Method and system for evaluating toughness of power-traffic coupling charging network

Abstract

The invention provides a power-traffic coupling charging network toughness evaluation method and system, and belongs to the technical field of power system analysis. According to the method, a power-traffic coupling charging network is constructed, a typhoon path and a Holland wind field model are simulated in combination with a Bezier curve to calculate a wind speed, a dynamic traffic flow distribution, alternating current optimal power flow and cascade fault model is established, a fault data set is generated by utilizing Monte Carlo simulation, a fault prediction model is trained through a dynamic Bayesian network, and a fault prediction result is obtained. Inputting the actually measured wind speed, outputting a dynamic probability graph, and calculating a dynamic toughness index to evaluate the toughness of the power-traffic coupling charging network. According to the method, the problem of insufficient fault propagation and dynamic evaluation between systems is considered, and the toughness evaluation precision of the power-traffic coupling charging network under typhoon disasters is improved.

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-traffic coupled charging network. Background Art

[0002] Power and transportation are two cores of the urban infrastructure system. The urban distribution network lines are at the end of the main power grid. The network scale is large, there are numerous lines and a large number of them are installed outdoors, making them extremely vulnerable to 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-traffic coupled charging network to extreme events is one of the keys to ensuring the safe and stable development of the city.

[0003] At present, the 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 loads after the failure of electric vehicle charging stations. The research on the resilience evaluation of the power-traffic coupled charging network only considers the connection between single systems and does not consider the problem of the spread of failures between different systems. At the same time, the existing resilience evaluation of the power-traffic coupled system has static limitations and cannot reasonably evaluate the dynamic ability of the power-traffic 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-traffic coupled charging network to solve the problems in the prior art that only the connection between single systems is considered, the problem of the spread of failures between different systems is not considered, and the dynamic ability of the power-traffic coupled system to resist typhoon disasters cannot be reasonably evaluated.

[0005] In a first aspect, the present invention provides a method for evaluating the resilience of a power-traffic coupled charging network, including: Constructing a power-traffic coupled charging network according to the power system, transportation system, and electric vehicle charging stations in the target area; 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 of each node on all transmission lines at different times affected by the typhoon according to the Holland wind field model, so as to construct an initial fault model of the power system transmission line; 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; 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; Construct a cascading fault model according to the dynamic traffic flow allocation model, the AC optimal power flow convergence model, and the situation of the charging station out of operation; 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; Construct a fault prediction model based on the 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; 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.

[0006] 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 the Bessel curve to simulate the random path of the typhoon operation, and determining the wind field wind speed of each node on all transmission lines at different times according to the Holland wind field model, includes: Calculate the power outage probability of each transmission line at the target moment according to the following formula: ; ; ; ; ; where, P L (H L (t)) is the power outage 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; 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.

[0007] Optionally, according to the initial fault model and the cascading fault model, for each node of the transmission lines, the wind speed data sets of each transmission line at different times are determined based on the historical typhoon wind speed, and the fault state data sets of each transmission line and each charging station are determined according to the Monte Carlo simulation method, including: Obtain the wind field wind speed of the target transmission line affected by the typhoon at the target time; Determine the power-off 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; Generate a random number of the power-off probability of the target transmission line at the target time by using the Monte Carlo simulation method; In the case where the random number is greater than or equal to the power-off 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-off 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; 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 fault state data set of the charging stations in a single typhoon accident; 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 cut off other transmission lines, go to the next time and re-obtain the power-off probability of the target transmission line at the target time; if the power system converges and the cascading fault model cuts off other transmission lines, update the fault state data set of the transmission lines and the fault state data set of the charging stations until the current time is the preset total number of times, stop updating, and obtain the final fault state data set of the transmission lines and the fault state data set of the charging stations.

[0008] Optionally, construct a fault prediction model based on the dynamic Bayesian network, and use the wind speed data sets, fault state data sets of each transmission line and the fault state data sets of each charging station as inputs, and use the conditional probability parameters of the dynamic Bayesian network model as outputs for training to obtain a trained fault prediction model, including: Construct the fault prediction model G based on the dynamic Bayesian network B Expression: ; ; ; ; ; ; ; 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 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 has a fault; 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 has a fault; 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; 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: ; 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 dataset of the failure status 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 failure prediction model G B to obtain a trained failure prediction model.

[0009] Optionally, input the measured typhoon wind speed into the trained failure prediction model to obtain the dynamic probability map of the failure status 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, including: Calculate the resilience index of the target charging station at the target time according to the following formula: ; where, 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 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: ; where, 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.

[0010] In a second aspect, the present invention provides a power-traffic coupled charging network resilience evaluation system, 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 typhoon operation by 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 determine the wind field wind speed affected by typhoon on each node of all transmission lines at different times according to the Holland wind field model, so as 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 in the power system; The fifth construction module 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 charging stations stopping operation; The first determination module is used to determine the wind speed data set of each transmission line at different times according to the historical typhoon wind speed for each node of each transmission line based on 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; The sixth construction module is used to construct a fault prediction model based on a dynamic Bayesian network, and train it with 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 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 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.

[0011] Optionally, the second construction module 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: ; ; ; ; ; where, P L (H L (t)) is the power outage 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 geographical coordinate map at time t; x bx(t) 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 path control points on the Bessel curve; represents the combination number operation;! represents the factorial operator.

[0012] Optionally, the first determination module includes: An acquisition unit for acquiring the wind field wind speed of the target transmission line affected by the typhoon at the target time; A first determination unit for determining 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; A random number generation unit for generating a random number of the power outage probability of the target transmission line at the target time by using the Monte Carlo simulation method; A second determination unit for determining 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 determining that the target transmission line does not fail at the target time when the random number is less than the power outage probability; traversing all transmission lines to obtain a transmission line fault state data set in a single typhoon accident; A third determination unit for determining that the target charging station fails when all the transmission lines connected to the target charging station fail; traversing all charging stations to obtain a charging station fault state data set in a single typhoon accident; A fourth determination unit for, when at least one charging station fails, invoking the cascading fault model. If the power system converges and the cascading fault model does not trip other transmission lines, then enter the next time and re-acquire 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, then 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, stop updating, and obtain the final transmission line fault state data set and the charging station fault state data set.

[0013] Optionally, the sixth construction module includes: A first construction unit for constructing an expression of the fault prediction model G B based on the dynamic Bayesian network: ; ; ; ; ; ; ; 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 transmission line L; S L = 1 indicates that transmission line L is normal; S L = 0 indicates that transmission line L has a fault; S c is the fault state of charging station c; S c = 1 indicates that charging station c is normal; S c = 0 indicates that charging station c has a fault; c = 1, 2,..., C'; C' is the total number of charging stations; H L represents the wind speed state of 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 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 power system nodes described by the Bayesian network; 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: ; 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; 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.

[0014] Optionally, the second determination module includes: A second calculation unit, configured to calculate a resilience index of a target charging station at a target moment according to the following formula: ; wherein, R c (t) is the resilience index of charging station c at moment t; P c (S t =0) represents the probability that charging station c fails at moment t; S t =0 means that the state value at moment t is 0; T is the total number of preset moments; A third calculation unit, configured to calculate a resilience index of the power-transportation coupled charging network at different moments according to the following formula: ; wherein, R NET (t) is the resilience index of the power-transportation coupled charging network at moment 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.

[0015] The present invention provides a method and system for evaluating the resilience of a power-transportation coupled charging network. In the method, the coupling relationship between the power network, the transportation network, and the charging network is comprehensively considered, and the interdependent systems can be integrated into a whole for studying their characteristics; the cascading 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 cascading failure model considering the dynamic traffic flow redistribution, considering the constraint of the bus voltage and the safety constraint of the line power flow; based on the dynamic Bayesian network, the present invention studies the operation change 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

[0016] 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.

[0017] Figure 1 It 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; Figure 2 It is a schematic diagram of a fault simulation of a typhoon on a power distribution network provided by an embodiment of the present invention; Figure 3 It is a prediction result diagram of the dynamic power outage probability of a transmission line under a low-voltage wind speed level provided by an embodiment of the present invention; Figure 4The figure of the predicted 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; Figure 5 The figure of the predicted 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; Figure 6 The figure of the predicted 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; Figure 7 The figure of the predicted result of the power outage probability of the charging station under the storm wind speed level provided by the embodiment of the present invention; Figure 8 The figure of the predicted 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; Figure 9 The schematic diagram of the resilience assessment calculation provided by the embodiment of the present invention; Figure 10 The curve graph of the resilience change of the charging network provided by the embodiment of the present invention; Figure 11 The structural schematic diagram of a power - traffic coupled charging network resilience assessment system provided by the embodiment of the present invention. Detailed implementation manners

[0018] 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 the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without making creative efforts belong to the scope of protection of the present invention.

[0019] Embodiment 1

[0020] As Figure 1 shown, the embodiment of the present invention provides a power - traffic coupled charging network resilience assessment method, including: 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.

[0021] In this step, exemplarily, construct a power - traffic coupled charging network according to formulas (1) - (10).

[0022] G r (V r ,E r ,W r )(1) G e (V e ,E e ,W e )(2) Gp (V p , E p , W p )(3) G(V, E, W)(4) (5) (6) (7) (8) (9) (10) In formula (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 each entropy weight in the entropy matrix, and each entropy weight is the time required for the corresponding road; in formula (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 each entropy weight in the entropy matrix is expressed as ; in formula (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 p represents the entropy matrix of the power system, and each entropy weight in the entropy matrix is expressed as the percentage of the remaining capacity of the power system; in formula (4), G represents the power-transportation charging coupling network; V represents the set of nodes of the power-transportation coupling charging network, which is composed of V e and V p ; E represents the set of lines of the power-transportation coupling network, which is composed of E r , E e and E c ; E c is the line coupling link; W represents the entropy matrix of the power-transportation coupling network, which is composed of W p , W e and W c ; W c is the entropy matrix of the coupling link; formulas (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; cr 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 node m and node n in the charging station network; is the Laplacian matrix of the power-traffic 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-traffic 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 of the power-traffic-coupling charging network; φ max is the maximum charging power of the charging station; φ k' is the actual charging power.

[0023] 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 operation, and determine 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, so as to construct the initial fault model of the power system transmission line.

[0024] 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.

[0025] For the transmission lines of the power system in the target area, it is necessary to strictly calculate the wind speed impact on the area where the lines are located at different times according to the actual distances between the lines and the nodes on each line.

[0026] The running path deviation of a typhoon in a city is not too large. Therefore, when using the Bessel curve mathematical model to simulate the random path of a typhoon in the urban power 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.

[0027] Exemplarily, calculate the power outage probability of each transmission line at the target time according to the following formula: (11) (12) (13) (14) (15) 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 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 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 point 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.

[0028] Step 103, construct a dynamic traffic flow allocation model based on cellular transmission according to the time-coupled flow propagation characteristics of traffic assignment and the electric vehicle reallocation principle.

[0029] In this step, when constructing a dynamic traffic flow assignment model based on cellular transmission, the dynamic system optimal scheme should be used, that is, user participants need to obey the dispatcher's command to minimize the total travel cost of all users, which is used to determine the traffic flow characteristics after the charging station stops operating.

[0030] The construction formula of the cellular transmission model is as shown in Equations (16)-(26): (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) 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 at cell i at time t and the receiving volume at cell i + 1 ; Equation (18) represents the sending volume at 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 at cell i + 1 at time t depending on the smaller value between the outflow capacity at 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 diverging cell, where α is the diverging 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 having the diverging coefficient; represents the number of vehicles flowing out of cell i; represents the sending volume of cell i; represents the receiving volume of cell j; represents the received volume of cell k; Equations (23)-(26) are the mathematical models of the merging cell, where β is the merging coefficient, which is the set value of the externally input parameter; Equations (23)-(24) represent and the number of vehicles flowing out 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 receiving volume of cell i at time t; Equation (25) represents that cell i is in an unobstructed state when the sum of the sending volumes of cells j and k is less than the receiving volume 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 volumes of cells j and k is greater than the receiving volume of cell i; represents the vehicle occupancy rate of cell i at time t.

[0031] 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. The charging cell constraints are as follows: (27) (28) (29) (30) (31) (32) (33) (34) (35) (36) (37) (38) (39) (40) Among them, C C is the set of charging cells; R w is the set of all paths in OD pair (origin and destination, starting point - ending point) w; T s is the starting time of charging; T f is the transition time from the non - queuing state to the 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 the non - queuing case to the 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 (32) means that when transitioning from the non - queuing case to 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 (33) means that T s +T cThe traffic flow out of the charging cell at a certain moment is equal to the charging start time T s The number of vehicles that can just be charged; Equations (34) and (35) are the constraints on the vehicle occupancy rate when the queuing situation transitions to the non-queuing situation; Equation (36) represents T s +T c The traffic flow out of the charging cell at a certain moment is equal to the charging start time T s The number of vehicles that can just be charged; Equation (37) indicates 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) represents the traffic flow conservation of charging station i; Equation (39) represents that the inflow and outflow need to 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 are no vehicles with charging demands in the end cell i.

[0032] Expand the cell transmission model into a dynamic traffic assignment model, making its merging and diverging coefficients not affect the optimality of the dynamic traffic flow, introducing a charging cell to capture the electric vehicle charging load, and simultaneously satisfying the accurate simulation of traffic flow at the path level. Construct the dynamic traffic assignment model according to Equations (41)-(55).

[0033] (41) (42) (43) (44) (45) (46) (47) (48) (49) (50) (51) (52) (53) (54) (55) 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, 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 time moments; C is the total number of ordinary cells; C s is 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 on 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 traffic network flow demand constraint; 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 at time t in state a; 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 of 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 passing through path r from cell k to cell i at time t - 1 in state a; Γ(i) represents the set of the next cells of cell i; represents the number of vehicles passing through path r from cell i to cell j 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 passing through path r from cell k to cell i 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 in cell i at time t; Equations (52)-(53) are the initial values of cell occupancy and link flow; Equations (54)-(55) are the non-negativity constraints of cell occupancy and link flow.

[0034] Step 104: Construct an AC optimal power flow convergence model based on the bus voltage phase angle, voltage amplitude, generator active power, and generator reactive power of each transmission line in the power system.

[0035] In this step, the construction of the AC optimal power flow convergence model is specifically manifested as a model that can be cyclically called by the initial faults of the power system transmission lines to shed 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 continuously trimmed until the system power flow converges. The expression of the AC optimal power flow convergence model is: (56) 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 Injected cost function; is the reactive power of generator A Injected cost function; n g is the total number of generators; θ is the voltage phase angle; V m is the voltage amplitude.

[0036] The active power balance constraint and reactive power balance constraint represented by Equation (56) are as shown in Equations (57)-(58): P p (θ, V m , P g ) = P bus (θ, V m ) + P d - C g P g = 0 (57) Q p (θ, V m , Q g ) = Q bus (θ, V m ) + Q d - C g Q g = 0 (58) P p (θ, V m , P g ) is the active power balance constraint of the generator; P bus (θ, V m ) is the node active power; Pd is the active power of the line; C g is an n b ×n g matrix, where n b is the total number of nodes. For C g the internal element (M, A), if generator A is on bus M, then this element is 1, otherwise it is 0; in Equation (58), Q p (θ, V m , Q g ) is the reactive power balance constraint of the generator; Q bus (θ, V m ) is the reactive power of the node; Q d is the reactive power of the line.

[0037] The non - linear functions of the bus voltage phase angle and voltage amplitude are specifically as shown in Equations (59) - (64) h f (θ, V m ) = |F f (θ, V m )| - F max ≤0 (59) h t (θ, V m ) = |F t (θ, V m )| - F max ≤0 (60) (61) (62) (63) (64) 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 non - linear function of the starting - end branch flow; F f (θ, V m ) is the starting - end branch flow vector; F max is the branch flow limit vector; h t (θ, V m ) is the non - linear function of the ending - end branch flow; F t (θ, V m ) is the ending - end branch flow vector; S f (θ, V m ) is the apparent power; P f (θ, V m ) is the real active power; I f (θ, Vm ) 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 matrix; 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 matrix; 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.

[0038] The constraint equations are specifically as shown in equations (65) - (68): (65) (66) (67) (68) 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 maximum and minimum limits of the node voltage; 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 injected reactive power; is the injected reactive power; is the maximum value of the injected reactive power.

[0039] 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.

[0040] The power system and the traffic system are highly dependent and interact with each other over time, and are coupled through electric vehicle charging stations. When the power system suffers from typhoon disasters, initial faults will occur in the transmission lines. The disturbances in the power system may spread to the electric vehicle charging stations, causing the charging stations to be interrupted. The interruption or unavailability of the charging stations may cause the affected electric vehicles to seek 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 lines are 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.

[0041] The cycle process of cascading faults is as follows: 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 the 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 perform convergence calculation again. Repeat 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): (69) 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 stations i and j in the charging station network; DE i is the power load at charging station i; V e represents the set of charging stations.

[0042] 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 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.

[0043] By using random number generation, the fault probability of the distribution network line affected by the typhoon is compared with the generated random number to determine the fault state of the distribution network line equipment. The Monte Carlo simulation method approximates the state distribution of the system by generating a large number of random samples, and gradually updates these samples to reflect the evolution of the system over time, which can effectively estimate the state and parameters of the system.

[0044] Exemplarily, this step includes: Obtain the wind speed of the target transmission line in the wind field affected by the typhoon at the target time.

[0045] The power outage probability of the target transmission line at the target time is determined according to the wind speed of the wind field affected by the typhoon at the target transmission line at the target time.

[0046] The 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.

[0047] When the random number is greater than or equal to the power outage probability, the target transmission line is judged to have failed at the target time; when the random number is less than the power outage probability, the target transmission line is judged not to have failed at the target time; all transmission lines are traversed to obtain the transmission line fault status data set in a single typhoon accident.

[0048] When all the transmission lines connected to the target charging station fail, the target charging station is determined to be faulty; all charging stations are traversed to obtain a charging station fault status dataset in a single typhoon accident.

[0049] In the case of at least one charging station failure, the cascading failure model is called. If the power system converges and the cascading failure model does not cut off other transmission lines, the next moment is entered and the power outage probability of the target transmission line at the target moment is re-obtained; if the power system converges and the cascading failure model cuts off other transmission lines, the transmission line fault status dataset and the charging station fault status dataset are updated until the current moment is the preset total number of moments (that is, the value of the current moment is equal to the preset total number of moments), and the update is stopped to obtain the final transmission line fault status dataset and the charging station fault status dataset.

[0050] The simulation results of a single accident are as follows: Figure 2 As shown, red represents a line fault and green represents a normal line operation: 1) Typhoon wind field simulation: At time t, for each line node, according to the wind speed of historical typhoons, the wind field speed of each line is calculated by equations (11) to (15) to obtain the single simulated wind field speed H L (t). For each wind speed H L (t), the empirical power outage probability P of each line can be calculated by formula (11): L (HL (t)).

[0051] 2) Monte Carlo simulation: For each P L (H L (t)), generate a random number to compare 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 are faulty, the charging station is also damaged, and the electric vehicle charging station fault status data S C in this typhoon accident is obtained.

[0052] 3) When there is a damaged electric vehicle charging station, call the cascading fault model. If the system converges and no other lines are disconnected, proceed to the next step; if the cascading fault model disconnects other lines, update the transmission line fault status data S L' and the electric vehicle charging station fault status data S C .

[0053] 4) Return to step 1) and enter time t+1. When t is equal to the set time slice T, this simulation ends.

[0054] 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 different levels of typhoon wind speeds in the history of the target area, each typhoon is simulated 1000 times in each direction to obtain the data sets of all simulation results.

[0055] Step 107, construct a fault prediction model based on the 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 input, and use the conditional probability parameters of the dynamic Bayesian network model as output for training to obtain a trained fault prediction model.

[0056] Exemplarily, the expression for constructing the fault prediction model G B is: (70) (71) (72) (73) (74) (75) (76) 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; 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 speeds; H L = 2 indicates typhoon, severe typhoon, and super typhoon wind speeds; 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 status of transmission line L at time t; is the fault status of transmission line L at time t; is the fault status of transmission line L at time t; P(·) is the conditional probability of the power system nodes described by the Bayesian network.

[0057] Table 1 Results of various typhoon types and their wind speeds

[0058] Equation (75) represents the directed line of the causal relationship of the dynamic Bayesian network, indicating that in each time slice, the line wind speed represents the parent node of the transmission line state, and the transmission line state represents 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.

[0059] Construct the expression of the set D of the wind speed dataset, fault status dataset of each transmission line, and fault status dataset of each charging station: (77) where 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.

[0060] 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.

[0061] Exemplarily, according to Equations (78)-(84), the maximum a posteriori estimation method and the EM parameter learning method are combined for model training: (78) (79) (80) (81) (82) (83) (84) Equation (78) is the basic formula for the conditional probability of a Bayesian network. P(θ|D) is the probability of parameter θ given the known data; 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 normalizing 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 EM parameter learning; Equation (81) represents the transition probability a for different time periods o, h, and g oh , 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, and the calculated value of the expected sufficient statistic M hg ; Equations (83)-(84) are the M-step of EM parameter learning, representing updating the transition probability and 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.

[0062] 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.

[0063] In this step, the calculation formula of the dynamic probability is as shown in Equations (85)-(86): (85) (86) where is the joint probability distribution, representing the probability of all N variables X from time 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 1 to N variables) at T time instants under this observed data; is the observed probability after the states of all variables at time 0; is the observation probability after all variable states 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 period when the actual typhoon approaches, 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 , and as shown in Figure 6 , Figure 7 and Figure 8 , where t0 to t5 are the time numbers.

[0064] 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 , 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: (87) 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.

[0065] After calculating the resilience of the 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-transportation coupled charging network at different times is calculated according to the following formula: (88) 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; the change in the resilience level of the charging network is calculated as shown in Figure 10 .

[0066] In summary, the power-transportation coupled charging network resilience evaluation method provided in this embodiment comprehensively considers the coupling relationship among the power network, transportation network, and charging network, and can integrate the interdependent systems into a whole for studying their characteristics. By using a cascading failure model that takes into account the redistribution of dynamic traffic flow, considering the constraints of bus voltage and the security constraints of line power flow, this embodiment can accurately analyze the secondary impact of the transportation system on the power system. Based on the dynamic Bayesian network, this embodiment studies the operation changes of the coupled system under disturbances or faults, and can accurately predict possible fault conditions at different times.

[0067] Embodiment 2

[0068] Based on the same inventive concept as Embodiment 1, this embodiment also provides a power-transportation coupled charging network resilience evaluation system. Since the principle of solving problems by this system is similar to that of the aforementioned power-transportation coupled charging network resilience evaluation method, the implementation of this system can refer to the implementation of the power-transportation coupled charging network resilience evaluation method.

[0069] As Figure 11 shown, the power-transportation coupled charging network resilience evaluation system includes: The first construction module 10 is used to construct a power-transportation coupled charging network according to the power system, transportation system, and electric vehicle charging stations in the target area.

[0070] 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 wind speed of each node on all transmission lines affected by the typhoon at different times according to the Holland wind field model, so as to construct the initial fault model of the power system transmission line.

[0071] 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 distribution and the electric vehicle redistribution principle.

[0072] 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.

[0073] The fifth construction module 50 is used to construct a cascading failure 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.

[0074] The first determination module 60 is configured to, according to the initial fault model and the cascading fault model, respectively 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.

[0075] The sixth construction module 70 is configured to construct a fault prediction model based on a dynamic Bayesian network, and use the wind speed data sets, fault state data sets of each transmission line, and the fault state data sets of each charging station as inputs, and use the conditional probability parameters of the dynamic Bayesian network model as outputs for training to obtain a trained fault prediction model.

[0076] The second determination module 80 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.

[0077] Exemplarily, the second construction module includes: The first calculation unit is configured to calculate the power outage probability of each transmission line at the target time according to the following formula: ; ; ; ; .

[0078] Where, P L (H L (t)) is the power outage 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 abscissa of the randomly generated control point; y(t) is the ordinate of the typhoon on the geographic 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; The symbol "C" represents the combination operation; the symbol "!" represents the factorial operator.

[0079] Exemplarily, the first determination module includes: An acquisition unit, configured to acquire the wind field wind speed of the target transmission line affected by a typhoon at a target moment.

[0080] A first determination unit, 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.

[0081] A random number generation unit, 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.

[0082] A second determination unit, 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; 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.

[0083] 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 fault status of the charging stations in a single typhoon accident.

[0084] A fourth determination unit, 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, 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 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 is 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.

[0085] Exemplarily, the sixth construction module includes: A first construction unit, configured to construct an expression of a fault prediction model G based on a dynamic Bayesian network B : ; ; ; ; ; ; .

[0086] Among them, 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; 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.

[0087] 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: .

[0088] 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.

[0089] 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.

[0090] Exemplarily, the second determination module includes: A second calculation unit, configured to calculate the resilience index of the target charging station at the target moment according to the following formula: .

[0091] Wherein, R c (t) is the resilience index of charging station c at moment t; P c (S t = 0) represents the probability that charging station c fails at moment t; S t = 0 indicates that the state value at moment t is 0; T is the total number of preset moments.

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

[0093] Wherein, R NET (t) is the resilience index of the power-traffic coupled charging network at moment 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.

[0094] 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.

[0095] Embodiment 3

[0096] 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-traffic coupled charging network resilience evaluation method described in Embodiment 1 are implemented.

[0097] For the more specific process of the above method, reference can be made to the corresponding content disclosed in Embodiment 1, and details will not be elaborated here.

[0098] Embodiment 4

[0099] This embodiment provides a computer-readable storage medium, used to store a computer program; when the computer program is executed by a processor, the steps of the power-traffic coupled charging network resilience evaluation method described in Embodiment 1 are implemented.

[0100] For the more specific process of the above method, reference can be made to the corresponding content disclosed in Embodiment 1, and details will not be elaborated here.

[0101] Embodiment 5

[0102] 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.

[0103] For a more specific process of the above method, reference can be made to the corresponding content disclosed in Embodiment 1, which will not be elaborated here.

[0104] In this specification, the various embodiments are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts among the various embodiments, reference can be made 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 descriptions are relatively simple. For the relevant parts, reference can be made to the description in the method section.

[0105] Those skilled in the art can clearly understand that the technologies 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 parts that contribute 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 disc, etc., and includes several instructions to enable 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.

[0106] In some embodiments, the computer-executable instructions can be in the form of a program, software, software module, script, or code, and can be 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.

[0107] As an example, the computer-executable instructions may or may not correspond to files in the file system, and can be stored as a part of a file that stores other programs or data. For example, they can 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 cooperative files (for example, files that store one or more modules, subroutines, or code portions).

[0108] As an example, the computer-executable instructions may be deployed to execute on one electronic device, or on multiple electronic devices located at one location, or, on multiple electronic devices distributed at multiple locations and interconnected through a communication network.

[0109] The present invention has been described in detail above in conjunction with specific embodiments and exemplary examples, but 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 substitutions, modifications or improvements can be made to the technical solutions and implementation manners of the present invention, and these all fall within the scope of the present invention. The protection scope of the present invention shall be subject to the appended claims.

Claims

1. A method for evaluating the resilience of a power-transportation coupled charging network, characterized in that, Including: Construct a power - traffic coupled charging network based on the power system, transportation system, and charging stations for electric vehicles within the target area; According to the nodes on each transmission line of the power system within the target area and the historical typhoon wind speed at the initial moment, use Bessel curves to simulate the random path of typhoon movement, and determine 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, so as to construct the initial fault model of the power system transmission lines; Construct a dynamic traffic flow assignment model based on cellular transmission according to the time - coupled flow propagation characteristics of traffic assignment and the principle of electric vehicle redistribution; 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; Construct a cascading fault model according to the dynamic traffic flow assignment model, AC optimal power flow convergence model, and the situation of charging stations stopping operation; 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 with 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; 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 use the conditional probability parameters of the dynamic Bayesian network model as outputs for training to obtain a trained fault prediction model; 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.

2. The power-transportation coupled charging network resilience evaluation method according to claim 1, wherein The step of using Bessel curves to simulate the random path of typhoon movement 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 to construct the initial fault model of the power system transmission lines, including: Calculate the power outage probability of each transmission line at the target moment 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.

3. The power-transportation coupled charging network resilience evaluation method according to claim 1, wherein The step of determining the wind speed data set of each transmission line at different times with the historical typhoon wind speed and determining the fault state data set of each transmission line and each charging station according to the Monte Carlo simulation method according to the initial fault model and the cascading fault model, including: Obtain the wind field wind speed affected by the typhoon of the target transmission line at the target moment; Determine the power outage probability of the target transmission line at the target moment according to the wind field wind speed affected by the typhoon of the target transmission line at the target moment; Use the Monte Carlo simulation method to generate a random number of the power outage probability of the target transmission line at the target moment; 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.

4. 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 taking 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 taking the conditional probability parameters of the dynamic Bayesian network model as outputs for training to obtain a trained fault prediction model, includes: Construct a fault prediction model G based on the dynamic Bayesian network B 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 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 state 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 L represents the wind speed state of 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 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 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.

5. 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 diagram of the fault status of each transmission line and each charging station, 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: 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; 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.

6. 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-transportation coupled charging network according to the power system, transportation 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, so as 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, the AC optimal power flow convergence model, and the situation of the charging station stopping operation; The first determination module is used to determine the wind speed data sets of each transmission line at different times based on the historical typhoon wind speed for the nodes of each transmission line according to the initial fault model and the cascading fault model, and determine the fault status data sets of each transmission line and each charging station according to the Monte Carlo simulation method; The sixth construction module is used to 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 the 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; The second determination module is used to input the measured typhoon wind speed into the trained fault prediction model to obtain the dynamic probability diagram of the fault status 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.

7. The power-transportation coupled charging network resilience assessment system according to claim 6, wherein The second construction module 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 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 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.

8. The power-transportation coupled charging network resilience evaluation system according to claim 6, wherein The first determination module includes: The acquisition unit is used to acquire the wind field wind speed of the target transmission line affected by the typhoon at the target time; The first determination unit is used 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; The random number generation unit is used 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; The second determination unit is used 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; 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 the transmission line fault status data set in a single typhoon accident; The third determination unit is used 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 the charging station fault status data set in a single typhoon accident; The fourth determination unit is used to, when at least one charging station fails, call the cascading fault model. If the power system converges and the cascading fault model does not disconnect other transmission lines, enter the next time and re-acquire the power outage probability of the target transmission line at the target time; 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 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.

9. The power-transportation coupled charging network resilience evaluation system according to claim 6, wherein The sixth construction module 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 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 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 state 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 L represents the wind speed state of 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 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 power system nodes described by the Bayesian network; The second construction unit is used to construct the expression of the set D of the wind speed data set, fault status data set of each transmission line, and the 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; 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.

10. The power-transportation coupled charging network resilience assessment system according to claim 6, wherein The second determination module includes: 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: ； Among them, R c (t) is the resilience index of the charging station c at time t; P c (S t = 0) represents the probability of the 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