A vulnerability assessment method based on hybrid transport allocation factors

Abstract

This invention discloses a vulnerability assessment method based on a hybrid transmission allocation factor, belonging to the field of vulnerability assessment technology for interdependent networks. The method includes: constructing a coupled system model comprising a power network layer, a hydrogen energy network layer, and a logistics network layer; calculating the transmission allocation factors for each of the power network layer, hydrogen energy network layer, and logistics network layer; and establishing inter-layer sensitivity coupling relationships through a coupling weight matrix. This vulnerability assessment method based on a hybrid transmission allocation factor employs layered sensitivity calculation and cross-layer coupling mapping techniques to form a global hybrid transmission allocation factor (MTDF) matrix. This enables the quantification of the cross-layer impact of new energy output fluctuations, automatic identification of key vulnerable nodes, and the establishment of a cascading fault propagation model. By calculating the L2 norm before and after a fault, a vulnerability assessment index is proposed, solving the technical challenge of assessing the coupled vulnerability of the three-layer network of power-hydrogen-logistics in Green Harbor.

Description

Technical Field

[0001] This invention belongs to the field of vulnerability assessment technology for interdependent networks, and specifically relates to a vulnerability assessment method based on a hybrid transmission allocation factor. Background Technology

[0002] Energy sectors should effectively tap their own potential to improve energy efficiency. Common technical means include using alternative clean energy sources and equipment-based energy efficiency enhancement technologies. As important transportation hubs, these energy efficiency enhancement methods will also become key technologies for ports to achieve green development. The transportation capacity of a port logistics network depends on weak transportation nodes. By comprehensively utilizing multiple energy forms such as electricity and hydrogen, and achieving coordinated optimization among multiple energy sources, energy utilization efficiency and the capacity of port logistics transportation nodes can be effectively improved. With the deepening of port electrification, the coupling between logistics and energy systems is becoming increasingly close. Energy-logistics coupling is a significant difference between integrated port energy systems and traditional land-based applications.

[0003] As the backbone of maritime transport networks connecting value chains and markets worldwide, ports are vulnerable to various issues such as extreme weather events and environmental hazards, making their transport capacity particularly crucial after disruptions. Vulnerability assessments of single networks like power grids are well-established, and analytical methods based on complex network theory are considered essential tools for studying complex systems. However, for the multi-layered coupled networks of ports, after a disruption, the fault propagates across multiple layers due to the cascading coupling relationships, leading to a larger-area failure. Yet, assessing the vulnerability of hydrogen-containing green port energy-material coupled networks presents a more complex and multi-dimensional challenge compared to single-network assessments, and lacks precise evaluation metrics. Therefore, a novel vulnerability assessment method based on hybrid transmission allocation factors is needed to address these existing problems. Summary of the Invention

[0004] The purpose of this invention is to provide a vulnerability assessment method based on a hybrid transmission allocation factor to solve the above-mentioned problems.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a vulnerability assessment method based on a hybrid transmission allocation factor, which performs the following steps S1-S6 to complete the vulnerability assessment for the power grid, hydrogen energy network, and logistics network in a port area:

[0006] Step S1: Construct a coupled system model based on complex network theory, including an electric network layer, a hydrogen energy network layer, and a logistics network layer;

[0007] Step S2: Calculate the transmission allocation factor for each layer:

[0008] Power network layer: Based on power transfer distribution factor (PTDF)

[0009] Hydrogen Energy Network Layer: Transport Sensitivity Factor (HTDF) Based on Pipeline Flow Model

[0010] Logistics network layer: Logistics transfer allocation factor (LTDF) based on transport flow.

[0011] Step S3: Establish inter-layer sensitivity coupling relationships through the coupling weight matrix and construct the global hybrid transmission allocation factor (MTDF) matrix;

[0012] Step S4: Solve for the norm of each node column vector based on the MTDF matrix;

[0013] Step S5: Perform Monte Carlo simulation that includes fluctuations in new energy output to realize cascade fault propagation model analysis.

[0014] Step S6: Calculate the relative rate of change of the spectral norm of the sensitivity matrix based on the column vector norms of the MTDF matrix before and after the fault.

[0015] Preferably, the method for establishing the multi-layer network coupled system model in step S1 is as follows:

[0016] The coupled system model is abstracted as a connected graph G = {D, L, W}. Each layer represents a transmission network, and vertices represent corresponding buses or actual facilities. Vertices are connected by links, representing the transmission path of that layer's network. Different networks are connected by coupling links, representing the energy transmission path between networks. Specifically, vertices and links can be described as: power nodes D... e Hydrogen node D h Logistics equipment node D c Transmission line L e Gas transmission line L h Container transport route L c The network consists of coupled lines; each coupled line runs from the starting node to the ending node. The weight W of each link represents the transmission capacity of that path. Based on the transmission characteristics of electricity and hydrogen, the power transmission lines and gas transmission lines are bidirectional. Due to the special nature of port logistics transportation, the direction of container transmission lines remains constant for a period of time. The berth is set as the starting point, and the berth is considered a type of logistics equipment node. Therefore, the coupled system model is represented by both undirected and directed graphs.

[0017] Preferably, the method for establishing the sensitivity matrices of the hydrogen energy network layer and the logistics network layer in step S2 is as follows:

[0018] Following the method for solving the PTDF matrix of a power system, the hydrogen energy network layer is considered to consist of nodes and pipes, and the solution is obtained through the Weymouth equation:

[0019]

[0020] in, P i0 For steady-state pressure, For steady-state flow, C ij P is the pipe constant. i P j This represents the pressure at the starting node i and the ending node j;

[0021] The transmission sensitivity factor matrix of the hydrogen energy network layer is derived by analogy with the power transmission allocation factor:

[0022] Construct the admittance matrix: Y ij =-K ij (i≠j);

[0023] Y ij K represents the equivalent admittance between node i and node j. ij Indicates the branch admittance coefficient;

[0024] Constructing the association matrix:

[0025] M ij This represents the correlation coefficient matrix of the pipeline nodes.

[0026] Preferably, the construction of the MTDF matrix in step S3 satisfies the following relationship:

[0027]

[0028] Among them, W ph W hl W pl These represent the weight matrices for the electro-hydrogen coupling, hydrogen-logistics coupling, and power-logistics coupling, respectively. The rows of the matrices represent the paths in the coupled system model, the columns represent the nodes in the coupled network, and the elements on the diagonal represent the sensitivity matrices for each layer, denoted by Felectricity-hydrogen coupling weight matrix. p For example, each element in the matrix represents the sensitivity of power lines to the flow at power nodes, and the power network layer matrix F p These are the elements on the diagonal of the matrix;

[0029] The other elements are the coupling weight matrices between different layers, and their calculation methods include:

[0030] Electro-hydrogen coupling weight matrix:

[0031] W ph (i,j)=P elec,j

[0032] In the formula, i represents a node in the power grid layer, j represents a node in the hydrogen energy network layer, and P elec,jPower requirements of electrolytic cell nodes;

[0033] Hydrogen energy-logistics coupling weight matrix:

[0034]

[0035] In the formula, k represents the hydrogen fuel cell vehicle node in the logistics network layer. This indicates the power requirements of the hydrogen fuel cell vehicle.

[0036] Power-logistics coupling weight matrix:

[0037] W pl (i,n)=P logistic,n

[0038] In the formula, n represents the electrical equipment node in the logistics network layer, and P logistic,n Power requirements for logistics equipment.

[0039] As a preferred embodiment of the present invention, step S4 involves calculating the norm of the column vector of each node:

[0040] When assessing network vulnerability, the L2 norm is typically calculated for the PTDF column vector (i.e., the vector composed of the sensitivities of all nodes to that line) of a specific path, such as line l.

[0041]

[0042] Wherein: F l Let represent the PTDF column vector corresponding to the l-th line, k represent the nodes in the network, and N represent the total number of nodes in the network.

[0043] Preferably, the cascading fault propagation model constructed in step S5 is as follows:

[0044] Intra-layer fault triggering

[0045] Power network: Line power over-limit triggered interruption based on thermal stability constraints

[0046] Disconnection conditions:

[0047] Among them, P ij Indicates the actual power of the line. Indicates the rated power of the line;

[0048] Hydrogen network layer: Isolation triggered by excessive pipeline flow.

[0049] Disconnection conditions:

[0050] Among them, Q pq Indicates the actual flow rate of the pipeline. Indicates the rated flow rate of the pipeline;

[0051] Logistics network layer: Path overload triggers blocking

[0052] Disconnection conditions:

[0053] F k This represents the actual material flow rate of path k. This represents the rated material flow rate for path k;

[0054] Cross-layer fault mapping

[0055] Through the electro-hydrogen coupling weight matrix W ph Hydrogen energy-logistics coupling weight matrix W pl Power-logistics coupling weight matrix W hl Transmit the impact of the failure to the global network:

[0056] Electro-hydrogen cascade effect:

[0057] Q H2,prod =η elec ·∑W ph (i,j)·P bus,i ;

[0058] Among them, Q elec,j η represents the power supply to the electrolytic cell. elec P represents the electrolysis efficiency. bus,i This represents the power of power node i;

[0059] Hydrogen logistics cascade effect:

[0060]

[0061] in, Indicates the operating power of the transfer vehicle. This indicates the actual power supplied by the transfer vehicle;

[0062] Electro-physical cascade effect:

[0063]

[0064] Among them, P bus,i P represents the power of power node i. req This indicates the rated power of the coupled line.

[0065] The termination conditions for cascade iterations include:

[0066] Convergence condition: No new faulty components are added in two consecutive iterations;

[0067] Forced termination: Reaching the preset maximum number of iterations to prevent infinite loops.

[0068] Preferably, the relative rate of change of the spectral norm of the sensitivity matrix calculated in step S6 is as follows:

[0069] Calculation formula:

[0070]

[0071] Among them, M pre It is the global sensitivity matrix before the fault, namely the coupling sensitivity of the power network layer, hydrogen energy network layer, and logistics network layer, M. post It is the global sensitivity matrix after a fault, and ||·||2 is the spectral norm of the matrix, i.e. the maximum singular value, which reflects the direction in which the system is most sensitive to disturbances.

[0072] Implementation steps:

[0073] Pre-failure calculation: Run the pre-failure electro-hydrogen logistics coupled system model and construct the pre-failure global sensitivity matrix M. pre Calculate its spectral norm;

[0074] Post-fault calculation: After cascading fault simulation, construct the global sensitivity matrix M after the fault. post Calculate its spectral norm;

[0075] Indicator Calculation: Calculate the RSNV value according to the above formula to assess changes in network vulnerability.

[0076] Considering the coupling characteristics of the electricity-hydrogen-logistics network in the Green Harbor electricity-hydrogen-logistics network, this invention takes into account the cascading effects of the interruption event on the power network, hydrogen energy network and logistics network when assessing the vulnerability of the system under an interruption event.

[0077] The technical effects and advantages of this invention are as follows: This vulnerability assessment method based on hybrid transmission allocation factors selects the green port's electricity-hydrogen-logistics coupled network as the object of vulnerability assessment. Building upon existing research on the vulnerability assessment of interdependent networks, it comprehensively considers the impact of cascading failure propagation in the network and proposes a comprehensive vulnerability assessment index. Based on the multi-energy flow synergy characteristics of the green port, it innovatively constructs an electricity-hydrogen-logistics coupled network vulnerability assessment model. By quantifying the cascading effect of hydrogen energy transmission fluctuations and logistics capacity reduction caused by power outages, it overcomes the limitation of neglecting cross-energy coupling vulnerability points in single-energy network vulnerability assessments, achieving a synergistic assessment of port energy supply stability and logistics transportation capacity under outage scenarios. The method is based on the L2 norm change rate of the MTDF matrix. The cascading effect index RSNV quantifies the cascading fault propagation of interruption events across multi-layer networks. Based on the sensitivity of lines to node traffic, vulnerability assessment is conducted, accurately reflecting the vulnerability of the port's coupled network and laying a solid foundation for guiding the construction of the network topology of hydrogen-containing green ports. By constructing a three-layer network coupling model of power-hydrogen-logistics, and employing hierarchical sensitivity calculation and cross-layer coupling mapping techniques, a global hybrid transmission distribution factor (MTDF) matrix is ​​formed to quantify the cross-layer impact of new energy output fluctuations, automatically identify key vulnerable nodes, and establish a cascading fault propagation model. By calculating the L2 norm before and after the fault, a vulnerability assessment index is proposed, solving the technical challenge of assessing the coupling vulnerability of the three-layer network of power-hydrogen-logistics in green ports. Attached Figure Description

[0078] Figure 1 This is a schematic flowchart of the method of the present invention;

[0079] Figure 2 This is a schematic diagram of the green port electricity-hydrogen-logistics coupling of the present invention;

[0080] Figure 3 500 wind scene images generated by Latin hypercube provided in this embodiment of the invention;

[0081] Figure 4 This invention provides 500 photovoltaic scene images generated by Latin hypercube in an embodiment of the invention.

[0082] Figure 5 Five landscape scene images after K-means clustering reduction provided in this embodiment of the invention;

[0083] Figure 6 Five photovoltaic scene diagrams after K-means clustering reduction provided in this embodiment of the invention;

[0084] Figure 7 Norm graph of the column vector of the node before the fault is provided in the embodiment of the present invention;

[0085] Figure 8 The fault-caused node column vector norm graph provided in this embodiment of the invention. Detailed Implementation

[0086] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0087] This invention provides, for example Figure 1 The vulnerability assessment method shown is based on a hybrid transmission allocation factor. For the Green Harbor Electricity-Hydrogen-Logistics Coupled Network (a composite of an electricity network, hydrogen energy network, and logistics network), the following steps S1-S6 are executed to complete a comprehensive vulnerability assessment of the coupled, cascaded, interdependent network system:

[0088] Step S1: Construct a coupled network model. The specific steps are as follows:

[0089] The Green Harbor Electricity-Hydrogen-Logistics Coupling Network includes a power network topology diagram, a hydrogen energy transmission network structure diagram, and a container logistics transmission network structure diagram. Specifically, the vertices and links can be described as: Power node D e Hydrogen node D h Logistics equipment node D c Transmission line L e Gas transmission line L h Container transport route L c And the coupling lines between networks, which run from the starting node to the ending node.

[0090] Step S2: In the power network, the PTDF matrix reflecting the sensitivity of line flow to node flow can be constructed from the admittance matrix and the correlation matrix. By analogy, the HTDF matrix of the hydrogen energy network and the LTDF matrix of the logistics network are solved respectively.

[0091] Step S2.1: Using a 30-node power network, the generator data is modified to simulate the actual port conditions. Wind turbines and photovoltaic power generation can meet the daily operation needs of the port, achieving off-grid operation. The PTDF matrix of the power network is solved using the admittance matrix and correlation matrix of the power network.

[0092] Step S2.2: Solve using the Weymouth equation:

[0093]

[0094] in, P i0 For steady-state pressure, For steady-state flow, C ij Pipe constant P i P j This represents the pressure at the starting node i and the ending node j;

[0095] The transmission sensitivity factor matrix of the hydrogen energy network layer is derived by analogy with the power transmission allocation factor:

[0096] 1. Construct the admittance matrix: Y ij =-K ij (i≠j)

[0097] Y ij K represents the equivalent admittance between node i and node j. ij Indicates the branch admittance coefficient;

[0098] 2. Construct the association matrix: M ij The correlation coefficient matrix representing the pipeline nodes;

[0099] Finally, substituting into the formula, we get: HTDF = M · (Y')-1

[0100] Step S2.3: The port logistics network uses berths, quay cranes, transfer vehicles, and yard cranes as container transfer nodes. The container transfer direction is berth → quay crane → transfer vehicle → yard crane. Different types of equipment are connected in a fully connected manner. Then, the admittance matrix of the logistics network is constructed and the LTDF matrix is ​​calculated.

[0101] Improved BPR function:

[0102]

[0103] In the formula, R represents the path impedance to be solved, Q is the current flow rate, and Q0 is the current flow rate. cap The path capacity is α, and β are BPR parameters, with the default values ​​being α = 0.15 and β = 4.

[0104] Step S3: Port logistics systems differ from traditional transportation systems. Logistics networks transport containers and do not possess energy carrier characteristics; therefore, energy demand in the logistics network is fixed at the network nodes. Quay cranes and yard cranes are powered by the power grid, berths are also powered by the power grid due to shore power facilities, and transfer vehicles are refueled by a hydrogen network. This leads to the coupling relationship of the three-layer network, calculated using the following method: Electro-hydrogen coupling weight:

[0105] W ph (i,j)=P elec,j

[0106] i represents a node in the power network layer, j represents a node in the hydrogen energy network layer, and P elec,j Power requirements of electrolytic cell nodes;

[0107] Hydrogen energy logistics coupling weights:

[0108]

[0109] In the formula, k represents the hydrogen fuel cell vehicle node in the logistics network layer. This indicates the power requirements of the hydrogen fuel cell vehicle.

[0110] Power logistics coupling weights:

[0111] W pl (i,n)=P logistic,n

[0112] In the formula, n represents the electrical equipment node in the logistics network layer, and P logistic,n Power requirements for logistics equipment.

[0113] Step S3.1: Based on the sensitivity matrix of each network layer and the network coupling relationship, the global hybrid transmission allocation factor (MTDF) matrix can be constructed as follows:

[0114]

[0115] Among them, W ph W hl W pl These represent the weight matrices for the electro-hydrogen coupling, hydrogen-logistics coupling, and electricity-logistics coupling, respectively. The rows of the matrices represent the lines of the coupling network, and the columns represent the nodes of the coupling network. F p F h F l The elements on the diagonal of the sensitivity matrices for the power network layer, hydrogen energy network layer, and logistics network layer represent the sensitivity matrices for each layer, respectively. The remaining elements are derived from the sensitivity matrices for each layer and the coupling relationships.

[0116] Step S4: For each node, calculate the norm of its corresponding column in the MTDF, which serves as the vulnerability index for that node. Based on the vulnerability indices, identify the critical nodes that have the greatest impact on the entire coupled network.

[0117] The method for calculating the L2 norm of column nodes is as follows:

[0118]

[0119] In the formula, F l Let represent the PTDF column vector corresponding to the l-th line, k represent the nodes in the network, and N represent the total number of nodes in the network.

[0120] Step S5: After an interruption event occurs, some nodes in the Green Harbor Electricity-Hydrogen-Logistics Coupled Network will fail first. The initial fault is simulated using Monte Carlo simulation, and the propagation process of the fault between the power network layer, the hydrogen energy network layer, and the logistics network layer is simulated using a cascaded fault propagation model. The specific process is shown in the figure.

[0121] Step S5.1: The initial fault simulation process in Monte Carlo is as follows:

[0122] 1. Latin hypercube sampling to generate scenes

[0123] (1) Determine the number of samples m. In this paper, we set it to generate 500 scenarios.

[0124] (2) Determine the distribution function of power generation.

[0125] (3) Divide the equivalent region of the power generation distribution function into m intervals, and each interval has the same probability.

[0126] (4) Independently and randomly draw samples from different intervals and record them. Ensure that only one sample appears in each interval.

[0127] The shuffled samples and generated results are shown in the figure.

[0128] The wind turbine scene generation uses the Weibull distribution, while the photovoltaic scene generation uses the Beta distribution.

[0129] 2. K-means clustering reduction scenario

[0130] (1) Select the number of clusters n to be divided into. In this paper, the clustering is reduced to 5 representative scenarios.

[0131] (2) Randomly select 5 data points from the scene dataset as the initial centroids.

[0132] (3) For each scene point, calculate the distance between it and each centroid, and then assign it to the nearest cluster. The distance is calculated by Euclidean distance.

[0133] (4) For each cluster, calculate the average value of all data points in the cluster and use the average value as the new centroid.

[0134] (5) Repeat steps (3) and (4) until the cluster assignment no longer changes or the maximum number of iterations is reached.

[0135] The five cluster centers are output, representing five representative scenarios. The five clustered wind turbine and photovoltaic scenarios are shown in the figure.

[0136] Step S5.2: As a preferred technical solution of the present invention, the constructed cascading fault propagation model is as follows:

[0137] 1. Intra-layer fault triggering

[0138] Power network layer: Line power over-limit triggered disconnection (based on thermal stability constraints)

[0139] Disconnection conditions:

[0140] Among them, P ij Indicates the actual power of the line. Indicates the rated power of the line;

[0141] Hydrogen Network Layer: Isolation Triggered by Excessive Pipe Flow

[0142] Disconnection conditions:

[0143] Among them, Q pq Indicates the actual flow rate of the pipeline. Indicates the rated flow rate of the pipeline;

[0144] Logistics network layer: Path overload triggers blocking

[0145] Disconnection conditions:

[0146] Among them, F k This represents the actual material flow rate of path k. Indicates the rated material flow rate of path k

[0147] 2. Cross-layer fault mapping

[0148] Through the electro-hydrogen coupling weight matrix W ph Hydrogen energy-logistics coupling weight matrix W pl Power-logistics coupling weight matrix W hl Propagate the impact of the failure to other layers:

[0149] Electro-hydrogen cascade effect:

[0150] Q elec,k =η elec ·∑W ph (i,j)·P bus,i

[0151] Among them, Q elec,j η represents the power supply to the electrolytic cell. elec P represents the electrolysis efficiency. bus,i This represents the power of power node i;

[0152] Hydrogen logistics cascade effect:

[0153]

[0154] in, Indicates the operating power of the transfer vehicle. This indicates the actual power supplied by the transfer vehicle;

[0155] Electro-physical cascade effect:

[0156]

[0157] Among them, P bus,i P represents the power of power node i. req Indicates the rated power of the coupling line;

[0158] 3. Termination conditions for cascade iterations include:

[0159] Convergence condition: No new faulty components are added in two consecutive iterations;

[0160] Forced termination: Reaching the preset maximum number of iterations to prevent infinite loops;

[0161] Step S6: Calculate the relative rate of change of the sensitivity matrix spectral norm based on the column vector norms before and after the cascaded fault. The calculation formula is as follows:

[0162]

[0163] Among them, M pre It is the global sensitivity matrix before the fault, namely the coupling sensitivity of the power network layer, hydrogen energy network layer, and logistics network layer, M. post It is the global sensitivity matrix after a fault, and ||·||2 is the spectral norm of the matrix, i.e. the maximum singular value, which reflects the direction in which the system is most sensitive to disturbances.

[0164] Implementation steps:

[0165] 1. Pre-failure calculation: Run the pre-failure electro-hydrogen logistics coupled system model and construct the pre-failure global sensitivity matrix M. pre Calculate its spectral norm;

[0166] 2. Post-fault calculation: After cascading fault simulation, a global sensitivity matrix M is constructed. post Calculate its spectral norm;

[0167] 3. Indicator Calculation: Calculate the RSNV value according to the formula to assess changes in network vulnerability.

[0168] The following is an embodiment of the present invention, based on the vulnerability assessment method for the green port electricity-hydrogen-logistics coupling system based on the hybrid transmission allocation factor designed in this invention. This method comprehensively assesses the vulnerability of the green port electricity-hydrogen-logistics coupling network under interruption events from three levels: electricity, hydrogen energy, and logistics. This embodiment is analyzed and demonstrated in conjunction with specific data:

[0169] Figure 1 This is a flowchart of the vulnerability assessment method. Figure 2 This is a schematic diagram of the green port's electricity-hydrogen-logistics coupling. In the power grid, wind turbines and photovoltaics provide power support for the system. In the hydrogen energy network, electrolyzers use the power grid to produce hydrogen to support the hydrogen load within the system. Specifically, grid nodes 6 and 22 supply power to hydrogen grid nodes 1 and 2, hydrogen grid nodes 4-7 transport hydrogen to logistics network nodes 6-11, grid nodes 1 and 2 provide shore power to logistics network nodes 1 and 2, grid nodes 3-5 provide power to logistics network nodes 3-5, and grid nodes 12-15 provide power to logistics network nodes 12-15.

[0170] Step S1 abstracts the IEEE 30-node power network, 8-node hydrogen energy network, and 15-node port logistics network into a connected graph G = {D, L, W}. Specifically, the vertices and links can be described as: power node D e Hydrogen node D h Logistics equipment node D c Transmission line L e Gas transmission line L h Container transport route L c .

[0171] For the power network, the PTDF matrix is ​​solved using the method in step S2.1; for the hydrogen energy network, the flow rate of the hydrogen pipeline is calculated using the improved Weymouth equation flow resistance in step S2.2, and the node-pipeline correlation matrix and the admittance matrix of the hydrogen energy network are constructed, and finally the HTDF matrix is ​​constructed; for the logistics network, the corrected impedance is solved using the BPR function in step S2.3, the container transmission direction is berth → quay crane → transfer vehicle → yard crane, and a full connection method is adopted between different types of equipment. Then the admittance matrix of the logistics network is constructed and the LTDF matrix is ​​calculated.

[0172] In step S3, sensitivity matrices are constructed for the coupling relationships of electricity-hydrogen energy, electricity-logistics, and hydrogen-logistics. Subsequently, a global hybrid transmission allocation factor matrix (MTDF) is constructed. The rows of the matrix represent the lines of the coupled network, the columns represent the nodes of the coupled network, and the elements on the diagonal of the matrix represent the sensitivity matrices of each layer.

[0173] Next, step S4 is used to process the elements in the MTDF matrix. The norm of the column vector in the MTDF matrix is ​​calculated as the vulnerability index of each point. The nodes are sorted according to their vulnerability indices, and the key nodes with greater impact are identified as shown in Table 1.

[0174] Table 1. Key nodes with the highest vulnerability indices in each layer of the network.

[0175] Network layer Node number Vulnerability Indicators Power network layer 22 9.3838 Hydrogen network layer 2 14.5867 Logistics network layer 2 2.2700

[0176] After calculating the node norms of the initial network, the initial fault was simulated using Monte Carlo simulation in step S5, and the propagation process of the fault between the power network layer, the hydrogen energy network layer, and the logistics network layer was simulated using a cascaded fault propagation model. 500 wind power outputs following a Weibull distribution and photovoltaic outputs following a Beta distribution were generated using historical data. Figure 3 , Figure 4 As shown, K-means clustering was then used to reduce the number of scenarios to five, as follows: Figure 5 , Figure 6 As shown, the wind turbine and photovoltaic output are then added to the cascaded fault model to simulate fault propagation after the interruption event. The norm of the node column vector after the fault is obtained through steps S3 and S4, as shown below. Figure 7 , Figure 8 As shown.

[0177] Finally, based on the node column vector norms before and after the fault, the relative change rate of the sensitivity matrix spectral norm in step S6 was calculated to be 15.69, as shown in Table 2. The vulnerability of different greenport electricity-hydrogen-logistics coupling networks can be determined by the relative change rate of the sensitivity matrix spectral norm (RSNV).

[0178] Table 2 Rate of change of spectral norm before and after the fault

[0179]

[0180]

[0181]

[0182]

[0183] This vulnerability assessment method based on hybrid transmission allocation factors selects the green port's electricity-hydrogen-logistics coupled network as the object of vulnerability assessment. Building upon existing research on vulnerability assessment of interdependent networks, it comprehensively considers the impact of cascading failure propagation and proposes a comprehensive vulnerability assessment index. Based on the multi-energy flow synergy characteristics of the green port, it innovatively constructs an electricity-hydrogen-logistics coupled network vulnerability assessment model. By quantifying the cascading effect of hydrogen energy transmission fluctuations and logistics capacity reduction caused by power outages, it overcomes the limitation of neglecting cross-energy coupling vulnerability points in single-energy network vulnerability assessments, achieving a synergistic assessment of port energy supply stability and logistics transportation capacity under outage scenarios. The cascading effect index is based on the L2 norm change rate of the MTDF matrix. The RSNV (Resource-Side Variable Number) metric quantifies the cascading fault propagation of outage events across multi-layer networks. Based on the sensitivity of lines to node traffic, it performs vulnerability assessment, accurately reflecting the vulnerability of the port's coupled network and laying a solid foundation for guiding the construction of a network topology for hydrogen-containing green ports. By constructing a three-layer network coupling model of power, hydrogen energy, and logistics, and employing hierarchical sensitivity calculation and cross-layer coupling mapping techniques, a global hybrid transmission distribution factor (MTDF) matrix is ​​formed. This enables the quantification of the cross-layer impact of new energy output fluctuations, automatic identification of key vulnerable nodes, and the establishment of a cascading fault propagation model. By calculating the L2 norm before and after the fault, a vulnerability assessment index is proposed, solving the technical challenge of assessing the vulnerability of the three-layer network coupling of power, hydrogen energy, and logistics in green ports.

[0184] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A vulnerability assessment method based on a hybrid transmission allocation factor, characterized in that: include: Construct a coupled system model that includes an electric network layer, a hydrogen energy network layer, and a logistics network layer; Calculate the transmission allocation factors for the power network layer, hydrogen energy network layer, and logistics network layer respectively: Inter-layer sensitivity coupling relationships are established by coupling weight matrices, and a global hybrid transmission allocation factor matrix is ​​constructed. The norm of each node's column vector is calculated based on the global hybrid transmission allocation factor matrix. Simulate the probability of power generation failure in new energy sources; The relative rate of change of the spectral norm of the sensitivity matrix is ​​calculated based on the column vector norms of the global hybrid transmission allocation factor matrix before and after the fault. The method for establishing the sensitivity matrix of the hydrogen energy network layer and the logistics network layer includes: The hydrogen energy network layer is set as a node-pipeline, and the solution is obtained through the Weymouth equation: ;in, P i0 Indicates steady-state pressure. C represents steady-state flow. ij P represents the pipe constant. i P j This represents the pressure at the starting node i and the ending node j; The transmission sensitivity factor matrix of the hydrogen energy network layer is derived by analogy with the power transmission allocation factor, including: Construct the admittance matrix: Y ij =-K ij (i≠j); Y ij K represents the equivalent admittance between node i and node j. ij Indicates the branch admittance coefficient; Constructing the association matrix: M ij The correlation coefficient matrix representing the pipeline nodes; The construction of the global hybrid transmission allocation factor matrix satisfies the following relationship: Among them, W ph W hl W pl Let F represent the weight matrices for the electro-hydrogen coupling, hydrogen-logistics coupling, and electricity-logistics coupling, respectively. The rows of the matrices represent the paths in the coupled system model, and the columns represent the nodes in the coupled network. p F h F l These represent the sensitivity matrices for the power network layer, the hydrogen energy network layer, and the logistics network layer, respectively. Electro-hydrogen coupling weight matrix: W ph (i,j)=P elec,j i represents a node in the power network layer, j represents an electrolyzer node in the hydrogen energy network layer, and P elec,j This indicates the power requirements of each node in the electrolytic cell; Hydrogen energy-logistics coupling weight matrix: W hl (j,k)=Q H2,k In the formula, k represents the hydrogen fuel cell vehicle node in the logistics network layer, and Q... H2,k This indicates the power requirements of hydrogen fuel cell transport vehicles; Power-logistics coupling weight matrix: W pl (i,n)=P logistic,n In the formula, n represents the electrical equipment node in the logistics network layer, and P logistic,n Power requirements for logistics equipment.

2. The vulnerability assessment method based on hybrid transmission allocation factor according to claim 1, characterized in that: The calculation of transmission allocation factors for the power network layer, hydrogen energy network layer, and logistics network layer includes: the power network layer is calculated based on the power transmission distribution factor; the hydrogen energy network layer is calculated based on the transmission sensitivity factor of pipeline flow; and the logistics network layer is calculated based on the logistics transmission allocation factor of transportation flow.

3. The vulnerability assessment method based on hybrid transmission allocation factor according to claim 1, characterized in that: The method for constructing the coupled system model includes: constructing a connected graph G = {D, L, W}, where vertices G include: power nodes D e Hydrogen node D h and logistics equipment node D c Link L includes: transmission line L e Gas transmission line L h Container transport route L c And the coupling lines between networks; The coupled line runs from the starting node to the ending node; the weight W of each link represents the transmission capacity of that link, and the transmission line L... e With gas transmission line L h For bidirectional transmission, container transport line L c With the direction constant, set the berth as the starting point.

4. The vulnerability assessment method based on hybrid transmission allocation factor according to claim 1, characterized in that: The process of solving for the norm of each node's column vector includes: When assessing network vulnerability, the L2 norm is calculated for the PTDF column vector of the l-th line: Among them, F l Let represent the PTDF column vector corresponding to the l-th line, k represent the nodes in the network, and N represent the total number of nodes in the network.

5. The vulnerability assessment method based on a hybrid transmission allocation factor according to claim 4, characterized in that: The simulated new energy power output failure probability is used to realize the cascaded failure propagation model analysis, which includes: Intra-layer fault triggering: Power network layer: Line power over-limit triggers disconnection, disconnection condition: Among them, P ij Indicates the actual power of the line. Indicates the rated power of the line; Hydrogen network layer: Isolation triggered by excessive pipeline flow; disconnection condition: Among them, Q pq Indicates the actual flow rate of the pipeline. Indicates the rated flow rate of the pipeline; Logistics network layer: Path overload triggers blocking; disconnection condition: Among them, F k F represents the actual material flow rate of path k. k max This represents the rated material flow rate for path k.

6. The vulnerability assessment method based on hybrid transmission allocation factor according to claim 5, characterized in that: The cascaded fault propagation model also includes: cross-layer fault mapping: Through the electro-hydrogen coupling weight matrix W ph Hydrogen energy-logistics coupling weight matrix W pl Power-logistics coupling weight matrix W hl Transmit the impact of the failure to the global network: Electro-hydrogen cascade effect: Q elec,k =η elec ·∑W ph (i,j)·P bus,i； Among them, Q elec,k η represents the power supply to the electrolytic cell. elec P represents the electrolysis efficiency. bus,i This represents the power of power node i; Hydrogen logistics cascade effect: ;in, Q represents the operating power of the transfer vehicle. H2,k This indicates the actual power supplied by the transfer vehicle; Electro-physical cascade effect: Among them, P bus,i P represents the power of power node i. req This indicates the rated power of the coupled line.

7. The vulnerability assessment method based on a hybrid transmission allocation factor according to claim 6, characterized in that: The cascaded fault propagation model also includes: cascaded iteration termination condition: Convergence condition: No new faulty nodes are added in two consecutive iterations; Forced termination: The preset maximum number of iterations has been reached.

8. The vulnerability assessment method based on hybrid transmission allocation factor according to claim 1, characterized in that: The relative rate of change of the spectral norm of the calculated sensitivity matrix includes: Pre-failure calculation: Run the pre-failure electro-hydrogen logistics coupled system model and construct the pre-failure global sensitivity matrix M. pre Calculate its spectral norm; Post-fault calculation: After cascading fault simulation, construct the global sensitivity matrix M after the fault. post Calculate its spectral norm; Indicator Calculation: The RSNV value is calculated using the following formula to assess changes in network vulnerability; Among them, M pre It is the global sensitivity matrix before the fault, M post It is the global sensitivity matrix after the fault, and ||·||2 is the spectral norm of the matrix.

Images