Two-stage toughness evaluation method for electro-hydrogen comprehensive energy system under cross-regional resource sharing

Abstract

The invention discloses a two-stage toughness evaluation method for an electro-hydrogen comprehensive energy system under cross-regional resource sharing, and the method comprises the following steps: (1) fusing and embedding a cross-regional sharing mechanism of mobile emergency resources, such as hydrogen energy, based on a disaster response architecture of active deployment before a disaster and cooperative recovery after the disaster; and establishing a two-stage toughness evaluation framework of the electro-hydrogen comprehensive energy system under cross-regional resource sharing. And (2) fully considering the uncertainty of multi-region fault line distribution in the electricity-hydrogen integrated energy system and the cross-region mutual aid potential of the mobile power supply, and proposing the pre-disaster cross-region deployment method of the mobile power supply of the electricity-hydrogen integrated energy system based on the SAC (Soft Actor-Critic) algorithm. And (3) putting forward a post-disaster recovery strategy of the electro-hydrogen comprehensive energy system under cross-regional resource sharing by considering the cross-regional support potential and post-disaster cooperation potential of the mobile emergency resources. And (4) starting from the resistance and restoring force of the system under extreme disasters, establishing multi-dimensional toughness evaluation indexes of the electro-hydrogen comprehensive energy system, and proposing the toughness evaluation method of the electro-hydrogen comprehensive energy system based on Latin hypercube sampling. According to the method, the cross-regional mutual aid potential of the mobile emergency resources in the pre-disaster active deployment stage and the post-disaster collaborative recovery stage is fully considered, and the toughness level of the electric-hydrogen comprehensive energy system can be accurately evaluated.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of comprehensive energy system applications, in particular to a two-stage resilience evaluation method for an electric-hydrogen comprehensive energy system under cross-regional resource sharing. BACKGROUND

[0002] Hydrogen energy has advantages such as abundant sources and high energy density. Combined with its cross-time and space flexible adjustment characteristics, it can not only enhance the flexibility of the power distribution network, but also improve the resilience of the power distribution network through hydrogen fuel power generation vehicles and hydrogen long tube trailers and other hydrogen energy equipment. At the same time, due to the high redundancy of the transportation network, hydrogen energy can still be reliably transferred between different load nodes and regions under extreme disasters. At present, hydrogen energy technology is constantly maturing and improving. The installed capacity of electrolytic cells in China has broken through 47.7GW, and more than 540 hydrogen refueling stations have been built. With the development of hydrogen energy equipment technology, the coupling between the power grid and the hydrogen energy system will be closer, thereby further strengthening the resilience support capability of hydrogen energy to the power grid.

[0003] Researching the resilience evaluation method of the electric-hydrogen comprehensive energy system can scientifically and quantitatively evaluate the resilience level of the EH-IES under extreme disasters and fully tap its disaster resistance potential. This not only helps the resource allocation of the electric-hydrogen comprehensive energy system, but also provides theoretical support and decision basis for the optimal operation of the electric-hydrogen comprehensive energy system under extreme disasters. Therefore, researching the resilience evaluation method of the electric-hydrogen comprehensive energy system responds to the demand of the national energy security strategy and has important theoretical research value and engineering application prospect.

[0004] However, the existing resilience evaluation method only focuses on a single region and cannot quantify the improvement effect of cross-regional mutual aid of mobile emergency resources on resilience. At the same time, the existing method is based on the post-disaster coordinated scheduling of distributed power sources, mobile power sources, and maintenance personnel and other emergency resources, and constructs a resilience evaluation model with the minimum system load shedding loss as the target, focusing on the evaluation of post-disaster response capability, generally ignoring the improvement effect of pre-disaster deployment of emergency resources on system resilience, which greatly affects the accuracy of the resilience evaluation of the electric-hydrogen comprehensive energy system. SUMMARY

[0005] The purpose of the present application is to provide a two-stage resilience evaluation method for an electric-hydrogen comprehensive energy system under cross-regional resource sharing, comprising the following steps:

[0006] 1) Fully considering the uncertainty of the multi-region fault line distribution in the electric-hydrogen comprehensive energy system and the cross-regional mutual aid potential of the mobile power source, a pre-disaster deployment strategy for the multi-region electric-hydrogen comprehensive energy system based on the discrete SAC algorithm is generated;

[0007] 2) Considering the cross-regional support potential and post-disaster coordination potential of mobile emergency resources, a post-disaster recovery strategy for the electric-hydrogen comprehensive energy system under cross-regional resource sharing is generated;

[0008] 3) Based on the pre-disaster deployment strategy of multi-region electricity-hydrogen integrated energy system, the post-disaster recovery strategy of electricity-hydrogen integrated energy system under cross-region resource sharing, the resilience of electricity-hydrogen integrated energy system is evaluated.

[0009] Further, in step 1), the step of generating a pre-disaster deployment strategy of multi-region electricity-hydrogen integrated energy system based on discrete SAC algorithm includes:

[0010] 1.1) Constructing a post-disaster emergency response model of electricity-hydrogen integrated energy system;

[0011] 1.2) Model the pre-disaster cross-region deployment decision-making process of electricity-hydrogen integrated energy system, and construct state space, action space and environment;

[0012] 1.3) Use the strategy network to generate the pre-disaster deployment position of the mobile power source of the multi-region electricity-hydrogen integrated energy system; the strategy network takes the state of the agent as the input and the action as the output;

[0013] 1.4) Based on the pre-disaster deployment position of the mobile power source of the multi-region electricity-hydrogen integrated energy system, solve the post-disaster emergency response model of the electricity-hydrogen integrated energy system to obtain the load shedding loss under the current deployment position, as the agent reward function;

[0014] 1.5) Use two comment networks to evaluate the value of the action output by the strategy network, and select the action with the maximum value as the pre-disaster deployment strategy of the multi-region electricity-hydrogen integrated energy system.

[0015] Further, the post-disaster emergency response model of the electricity-hydrogen integrated energy system aims to minimize the expected total operating cost of the regional electricity-hydrogen integrated energy system, that is:

[0016]

[0017] The constraint conditions of the post-disaster emergency response model of the electricity-hydrogen integrated energy system include mobile power source operation constraint, hydrogen energy system operation constraint, network reconstruction constraint, abandoned load balance constraint, and power grid power flow constraint;

[0018] The mobile power source operation constraint includes mobile power storage discharge power constraint, mobile power storage energy balance constraint, mobile power storage upper / lower limit constraint, hydrogen fuel power generation vehicle discharge power constraint, hydrogen fuel power generation vehicle energy balance constraint, hydrogen fuel power generation vehicle hydrogen storage upper / lower limit constraint, and grid access point mobile power source maximum grid access number limit constraint;

[0019] The hydrogen energy system operation constraint includes hydrogen fuel cell operation power constraint, hydrogen storage tank hydrogen storage upper / lower limit constraint, and hydrogen storage tank energy balance constraint;

[0020] Wherein, the mobile power source operation constraint is as follows:

[0021]

[0022] The hydrogen energy system operation constraints are as follows:

[0023]

[0024] The network reconfiguration constraints are as follows:

[0025]

[0026] The abandoned load balancing constraints are as follows:

[0027]

[0028]

[0029] The distribution network power flow constraints are as follows:

[0030]

[0031] Further, the state space is as follows:

[0032] o = [U branch,all , P load,all , P H,all , Q H,all , P EV,all , Q EV,all , P HEV,all , Q HEV,all ] (31)

[0033] In the formula, o represents the state space of the intelligent agent, including the topological state information U branch,all of all areas, the load prediction information P load,all , the maximum power generation information P H,all of the hydrogen energy system, the hydrogen energy storage capacity information Q H,all , the maximum power generation information P EV,all of all mobile electric energy storage, the mobile electric energy storage capacity information Q EV,all , the maximum power generation information P HEV,all of the hydrogen fuel power generation vehicle, and the hydrogen fuel power generation vehicle capacity information Q HEV,all .

[0034] The action space is as follows:

[0035]

[0036] In the formula, a is the action space of the intelligent agent; u i,n represents the position state variable of the mobile electric energy storage / hydrogen fuel power generation vehicle at different power grid access points; a n represents the deployment of the nth mobile power supply to the first an Grid access point, at this time the nth mobile power supply is deployed at a n location, the state variable of the location is set to 1, indicating that it is deployed at a n node; N M,all is the maximum number of grid access points; N EV ,all , N HEV,all is the maximum deployment number of mobile energy storage / hydrogen fuel power generation vehicles;

[0037] The environment includes a post-disaster emergency response model of an integrated electricity-hydrogen energy system;

[0038] The reward function is as follows:

[0039]

[0040] In the formula, the reward function r up of the agent is the sum of the load shedding losses of all areas, r i down is the reward function of the ith lower-level agent, is the total penalty of the cut-off load of area i; is the total penalty of the cut-off hydrogen load of area i.

[0041] Further, the policy network generates a softmax distribution to output the corresponding probabilities of all possible discrete actions;

[0042] The softmax distribution is as follows:

[0043]

[0044] In the formula, x n,j is the corresponding preference value of the jth action dimension in the nth action output by the policy network, reflecting the relative selection tendency of the nth action selecting the jth action dimension; p n,j is the selection probability of the action;

[0045] The loss function of the policy network is as follows:

[0046]

[0047] In the formula, D is the experience pool, a H is the temperature coefficient; is the entropy regularization term; Q θ (o, a) is the Q value function; E o,a～D is the expectation;

[0048] The entropy regularization term is as follows:

[0049] ​

[0050] where A is the total number of actions of the agent, K n is the total action latitude of the nth action, respectively;

[0051] The Q-value function is as follows:

[0052]

[0053] where, are two Q-value functions with the same structure;

[0054] The Q-value network loss function is as follows:

[0055]

[0056] where L(θ1), L(θ2) are the losses of the Q-value functions ; r is the reward; E (s,a,r)～D is the expectation;

[0057] The loss function L(α H ) of the temperature coefficient is as follows:

[0058]

[0059] where H tar is the target entropy of the agent; is the expectation;

[0060] The action network, the two evaluation networks, and the temperature coefficient are updated as follows:

[0061]

[0062] where α represents the learning rate of the agent; L(θ m ) represents the Q-value network loss function; θ m represents the evaluation network parameter; represents the action network parameter.

[0063] Further, in step 2), the step of generating the post-disaster recovery strategy of the cross-region resource sharing electric-hydrogen integrated energy system includes:

[0064] 2.1) constructing an electric-hydrogen integrated energy system post-disaster recovery model considering emergency resource collaborative scheduling;

[0065] 2.2) constructing a mobile emergency resource post-disaster allocation model considering cross-region resource sharing;

[0066] 2.3) solving the electric-hydrogen integrated energy system post-disaster recovery model and the mobile emergency resource post-disaster allocation model to obtain the post-disaster recovery strategy of the electric-hydrogen integrated energy system under cross-region resource sharing.

[0067] Further, the objective function C1 of the post-disaster recovery model of the integrated electricity-hydrogen energy system considering the coordinated dispatch of emergency resources is as follows:

[0068] min C1=C lose,E +C lose,H +C tran +C gen,E +C gen,H -E disp (46)

[0069]

[0070] In the formula, N T2 represents a set of post-disaster recovery scheduling periods in the region; c tran is the unit scheduling cost of mobile emergency resources; respectively represent the state variables of the nth group of mobile electric energy storage, hydrogen fuel power generation vehicles and repair personnel located on the road at time t; N RC represents a set of repair personnel; c disp is the unit subsidy of the mobile emergency resources participating in the inter-regional allocation; respectively represent the allocable state of the nth group of mobile electric energy storage / hydrogen fuel power generation vehicles / repair personnel at time t, if 1, it represents that it can participate in the inter-regional allocation at time t, otherwise, it cannot. C tran is the total scheduling cost of mobile emergency resources; E disp is the subsidy of the mobile emergency resources participating in the inter-regional allocation;

[0071] The constraint conditions of the post-disaster recovery model of the integrated electricity-hydrogen energy system considering the coordinated dispatch of emergency resources include the space-time scheduling constraints of mobile emergency resources, the post-disaster recovery operation constraints of mobile power sources, the post-disaster recovery operation constraints of hydrogen energy systems, the line repair state constraints, the allocable state constraints of mobile emergency resources, the distribution network reconfiguration constraints (16)-(19), the electricity-hydrogen load balance constraints (21)-(24) and the distribution network power flow constraints (25)-(31);

[0072] The space-time scheduling constraints of mobile emergency resources include the space-time position uniqueness constraints of mobile emergency resources, the position change state constraints of mobile emergency resources, the limitation constraints that the position transfer of mobile emergency resources among all nodes must pass through the road, and the space-time position transfer constraints of mobile emergency resources, that is:

[0073]

[0074] In the formula, u i,t,n is the state variable of the space-time position of mobile emergency resources, if 1, it represents that the nth group of mobile emergency resources is located at node i at time t, otherwise, it is not; u 0,t,nis the state variable of the mobile emergency resource at time t on the road; N R represents a set of nodes of the traffic network coupled with grid access nodes, fault lines, hydrogen energy systems, or hydrogen refueling stations; N MER represents a set of all mobile emergency resources; h i,t,n is the state variable of the position change of the nth group of mobile emergency resources at time t at node i; h i,t,n is 1 / -1, respectively, representing that the nth group of mobile emergency resources enters / leaves node i at time t; T ij is the time required to move from node i to node j; h 0,t,n represents that the nth group of mobile emergency resources still stays at node i after time t;

[0075] The post-disaster recovery operation constraints of the mobile power supply include the uniqueness constraints of the charging / discharging state of the mobile electric energy storage, the charging / discharging power constraints of the mobile electric energy storage, the energy balance constraints of the mobile electric energy storage, the upper / lower limit constraints of the energy storage of the mobile electric energy storage, the hydrogen refueling / discharging state uniqueness constraints of the hydrogen fuel power generation vehicle, the hydrogen refueling rate / discharging power constraints of the hydrogen fuel power generation vehicle, the energy balance constraints of the hydrogen fuel power generation vehicle, the upper / lower limit constraints of the hydrogen storage of the hydrogen fuel power generation vehicle, and the maximum number of mobile power supply access to the grid access point, i.e.

[0076]

[0077] In the formula, are the state variables of the charging / discharging of the nth mobile electric energy storage at node i at time t, and if 1, it represents that the nth mobile electric energy storage is in the charging / discharging state at node i at time t; is the charging power of the nth mobile electric energy storage at node i at time t; P EV,in,max is the upper limit of the charging power of the mobile electric energy storage; η EV,in , η EV,out is the charging / discharging efficiency of the mobile electric energy storage; are the state variables of the hydrogen refueling / discharging of the nth hydrogen fuel power generation vehicle at node i at time t, and if 1, it represents that the nth hydrogen fuel power generation vehicle is in the hydrogen refueling / discharging state at node i at time t; N H is a set of nodes of the hydrogen energy system or hydrogen refueling station; is the hydrogen refueling mass of the nth hydrogen fuel power generation vehicle at node i at time t; M HEV ,in,max is the upper limit of the hydrogen refueling rate of the hydrogen fuel power generation vehicle; is the state variable of the nth mobile electric energy storage at node i at time t; is the state variable of the nth hydrogen fuel power generation vehicle at node i at time t; is the discharging power of the nth mobile electric energy storage at node i at time t; PEV,out,max is the maximum discharging power; is the energy storage level of the nth mobile energy storage at time t; is the discharging power of the nth hydrogen fuel cell vehicle at time t at node i; P HEV,out,max is the upper limit of discharging power;

[0078] The post-disaster recovery operation constraints of the hydrogen energy system include the uniqueness constraint of the operation state of the electrolyzer and the hydrogen fuel cell, the operation power constraint of the electrolyzer, the operation power constraint of the hydrogen fuel cell, the upper / lower limit constraint of the hydrogen storage of the hydrogen storage tank, and the energy balance constraint of the hydrogen storage tank, i.e.

[0079]

[0080] In the formula, are the operation state variables of the electrolyzer / hydrogen fuel cell at time t, and if 1, it represents that the electrolyzer / hydrogen fuel cell is in working state at time t, and if only hydrogenation stations are deployed in the region, and are 0; is the operation power of the electrolyzer at time t; P ED,max is the upper limit of the operation power of the electrolyzer; η ED is the energy conversion efficiency of the electrolyzer; η HS,in / η HS,out are the hydrogen charging / discharging efficiencies of the hydrogen storage tank; N HTT is a set of hydrogen long tube trailers; are the hydrogen charging / discharging masses of the nth hydrogen long tube trailer at time t;

[0081] The line repair state constraint includes that the damaged line can be restored to the available state after the repair personnel arrive and stay on the line for the required repair time, the damaged line can only be repaired by a group of repair personnel, and the actual closed state of the branch is jointly constrained by the virtual power flow model and the available state of the line, i.e.

[0082]

[0083] In the formula, t0 is the starting time of post-disaster recovery; is the time required for repairing the line ij; N D is a set of damaged lines;

[0084] The mobile emergency resource allocatable state constraint includes that the region is considered to have been completely recovered when all the loads can be supplied without hydrogen energy system and mobile power supply, the mobile power supply can only participate in the inter-regional allocation when it has no charging or power supply task at the current time, and the repair personnel can only participate in the inter-regional allocation after the region is completely recovered, i.e.

[0085]

[0086]

[0087] wherein, is the regional recovery state variable, and if it is 1, it represents that the region has been completely recovered; is the assignable state of the nth group of repair personnel in the inter-regional allocation this time;

[0088] Further, the objective function C2 of the post-disaster allocation model of the cross-regional resource sharing mobile emergency resource is as follows:

[0089] min C2 = C lose,E,all + C lose,H,all + C tran,all + C gen,E,all + C gen,H,all + C re,R,all (80)

[0090]

[0091] wherein, C lose,E,all , C lose,H,all , C tran,all , C gen,E,all , C gen,H,all , C re,R,all are respectively the total penalty of the cut-off load, the total penalty of the cut-off hydrogen load, the total scheduling cost of the mobile emergency resource, the total operation cost of the mobile electric energy storage, the total operation cost of the hydrogen fuel cell in the hydrogen fuel power generation vehicle and the hydrogen energy system, and the total penalty of the remaining repair time of the key line in the electric-hydrogen integrated energy system; N MG represents a set of all microgrids in the electric-hydrogen integrated energy system; represents a set of all loads at all levels in the microgrid m; N T3 represents a set of inter-regional allocation scheduling periods; c lose,El is the unit cut-off load penalty of the lth level load; P lose,Em,l,t is the cut-off load amount of the lth level load at time t in the microgrid m; N H,all represents a set of all hydrogen energy systems or hydrogen filling station nodes in the electric-hydrogen integrated energy system; M lose,Hi,t is the cut-off hydrogen load amount of the hydrogen energy system or hydrogen filling station at node i at time t; N RC,all / N HTT,all respectively represent a set of all repair personnel / all hydrogen long tube trailers in the electric-hydrogen integrated energy system; is a state variable of the nth hydrogen long tube trailer on the road at time t; is the power generation power of the hydrogen fuel cell at node i at time t; c re is the unit penalty coefficient of the remaining repair time of the key line; T re,Rα is the remaining repair time of the key line at time t in the α region;

[0092] The constraint conditions of the post-disaster allocation model of mobile emergency resources considering cross-region resource sharing include a maintenance personnel allocation model based on critical line repair duration, a mobile power allocation model based on virtual node aggregation, an inter-regional hydrogen allocation model, and time and space scheduling constraints (53)-(56) of mobile emergency resources.

[0093] The maintenance personnel allocation model based on critical line repair duration is as follows:

[0094]

[0095] In the formula, denotes a set composed of critical fault lines in the α region; T re,initα is the critical line repair duration of the α region at the initial time of this inter-regional allocation; t1 is the initial time of this inter-regional allocation; u RC,Aα,τ,n is the allocation state of the nth group of maintenance personnel in the α region at time τ, and is 1, indicating that the nth group of maintenance personnel is allocated to the α region at time τ; n RCn is the allocatable state of the nth group of maintenance personnel at this inter-regional allocation, and its value is determined by the regional recovery plan.

[0096] The mobile power allocation model based on virtual node aggregation is as follows:

[0097]

[0098] In the formula, P G,allm,t is the total injection power of mobile power, hydrogen energy system, and upper grid in the microgrid m at time t, and if the microgrid does not contain the above devices, the corresponding power is 0; N MMm / N Hm / N UGm respectively represent a set composed of grid access points / hydrogen energy systems / upper grid connection points in the microgrid m; P Gridi,t is the injection power of the upper grid connected to node i at time t; is the power consumption of the electrolytic tank located at node i at time t; P load,Em,l,t is the load demand of the lth level load in the microgrid m at time t; N Mα / N Hα respectively represent a set composed of grid access points / hydrogen energy systems or hydrogen filling station nodes in the α region; u EV,Aα,t,n / u HEV,Aα,t,n respectively represent the allocation state of the nth mobile electric energy storage / hydrogen fuel electric vehicle, and is 1, indicating that the nth mobile electric energy storage / hydrogen fuel electric vehicle is allocated to the α region at time t; n EVn / n HEVn respectively represent the allocatable state of the nth mobile electric energy storage / hydrogen fuel electric vehicle at this inter-regional allocation.

[0099] The inter-regional hydrogen distribution model is shown as follows:

[0100]

[0101] In the formula, u HTT,ini,t,n / u HTT,outi,t,n are the hydrogen charging / discharging state variables of the nth hydrogen long tube trailer at the i node t; u HTTi,t,n is the position state variable of the nth hydrogen long tube trailer at the i node t; M HTT,in,max / M HTT,out,min are the maximum hydrogen charging / discharging rates of the hydrogen long tube trailer; is the energy storage level of the nth long tube trailer at t; S HTT,max / S HTT ,min are the upper and lower limits of the energy storage of the hydrogen long tube trailer; M HTT,ini,t,n / M HTT,outi,t,n are the hydrogen charging / discharging mass of hydrogen energy of the nth hydrogen long tube trailer at the i node t, i.e., the inter-regional distribution plan of hydrogen energy.

[0102] Further, the steps of solving the post-disaster recovery model of the electric-hydrogen integrated energy system and the post-disaster distribution model of mobile emergency resources include:

[0103] 2.3.1) Linearizing the post-disaster recovery model of the electric-hydrogen integrated energy system and the post-disaster distribution model of mobile emergency resources, including the steps of:

[0104] For the absolute value term in formula (54), a 0-1 auxiliary variable h absi,t,n , linearizes the original constraint to obtain:

[0105]

[0106] By Big-M method, the conditional constraint (76) is converted to obtain:

[0107]

[0108] For the maximum function in formula (88), define T re,R,maxα as the equivalent output after linearization, and introduce a 0-1 auxiliary variable linearizes the original constraint by Big-M method to obtain:

[0109]

[0110] 2.3.2) After the disaster occurs, obtain the passing time matrix of each node in the region and between regions, and count the state information of mobile emergency resources, damaged line information, and electric and hydrogen load demand information;

[0111] 2.3.3) Each region calculates the linearized post-disaster recovery model of the integrated electricity-hydrogen energy system considering the coordinated scheduling of emergency resources, and reports the recovery plan and mobile emergency resource state information to the joint disaster center according to the calculation results;

[0112] 2.3.4) The joint disaster center calculates the linearized post-disaster allocation model of mobile emergency resources considering cross-regional resource sharing according to the recovery plan and mobile emergency resource state information reported by each region, and issues a mobile emergency resource allocation plan to each region;

[0113] 2.3.5) Each region calculates the linearized post-disaster recovery model of the integrated electricity-hydrogen energy system considering the coordinated scheduling of emergency resources according to the mobile emergency resource allocation results, executes the recovery plan for the previous t c period, and uploads the recovery plan and mobile emergency resource state information after the t c period;

[0114] 2.3.6) If all regions have been completely recovered, stop calculating. Otherwise, T = T + t c , go to step 2.3.4).

[0115] Further, the steps of evaluating the resilience of the integrated electricity-hydrogen energy system include:

[0116] 3.1) Determine the multi-dimensional resilience indicators of the integrated electricity-hydrogen energy system, including the integrated electricity-hydrogen load loss expectation EH-ILCLE, the integrated electricity-hydrogen minimum energy supply proportion expectation EH-IMSPE, and the integrated electricity-hydrogen recovery time expectation EH-IPRTE;

[0117] wherein the integrated electricity-hydrogen load loss expectation EH-ILCLE is as follows:

[0118]

[0119] wherein N S is the total number of simulation scenarios; is a set of nodes of the power distribution network in region a; is the time when the performance of region a is restored to the pre-disaster level without relying on the hydrogen energy system and mobile power supply in scenario s; is the start time of post-disaster recovery of region a in scenario s; is the unit load shedding penalty of node i in region a; is the load shedding amount of node i in region a at time t in scenario s; is the unit hydrogen load shedding penalty of region a; is the hydrogen load shedding amount of region a at time t in scenario s.

[0120] The minimum supply ratio expectation of the integrated electricity and hydrogen EH-IMSPE is shown as follows:

[0121]

[0122] In the formula, is the electricity load demand of node t of region i at time in scenario s; is the hydrogen load demand of region t at time in scenario s. A is the number of regions in scenario s;

[0123] The recovery time expectation of the integrated electricity and hydrogen EH-IPRTE is shown as follows:

[0124]

[0125] 3.2) Input the operation parameters of the integrated electricity and hydrogen energy system and the equipment parameters of various emergency resources, and set the maximum number of simulation scenarios N S , and let s = 1.

[0126] 3.3) Based on the disaster prediction information and the element failure probability model under the corresponding disaster, the failure probability of each line is calculated.

[0127] 3.4) The line failure scenario set of the integrated electricity and hydrogen energy system is constructed by the Latin hypercube sampling technology; the failure scenario set is represented by a random variable u ij , which represents the operation state of line ij, and if it is 1, it represents that the line is normally operated, otherwise, it represents that the line is affected by an extreme disaster and fails;

[0128] The random variable u ij is shown as follows:

[0129]

[0130] In the formula, is the inverse function of the cumulative distribution function of the failure probability of line ij; is the failure probability of line ij; y ij is a random value;

[0131] 3.5) According to the line failure probability, load characteristics, hydrogen energy system and mobile power state information of each region, the SAC algorithm is used to solve the pre-disaster cross-region deployment model of the mobile power of the integrated electricity and hydrogen energy system, and the pre-disaster cross-region deployment decision of the mobile power is obtained.

[0132] 3.6) The post-disaster recovery model of the integrated electricity and hydrogen energy system under cross-region resource sharing in the s-th scenario is solved, and the steps include:

[0133] 3.6.1) Import the line damage situation under the s th scene, obtain the node travel time matrix in the region and between regions, and import the operation parameters of the electric hydrogen comprehensive energy system and the pre-disaster deployment position of the mobile power supply.

[0134] 3.6.2) Each region calculates the post-disaster recovery model of the electric hydrogen comprehensive energy system considering the coordinated scheduling of emergency resources, and reports the recovery plan and the state information of the mobile emergency resources to the joint disaster center according to the calculation results.

[0135] 3.6.3) The joint disaster center calculates the post-disaster allocation model of the mobile emergency resources considering the cross-region resource sharing according to the recovery plan and the state information of the mobile emergency resources reported by each region, and issues the mobile emergency resource allocation plan to each region.

[0136] 3.6.4) According to the mobile emergency resource allocation result, each region calculates the post-disaster recovery model of the electric hydrogen comprehensive energy system considering the coordinated scheduling of emergency resources, executes the recovery plan in the first t period of time, and uploads the recovery plan and the state information of the mobile emergency resources after the t period of time. c c

[0137] 3.6.5) If all regions of the electric hydrogen comprehensive energy system have completed the recovery, go to step 3.7), otherwise, T=T+t, and go to step 3.6.3). c

[0138] 3.7) Store the post-disaster cross-region recovery operation state of the electric hydrogen comprehensive energy system under the s th scene.

[0139] 3.8) If the maximum number of simulation scenes is reached, go to step 3.6), otherwise, s=s+1, and go to step 3.9).

[0140] 3.9) Calculate the multi-dimensional resilience index of the electric hydrogen comprehensive energy system.

[0141] The technical effect of the present application is self-evident, and the present application proposes a two-stage resilience evaluation method of the electric hydrogen comprehensive energy system under cross-region resource sharing; the method fully considers the cross-region mutual aid potential of the mobile emergency resources in the pre-disaster active deployment and post-disaster coordinated recovery stage, and can accurately evaluate the resilience level of the electric hydrogen comprehensive energy system.

[0142] The present application considers the cross-region support capacity and multi-stage response capacity of the mobile power supply, and establishes a two-stage resilience evaluation method of the electric hydrogen comprehensive energy system under cross-region resource sharing. The results show that the present application forms a two-stage disaster response architecture of pre-disaster active deployment and post-disaster coordinated recovery in the time dimension, and fully considers the cross-region support potential of hydrogen energy and other mobile emergency resources in the space dimension, breaks through the single-region and post-disaster oriented resilience evaluation paradigm, and effectively solves the problem of low estimation of the resilience level of the electric hydrogen comprehensive energy system.​​​ BRIEF DESCRIPTION OF DRAWINGS

[0143] Figure 1 Two-stage resilience evaluation framework of electric-hydrogen integrated energy system under cross-regional resource sharing

[0144] Figure 2 Pre-disaster deployment strategy for multi-regional electric-hydrogen integrated energy system

[0145] Figure 3 Two-stage resilience evaluation process of electric-hydrogen integrated energy system under cross-regional resource sharing

[0146] Figure 4 Distribution of important load nodes, grid access points, hydrogen energy systems or hydrogen refueling station nodes in each region

[0147] Figure 5 Electric load curve

[0148] Figure 6 Hydrogen load curve

[0149] Figure 7 Load weighted recovery rate curve of EH-IES under fault scenario 38 in different cases

[0150] Figure 8 Resilience evaluation index of EH-IES under different configuration schemes

[0151] Figure 9 Resilience evaluation index of EH-IES under different configuration schemes DETAILED DESCRIPTION

[0152] The present application will be further described below in conjunction with the embodiments, but should not be understood as limiting the above-mentioned subject matter of the present application only to the following embodiments. According to ordinary technical knowledge and conventional means in the art, various substitutions and modifications can be made without departing from the above-mentioned technical idea of the present application, and all should be included in the protection scope of the present application.

[0153] Embodiment 1

[0154] A two-stage resilience evaluation method of electric-hydrogen integrated energy system under cross-regional resource sharing, specifically comprising the following steps:

[0155] 1. Based on the disaster response architecture of "proactive pre-disaster deployment-collaborative recovery after disaster", the cross-regional sharing mechanism of embedded mobile emergency resources such as hydrogen energy is fused, and a two-stage resilience evaluation framework of electric-hydrogen integrated energy system under cross-regional resource sharing is established, as shown in Figure 1

[0156] ​Pre-disaster deployment stage: Before the disaster, the joint disaster center calculates the line failure probability based on the disaster prediction information and the element failure probability model under the corresponding disaster. Then, based on the line failure probability of each region and the distribution of flexible resources such as hydrogen energy systems, the potential power demand of each region after the disaster is analyzed, the pre-disaster cross-regional support potential of mobile power is fully developed, and the pre-disaster cross-regional deployment strategy of mobile power of the electric-hydrogen integrated energy system is developed. Based on the above strategy, mobile power will complete cross-regional optimization deployment before the disaster, so that it can quickly provide emergency power support for the power grid after the failure, reducing the impact of extreme disasters on the power grid.

[0157] Post-disaster recovery stage: After the disaster, the regional distribution network operators and the joint disaster center develop a post-disaster cross-regional coordinated recovery plan for the electric-hydrogen integrated energy system based on the idea of "intra-regional autonomy, overall allocation, and inter-regional sharing", and quickly restore system performance according to the plan. Within the region, regional distribution network operators develop post-disaster recovery plans based on pre-disaster deployment results of mobile power, fully considering the coordination potential of various emergency resources, to minimize load loss within the region. In addition, regional distribution network operators also need to report regional recovery plans and mobile emergency resource status information to the joint disaster center for unified deployment by the joint disaster center. Between regions, the joint disaster center coordinates the recovery plans and mobile emergency resource status information reported by each region, fully considers the differences in damage to each region, and leverages the cross-regional support potential of mobile power, hydrogen energy, and maintenance personnel to reasonably allocate mobile emergency resources to prioritize important load supply in each region. At the same time, the joint disaster center will continuously update the recovery status of each region and re-allocate mobile emergency resources based on this to improve the utilization rate of mobile emergency resources.

[0158] Finally, based on the simulation results of the system's operating state in the post-disaster recovery stage, the multi-dimensional resilience evaluation indicators of the electric-hydrogen integrated energy system, including resistance, recovery, and comprehensive resilience level, are calculated to comprehensively evaluate the resilience level of the electric-hydrogen integrated energy system, providing theoretical support and optimization path for the resilience improvement of the electric-hydrogen integrated energy system.

[0159] 2. Fully considering the uncertainty of multi-region fault line distribution in the electric-hydrogen integrated energy system and the cross-regional mutual aid potential of mobile power, a pre-disaster cross-regional deployment method for mobile power of the electric-hydrogen integrated energy system based on the SAC (Soft Actor-Critic) algorithm is proposed:

[0160] 1) Post-disaster emergency response model of electric-hydrogen integrated energy system

[0161] 1.1) Objective function

[0162] Firstly, the post-disaster emergency response model of distributed power and network reconfiguration coordinated scheduling is considered, and the minimum total operation cost of regional electric-hydrogen integrated energy system is taken as the objective, including the penalty of cutting electricity load, the penalty of cutting hydrogen load, and the operation cost of distributed power as follows:

[0163] min(C lose,E +C lose,H +C gen,E +C gen,H ) (1)

[0164] In the formula, C lose,E is the total penalty of cutting electricity load; C lose,H is the total penalty of cutting hydrogen load; C gen,E is the total operation cost of mobile electric energy storage; C gen,H is the total operation cost of hydrogen fuel generator and hydrogen fuel cell in hydrogen energy system, and the specific composition is as follows:

[0165]

[0166] In the formula, N E , N T1 represent the set of distribution network nodes and the set of scheduling periods in the region respectively; c lose,Ei is the unit penalty of cutting electricity load of i node; P lose,Ei,t is the cutting electricity load of i node at t moment; c lose,H is the unit penalty of cutting hydrogen load; M lose,Ht is the cutting hydrogen load at t moment; N EV , N HEV , N RC represent the set of mobile electric energy storage, the set of hydrogen fuel generator and the set of repair personnel respectively; N M represents the set of power grid access points; c gen,E is the unit operation cost of mobile electric energy storage, P EV,outi,t,n is the discharging power of the nth mobile electric energy storage at i node at t moment; c gen,H is the unit operation cost of hydrogen fuel cell, P HEV,outi,t,n is the discharging power of the nth hydrogen fuel generator at i node at t moment; P FCt is the discharging power of hydrogen fuel cell in hydrogen energy system at t moment.

[0167] 1.1) Mobile power operation constraint

[0168] The mobile power supply can provide power support to the power grid through the grid access node, thereby improving the system resilience. The mobile power supply operation constraints include mobile electric energy storage operation constraints and hydrogen fuel power generation vehicle operation constraints. Equation (6) is the mobile electric energy storage discharge power constraint; equation (7) is the mobile electric energy storage energy balance constraint; equation (8) is the upper / lower limit constraint of the mobile electric energy storage; equation (9) is the hydrogen fuel power generation vehicle discharge power constraint; equation (10) is the hydrogen fuel power generation vehicle energy balance constraint; equation (11) is the upper / lower limit constraint of the hydrogen storage of the hydrogen fuel power generation vehicle; and equation (12) is the maximum number of grid access points of the mobile power supply.

[0169]

[0170] In the formula, u EV,ini,t,n / u EV,outi,t,n are state variables of the nth mobile electric energy storage at time t at node i, and if 1, it represents that the nth mobile electric energy storage is in the charging / discharging state at time t at node i; P EV,ini,t,n / P EV,outi,t,n are the charging / discharging power of the nth mobile electric energy storage at time t at node i; P EV,in,max / P EV,out,max are the upper limits of the charging / discharging power of the mobile power supply; S EVt,n is the energy storage level of the nth mobile electric energy storage at time t; η EV,in / η EV,out are the charging / discharging efficiencies of the mobile electric energy storage; S EV,max / S EV,min is the upper / lower limit of the energy storage of the mobile electric energy storage; u HEV,ini,t,n / u HEV,outi,t,n are state variables of the nth hydrogen fuel power generation vehicle at time t at node i, and if 1, it represents that the nth hydrogen fuel power generation vehicle is in the charging / discharging state at time t at node i; M HEV,ini,t,n / P HEV,outi,t,n are the charging / discharging power of the nth hydrogen fuel power generation vehicle at time t at node i; M HEV,in,max / P HEV,out,max are the upper limits of the charging / discharging power of the hydrogen fuel power generation vehicle; S HEVt,n is the energy storage level of the nth hydrogen fuel power generation vehicle at time t; η HEV is the power generation efficiency of the hydrogen fuel power generation vehicle; S HEV,max / S HEV,min is the upper / lower limit of the energy storage of the hydrogen fuel power generation vehicle.

[0171] 1.2) Hydrogen energy system operation constraints

[0172] Hydrogen energy systems can leverage the flexibility of hydrogen energy over time through the coordinated operation of hydrogen fuel cells, electrolyzers, and hydrogen storage tanks. The operating model of a hydrogen energy system can be represented by equations (13)-(15). Equation (13) represents the operating power constraint of the hydrogen fuel cell; equation (14) represents the upper / lower limit constraint of hydrogen storage in the hydrogen storage tank; and equation (15) represents the energy balance constraint of the hydrogen storage tank.

[0173]

[0174] In the formula, P FCt P represents the operating power of the hydrogen fuel cell at time t; FC,max η is the maximum operating power of the hydrogen fuel cell. FC Energy conversion efficiency (LHV) of hydrogen fuel cells. H It has the low calorific value of hydrogen. M represents the amount of hydrogen stored in the hydrogen storage tank at time t; load,Ht S represents the hydrogen load at time t; HS,min / S HS,max Minimum / maximum hydrogen storage mass of each hydrogen storage tank; η HS,in / η HS,out These refer to the hydrogen filling / discharging efficiency of the hydrogen storage tank.

[0175] 1.3) Network Reconfiguration Constraints

[0176] Following a fault, the distribution network can be reconfigured into multiple microgrids by adjusting the tie switches and the operating status of distributed generation sources. Let N be the potential root node set formed by the nodes connected to the substation and distributed generation sources. G The set consisting of all the lines in the distribution network is N. BR The microgrid established during the network reconfiguration process must satisfy a radial topology, and the necessary and sufficient conditions are: ① the number of closed circuits is equal to the number of network nodes minus the number of subgraphs; ② each subgraph is internally connected, as shown in equations (16)-(19).

[0177]

[0178] In the formula, δ Fij,t δ represents the closed-loop state variable of line ij in time period t during virtual power flow. Fi,t Let P be the state variable indicating whether a potential root node i will become the root node of the microgrid during time period t; M is a sufficiently large constant; P Fij,t Let be the virtual power flow passing through branch ij at time t.

[0179] 1.4) Load shedding balance constraints

[0180] Equations (20)-(23) represent the relationship between the actual supply load, the abandoned load and the demand load that need to be satisfied.

[0181]

[0182] where P load,Ei,t is the demand power load of i-node at time t; P re,Ei,t is the actual supply power load of i-node at time t; M load,Ht is the demand hydrogen load at time t; M re,Ht is the actual supply hydrogen load at time t.

[0183] 1.5) Power distribution network flow constraints

[0184] In this paper, the improved LinDistFlow linear flow model is used to simulate the power flow distribution of the distribution network in post-disaster recovery, as shown in equations (24)-(30):

[0185]

[0186] where P ij,t and Q ij,t are the active power and reactive power of line ij at time t, respectively, Q maxij are the upper limits of active power and reactive power of line ij, respectively, r ij and x ij are the resistance and reactance of line ij, respectively, is the total active and reactive power injection of distributed power sources at j-node at time t; is the square of the voltage of i-node at time t.

[0187] 2) Modeling of pre-disaster cross-regional deployment decision process of integrated electricity-hydrogen energy system

[0188] 2.1) State space

[0189] The agent can obtain global observation variables, including global topology fault probability information, global agent power and capacity information, i.e.:

[0190] o = [U branch,all , P load,all , P H,all , Q H,all , P EV,all , Q EV,all , P HEV,all , Q HEV,all ] ( 31)

[0191] where o represents the state space of the agent, which includes the topology state information U branch,all of all regions, the load prediction information P load,all , the maximum power generation information P H,all of the hydrogen energy system, the hydrogen energy storage capacity information Q H,all , and the maximum power generation information PEV,all , mobile electric energy storage capacity information Q EV,all , hydrogen fuel power generation vehicle maximum power generation information P HEV,all , hydrogen fuel power generation vehicle capacity information Q HEV,all .

[0192] 2.2) Action space

[0193] The agent improves the resilience level of the electric-hydrogen integrated energy system by deploying mobile power sources across regions based on global observation variables.

[0194]

[0195] In the formula, a is the action space of the agent, which includes all deployment results of mobile electric energy storage a EVj and all deployment results of emergency hydrogen fuel power generation vehicles a HEVj , discrete action a u,EVj ∈{1, 2, …, N M}, and selection 1 means deploying the jth mobile electric energy storage to the first power grid access point, and the emergency hydrogen fuel power generation vehicle is the same.

[0196]

[0197] In addition, to represent the maximum number of mobile power sources accessed by the power grid access point as described in formula (2.12). For any power grid access point i, if the number of mobile power sources deployed before the disaster exceeds N G,maxi , only the first N G,maxi mobile power sources are accessed by the power grid. For mobile power sources exceeding this number, they are not deployed to the power grid access point, thereby limiting the maximum number of mobile power sources accessed by the power grid access point, as shown in formula (2.35).

[0198]

[0199] 2.3) Environment

[0200] The environment mainly consists of the electric-hydrogen integrated energy system emergency response model considering network reconstruction and distributed power coordinated scheduling, i.e., after the agent gives the mobile power source deployment decision, the deployed mobile power sources and fixed power sources in the network are used to recover the weighted load as much as possible with network reconstruction technology, thereby evaluating the resilience level of the electric-hydrogen integrated energy system after the agent's decision.

[0201] The electric-hydrogen integrated energy system emergency response model aims to minimize the weighted load reduction, considering mobile power source operation constraints, hydrogen energy system operation constraints, network reconstruction constraints, load balance constraints, and power flow constraints. The specific model is shown below.

[0202] Objective function: formula (1)

[0203] Constraints: Formulas (6)-(30)

[0204] 2.4) Reward function

[0205] After obtaining the deployment structure of all mobile electric energy storage and emergency hydrogen fuel power generation vehicles, the load shedding under this deployment result can be obtained by solving the electric-hydrogen integrated energy system disaster emergency response model described in the environment. The problem studied aims to maximize the multi-region weighted load recovery, so the reward function is set as follows.

[0206]

[0207] In the formula, the agent reward function r up is the sum of load shedding for all regions, r downi is the reward function of the i-th lower agent, and C lose,Ei is the load shedding of region i, C lose,Hi is the total penalty of the cut-off electrical load of region i; C n,j is the total penalty of the cut-off hydrogen load of region i, which can be calculated based on the electric-hydrogen integrated energy system disaster emergency response model proposed in Section 2.3.2.

[0208] 3) Multi-region electric-hydrogen integrated energy system pre-disaster deployment strategy based on discrete SAC algorithm

[0209] SAC is a reinforcement learning algorithm based on policy gradient, which belongs to off-policy algorithm, and introduces entropy regularization, aiming to improve the stability and efficiency of training by maximizing the expected return and exploration strategy. The policy network is used to generate the pre-disaster deployment location of the multi-region electric-hydrogen integrated energy system mobile power supply, and the value network is used to evaluate the pros and cons of the mobile power supply deployment plan, as shown in Figure 2 .

[0210] SAC is mainly used for continuous action control, while the pre-disaster deployment actions of the multi-region electric-hydrogen integrated energy system are discrete actions. In order to model this action feature, first generate a softmax distribution from the action network to output the corresponding probabilities of all possible discrete actions, and then sample the distribution to get the discrete action. Generally speaking, the formula of the softmax distribution is as follows.

[0211]

[0212] In the formula, x n,j is the preference value of the j-th action dimension in the n-th action output by the Actor network, reflecting the relative selection tendency of the n-th action selecting the j-th action dimension, but not directly representing the probability of the action. It can be expressed as the selection probability p n,jIt is worth noting that the deployment location of multiple mobile electric energy storage and emergency fuel generator vehicles needs to be output simultaneously in this paper, so multiple discrete actions need to be output simultaneously, and a softmax distribution needs to be constructed for each output discrete action.

[0213] The pre-disaster deployment problem of the multi-region electric-hydrogen integrated energy system studied in this paper has a large action space, and traditional reinforcement learning algorithms often have difficulty effectively exploring the entire action space. Therefore, this paper uses the SAC algorithm as the core optimization algorithm, which introduces the idea of policy optimization with maximum entropy, can maximize the reward while promoting more randomness and exploration, thereby avoiding falling into a local optimal solution. The loss function calculation formula of the policy network is as formula (38).

[0214]

[0215] In the formula, D is the experience pool, and a H is the temperature coefficient, which is used to control the weight of the entropy regularization term in the loss function; is the entropy regularization term, which reflects the randomness of the current policy. A higher entropy value means that the policy distribution of the agent is more dispersed, and it may assign a larger selection probability to multiple actions, reflecting strong exploration behavior. A lower entropy value means that the policy distribution of the agent is more concentrated, and the agent is more inclined to choose certain specific actions. The entropy regularization term calculation formula of discrete action is shown in formula (39):

[0216]

[0217] In the formula, A is the total number of actions of the agent, and K n are the total action latitudes of the nth action, respectively. In addition, to reduce the learning efficiency reduction problem caused by overestimation of Q values, this paper introduces two Q value functions with the same structure, and takes the smaller value to participate in the loss function calculation of the policy network.

[0218]

[0219] In the conventional SAC algorithm, Q value update depends on the immediate reward at the current time and the reward and state value at the next time step, but the pre-disaster deployment of mobile power supply is a one-time single-step decision, only the immediate reward obtained by the current decision needs to be considered. Therefore, the Q value network loss function calculation formula is simplified as shown in formula (41):

[0220]

[0221] To ensure that the agent fully explores the environment and selects the best strategy, the SAC algorithm adjusts the temperature coefficient, and the loss function calculation of the temperature coefficient is as shown in formula (42).

[0222]

[0223] wherein H is H tar The target entropy of the agent is set to keep the entropy value near the target entropy, so that the SAC algorithm can automatically adjust the temperature coefficient to maintain a certain level of exploration while effectively utilizing learned strategies to achieve higher returns.

[0224] Based on the above loss function calculation formula, one action network, two evaluation networks and temperature coefficient can be updated as follows:

[0225]

[0226] wherein, α represents the learning rate of the agent.

[0227] 4) Training process

[0228] In the training process of the SAC algorithm, the agent continuously interacts with the environment, collects state, action and reward information, and stores these experiences in the experience pool. Then, a small batch of data is randomly sampled from the experience pool for updating the evaluation network and the action network. The evaluation network evaluates the value of the action by minimizing the error, while the action network optimizes the decision by maximizing the reward and entropy. Through continuous repetition of this process, the agent can gradually learn the optimal strategy in a strong uncertain environment. The training framework of the proposed method is shown in Algorithm 3-1.

[0229]

[0230] 5) Test process

[0231] In the test process, we first collect the action network parameters of the agent trained by Algorithm 3-1, and the evaluation network is no longer needed in the test process. For each test scenario, the action network is used to obtain the mobile power deployment decision, which is then substituted into the environment for verification. In general, the test process of the SAC algorithm proposed in this paper is shown in Algorithm 3-2.

[0232]

[0233] 3. Considering the cross-regional support potential and post-disaster coordination potential of mobile emergency resources, a post-disaster recovery strategy for integrated electric-hydrogen energy systems under cross-regional resource sharing is proposed.

[0234] 1) Post-disaster recovery model of integrated electric-hydrogen energy system considering emergency resource collaborative scheduling

[0235] 1.1) Objective function

[0236] A post-disaster recovery model of an integrated electricity-hydrogen energy system considering the coordination of emergency resource scheduling is proposed, which aims to minimize the total cost of each region in the integrated electricity-hydrogen energy system, including the penalty of cutting electricity load, the penalty of cutting hydrogen load, the scheduling cost of mobile emergency resources, the operation cost of hydrogen energy system and mobile power supply, and the subsidy of mobile emergency resources participating in inter-regional distribution, as follows:

[0237] min C1=C lose,E +C lose,H +C tran +C gen,E + C gen,H -E disp (46)

[0238] In the formula, C tran is the total scheduling cost of mobile emergency resources; E disp is the subsidy of mobile emergency resources participating in inter-regional distribution, which guides the mobile emergency resources without recovery tasks to participate in cross-regional distribution. The specific composition is as follows:

[0239]

[0240]

[0241] In the formula, N T2 represents the set of post-disaster recovery scheduling periods in the region; c tran is the unit scheduling cost of mobile emergency resources; respectively represent the state variables of the nth group of mobile electric energy storage / hydrogen fuel power generation vehicles / maintenance personnel located on the road at time t; N RC represents the set of maintenance personnel; c disp is the unit subsidy of mobile emergency resources participating in inter-regional distribution; n EVt,n / n HEVt,n / n RCt,n respectively represent the assignable state of the nth group of mobile electric energy storage / hydrogen fuel power generation vehicles / maintenance personnel at time t. If it is 1, it represents that it can participate in inter-regional distribution at time t, otherwise, it cannot.

[0242] 1.2) Constraint conditions

[0243] 1.2.1) Time and space scheduling constraints of mobile emergency resources

[0244] The mobile emergency resources include mobile energy storage, hydrogen fuel power generation vehicles, maintenance personnel and hydrogen long pipe trailers, which can travel between the grid access nodes, hydrogen energy systems and fault lines in the electric-hydrogen integrated energy system to realize the cross-regional resource sharing of the electric-hydrogen integrated energy system. The traditional time-space network modeling method needs to establish 0-1 decision variables for the transfer arcs and parking arcs between all traffic nodes, which leads to the square growth of the variable scale with the number of traffic nodes, and it is difficult to adapt to large-scale traffic networks under multi-regional coupling. Therefore, based on the concept of equivalent time network, the spatial distance between different traffic nodes is equivalent to the time distance between traffic nodes in this section, and the space-time scheduling constraints of mobile emergency resources are established as shown in equations (53)-(56), which effectively reduces the complexity of the traffic model. Equation (53) is the space-time position uniqueness constraint of mobile emergency resources; equation (54) represents the position change state of mobile emergency resources; equation (55) limits the position transfer between all nodes of mobile emergency resources to pass through the road; and equation (56) is the space-time position transfer constraint of mobile emergency resources.

[0245]

[0246] In the formula, u i,t,n is the state variable of the space-time position of mobile emergency resources, and if it is 1, it represents that the nth group of mobile emergency resources is located at node i at time t, otherwise, it is not; u 0,t,n is the state variable of the position of mobile emergency resources on the road at time t; N R represents a set of traffic network nodes coupled with grid access nodes, fault lines, hydrogen energy systems or hydrogen refueling stations; N MER represents a set of all mobile emergency resources; h i,t,n is the position change state variable of the nth group of mobile emergency resources at node i at time t, h i,t,n is 1 / -1, which represents that the nth group of mobile emergency resources enters / leaves node i at time t, and if it is 0, it represents that the nth group of mobile emergency resources still stays in node i at time t; T ij is the time required to move from node i to node j.

[0247] 1.2.2) Mobile power supply post-disaster recovery operation constraints

[0248] In the post-disaster recovery strategy of the integrated energy system of electric hydrogen proposed in this chapter, the mobile power supply is allowed to make dynamic decisions on its routing behavior between various grid access points or charging facilities, while fully considering the differences in the operating characteristics of mobile electric energy storage and hydrogen fuel power generation vehicles. The mobile power supply post-disaster recovery operation constraints are established, as shown in equations (57)-(67). Equation (57) is the uniqueness constraint of the state of the mobile electric energy storage charge / discharge; equations (58)-(59) are the mobile electric energy storage charge / discharge power constraints, which can charge / discharge at any grid access point; equation (60) is the energy balance constraint of the mobile electric energy storage; equation (61) is the upper / lower limit constraint of the energy storage of the mobile electric energy storage; equation (62) is the uniqueness constraint of the state of the hydrogen fuel power generation vehicle hydrogen charging / discharge; equations (63)-(64) are the hydrogen charging rate / discharge power constraints of the hydrogen fuel power generation vehicle, which can only charge hydrogen at a hydrogen refueling station or a hydrogen energy system; equation (65) is the energy balance constraint of the hydrogen fuel power generation vehicle; equation (66) is the upper / lower limit constraint of the hydrogen storage of the hydrogen fuel power generation vehicle; and equation (67) limits the maximum number of mobile power supplies connected to the grid access point.

[0249]

[0250]

[0251] wherein u EV,ini,t,n / u EV,outi,t,n are the state variables of the nth mobile electric energy storage charging / discharging at node i at time t, and if 1 represents that the nth mobile electric energy storage is in the charging / discharging state at node i at time t; P EV,ini,t,n is the charging power of the nth mobile electric energy storage at node i at time t; P EV,in,max is the upper limit of the charging power of the mobile electric energy storage; η EV,in is the charging efficiency of the mobile electric energy storage; u HEV,ini,t,n / u HEV,outi,t,n are the state variables of the nth hydrogen fuel power generation vehicle hydrogen charging / discharge at node i at time t, and if 1 represents that the nth hydrogen fuel power generation vehicle is in the hydrogen charging / discharge state at node i at time t; N H is a set consisting of hydrogen energy system or hydrogen refueling station nodes; M HEV,ini,t,n is the hydrogen charging mass of the nth hydrogen fuel power generation vehicle at node i at time t; M HEV,in,max is the upper limit of the hydrogen charging rate of the hydrogen fuel power generation vehicle.

[0252] 1.2.3) Hydrogen energy system post-disaster recovery operation constraints

[0253] In the post-disaster topology recovery stage, the hydrogen energy system can play the flexibility of hydrogen energy in time dimension through the coordination of hydrogen fuel cells, electrolyzers and hydrogen storage tanks, and can also play the flexibility of hydrogen energy in space dimension through the coordination of hydrogen long tube trailers. To depict the cross-time and space flexibility of the hydrogen energy system in the post-disaster recovery stage, the post-disaster recovery operation constraints of the hydrogen energy system are established, as shown in formulas (68)-(72). Formula (68) is a uniqueness constraint of the operation state of the electrolyzer and the hydrogen fuel cell; formula (69) is an operation power constraint of the electrolyzer; formula (70) is an operation power constraint of the hydrogen fuel cell; formula (71) is a hydrogen storage upper / lower limit constraint of the hydrogen storage tank; and formula (72) is an energy balance constraint of the hydrogen storage tank.

[0254]

[0255] In the formulas, u EDt / u FCt are operation state variables of the electrolyzer / hydrogen fuel cell at time t, and if 1, it represents that the electrolyzer / hydrogen fuel cell is in working state at time t, and if only hydrogenation stations are deployed in the region, u EDt and u FCt are both 0; is the operation power of the electrolyzer at time t; P ED,max is the upper limit of the operation power of the electrolyzer; η ED is the energy conversion efficiency of the electrolyzer; η HS,in / η HS,out are hydrogen charging / discharging efficiencies of the hydrogen storage tank; N HTT is a set of hydrogen long tube trailers; M HTT,int,n / M HTT,outt,n are charging / discharging masses of the nth hydrogen long tube trailer at time t, and the values are determined by the inter-regional mobile emergency resource allocation plan, which are constants here.

[0256] 1.2.4) Line repair state constraint

[0257] Considering the dynamic coupling characteristics of repair personnel scheduling and line operation state, the line repair state constraint is established, as shown in formulas (73)-(75). Formula (73) represents that the damaged line can be restored to the available state after the repair personnel arrives and stays on the line for the required repair time; formula (74) represents that the damaged line can only be repaired by a group of repair personnel; and formula (75) represents that the actual closed state of the branch is jointly constrained by the virtual power flow model and the available state of the line.

[0258]

[0259] In the formulas, t0 is the starting time of post-disaster recovery; is the time required for repairing the line ij; N D is a set of damaged lines.

[0260] 1.2.5) Mobile emergency resource allocable state constraints

[0261] To avoid the joint disaster center assigning mobile emergency resources that are performing recovery tasks to other areas, causing the current recovery process to be interrupted, this section establishes mobile emergency resource allocable state constraints, as shown in equations (76)-(79), which prohibit mobile emergency resources in the charging, power supply, or maintenance state from participating in inter-regional allocation, thereby ensuring the stable advancement of regional recovery. Equation (76) indicates that an area can be considered fully recovered when it can supply all loads without hydrogen energy systems and mobile power supplies; equations (77)-(78) indicate that mobile power supplies can only participate in inter-regional allocation when they have no charging or power supply tasks at the current time; and equation (79) indicates that maintenance personnel can only participate in inter-regional allocation after the area is fully recovered.

[0262]

[0263] In the formula, is the regional recovery state variable, and if it is 1, it represents that the area has been fully recovered.

[0264] 1.2.6) Other constraints

[0265] In addition to the above constraints, network reconfiguration constraints, load balance constraints, and power grid flow constraints, such as equations (16)-(19), (21)-(24), and (25)-(31), also need to be met.

[0266] 2) Mobile emergency resource post-disaster allocation model considering cross-regional resource sharing

[0267] 2.1) Objective function

[0268] The mobile emergency resource post-disaster allocation model considering cross-regional resource sharing takes the minimum total load loss, total power supply cost, and total maintenance time of all areas in the electric-hydrogen integrated energy system as the goal, including total cut-off electrical load penalty, total cut-off hydrogen load penalty, total mobile emergency resource scheduling cost, total hydrogen energy system and mobile power supply operation cost, and total key line remaining maintenance time penalty, as follows:

[0269] minC2=C lose,E,all +C lose,H,all +C tran,all +C gen,E,all +C gen,H,all +C re,R,all (80)

[0270] In the formula, C lose,E,all , C lose,H,all , C tran,all , C gen,E,all , C gen,H,all , C re,R,allrespectively represent the total penalty of cutting load, the total penalty of cutting hydrogen load, the total scheduling cost of mobile emergency resources, the total operation cost of mobile electric energy storage, the total operation cost of hydrogen fuel cell in hydrogen fuel power generation vehicle and hydrogen energy system, and the total penalty of remaining repair time of critical line in the whole electric-hydrogen integrated energy system, and the specific composition is as follows:

[0271]

[0272] In the formula, N MG represents a set of all microgrids in the electric-hydrogen integrated energy system; represents a set of loads at all levels in the microgrid m; N T3 represents a set of scheduling periods for inter-regional allocation; c lose,El is the unit cutting load penalty of the lth load; P lose,Em,l,t is the cutting load amount of the lth load at time t in the microgrid m; N H,all represents a set of all hydrogen energy systems or hydrogen filling station nodes in the electric-hydrogen integrated energy system; M lose,Hi,t is the cutting hydrogen load amount of the hydrogen energy system or hydrogen filling station at node i at time t; N RC,all / N HTT,all respectively represent a set of all repair personnel / a set of all hydrogen long tube trailers in the electric-hydrogen integrated energy system; represents the state variable of the nth hydrogen long tube trailer on the road at time t; is the power generation power of the hydrogen fuel cell at node i at time t; c re is the unit penalty coefficient of the repair time of the critical line; T re,Rα is the repair time of the critical line in the a region at time t.

[0273] 2.2) Constraint conditions

[0274] 2.2.1) Repair personnel allocation model based on repair time of critical line

[0275] Based on the recovery plan of each region, the repair time of the critical line of the power grid of each region is obtained, and a model for inter-regional allocation of repair personnel is established. Formula (87) is a calculation formula for the repair time of the critical line of each region at the initial time of this inter-regional allocation, wherein the required repair time of the damaged critical line and the scheduling result of the repair personnel can be obtained from the recovery plan of the region, and if the region has been completely recovered, the repair time of the critical line is 0; formula (88) is a dynamic updating formula for the repair time of the critical line of each region, which ensures that the repair time of the critical line of each region is a non-negative value through the maximum function; formula (89) indicates that only the repair personnel participating in the inter-regional allocation can be cross-regionally scheduled.

[0276]

[0277] In the formula, denotes the set of critical lines in the α region; T re,initα denotes the critical line maintenance time length of the α region at the initial inter-regional allocation time; t1 denotes the initial inter-regional allocation time; u RC,Aα,τ,n denotes the allocation state of the nth group of maintenance personnel in the α region at time τ, and is 1, indicating that the nth group of maintenance personnel is allocated to the α region at time τ; n RCn denotes the allocatable state of the nth group of maintenance personnel at the current inter-regional allocation, and its value is determined by the intra-regional recovery plan.

[0278] 2.2.2) Mobile power supply allocation model based on virtual node aggregation

[0279] Based on the post-disaster recovery plan of each region, the microgrid distribution information in each region is obtained, and the load of each microgrid is concentrated on the virtual node, and the inter-regional allocation model of the mobile power supply is established. Equation (90) represents the injection power of all power supplies in each microgrid; equation (91) is the power balance constraint of each microgrid; equation (92) represents that the load shedding amount of each level of load in each microgrid should be less than its demand; equations (93)-(94) represent the inter-regional allocation state of the mobile power supply; equations (95)-(96) represent that only the mobile power supply participating in the inter-regional allocation can be cross-region dispatched. In addition, the post-disaster recovery operation constraints of the mobile power supply such as equations (57)-(67) need to be considered, which are consistent with the post-disaster recovery model of the electricity-hydrogen integrated energy system.

[0280]

[0281] In the formula, P G,allm,t is the total injection power of the mobile power supply, hydrogen energy system and upper grid in microgrid m at time t, and if the microgrid does not contain the above devices, the corresponding power is 0; N MMm / N Hm / N UGm respectively represent the set of grid access points / hydrogen energy systems / upper grid connection points in microgrid m; P Gridi,t is the injection power of the upper grid connected to node i at time t; is the power consumption of the electrolyzer located at node i at time t; P load,Em,l,t is the load demand of the lth level of load in microgrid m at time t; N Mα / N Hα respectively represent the set of grid access points / hydrogen energy systems or hydrogen refueling station nodes in the α region; u EV,Aα,t,n / u HEV,Aα,t,n respectively represent the allocation state of the nth mobile electric energy storage / hydrogen fuel power generation vehicle, and is 1, indicating that the nth mobile electric energy storage / hydrogen fuel power generation vehicle is allocated to the α region at time t; n EVn / n HEVnis the allocatable state of the nth mobile electric energy storage / hydrogen fuel electric vehicle when it is allocated in this region, whose value is determined by the regional recovery plan.

[0282] 2.2.3) Inter-regional hydrogen allocation model

[0283] According to the operating state of the hydrogen energy system or hydrogen refueling station, the hydrogen refueling demand of the hydrogen fuel power vehicle and hydrogen load, the regional topology recovery information, and the hydrogen long tube trailer state information, an inter-regional hydrogen allocation model is established. Equation (97) is the uniqueness constraint of the hydrogen charging / discharging state of the hydrogen long tube trailer; equations (98)-(99) are the hydrogen charging / discharging rate constraints of the hydrogen long tube trailer; equation (100) is the upper / lower limit constraint of the hydrogen storage of the hydrogen long tube trailer; and equation (101) is the energy balance constraint of the hydrogen long tube trailer. In addition, the post-disaster recovery operation constraints of the hydrogen energy system, such as equations (68)-(72), are also considered, which are consistent with the post-disaster recovery model of the integrated electric-hydrogen energy system.

[0284]

[0285] wherein u HTT,ini,t,n / u HTT,outi,t,n are the charging / discharging state variables of the nth hydrogen long tube trailer at i node t; u HTTi,t,n is the position state variable of the nth hydrogen long tube trailer at i node t; M HTT,in,max / M HTT,out,min are the maximum charging / discharging rates of the hydrogen long tube trailer; is the energy storage level of the nth long tube trailer at t; S HTT,max / S HTT ,min are the upper / lower limits of the energy storage of the hydrogen long tube trailer; M HTT,ini,t,n / M HTT,outi,t,n are the charging / discharging mass of hydrogen energy of the nth hydrogen long tube trailer at i node t, i.e., the inter-regional allocation plan of hydrogen energy.

[0286] 2.2.4) Temporal and spatial scheduling constraints of mobile emergency resources

[0287] The temporal and spatial scheduling constraints of mobile emergency resources are consistent with the post-disaster recovery model of the integrated electric-hydrogen energy system, as shown in the foregoing.

[0288] 3) Linearization solving strategy of the two-layer post-disaster recovery model of the integrated electric-hydrogen energy system

[0289] 3.1) Model transformation method based on Big-M method

[0290] The two-layer post-disaster recovery model of the integrated electric-hydrogen energy system under cross-regional resource sharing is a mixed integer nonlinear programming problem. In order to improve the solving efficiency, the Big-M method is adopted in this section, and auxiliary variables are introduced to linearize the nonlinear terms in the model.

[0291] Introduce 0-1 auxiliary variable h absi,t,n 、 Linearize the original constraint into the following form.

[0292]

[0293] The original model formula (76) is a conditional constraint, which can be converted into the following form by Big-M method.

[0294]

[0295] For the maximum value function existing in formula (88), define T re,R,maxα as the equivalent output after linearization, and introduce 0-1 auxiliary variable Linearize the original constraint into the following form by Big-M method.

[0296]

[0297] Based on the above method, the double-layer post-disaster recovery model of the cross-regional resource sharing electric and hydrogen comprehensive energy system can be converted into a mixed integer linear programming problem, which can be solved efficiently by commercial solvers in turn.

[0298] 3.2) Algorithm flow

[0299] (1) After the disaster occurs, obtain the travel time matrix of each node in the region and between regions, and count the state information of mobile emergency resources, damaged line information, and electric and hydrogen load demand information.

[0300] (2) Each region calculates the post-disaster recovery model of the electric and hydrogen comprehensive energy system considering the coordinated scheduling of emergency resources, and reports the recovery plan and mobile emergency resource state information to the joint disaster center according to the calculation results.

[0301] (3) The joint disaster center calculates the post-disaster allocation model of mobile emergency resources considering cross-regional resource sharing according to the recovery plan and mobile emergency resource state information reported by each region, and issues mobile emergency resource allocation plan to each region.

[0302] (4) Each region calculates the post-disaster recovery model of the electric and hydrogen comprehensive energy system considering the coordinated scheduling of emergency resources according to the mobile emergency resource allocation result, executes the recovery plan in the first t c period, and uploads the recovery plan and mobile emergency resource state information after t c period.

[0303] (5) If all regions have been completely recovered, stop calculating. Otherwise, T = T + t c , go to step (3).

[0304] 4. From the perspective of resistance and resilience of systems under extreme disasters, the multi-dimensional resilience evaluation index of electric hydrogen integrated energy system is established, and the resilience evaluation method of electric hydrogen integrated energy system based on Latin hypercube sampling is proposed.

[0305] 1) Multi-dimensional resilience index of electric hydrogen integrated energy system

[0306] 1.1) Electric hydrogen integrated load curtailment loss expectation (EH-ILCLE): Calculate the scenario expectation of the total load curtailment loss of all regions in the electric hydrogen integrated energy system under all simulation scenarios.

[0307]

[0308] Where, N S is the total number of simulation scenarios; is the set of distribution network nodes in region α; is the time when the performance of region α in scenario s is restored to the pre-disaster level without relying on hydrogen energy systems and mobile power supplies, is the time when the post-disaster recovery of region α in scenario s starts; c lose,Eα,i is the unit load curtailment penalty of node i in region α; P lose,Es,α,i,t is the load curtailment amount of node i in region α at time t in scenario s; c lose,Hα is the unit hydrogen load curtailment penalty of region α; M lose,Hs,α,t is the hydrogen load curtailment amount of region α at time t in scenario s.

[0309] 1.2) Electric hydrogen integrated minimum supply proportion expectation (EH-IMSPE): Calculate the scenario expectation of the average minimum integrated energy supply level of all regions in the electric hydrogen integrated energy system under all simulation scenarios.

[0310]

[0311] Where, P load,Es,α,i,t is the electricity load demand of node i in region α at time t in scenario s; M load,Hs,α,t is the hydrogen load demand of region α at time t in scenario s.

[0312] 1.3) Electric Hydrogen Integrated Recovery Time Expectation (EH-IPRTE): The scenario expectation of the average recovery time of all regions in the electric hydrogen integrated energy system under all simulation scenarios is calculated.

[0313]

[0314] 2) Electric Hydrogen Integrated Energy System Resilience Assessment Method Based on Latin Hypercube Sampling

[0315] First, the line fault scenario set of the electric hydrogen integrated energy system is constructed by the Latin hypercube sampling technology. Define the random variable u ij characterizes the operating state of the line ij. If it is 1, it represents that the line is normally operating, otherwise it means that the line is affected by extreme disasters and fails. For the sampling size N S , the value range of the cumulative distribution function of each line fault probability is divided into N S equal and non-overlapping subintervals, and a value y ij is randomly selected in each interval, and the sampling value is calculated through the inverse function of its cumulative probability distribution function, as shown in equation (4.4).

[0316]

[0317] In the formula, is the inverse function of the cumulative distribution function of the line ij fault probability; p Lij is the fault probability of the line ij, which can be calculated based on the disaster prediction information and the element failure probability model under the corresponding disaster.

[0318] Then, based on the multi-region line fault scenario set, the electric hydrogen integrated energy system executes a two-stage disaster response strategy to simulate the systematic disaster response process of the electric hydrogen integrated energy system under extreme disasters: before the disaster, the SAC algorithm is used to optimize the pre-disaster cross-regional deployment scheme of mobile power; After the disaster, based on the idea of "independent within the region, overall planning of resources, and sharing between regions", the post-disaster cross-regional collaborative recovery of the electric hydrogen integrated energy system is carried out. Finally, according to the dynamic recovery trajectory of the system, the multi-dimensional resilience indicators of the electric hydrogen integrated energy system are calculated, which provides a quantitative basis for the optimization of resource allocation and defense strategies of the electric hydrogen integrated energy system.

[0319] 3) Two-stage resilience evaluation process of electric hydrogen integrated energy system under cross-regional resource sharing

[0320] The two-stage resilience evaluation process of the electric hydrogen integrated energy system under cross-regional resource sharing is shown in Figure 3 , and the specific steps are as follows.

[0321] (1) Input the operation parameters of the electric-hydrogen integrated energy system and the equipment parameters of various emergency resources, and set the maximum number of simulation scenarios N S and let s = 1.

[0322] (2) Based on the disaster prediction information and the element failure probability model under the corresponding disaster, the line failure probability of each region is calculated.

[0323] (3) According to the line failure probability, a multi-region line failure scenario set is established by using Latin hypercube sampling.

[0324] (4) According to the line failure probability of each region, the load characteristics, the state information of the hydrogen energy system and the mobile power supply, the SAC algorithm is used to solve the pre-disaster cross-region deployment model of the mobile power supply of the electric-hydrogen integrated energy system, and the pre-disaster cross-region deployment decision of the mobile power supply is obtained.

[0325] (5) Solve the post-disaster recovery model of the electric-hydrogen integrated energy system under cross-region resource sharing in the s th scenario.

[0326] ① Import the line damage situation under the s th scenario, obtain the node travel time matrix in the region and between regions, and import the operation parameters of the electric-hydrogen integrated energy system and the pre-disaster deployment position of the mobile power supply.

[0327] ② Each region calculates the post-disaster recovery model of the electric-hydrogen integrated energy system considering the cooperative scheduling of emergency resources, and reports the recovery plan and the state information of the mobile emergency resources to the joint disaster center according to the calculation results.

[0328] ③ The joint disaster center calculates the post-disaster allocation model of the mobile emergency resources considering cross-region resource sharing according to the recovery plan and the state information of the mobile emergency resources reported by each region, and issues the mobile emergency resource allocation plan to each region.

[0329] ④ Each region calculates the post-disaster recovery model of the electric-hydrogen integrated energy system considering the cooperative scheduling of emergency resources according to the mobile emergency resource allocation results, executes the recovery plan in the first t c period, and uploads the recovery plan and the state information of the mobile emergency resources after the t c period.

[0330] ⑤ If all regions of the electric-hydrogen integrated energy system have completed the recovery, go to step (6). Otherwise, T = T + t c , go to step ③.

[0331] (6) Store the post-disaster cross-region recovery operation state of the electric-hydrogen integrated energy system under the s th scenario.

[0332] (7) If the maximum number of simulation scenarios is reached, go to step (8), otherwise s = s + 1 go to step (5).

[0333] (8) Calculate the multi-dimensional resilience index of the electric-hydrogen integrated energy system.

[0334] The electric-hydrogen integrated energy system constructed in this section is composed of 4 improved IEEE 33-node distribution networks. Region 1-2 is equipped with a hydrogen refueling station (including a hydrogen storage tank and a hydrogen refueling machine), and region 3-4 is equipped with a hydrogen energy system (including an electrolyzer, a hydrogen fuel cell, a hydrogen storage tank, and a hydrogen refueling machine), both of which are configured at node 10. In addition, each region contains 4 mobile power grid access points, respectively configured at nodes 4, 13, 26, and 32. The distribution of important load nodes, grid access points, hydrogen energy system or hydrogen refueling station nodes in each region is shown in Figure 4 , and the electric load and hydrogen load curves are shown in Figures 5-6 . The electric load is divided into 3 levels according to the importance

[99] , and the load shedding penalty coefficients of each level are 1000, 100, and 5, respectively, as shown in Table 1. Each region is equipped with 2 mobile electric energy storage vehicles, 2 hydrogen fuel cell vehicles, and 1 maintenance team, and the relevant operating parameters of the mobile power supply and hydrogen energy equipment are shown in Table 2. This section assumes that the EH-IES is affected by a typhoon disaster at 8:00, with an average wind speed of 31 m / s in region 1 and region 3, and an average wind speed of 35 m / s in region 2 and region 4. The maximum number of simulation scenarios N S is set to 1000. The SAC algorithm is based on Python 3.8, the neural network framework is built by tensorflow 2.6.0, and Gurobi is used to model and solve the optimization model. The computer CPU model is Intel Core I9 14900@3.2GHz, the GPU model is NVIDIA 4070, and the memory is 32GB.

[0335] Table 1

[0336]

[0337] Table 2

[0338]

[0339]

[0340] To verify the effectiveness of the two-stage resilience evaluation method of the EH-IES under cross-regional resource sharing proposed in this paper, the following 4 simulation schemes are set up for comparative analysis.

[0341] Case 1: In the EH-IES, each region only uses regional emergency resources for post-disaster coordinated recovery, without considering the pre-disaster deployment of mobile power supply.

[0342] Case 2: In the EH-IES, each region only uses regional emergency resources for pre-disaster deployment and post-disaster coordinated recovery of mobile power supply.

[0343] Case 3: Considering cross-region resource sharing, EH-IES utilizes emergency resources for post-disaster cross-region collaborative recovery, without considering pre-disaster deployment of mobile power supply.

[0344] Case 4: The method in this paper, considering cross-region resource sharing, EH-IES performs pre-disaster cross-region deployment of mobile power supply and post-disaster cross-region collaborative recovery.

[0345] The resilience evaluation results of EH-IES under different cases are shown in Table 4.1. As can be seen from Table 4.1, compared with Case 1 and Case 3, the EH-ILCLE of Case 2 and Case 4 decreases by 53.37% and 61.31% respectively, and the EH-IMSPE increases by 151.74% and 155.89% respectively. The reason is that the pre-disaster deployment of mobile power supply can quickly cooperate with network reconstruction technology after the line failure to form multiple active microgrids to provide emergency power support for important loads in the system. On the contrary, the scheme that does not consider pre-disaster deployment of mobile power supply needs to dispatch mobile power supply from the warehouse to the grid access point in the post-disaster recovery phase. Limited by the transportation delay of mobile power supply, the system power support capability is insufficient in the initial failure stage, leading to an increase in system load shedding loss and a decrease in the minimum energy supply ratio. The above simulation analysis confirms that the pre-disaster deployment of mobile power supply is of great significance to the resilience improvement of EH-IES.

[0346] Table 3 Resilience evaluation results of EH-IES under different cases

[0347]

[0348] In addition, compared with Case 1, the EH-ILCLE of Case 3 decreases by 12.26%, and the EH-IRTE decreases by 14.89%. The reason is that after considering cross-region resource sharing, the joint disaster center can reasonably allocate mobile emergency resources in post-disaster recovery according to the distribution of failed lines to preferentially meet the supply of important loads. And it can reasonably allocate mobile power supply without energy supply task according to the real-time repair situation of each region to provide more energy support for EH-IES. At the same time, the maintenance personnel of the repaired region are allocated to the to-be-recovered region to speed up the post-disaster recovery process. Compared with Case 2, the EH-ILCLE of Case 4 decreases by 27.21%, the EH-IMSPE increases by 1.43%, and the EH-IRTE decreases by 14.89%. This shows that after considering cross-region resource sharing in both pre-disaster deployment and post-disaster recovery stages, the joint disaster center can not only allocate mobile emergency resources based on the damage situation and recovery plan of each region after the disaster, but also allocate mobile power supply between regions and deploy it within the region before the disaster, so as to provide more emergency power support for the system at the first time after the disaster, further improving the overall resilience level of EH-IES.

[0349] In summary, the two-stage disaster response model of EH-IES under cross-regional resource sharing proposed in this paper forms a two-stage disaster response architecture in the time dimension, with proactive deployment before disaster and coordinated recovery after disaster, and fully considers the cross-regional support potential of mobile emergency resources such as hydrogen energy in the spatial dimension, breaks through the single-region and post-disaster oriented resilience evaluation paradigm, and effectively solves the problem of low estimation of EH-IES resilience level.

[0350] Figure 7 The load-weighted recovery rate curves of EH-IES under different cases in the failure scenario 38 (region 1, 3 failure: line 1-2, line 4-5, line 15-16, line 32-33, line 25-29; region 2, 4 failure: line 1-2, line 2-3, line 7-8, line 15-16, line 3-23, line 29-30, line 32-33, line 12-22) are shown.

[0351] As can be seen from Figure 7 Compared with Case 1 and Case 3, the load-weighted recovery rates of Case 2 and Case 4 at 8:00 are increased from 29.98% and 29.87% to 93.13% and 94.53%, respectively, which again verifies the significant improvement of system resilience by pre-disaster deployment of mobile power supply. It is worth noting that compared with Case 1 and Case 3, the load-weighted recovery rates of Case 2 and Case 4 at 9:00-14:00 are decreased, with an average decrease of 1.22% and 0.79%, respectively. This is because the mobile power supply has small capacity and is difficult to support long-time high-power discharge. After providing power support to the system at 8:00, the state of charge is significantly reduced. In order to continuously guarantee the supply of important loads, the mobile power supply needs to reduce the operating power or charge in advance, resulting in a small decrease in the power support capacity of the system at 9:00-14:00. However, the pre-disaster deployment of mobile power supply can still significantly improve the resilience level of the system under extreme disasters.

[0352] Compared to Cases 1 and 2, Cases 3 and 4 showed a comprehensive improvement in load-weighted recovery rates from 8:00 to 14:00, with average increases of 1.06% and 1.58%, respectively. This improvement stemmed from the Joint Disaster Relief Center's coordination and allocation of mobile emergency resources, such as mobile power supplies, to areas with larger load-weighted power deficits, prioritizing the supply of critical loads, based on the damage situation and recovery plans in each region. Furthermore, the system recovery time for both Cases 3 and 4 was reduced by 3 hours, with all critical lines repaired by 20:00, restoring EH-IES performance to pre-disaster levels. This was because, as the repair process progressed, some critical lines in certain areas were repaired. At this point, the Joint Disaster Relief Center allocated surplus maintenance personnel to areas with more critical lines awaiting repair, effectively accelerating system recovery.

[0353] To investigate the impact of initial hydrogen storage capacity on the toughness of EH-IES, this section sets up different initial hydrogen storage capacity configurations for EH-IES based on Case 4, and calculates the toughness evaluation index of EH-IES under different configurations. The results are as follows: Figure 8 As shown. From Figure 8 As can be seen, increasing the initial hydrogen storage capacity of a hydrogen energy system can effectively improve the resilience level of EH-IES. As the initial hydrogen storage capacity increases from 600 kg to 1400 kg, the EH-ILCLE gradually increases from 459.73 × 10⁻⁶. 3 The amount decreased to 289.04 × 10. 3 The EH-IMSPE ratio gradually increased from 90.39% to 93.87%. This is because, before a disaster, increasing the hydrogen storage capacity of the hydrogen energy system allows hydrogen fuel cells to provide higher power and longer-lasting electrical support to the grid after a disaster. Simultaneously, it avoids hydrogen shedding load caused by the electrolyzer losing power and being unable to produce hydrogen.

[0354] However, as the initial hydrogen storage capacity of the hydrogen energy system increases, its impact on the EH-IES toughness gradually decreases. Under Case 4, for every 200 kg increase in the initial hydrogen storage capacity of the hydrogen energy system, the EH-ILCLE decreases by 96.49 × 10⁻⁶. 3 Yuan, 49.76×10 3 Yuan, 22.75×10 3 Yuan, 1.66×10 3The EH-IMSPE increased by 1.85%, 1.03%, 0.58%, and 0.01%, respectively. This is because the resilience improvement effect of the initial hydrogen storage capacity of the hydrogen energy system is limited by the dual energy supply potential on both the hydrogen supply side and the power supply side: On the hydrogen supply side, as the initial hydrogen storage capacity continuously increases, the hydrogen load shedding loss during post-disaster recovery of the EH-IES continuously decreases. When the initial hydrogen storage capacity is 1000 kg, the system can basically achieve full supply of hydrogen load after a disaster, and its hydrogen supply potential has been fully released. On the power supply side, hydrogen cannot directly supply energy to the grid, and its ability to support grid power is limited by the power generation capacity of the hydrogen fuel cell. In addition, the location of the hydrogen energy system in the grid topology is fixed, and its energy supply potential is greatly affected by the line operating status. Under extreme disasters, the power supply range of the hydrogen energy system may be small. As the initial hydrogen storage capacity of the hydrogen energy system increases, the power supply potential of the hydrogen energy system has also been fully utilized. At this point, further increasing the initial hydrogen storage capacity of the hydrogen energy system has little effect on improving the resilience level of the EH-IES.

[0355] To investigate the impact of the number of portable power banks on the resilience of EH-IES, this section sets up different numbers of portable power bank configurations based on Case 4, and calculates the EH-IES resilience evaluation index under different configurations. The results are as follows: Figure 9 As shown. From Figure 9 As can be seen, as the number of EH-IES mobile power supplies gradually increases from 12 (6 mobile energy storage vehicles and 6 hydrogen fuel cell generators) to 20 (10 mobile energy storage vehicles and 10 hydrogen fuel cell generators), the EH-ILCLE gradually increases from 419.26 × 10 3 The amount decreased to 213.63 × 10. 3 The percentage of EH-IMSPE gradually increased from 91.64% to 95.55%. This is because the increased number of mobile power sources not only improved the system's maximum power output but also added significant energy reserves to support a longer-lasting power supply. Furthermore, compared to hydrogen energy systems, which are limited in number and fixed location within the network topology, mobile power sources are more numerous and can be flexibly deployed at more dispersed grid access points, making their power supply range less susceptible to topology fluctuations.

[0356] However, the correlation between the number of mobile power supply units deployed and the improvement in system resilience also exhibits a diminishing marginal effect. Under Case 4, for every two additional mobile power supply units (one mobile energy storage unit and one hydrogen fuel cell generator), the EH-ILCLE decreases by 77.71 × 10⁻⁶. 3 Yuan, 50.85×10 3 Yuan, 43.31×10 3 Yuan, 33.76×10 3The EH-IMSPE and the EH-IMSPE are respectively increased by 1.20%, 1.01%, 0.92% and 0.78%. This is because, with the increase in the number of mobile power supplies, the unsupplied amount of important loads in the system gradually decreases, and the important loads in some scenarios can gradually achieve full supply, thereby causing the resilience improvement effect of the newly added mobile power supplies to decrease. In addition, in extreme disasters, important loads may be located in an island without a power grid access point, and it is difficult to effectively enhance the system resilience level by only increasing the number of mobile power supplies. Therefore, in order to guarantee the resilience and economy of the EH-IES, it is necessary to accurately evaluate the resilience improvement effect brought by the mobile power supplies, so as to realize the reasonable configuration of the mobile power supplies.

[0357] The application provides a two-stage resilience evaluation method for an electric-hydrogen integrated energy system under cross-region resource sharing, which comprises the following steps: establishing a two-stage resilience evaluation framework for the electric-hydrogen integrated energy system under cross-region resource sharing; establishing a pre-disaster cross-region deployment method for mobile power supplies of the electric-hydrogen integrated energy system based on the SAC algorithm; and establishing a post-disaster recovery strategy for the electric-hydrogen integrated energy system under cross-region resource sharing. The method can accurately evaluate the resilience level of the electric-hydrogen integrated energy system by fully considering the cross-region mutual aid potential of mobile emergency resources in the pre-disaster active deployment and post-disaster collaborative recovery stages.

Claims

1. A two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing, characterized in that, Includes the following steps: 1) Taking into full account the uncertainty of the distribution of faulty lines in multiple regions of the integrated electric-hydrogen energy system and the cross-regional mutual assistance potential of mobile power sources, a pre-disaster deployment strategy for the multi-region integrated electric-hydrogen energy system based on the discrete SAC algorithm is generated. 2) Considering the cross-regional support potential and post-disaster collaborative cooperation potential of mobile emergency resources, a post-disaster recovery strategy for the integrated electric-hydrogen energy system under cross-regional resource sharing is generated. 3) The resilience of the integrated electric-hydrogen energy system is assessed based on the pre-disaster deployment strategy of the multi-regional electric-hydrogen integrated energy system and the post-disaster recovery strategy of the integrated electric-hydrogen energy system under cross-regional resource sharing.

2. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 1, characterized in that, Step 1) involves generating a pre-disaster deployment strategy for a multi-regional integrated electric-hydrogen energy system based on the discrete SAC algorithm, including the following steps: 1.1) Construct a post-disaster emergency response model for an integrated electric-hydrogen energy system; 1.2) Model the decision-making process for the pre-disaster cross-regional deployment of the integrated electric-hydrogen energy system, and construct the state space, action space, and environment; 1.3) Utilize a policy network to generate pre-disaster deployment locations for mobile power supplies in multi-regional electric-hydrogen integrated energy systems; the policy network takes agent states as input and actions as output. 1.4) Based on the pre-disaster deployment location of mobile power sources in the multi-regional electric-hydrogen integrated energy system, solve the post-disaster emergency response model of the electric-hydrogen integrated energy system to obtain the load reduction loss at the current deployment location, which can be used as the reward function of the intelligent agent. 1.5) Utilize two comment networks to evaluate the value of the strategy network output actions, and select the action with the highest value as the pre-disaster deployment strategy for the multi-regional electric-hydrogen integrated energy system.

3. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 2, characterized in that, The post-disaster emergency response model for the integrated electric-hydrogen energy system aims to minimize the expected total operating cost of the integrated electric-hydrogen energy system within the region, namely: min(C lose,E +C lose,H +C gen,E +C gen,H ) (1) In the formula, N E N T1 These represent the distribution network nodes and the set of dispatching time periods within the region, respectively; c lose,Ei Penalty for unit load shedding at node i; Let c be the load amount cut off at node i at time t; lose,H The penalty is applied to the unit hydrogen load. The hydrogen loading at time t; N EV N HEV N RC These represent the groups consisting of mobile energy storage, hydrogen fuel cell generators, and maintenance personnel, respectively; N M Represents the set of grid connection points; c gen,E For the unit operating cost of mobile energy storage, c represents the discharge power of the nth mobile energy storage unit located at node i at time t; gen,H For the unit operating cost of hydrogen fuel cells, P represents the discharge power of the nth hydrogen fuel cell generator at node i at time t; t FC C represents the discharge power of the hydrogen fuel cell within the hydrogen energy system at time t. lose,E The total penalty for disconnecting the load; C lose,H The total penalty for hydrogen load reduction; C gen,E The total operating cost of mobile energy storage; C gen,H This refers to the total operating cost of hydrogen fuel cell vehicles and hydrogen energy systems. The constraints of the post-disaster emergency response model for the integrated electric-hydrogen energy system include mobile power supply operation constraints, hydrogen energy system operation constraints, network reconfiguration constraints, load shedding balance constraints, and power flow constraints of the distribution network. The constraints on mobile power supply operation include mobile energy storage discharge power constraints, mobile energy storage energy balance constraints, mobile energy storage upper / lower limit constraints, hydrogen fuel cell vehicle discharge power constraints, hydrogen fuel cell vehicle energy balance constraints, hydrogen fuel cell vehicle hydrogen storage upper / lower limit constraints, and the maximum number of mobile power supplies that can be connected to the grid at the grid access point. The constraints on the operation of hydrogen energy systems include the operating power constraints of hydrogen fuel cells, the upper / lower limits of hydrogen storage in hydrogen storage tanks, and the energy balance constraints of hydrogen storage tanks. The operating constraints of the power bank are as follows: In the formula, Let be the state variable of the nth mobile energy storage unit releasing energy at node i at time t. If it is 1, it means that the nth mobile energy storage unit is in the energy releasing state at node i at time t. P represents the energy released by the nth mobile energy storage unit at node i at time t. EV,out,max These are the upper limits of the power output of the power bank; Let η be the energy storage level of the nth mobile energy storage unit at time t; EV,out These are the energy release efficiency of mobile energy storage; S EV,max S EV,min Upper / lower limits for mobile energy storage; Let P be the state variable of the nth hydrogen fuel cell generator at time t when it releases energy at node i. If P is 1, it means that the nth hydrogen fuel cell generator is in a charging / discharging state at node i at time t. HEV,outi,t,n P represents the energy released by the nth hydrogen fuel cell generator at node i at time t; HEV,out,max These are the upper limits of the energy output of hydrogen fuel cell vehicles; η represents the energy storage level of the nth hydrogen fuel cell generator at time t; HEV For the power generation efficiency of hydrogen fuel cell vehicles; S HEV,max / S HEV,min The upper / lower limits for energy storage in hydrogen fuel cell vehicles; The maximum number of mobile power supplies that can be connected to the power grid at the grid access point; LHV H It has the lowest calorific value of hydrogen. The operating constraints of hydrogen energy systems are as follows: In the formula, P t FC P represents the operating power of the hydrogen fuel cell at time t. FC,max η is the maximum operating power of the hydrogen fuel cell. FC Energy conversion efficiency (LHV) of hydrogen fuel cells. H It has the lowest calorific value of hydrogen. S represents the amount of hydrogen stored in the hydrogen storage tank at time t; HS,min / S HS ,max Minimum / maximum hydrogen storage capacity of each hydrogen storage tank; The actual hydrogen supply load at time t; The network reconstruction constraints are as follows: In the formula, Let be the closed-loop state variable of line ij in time period t during the virtual power flow; Let be the state variable indicating whether potential root node i is the root node of the microgrid during time period t; M is a sufficiently large constant. Let be the virtual power flow passing through branch ij at time t. Let t be the virtual power flow passing through branch jh at time t. The load shedding balancing constraints are as follows: In the formula, Let i be the electrical load demand at time t. The actual active power load supplied to node i at time t; The required hydrogen load at time t; The actual hydrogen supply load at time t; Let be the amount of power cut-off at node i at time t; Let be the hydrogen shear load at time t; The power flow constraints of the distribution network are shown below: In the formula, P ij,t Q ij,t Let represent the active power and reactive power of line ij at time t, respectively. R represents the upper limits of active and reactive power of line ij, respectively. ij x ij Let be the resistance and reactance of line ij, respectively. Inject the sum of active and reactive power into the distributed power source at node j at time t; Let be the square of the voltage at node i at time t; Let P be the closed-loop state variable of distribution network line ij in time period t; jk,t Q jk,t Let be the active power and reactive power of line jk at time t; These are the lower and upper limits of the square of the voltage.

4. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 2, characterized in that, The state space is shown below: o=[U branch,all ,P load,all ,P H,all ,Q H,all ,P EV,all ,Q EV,all ,P HEV,all ,Q HEV,all ] (31) In the formula, o represents the state space of the agent, including the topological state information U of all regions. branch,all Load forecasting information P load,all Maximum power generation information P of hydrogen energy system H,all Hydrogen storage capacity information Q H,all And information on the maximum power generation of all mobile energy storage devices, P. EV,all Mobile energy storage capacity information Q EV,all Maximum power generation information for hydrogen fuel cell vehicles (P) HEV,all Hydrogen fuel cell vehicle capacity information Q HEV,all . The motion space is shown below: In the formula, a represents the action space of the agent; u i,n Characterizing the location state variables of mobile energy storage / hydrogen fuel cell vehicles at different grid connection points; a n The representative will deploy the nth mobile power supply to the th a. n At the power grid connection point, the nth mobile power supply is located at a. n The location's state variable is set to 1 to indicate that it is deployed in a. n Node; N M,all N represents the maximum number of grid connection points. EV,all N HEV ,all This represents the maximum number of mobile energy storage / hydrogen fuel cell vehicles that can be deployed. The environment includes a post-disaster emergency response model for an integrated electric-hydrogen energy system; The reward function is as follows: In the formula, the agent's reward function r up r is the sum of load shedding losses in all regions. i down Let be the reward function for the i-th lower-level agent. The total penalty for cutting off the electrical load in area i; The total penalty for cutting hydrogen load in region i.

5. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 2, characterized in that, The policy network generates a softmax distribution to output the corresponding probabilities of all possible discrete actions; The softmax distribution is shown below: In the formula, x n,j The preference value corresponding to the j-th action dimension in the n-th action output by the policy network reflects the relative preference for choosing the j-th action dimension in the n-th action; p n,j The probability of choosing an action; Loss function of policy network As shown below: In the formula, D is the experience pool, and α H Temperature coefficient; Q is the entropy regularization term; θ (o,a) is the Q-valued function; E o,a～D For expectations; Among them, the entropy regularization term As shown below: In the formula, A represents the total number of actions of the agent, and K... n These represent the total action dimensions of the nth action; The Q-value function is shown below: In the formula, These are two Q-value functions with identical structures; The Q-value network loss function is shown below: In the formula, L(θ1) and L(θ2) are Q-value functions. The loss; r is the reward; E (s,a,r)～D For expectations; Loss function of temperature coefficient L(α) H As shown below: In the formula, H tar The target entropy of the intelligent agent; For expectations; The action network, two evaluation networks, and temperature coefficient are updated as follows: In the formula, α represents the learning rate of the agent; L(θ) m ) represents the Q-value network loss function; θ m Indicates the evaluation of network parameters; This represents the parameters of the action network.

6. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 1, characterized in that, Step 2) involves generating a post-disaster recovery strategy for a cross-regional resource-sharing integrated electric-hydrogen energy system, including the following steps: 2.1) Construct a post-disaster recovery model for an integrated electric-hydrogen energy system that considers the coordinated scheduling of emergency resources; 2.2) Construct a post-disaster allocation model for mobile emergency resources that considers cross-regional resource sharing; 2.3) Solve the post-disaster recovery model of the integrated electric-hydrogen energy system and the post-disaster allocation model of mobile emergency resources to obtain the post-disaster recovery strategy of the integrated electric-hydrogen energy system under cross-regional resource sharing.

7. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 6, characterized in that, The objective function C1 of the post-disaster recovery model for an integrated electric-hydrogen energy system considering the coordinated scheduling of emergency resources is shown below: min C1=C lose,E +C lose,H +C tran +C gen,E +C gen,H -E disp (46) In the formula, N T2 c represents the set of time periods for post-disaster recovery scheduling within the region. tran For the unit dispatch cost of mobile emergency resources; Let N represent the state variables of the nth group of mobile energy storage, hydrogen fuel cell vehicle, and maintenance personnel located on the road at time t; RC c represents the group of maintenance personnel; disp Subsidies for units participating in the inter-regional allocation of mobile emergency resources; Let C represent the allocatable state of the nth group of mobile energy storage / hydrogen fuel cell vehicle / maintenance personnel at time t. A value of 1 indicates that the vehicle / person can participate in inter-regional allocation at time t; otherwise, they cannot. tran E represents the total dispatch cost of mobile emergency resources. disp Subsidies for mobile emergency resources participating in inter-regional allocation; The constraints of the post-disaster recovery model of the integrated electric-hydrogen energy system considering the coordinated scheduling of emergency resources include the time and space constraints of mobile emergency resources, the post-disaster recovery operation constraints of mobile power supply, the post-disaster recovery operation constraints of hydrogen energy system, the line maintenance status constraints, the allocable status constraints of mobile emergency resources, the distribution network reconfiguration constraints (16)-(19), the electric-hydrogen load balance constraints (21)-(24) and the distribution network power flow constraints (25)-(31); Among them, the spatiotemporal scheduling constraints of mobile emergency resources include the uniqueness constraint of the spatiotemporal location of mobile emergency resources, the location change state constraint of mobile emergency resources, the restriction constraint that the location transfer of mobile emergency resources between all nodes must pass through roads, and the spatiotemporal location transfer constraint of mobile emergency resources, namely: In the formula, u i,t,n This is a state variable representing the spatiotemporal location of mobile emergency resources. If it is 1, it means that the nth group of mobile emergency resources is located at node i at time t; otherwise, it is not. 0,t,n Let N be the state variable of the mobile emergency resource located on the road at time t; R N represents the set of transportation network nodes coupled to grid connection nodes, faulty lines, hydrogen energy systems, or hydrogen refueling stations; MER Represents the set of all mobile emergency resources; h i,t,n Let h be the state variable representing the change in position of node i at time t for the nth group of mobile emergency resources. i,t,n When T is 1 or -1, it represents the nth group of mobile emergency resources entering / leaving node i after time t, respectively; ij h is the time required to move from node i to node j. 0,t,n This indicates that the nth group of mobile emergency resources remains at node i after time t. The constraints for the post-disaster recovery and operation of mobile power sources include: the uniqueness of the charging / discharging state of mobile energy storage, the charging / discharging power of mobile energy storage, the energy balance of mobile energy storage, the upper / lower limit constraints of the energy storage capacity of mobile energy storage, the uniqueness of the hydrogen charging / discharging state of hydrogen fuel cell generators, the hydrogen charging rate / discharging power constraints of hydrogen fuel cell generators, the energy balance constraints of hydrogen fuel cell generators, the upper / lower limit constraints of the hydrogen storage capacity of hydrogen fuel cell generators, and the maximum number of mobile power sources that can be connected to the grid. In the formula, These are the state variables of the nth mobile energy storage vehicle charging / discharging at node i at time t. If the value is 1, it means that the nth mobile energy storage vehicle is in a charging / discharging state at node i at time t. P represents the charging power of the nth mobile energy storage vehicle at node i at time t. EV,in,max Maximum charging power for mobile energy storage; η EV,in η EV,out Improve the charging and discharging efficiency of mobile energy storage; Let N represent the state variables of the nth hydrogen fuel cell vehicle at time t, where it is either charging or discharging hydrogen at node i. A value of 1 indicates that the nth hydrogen fuel cell vehicle is in a charging / discharging state at node i at time t. H A collection consisting of nodes of a hydrogen energy system or hydrogen refueling station; M represents the hydrogen mass of the nth hydrogen fuel cell vehicle at node i at time t; HEV,in,max The upper limit for hydrogen refueling rate of hydrogen fuel cell vehicles; Let be the state variable of the nth mobile energy storage vehicle at node i at time t; Let be the state variable of the nth hydrogen fuel cell vehicle at node i at time t; P represents the discharge power of the nth mobile energy storage vehicle at node i at time t. EV,out,max This represents the maximum discharge power. Let be the energy storage level of the nth mobile energy storage unit at time t; Let P be the energy output of the nth hydrogen fuel cell generator at node i at time t; HEV,out,max This is the upper limit of the energy release power; The constraints for post-disaster recovery operation of hydrogen energy systems include the uniqueness constraint of the operating status of the electrolyzer and hydrogen fuel cell, the operating power constraint of the electrolyzer, the operating power constraint of the hydrogen fuel cell, the upper / lower limit constraint of hydrogen storage tank, and the energy balance constraint of the hydrogen storage tank, namely: In the formula, These are the operating status variables of the electrolyzer / hydrogen fuel cell at time t. A value of 1 indicates that the electrolyzer / hydrogen fuel cell is in operation at time t. If only hydrogen refueling stations are deployed in the area, then... and All are 0; P represents the operating power of the electrolytic cell at time t. ED,max This represents the upper limit of the operating power of the electrolytic cell; η ED The energy conversion efficiency of the electrolyzer; η HS,in / η HS,out These represent the hydrogen storage tank charging / discharging efficiency; N HTT A collection consisting of hydrogen-powered long-tube trailers; These represent the hydrogen charge / discharge mass of the nth hydrogen long-tube trailer at time t; Line maintenance status constraints include: a damaged line can be restored to an available state after the required repair time for maintenance personnel to arrive and remain on the line; a damaged line can only be repaired by one group of maintenance personnel; and the actual closed state of a branch is jointly constrained by the virtual power flow model and the line's available state. In the formula, t0 is the start time of post-disaster recovery; The time required to repair line ij; N D A collection of damaged lines; The constraints on the allocability of mobile emergency resources include: a region is considered fully restored when it can supply the full load without the need for hydrogen energy systems and mobile power supplies; mobile power supplies can only participate in inter-regional allocation when they have no charging or power supply tasks at the current moment; and maintenance personnel can only participate in inter-regional allocation after the region has been fully restored. In the formula, This is a region recovery status variable; a value of 1 indicates that the region has been fully recovered. This represents the assignability status of the nth group of maintenance personnel during this inter-regional allocation.

8. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 6, characterized in that, The objective function C2 of the mobile emergency resource post-disaster allocation model considering cross-regional resource sharing is shown below: my C2=C lose,E,all +C lose,H,all +C tran,all +C gen,E,all +C gen,H,all +C re,R,all (80) In the formula, C lose,E,all C lose,H,all C tran,all C gen,E,all C gen,H,all C re,R,all These are, respectively, the total penalty for power load shedding in all areas of the integrated electric-hydrogen energy system, the total penalty for hydrogen load shedding, the total dispatch cost of mobile emergency resources, the total operating cost of mobile energy storage, the total operating cost of hydrogen fuel cell vehicles and hydrogen fuel cells in the hydrogen energy system, and the total penalty for remaining critical line maintenance time; N MG It represents the collection of all microgrids in an integrated electric-hydrogen energy system; N represents the set of loads at all levels within a microgrid m; T3 Represents the set of scheduling time slots allocated between regions; c lose,El The unit load shedding penalty for level l load; P lose,Em,l,t N represents the load shedding amount of the l-th level load within the microgrid at time t; H,all M represents the set of all hydrogen energy systems or hydrogen refueling station nodes in an integrated electric-hydrogen energy system; lose,Hi,t N represents the hydrogen load shedding rate at time t for the hydrogen energy system or hydrogen refueling station located at node i; RC ,all / N HTT,all These respectively represent the collection of all maintenance personnel and all hydrogen long-tube trailers in the integrated electric-hydrogen energy system; This represents the state variable of the nth hydrogen long-tube trailer at time t, where the trailer is located on the road. c represents the power generation of the hydrogen fuel cell located at node i at time t; re T is the penalty coefficient for the unit waiting time for maintenance on the critical path; re,Rα The critical path maintenance time at time t in region α; The constraints of the mobile emergency resource post-disaster allocation model considering cross-regional resource sharing include the maintenance personnel allocation model based on critical path maintenance time, the mobile power allocation model based on virtual node aggregation, the inter-regional hydrogen allocation model, and the time and space constraints of mobile emergency resources (53)-(56). The maintenance personnel allocation model based on critical path maintenance time is shown below: In the formula, T represents the set of critical faulty lines within region α; re,initα The critical path maintenance duration for region α at the initial time of inter-region allocation; t1 is the initial time of inter-region allocation; u RC,Aα,τ,n This represents the allocation status of the nth group of maintenance personnel in region α at time τ. A value of 1 indicates that the nth group of maintenance personnel was allocated to region α at time τ. RCn This represents the assignability status of the nth group of maintenance personnel during this inter-regional allocation, and its value is determined by the regional recovery plan. The mobile power distribution model based on virtual node aggregation is shown below: In the formula, P G,allm,t N represents the total injected power from mobile power sources, hydrogen energy systems, and the upstream power grid within the microgrid at time t. If the microgrid does not contain any of these devices, its corresponding power is 0. MMm / N Hm / N UGm These represent the sets consisting of the grid connection points within the microgrid m, the hydrogen energy system, and the connection points to the upstream grid, respectively; P Gridi,t Let be the injected power of the upstream power grid connected to node i at time t; P represents the power consumption of the electrolytic cell located at node i at time t. load,Em,l,t N represents the load demand of the l-th level load within the microgrid at time t; Mα / N Hα These represent the sets of grid connection points / hydrogen energy systems or hydrogen refueling station nodes within region α; u EV,Aα,t,n / u HEV,Aα,t,n These represent the allocation status of the nth mobile energy storage / hydrogen fuel cell vehicle. A value of 1 indicates that the nth mobile energy storage / hydrogen fuel cell vehicle is allocated to region α at time t; n EVn / n HEVn These represent the allocatable states of the nth mobile energy storage / hydrogen fuel cell vehicle during this inter-regional allocation process. The inter-regional hydrogen distribution model is shown below: In the formula, u HTT,ini,t,n / u HTT,outi,t,n These are the hydrogen charging / discharging state variables of the nth hydrogen long-tube trailer at node i at time t; u HTTi,t,n Let M be the position state variable of the nth hydrogen long-tube trailer at node i at time t; HTT,in,max / M HTT,out,min These are the maximum hydrogen charging / discharging rates for the hydrogen-powered long-tube trailer; S represents the energy storage level of the nth long-tube trailer at time t; HTT,max / S HTT,min These represent the upper and lower limits of hydrogen long-tube trailer energy storage; M HTT,ini,t,n / M HTT,outi,t,n Let represent the hydrogen charging / discharging mass of the nth hydrogen long-tube trailer at time t at node i, i.e., the hydrogen energy allocation plan between regions.

9. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 6, characterized in that, The steps for solving the post-disaster recovery model of the integrated electric-hydrogen energy system and the post-disaster allocation model of mobile emergency resources include: 2.3.1) Linearize the post-disaster recovery model of the integrated electric-hydrogen energy system and the post-disaster allocation model of mobile emergency resources. The steps include: For the absolute value term in equation (54), a 0-1 auxiliary variable h is introduced. absi,t,n , Linearizing the original constraints, we get: By transforming the condition constraint (76) using the Big-M method, we obtain: For the maximum value function in equation (88), define T re,R,maxα The equivalent output after linearization is obtained, and 0-1 auxiliary variables are introduced. By linearizing the original constraints using the Big-M method, we obtain: 2.3.2) After a disaster occurs, obtain the travel time matrix of each node within and between regions, and statistically analyze the status information of mobile emergency resources, information on damaged lines, and information on electricity and hydrogen load demand; 2.3.3) The post-disaster recovery model of the integrated electric-hydrogen energy system, which takes into account the coordinated scheduling of emergency resources, is calculated for each region. The recovery plan and mobile emergency resource status information are reported to the joint disaster relief center based on the calculation results. 2.3.4) Based on the recovery plans and mobile emergency resource status information reported by each region, the Joint Disaster Relief Center calculates a linearized mobile emergency resource post-disaster allocation model that considers cross-regional resource sharing, and issues mobile emergency resource allocation plans to each region. 2.3.5) Based on the allocation results of mobile emergency resources, each region calculates the linearized post-disaster recovery model of the integrated electric-hydrogen energy system considering the coordinated scheduling of emergency resources, and executes the model before t. c The recovery plan for the specified time period, and upload the t c Post-period recovery plans and mobile emergency resource status information; 2.3.6) If all regions have been fully restored, stop the calculation. Otherwise, T = T + t c Proceed to step 2.3.4).

10. The two-stage resilience assessment method for an integrated electric-hydrogen energy system under cross-regional resource sharing as described in claim 1, characterized in that, The steps for assessing the resilience of an integrated electric-hydrogen energy system include: 3.1) Determine the multidimensional resilience indicators of the integrated electric-hydrogen energy system, including the expected integrated electric-hydrogen load shedding loss (EH-ILCLE), the expected integrated electric-hydrogen minimum energy supply ratio (EH-IMSPE), and the expected integrated electric-hydrogen recovery time (EH-IPRTE). The expected EH-ILCLE of the combined load shedding loss for hydrogen-electricity combination is as follows: In the formula, N S This represents the total number of simulated scenarios; The set of distribution network nodes within region α; This refers to the moment when region α, under scenario s, does not rely on hydrogen energy systems or mobile power sources and its performance recovers to pre-disaster levels. This represents the start time of post-disaster recovery in region α under scenario s. The unit load cut-off penalty for node i in region α; Let t be the amount of power cut-off load at node i in region α under scenario s; The unit hydrogen shear load penalty applies to the α region; Let be the hydrogen shearing load in region α at time t under scenario s. The expected minimum energy supply ratio of combined hydrogen and electricity (EH-IMSPE) is shown below: In the formula, Let t be the electrical load demand at node i in region α under scenario s; Let N be the hydrogen load demand in region α at time t under scenario s. A Let 's' be the number of regions in scenario 's'. The expected electro-hydrogen integrated recovery time (EH-IPRTE) is shown below: 3.2) Input the operating parameters of the integrated electric-hydrogen energy system and the equipment parameters of various emergency resources, and set the maximum number of simulation scenarios N. S And let s = 1. 3.3) Based on disaster prediction information and the component failure probability model under the corresponding disaster, calculate the failure probability of each line. 3.4) A fault scenario set for the integrated electric-hydrogen energy system is constructed using Latin hypercube sampling technology; the fault scenario set is determined by random variable u. ij This characterizes the operating status of line ij. If it is 1, it means that the line is operating normally. Otherwise, it means that the line has failed due to extreme disasters. random variable u ij As shown below: In the formula, It is the inverse function of the cumulative distribution function of the fault probability of line ij; Let y be the fault probability of line ij; ij It is a random value; 3.5) Based on the line fault probability, load characteristics, and status information of hydrogen energy system and mobile power supply in each region, the SAC algorithm is used to solve the pre-disaster cross-regional deployment model of mobile power supply for integrated electric-hydrogen energy system, and obtain the pre-disaster cross-regional deployment decision of mobile power supply. 3.6) Solve the post-disaster recovery model of the integrated electric-hydrogen energy system under cross-regional resource sharing in the s-th scenario. The steps include: 3.6.1) Import the line damage information for the s-th scenario, obtain the travel time matrix of each node within and between regions, and import the operating parameters of the integrated electric-hydrogen energy system and the pre-disaster deployment location of the mobile power supply. 3.6.2) Each region calculates a post-disaster recovery model for an integrated electric-hydrogen energy system that takes into account the coordinated scheduling of emergency resources, and reports the recovery plan and mobile emergency resource status information to the joint disaster relief center based on the calculation results. 3.6.3) Based on the recovery plans and mobile emergency resource status information reported by each region, the Joint Disaster Relief Center calculates a post-disaster allocation model for mobile emergency resources that takes into account cross-regional resource sharing, and issues mobile emergency resource allocation plans to each region. 3.6.4) Based on the allocation results of mobile emergency resources, each region calculates a post-disaster recovery model for the integrated electric-hydrogen energy system, taking into account the coordinated scheduling of emergency resources, and executes the model before t. c The recovery plan for the specified time period, and upload the t c Recovery plans and mobile emergency resource status information after the designated period. 3.6.5) If all areas of the integrated electric-hydrogen energy system have been restored, proceed to step 3.7); otherwise, T = T + t c Proceed to step 3.6.3). 3.7) Store the post-disaster cross-regional recovery status of the integrated electric-hydrogen energy system under the s-th scenario. 3.8) If the maximum number of simulated scenarios is reached, proceed to step 3.6); otherwise, s = s + 1 and proceed to step 3.

9. 3.9) Calculate the multidimensional resilience index of the integrated electric-hydrogen energy system.

Images