed of the wind field affected by the typhoon at the target time; Monte Carlo simulation method is used to generate random numbers for the probability of power outage of the target transmission line at the target time; When the random number is greater than or equal to the power outage probability, it is determined that the target transmission line fails at the target time; when the random number is less than the power outage probability, it is determined that the target transmission line does not fail at the target time; all transmission lines are traversed to obtain the transmission line fault status data set in a single typhoon accident; When all the transmission lines connected to the target charging station fail, it is determined that the target charging station fails; all charging stations are traversed to obtain the charging station fault status data set in a single typhoon accident; When at least one charging station fails, the cascading fault model is called. If the power system converges and the cascading fault model does not trip other transmission lines, go to the next time and re-obtain the power outage probability of the target transmission line at the target time; if the power system converges and the cascading fault model trips other transmission lines, update the transmission line fault status data set and the charging station fault status data set until the current time is the preset total number of times, stop updating, and obtain the final transmission line fault status data set and the charging station fault status data set.

3. The power-transportation coupled charging network resilience evaluation method according to claim 1, wherein The construction of a fault prediction model based on a dynamic Bayesian network, and training with the wind speed data set, fault status data set of each transmission line, and fault status data set of each charging station as input and the conditional probability parameters of the dynamic Bayesian network model as output to obtain a trained fault prediction model, including: Construct a fault prediction model G based on a dynamic Bayesian network B The expression of ； ； ； ； ； ； ； Among them, N B is the set of power system nodes described by the Bayesian network; L B is the set of transmission lines in the power system described by the Bayesian network; θ B is the conditional probability distribution of the power system nodes described by the Bayesian network; t = 0, 1, 2,..., T; T is the total number of preset time instants; S L is the fault state of the transmission line L; S L = 1 indicates that the transmission line L is normal; S L = 0 indicates that the transmission line L is faulty; S c is the fault state of the charging station c; S c = 1 indicates that the charging station c is normal; S c = 0 indicates that the charging station c is faulty; c = 1, 2,..., C'; C' is the total number of charging stations; H L represents the wind speed state of the transmission line L at the total number of preset time instants T; H L = 0 indicates tropical depression wind speed; H L = 1 indicates tropical storm and severe tropical storm wind speeds; H L = 2 indicates typhoon, severe typhoon and super typhoon wind speeds; L = 1, 2,..., L'; L' is the total number of transmission lines in the power system; is the wind speed state of the transmission line L at time t; is the fault state of the transmission line L at time t; is the fault state of the transmission line L at time t; P(·) is the conditional probability of the power system nodes described by the Bayesian network; Construct an expression for the set D of the wind speed data set, fault status data set of each transmission line, and fault status data set of each charging station: ； Among them, H L' is the wind speed data set of all transmission lines; S L' is the fault status data set of all transmission lines; S C' is the fault status data set of all charging stations; q is the number of typhoon accident simulations; Taking set D as the input and the conditional probability parameters of the dynamic Bayesian network model as the output, the maximum a posteriori estimation method and the EM parameter learning method are used to train the fault prediction model G B to obtain a trained fault prediction model.

4. The power-transportation coupled charging network resilience evaluation method according to claim 1, wherein Input the measured typhoon wind speed into the trained fault prediction model to obtain the dynamic probability map of the fault status of each transmission line and each charging station, determine the dynamic resilience index of the power-traffic coupled charging network, and evaluate the resilience of the power-traffic coupled charging network according to the dynamic resilience index, including: Calculate the resilience index of the target charging station at the target time according to the following formula: ； Among them, R c (t) is the resilience index of charging station c at time t; P c (S t = 0) represents the probability of failure of charging station c at time t; S t = 0 means that the state value at time t is 0; T is the total number of preset times; Calculate the resilience index of the power-traffic coupled charging network at different times according to the following formula: ； Among them, R NET (t) is the resilience index of the power-transportation coupled charging network at time t; w c is the weight of the resilience index of charging station c determined by the entropy weight method; C' is the total number of charging stations.

5. A resilience evaluation system for a power-transportation coupled charging network, characterized in that, Including: The first construction module is used to construct a power-traffic coupled charging network according to the power system, traffic system, and charging stations of electric vehicles in the target area; The second construction module is used to simulate the random path of the typhoon operation by using the Bessel curve according to the nodes on each transmission line of the power system in the target area and the historical typhoon wind speed at the initial time, and determine the wind field wind speed affected by the typhoon at each node on all transmission lines at different times according to the Holland wind field model to construct the initial fault model of the power system transmission line; The third construction module is used to construct a dynamic traffic flow distribution model based on cellular transmission according to the time-coupled flow propagation characteristics of traffic distribution and the electric vehicle reallocation principle; The fourth construction module is used to construct an AC optimal power flow convergence model according to the bus voltage phase angle, voltage amplitude, generator active power, and generator reactive power of each transmission line of the power system; The fifth construction module is used to construct a cascading fault model according to the dynamic traffic flow distribution model, AC optimal power flow convergence model, and the situation of charging stations stopping operation; A first determination module is configured to determine, based on the initial fault model and the cascading fault model, wind speed datasets of each transmission line at different times using historical typhoon wind speeds for each node of each transmission line, and to determine fault status datasets of each transmission line and each charging station using a Monte Carlo simulation method; The sixth construction module is used to build a fault prediction model based on a dynamic Bayesian network. The model takes the wind speed dataset, fault status dataset of each transmission line, and fault status dataset of each charging station as input, and uses the conditional probability parameters of the dynamic Bayesian network model as output for training to obtain a trained fault prediction model. The second determination module is used to input the measured typhoon wind speed into the trained fault prediction model to obtain a dynamic probability map of the fault status of each transmission line and each charging station, thereby determining the dynamic resilience index of the power-transportation coupled charging network and evaluating the resilience of the power-transportation coupled charging network based on the dynamic resilience index; The second building block includes: The first calculation unit is used to calculate the power outage probability of each transmission line at the target time according to the following formula: ； ； ； ； ； Among them, P L (H L (t)) is the power outage probability of the transmission line L at time t; H L (t) is the wind field wind speed of the transmission line L affected by the typhoon at time t; is the maximum wind resistance level of the transmission line L; exp(·) is the exponential function with the natural constant e as the base; V max is the maximum wind speed of the typhoon; R max is the maximum influence radius of the typhoon; B is the order of the Bessel curve; r(t) is the distance of the target transmission line from the typhoon center at time t; BH is the Holland coefficient; x(t) is the abscissa of the typhoon on the geographical coordinate map at time t; x b is the abscissa of the randomly generated intermediate control point; y(t) is the ordinate of the typhoon on the geographical coordinate map at time t; y b is the ordinate of the randomly generated intermediate control point; t b is the b-th power of t; b is the number of the path control points on the Bessel curve; represents the combination number operation;! represents the factorial operator.

6. The power-transportation coupled charging network resilience evaluation system according to claim 5, wherein The first determining module includes: An acquisition unit, configured to acquire the wind speed of a target transmission line in a wind field affected by a typhoon at a target time; A first determining unit is configured to determine a power outage probability of the target transmission line at a target time according to a wind speed of a wind field affected by a typhoon on the target transmission line at a target time; A random number generation unit, used to generate a random number for the probability of power outage of a target transmission line at a target time using a Monte Carlo simulation method; The second determining unit is configured to determine that a fault has occurred on the target transmission line at a target time if the random number is greater than or equal to the power outage probability; and to determine that no fault has occurred on the target transmission line at the target time if the random number is less than the power outage probability; and to traverse all transmission lines to obtain a data set of transmission line fault states in a single typhoon event; The third determining unit is configured to determine that the target charging station is faulty when all transmission lines connected to the target charging station are faulty; traverse all charging stations to obtain a charging station fault status dataset in a single typhoon event; The fourth determination unit is configured to call the cascading failure model when at least one charging station fails. If the power system converges and the cascading failure model does not cut off other transmission lines, the fourth determination unit is configured to enter the next moment and re-acquire the power outage probability of the target transmission line at the target moment. If the power system converges and the cascading failure model cuts off other transmission lines, the fourth determination unit is configured to update the transmission line fault status dataset and the charging station fault status dataset until the current moment reaches the preset total number of moments, then stop updating, and obtain the final transmission line fault status dataset and the charging station fault status dataset.

7. The power-transportation coupled charging network resilience assessment system according to claim 5, wherein The sixth building block includes: The first construction unit is used to construct a fault prediction model G based on a dynamic Bayesian network B The expression of ； ； ； ； ； ； ； Among them, N B is the set of power system nodes described by the Bayesian network; L B is the set of power system transmission lines described by the Bayesian network; θ B is the conditional probability distribution of the power system nodes described by the Bayesian network; t = 0, 1, 2,..., T; T is the total number of preset moments; S L is the fault state of the transmission line L; S L = 1 indicates that the transmission line L is normal; S L = 0 indicates that the transmission line L is faulty; S c is the fault state of the charging station c; S c = 1 indicates that the charging station c is normal; S c = 0 indicates that the charging station c is faulty; c = 1, 2,..., C'; C' is the total number of charging stations; H L represents the wind speed state of the transmission line L at the total number of preset moments T; H L = 0 indicates tropical depression wind speed; H L = 1 indicates tropical storm and severe tropical storm wind speed; H L = 2 indicates typhoon, severe typhoon and super typhoon wind speed; L = 1, 2,..., L'; L' is the total number of transmission lines in the power system; is the wind speed state of the transmission line L at time t; is the fault state of the transmission line L at time t; is the fault state of the transmission line L at time t; P(·) is the conditional probability of the power system nodes described by the Bayesian network; The second construction unit is used to construct an expression for the set D of wind speed datasets, fault status datasets of each transmission line, and fault status datasets of each charging station: ； Among them, H L' is the wind speed dataset of all transmission lines; S L' is the fault status dataset of all transmission lines; S C' is the fault status dataset of all charging stations; q is the number of typhoon accident simulations; A model training unit, which takes the set D as the input and the conditional probability parameters of the dynamic Bayesian network model as the output, and uses the maximum a posteriori estimation method and the EM parameter learning method to train the fault prediction model G B to obtain a trained fault prediction model.

8. The power-transportation coupled charging network resilience assessment system according to claim 5, characterized in that: The second determining module includes: The second calculation unit is configured to calculate the resilience index of the target charging station at the target time according to the following formula: ； Among them, R c (t) is the resilience index of charging station c at time t; P c (S t = 0) represents the probability of charging station c failing at time t; S t = 0 means that the state value at time t is 0; T is the total number of preset times; The third calculation unit is used to calculate the resilience index of the power-transportation coupled charging network at different times according to the following formula: ； Among them, R NET (t) is the resilience index of the power-transportation coupled charging network at time t; w c is the weight of the resilience index of charging station c determined by the entropy weight method; C' is the total number of charging stations.

Images