A modeling and time efficiency analysis method for forest fire air-ground cooperative emergency rescue process

Abstract

The application provides a modeling and time efficiency analysis method for a forest fire air-ground cooperative emergency rescue process, first constructs a 'library-place-transition-arc' process chart for the forest fire air-ground cooperative emergency rescue, and builds an SPN model according to the process chart; builds an isomorphic Markov chain of the SPN model and obtains the steady-state probability of each state; calculates the library busy probability and transition utilization rate and determines the total time of the emergency rescue process according to the library busy probability and the transition utilization rate; finally, the transition and link with higher utilization rate are analyzed and expanded to find further optimization space. The application can model and analyze the time efficiency of the forest fire air-ground cooperative emergency rescue process, can propose a more widely covered and more reliable air-ground cooperative rescue process system based on the existing rescue process, and can deeply analyze the rescue link with high time proportion and analyze the internal reasons, so that more comprehensive scheme assistance and decision basis are provided for the emergency management department.

Description

Technical Field

[0001] This invention relates to the field of emergency rescue technology, and in particular to a modeling and timeliness analysis method for a forest fire air-ground coordinated emergency rescue process. Background Technology

[0002] Forest fires have a severe impact on human society and the ecological environment. They are extremely destructive natural disasters, burning large amounts of forest vegetation, producing dense smoke and harmful gases, severely disrupting the ecological balance, and drastically reducing biodiversity. At the same time, the soil erosion and compaction caused by forest fires result in enormous losses to agricultural production and infrastructure construction.

[0003] Due to the disruption of ground transportation in disaster environments, relying solely on traditional ground-based disaster relief methods is no longer sufficient to improve the efficiency of complex and ever-changing fire rescue operations. How to leverage the flexibility, maneuverability, and less geographically restrictive characteristics of aircraft to evolve from a single ground-based disaster relief model to air-ground coordinated rescue, and to establish an efficient "air-ground coordinated" emergency fire rescue process, is a pressing scientific problem and a practical need. Therefore, building an air-ground coordinated rescue model based on the existing disaster relief system is of great significance for significantly shortening the operation time of the rescue process and for the command work of decision-makers. Summary of the Invention

[0004] This invention proposes a modeling and timeliness analysis method for air-ground coordinated emergency rescue processes in forest fires. Based on existing rescue processes, it can propose a more comprehensive and reliable air-ground coordinated rescue process system, and conduct timeliness analysis on the air-ground coordinated emergency rescue process to determine the overall rescue process time and conduct in-depth evaluation of rescue links with high time proportions. This provides emergency management departments with more comprehensive scheme assistance and decision-making basis.

[0005] The technical solution of this invention is implemented as follows: a method for modeling and timeliness analysis of a forest fire air-ground coordinated emergency rescue process, comprising the following steps:

[0006] Step 1: Using the rescue stage as the reservoir node, the tasks completed in the rescue stage as the transitions, and the arcs with arrows to represent the sequence of implementation between the various rescue stages, construct a "reservoir-transition-arc" flowchart for the air-ground coordinated emergency rescue process for forest fires.

[0007] Step 2: According to the "Place-Transition-Arc" flowchart, the place node is H. x The transition to T y Given that x and y are both positive integers, construct an SPN model for the air-ground coordinated emergency rescue process for forest fires;

[0008] Step 3: Use PIPE simulation software to test the SPN model and obtain the reachability graph of the SPN model. The reachability graph contains the identifier M. i For all warehouses H x The set of token states contained therein, 0≤i≤n-1, where n is the total number of identifiers;

[0009] Step 4: Based on the identifier M in the reachability graph i Construct an isomorphic Markov chain for the SPN model, and calculate each identifier M based on this Markov chain. i The steady-state probability;

[0010] Step 5: Calculate H for each storage location node x Busy probability and various transitions T y The utilization rate is then calculated, and the total runtime t of the SPN model is determined. The transitions with higher utilization rates, T, are then selected. y As a key factor restricting the efficiency of rescue efforts;

[0011] Step 6: By changing the transition T of key links y The implementation rate is determined, and its relationship with the total running time t of the SPN model is fitted to determine the transition T of key links. y The optimal implementation rate provides a relevant entry point for decision-makers to further improve the efficiency of emergency rescue.

[0012] Furthermore, in step 4, the identifier M is... i The steady-state probability matrix equation is as follows:

[0013]

[0014] Where A = [P(M0), P(M1), ..., P(M... i ),…,P(M n-1 Let P(M0), P(M1), ..., P(M2) represent the set of all identifiers of steady-state probabilities. i ),…,P(M n-1 ) represents the steady-state probability of the corresponding rescue stage; Let ξ be the transition excitation rate matrix. ij For M i To M j The transfer rate, M i M represents the i-th identifier. j Let i represent the j-th identifier, 0 ≤ j ≤ n-1; when i ≠ j, and M is a Markov chain... i To M j When there is a directed arc, ξ ij Equal to the transition excitation rate λ on the arc y Transition excitation rate λ y For the change Ty The average implementation rate (d / time) was determined by analyzing data from previous forest fire cases to identify various changes in T. y The average implementation rate; if there is no directed arc between them, then ξ ij The value is 0; when i = j, the value corresponding to the diagonal of the Q matrix is ​​0. ξ ii The value is M i With M j The excitation rate λ of strain transfer on the directed arc between them y The negative of the sum.

[0015] Furthermore, in step 5, the H of each storage node is calculated according to formula (2). x The probability of being busy is given by formula (2) as follows:

[0016]

[0017] Among them, P[M(H x [) = 1] represents different identifiers of the internal storage node H x The probability of being busy in an excited state, H x Being in the activated state means having one token. Indicates the identifier M j Contains the activated reservoir node H x The cumulative probability value.

[0018] Furthermore, in step 5, each transition T is calculated according to formula (3). y The utilization rate is given by formula (3) as follows:

[0019]

[0020] Among them, U(T) y ) is able to trigger transition T y All identifiers M i The sum of the steady-state probabilities is the transition T. y Utilization rate; E is for allowing changes T y All implementable identifiers M i A set of.

[0021] Further, in step 5, the total running time t of the SPN model is calculated according to formula (4), which is as follows:

[0022]

[0023] in, This represents the average number of tags in the library under stable system conditions. ∑U(T y For all changes T yThe sum of utilization rates, where R is the marked flow rate flowing into the system; R(T) y H x )=W(T y H x )U(T1)λ1, where W(T y H x ) represents the transition T y Pointing to the storage node H x The directed arc weights are given by U(T1), where U(T1) is the utilization rate of transition T1, and λ1 is the average implementation rate of transition T1.

[0024] Furthermore, in step 1, following the logical framework and emergency response approach of "fire monitoring → information transmission → situation assessment → dispatching rescue → extinguishing open flames → clearing remaining fires", and combining the operational mechanism of aviation rescue in forest fires, a "location-change-arc" flowchart for the air-ground coordinated emergency rescue process for forest fires is constructed, with the rescue stage as the storage node, the tasks completed in the rescue stage as the change, and the arc with arrows representing the sequence of implementation between each rescue stage.

[0025] The beneficial effects of this invention are:

[0026] This invention constructs a SPN model for the air-ground coordinated emergency rescue process for forest fires, integrating ground and air rescue forces. This model can optimize the current forest fire rescue system through air-ground coordination, estimate the time required for the rescue process, and further analyze the causes of key time-consuming steps within the system. It provides effective solutions and decision-making basis for emergency management departments to further improve rescue efficiency after adopting this system. Based on the emergency rescue process of the 2018 Yiyuan County "4.6" forest fire, the rescue time can be shortened from 18 hours to 13.65 hours, improving overall efficiency by 24.2%.

[0027] This invention takes the changes with high utilization rates as the key links and the links to be optimized in the forest fire air-ground linkage emergency rescue process model, fits the relationship between the implementation rate of the key links and the overall operation time, and determines the optimal implementation rate of the changes of the key links.

[0028] The SPN model of this invention reveals that the rescue aircraft approval process has a relatively weak impact on the system compared to the other two key changes. However, because its changes can improve the efficiency of bucket operations in controlling fires and the approval process for rescue aircraft, it can further reduce the process time by 1.92 hours, and therefore deserves more attention. This invention achieves better results with relatively small changes by improving the implementation speed of key processes. Therefore, when analyzing key processes using the SPN model, it is also important to pay attention to the relationships between them. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 A "reservoir-change-arc" flowchart for the air-ground coordinated emergency rescue process for forest fires;

[0031] Figure 2 SPN model for air-ground coordinated emergency rescue process for forest fires;

[0032] Figure 3 This is the reachability graph for the SPN model;

[0033] Figure 4 The relationship between the implementation rate of the three key rescue stages and the total operation time t;

[0034] Figure 5 For λ 14 The variation of total runtime t under different implementation rates;

[0035] Figure 6 For λ 17 Variation of total runtime t under different implementation rates. Detailed Implementation

[0036] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] A method for modeling and timeliness analysis of a coordinated air-ground emergency rescue process for forest fires includes the following steps:

[0038] Step 1: Following the logical framework and emergency response approach of "fire monitoring → information transmission → situation assessment → rescue deployment → open flame extinguishing → remaining fire cleanup," and combining the operational mechanism of aerial rescue in forest fires, this step constructs a "location-change-arc" flowchart for the air-ground coordinated emergency rescue process for forest fires. The flowchart uses rescue stages as nodes, tasks completed in each rescue stage as transitions, and arrowed arcs to represent the sequence of implementation between each rescue stage. Figure 1 As shown.

[0039] After a forest fire breaks out, the forest fire prevention command organizes nearby rescue forces to the affected area. Simultaneously, drones are deployed to monitor the fire and provide situational feedback. If the situation escalates, an emergency rescue command center is established to assess the fire situation using drone monitoring and other methods, formulate an air-ground coordinated emergency rescue plan, and organize rescue forces. Ground rescue forces, upon arrival at the fire site, use machinery to dig firebreaks to prevent the fire from spreading. Meanwhile, air rescue forces, after obtaining airspace approvals, proceed to the site to conduct bucket operations, controlling the fire from the center. Once the fire weakens, rescue personnel are deployed to the fire area via aerial slings to fight the fire from the inside out, ultimately integrating with ground rescue forces fighting the fire from the outside in to achieve multi-point breakthroughs and segmented firefighting. After the firefighting operation is completed, aircraft equipped with infrared cameras continuously monitor the remaining fire and guard the fire site.

[0040] Step 2: Based on the "Location-Transition-Arc" flowchart, and according to the input and output relationships between key links in the forest fire emergency rescue process and the SPN model, represent task resources, information, and conditions as a finite set of locations, and represent the time, actions, and information transmission and reception during the rescue activity as a finite set of transitions. The location node is H. x The transition to T y Both x and y are positive integers. Furthermore, a stochastic SPN model is established based on the optimized relationships between the initial information feedback, the mid-stage fire rescue, and the later stage of fire scene management, such as... Figure 2 As shown.

[0041] This embodiment aims to explore the efficiency of air-ground coordinated emergency rescue in forest fires and the time-consuming rescue phases. Therefore, the weight of each transition in the emergency process is set to 1, and the initial repository node H1 has 1 token. The model means that when information about a forest fire is received, H1 begins to pass tokens down. T1 and T2 represent the dispatch of specialists to the scene by the emergency department after the forest fire occurs, and the deployment of drones to collect fire-related information. T3 represents the determination of the fire situation based on the information feedback from the on-site specialists and drones, and the commencement of air-ground rescue deployment. T4, T6, and T7 represent the determination of the fire level and the dispatch of rescue teams to the assigned rescue areas to carry out rescue activities. T5 represents the advance specialists remaining on-site to organize initial fire fighting and integrating with the subsequent main ground rescue force (T8). T8 to T 15 For various rescue missions in both air and ground rescue, and in T 15 Integration will be carried out at this location (dual-line coordinated operation). T 16 To T 18 To prepare for the later stages of the firefighting operation, the remaining embers were cleared and the fire site was monitored. 19 As the disaster worsens, reinforcements are needed. The specific meanings of the 22 locations and 19 transitions included in the model are shown in Table 1 below.

[0042] Table 1. Meaning of locations and changes in the model.

[0043]

[0044] Step 3: Use PIPE simulation software to test the SPN model and obtain the reachability map of the SPN model, such as... Figure 3 As shown in the figure, the analysis indicates that the token flow in the SPN model is normal, with only one token existing in each place node Hx, and no deadlock phenomenon occurs.

[0045] Depend on Figure 3 It can be seen that the forest fire air-ground coordinated emergency rescue process model constructed in this embodiment has safety, boundedness, dynamism, and accessibility. The marker M in the accessibility map... i For all place nodes H x The set of token states, 0≤i≤n-1, where n is the total number of tokens. The initial token M0=(1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0), indicates that in the initial state, the place node H1 has 1 token, while other places have no tokens, which is simplified to M0=(1); M1=(2) indicates that in state 1, only place H2 has one token, and all other places have no tokens. Similarly, simplifying all the identifiers in the SPN model, the specific distribution is as follows: M0 = (1), M1 = (2), M2 = (3,4), M3 = (5,6), M4 = (5,8), M5 = (6,7), M6 = (7,8), M7 = (6,9), M8 = (8,9), M9 = (6,10,11), M 10 = (8, 10, 11), M 11 = (6, 10, 14), M 12 = (8, 10, 14), M 13 = (11,12), M 14 = (6, 10, 16), M 15 = (8, 10, 16), M 16 = (12, 14), M 17 = (11, 13), M 18 = (6, 10, 18), M 19 = (8, 10, 18), M 20 = (12, 16), M 21 = (13,14), M 22 = (11, 15), M 23 = (12, 18), M 24 = (13, 16), M 25 = (14, 15), M 26 = (11, 17), M27 = (13, 17), M 28 = (15, 16), M 29 = (14, 17), M 30 = (15, 18), M 31 = (16, 17), M 32 = (17,18), M 33 =(19), M 34 =(20), M 35 =(21), M 36 = (22).

[0046] Step 4: Based on the identifier M in the reachability graph i Construct an isomorphic Markov chain for the SPN model, and calculate each identifier M based on this Markov chain. i The steady-state probability is obtained using the following method:

[0047] By combining the steady-state distribution of Markov chains with the Chepman-Kolmogorov equation, we can identify M. i The steady-state probability matrix equation is as follows:

[0048]

[0049] Where A = [P(M0), P(M1), ..., P(M... i ),…,P(M n-1 Let P(M0), P(M1), ..., P(M2) represent the set of all identifiers of steady-state probabilities. i ),…,P(M n-1 ) represents the steady-state probability of the corresponding rescue stage; Let ξ be the transition excitation rate matrix. ij For M i To M j The transfer rate, M i M represents the i-th identifier. j Let i represent the j-th identifier, 0 ≤ j ≤ n-1; when i ≠ j, and M is a Markov chain... i To M j When there is a directed arc, ξ ij Equal to the transition excitation rate λ on the arc y Transition excitation rate λ y For the change T y The average implementation rate can be determined by analyzing data from past forest fire cases to identify the average implementation rate of each task, which is termed the change T. y The average implementation rate; if there is no directed arc between them, then ξ ij The value is 0; when i = j, the value corresponding to the diagonal of the Q matrix is ​​0. ξii The value is M i With M j The excitation rate λ of strain transfer on the directed arc between them y The negative of the sum.

[0050] Step 5: Calculate H for each storage location node x Busy probability and various transitions T y The utilization rate is then calculated, and the total running time t of the SPN model is determined. The nodes H with higher probability are then selected. x As the subject of subsequent in-depth optimization research, the specific methods are as follows:

[0051] Calculate the H of each storage node according to formula (2). x The probability of being busy is given by formula (2) as follows:

[0052]

[0053] Among them, P[M(H x [) = 1] represents different identifiers of the internal storage node H x The probability of being busy in an excited state, H x Being in the activated state means having one token. Indicates the identifier M j Contains the activated reservoir node H x The cumulative probability value.

[0054] Calculate each transition T according to formula (3) y The utilization rate is given by formula (3) as follows:

[0055]

[0056] Among them, U(T) y ) is able to trigger transition T y All identifiers M i The sum of the steady-state probabilities is the transition T. y The high utilization rate of the change indicates that this rescue link takes a relatively long time in emergency decision-making and response within the "air-ground coordinated" rescue process for forest fires; E is to allow the change T y All implementable identifiers M i A set of.

[0057] The total running time t of the SPN model is calculated according to formula (4), which is as follows:

[0058]

[0059] in, This represents the average number of tags in the library under stable system conditions. ∑U(T yFor all changes T y The sum of utilization rates, where R is the marked flow rate flowing into the system (SPN model); R(T) y H x )=W(T y H x )U(T1)λ1, where W(T y H x ) represents the transition T y Pointing to the storage node H x The directed arc weights are given by U(T1), where U(T1) is the utilization rate of transition T1, and λ1 is the average implementation rate of transition T1.

[0060] The runtime of the SPN model is negatively correlated with the implementation rate of each rescue stage; that is, as the speed of rescue mission implementation increases, the overall rescue time decreases. Therefore, this embodiment selects the rescue stage with the highest transition utilization rate as the key stage, explores the impact of the key stage's implementation rate on the process runtime, and analyzes the changes in the overall rescue runtime by adjusting the implementation rate of the key stage.

[0061] Taking the 2018 Yiyuan County, Shandong Province "4.6" forest fire rescue process as an example, based on the SPN model of the forest fire air-ground coordinated emergency rescue process constructed in this embodiment, the overall rescue efficiency of the system is simulated and analyzed, key links are identified, and the key issues restricting rescue efficiency and their causes are discussed. Through reviewing relevant materials and literature, the overall rescue time in the Yiyuan County, Shandong Province "4.6" forest fire was determined to be 18 hours, with the implementation rates of various transitions shown in the table below:

[0062] Table 2. Changes T y Average implementation rate λ y

[0063]

[0064] The implementation rate λ of each change in Table 2 y Substituting into equation (1) yields all identifiers M. i The steady-state probabilities are shown in Table 3 below.

[0065] Table 3 Steady-state probability results for reachability identifiers

[0066] logo steady state probability logo steady state probability logo steady state probability logo steady state probability <![CDATA[P(M0)]]> 0.02853 <![CDATA[P(M 10 )]]> 0.03090 <![CDATA[P(M 20 )]]> 0.00587 <![CDATA[P(M 30 )]]> 0.04288 <![CDATA[P(M1)]]> 0.05705 <![CDATA[P(M 11 )]]> 0.00359 <![CDATA[P(M 21 )]]> 0.00182 <![CDATA[P(M 31 )]]> 0.08705 <![CDATA[P(M2)]]> 0.00571 <![CDATA[P(M 12 )]]> 0.00486 <![CDATA[P(M 22 )]]> 0.01602 <![CDATA[P(M 32 )]]> 0.08558 <![CDATA[P(M3)]]> 0.01223 <![CDATA[P(M 13 )]]> 0.00775 <![CDATA[P(M 23 )]]> 0.00399 <![CDATA[P(M 33 )]]> 0.08558 <![CDATA[P(M4)]]> 0.00204 <![CDATA[P(M 14 )]]> 0.01057 <![CDATA[P(M 24 )]]> 0.00981 <![CDATA[P(M 34 )]]> 0.19903 <![CDATA[P(M5)]]> 0.02445 <![CDATA[P(M 15 )]]> 0.01743 <![CDATA[P(M 25 )]]> 0.00337 <![CDATA[P(M 35 )]]> 0.05705 <![CDATA[P(M6)]]> 0.01834 <![CDATA[P(M 16 )]]> 0.00141 <![CDATA[P(M 26 )]]> 0.01842 <![CDATA[P(M 36 )]]> 0.05705 <![CDATA[P(M7)]]> 0.00445 <![CDATA[P(M 17 )]]> 0.00918 <![CDATA[P(M 27 )]]> 0.00772 <![CDATA[P(M8)]]> 0.00411 <![CDATA[P(M 18 )]]> 0.00560 <![CDATA[P(M 28 )]]> 0.03076 <![CDATA[P(M9)]]> 0.02470 <![CDATA[P(M 19 )]]> 0.01116 <![CDATA[P(M 29 )]]> 0.00396

[0067] The H of each storage node is calculated using equation (2). x The probability of containing a token is used to calculate the busy probability of the 22 supply nodes. Then, the transitions T are obtained using equation (3). y The utilization rates are shown in Table 4 below. The results indicate that the three transitions with the highest utilization rates in the SPN model are T... 17 The remaining fire has been extinguished, T14 The probability of success for the bucket-bombing operation to control the fire, and the approval of the T9 rescue aircraft application, are 0.19903, 0.16149, and 0.10697, respectively. The high utilization rate indicates that this step accounts for a significant proportion of the time in the rescue process, making it a key factor limiting rescue efficiency and a crucial area for optimization.

[0068] Table 4 Change T y Utilization results

[0069] change Utilization change Utilization change Utilization change Utilization T1 0.02853 T6 0.04279 T11 0.02853 T16 0.08558 T2 0.05705 T7 0.03535 T12 0.01901 T17 0.19903 T3 0.00571 T8 0.06435 T13 0.09303 T18 0.05705 T4 0.01427 T9 0.10697 T14 0.16149 T19 0.05705 T5 0.03668 T10 0.01902 T15 0.08558

[0070] The reason for the lengthy ember cleanup was that the fire in this case originated on a mountaintop and then spread downwards, becoming a downhill fire driven by wind. On that day, the wind force reached level 5-6, with gusts reaching level 7, resulting in high wind speeds. Furthermore, the dense vegetation in the area facilitated the rapid spread and intense combustion of crown fires. Additionally, the steep terrain and high altitude slowed the rescue personnel's progress to the ember locations, preventing immediate extinguishing of the fire. This led to multiple instances of reignition and even explosions that breached firebreaks, further complicating the ember cleanup efforts.

[0071] The reason why helicopter water-dropping operations took so long was that the wind was too strong on the day of the forest fire, increasing the difficulty of operating the rescue aircraft. To ensure flight safety, the pilots had to slow down the flight speed, thus prolonging the operation time. In addition, the helicopters involved in this firefighting effort consisted of one M-26 heavy helicopter (with a single water load of 15 tons) and one M-8 medium helicopter (with a single water load of 3 tons). The M-8 medium helicopter, with its smaller water load, needed to frequently collect water. Furthermore, the average flight distance between the water collection point and the fire site was 12 kilometers, and most of the helicopter's operating time was spent traveling to and from the water collection point, severely exacerbating the time required for water-dropping firefighting operations.

[0072] The reason why the application and approval process for rescue aircraft takes a long time is that some grassroots fire prevention command departments are not familiar with the application process for forest fire rescue helicopters, and airspace issues have always been a key constraint on my country's aviation emergency response. As a result, the coordination of mission approval, route planning, and crew coordination is not smooth, which delays the timeliness of helicopter rescue.

[0073] Simulation calculations of the air-ground coordinated rescue time during the "4.6" forest fire in Yiyuan County, Shandong Province in 2018 revealed that the average implementation time of the system (SPN model) was 13.65 hours, which is 4.35 hours shorter than the original plan (18 hours) in the real-world case, where ground and air rescue responded sequentially, thus improving efficiency by 24.2%. Specific calculations are as follows:

[0074]

[0075] To analyze the impact of the operational rate of key links on the overall rescue time, this embodiment fits λ using formulas (1)-(4). 17 , λ 14 The relationship between the implementation rate of the 93 key links and the overall rescue time (total running time t of the SPN model) Figure 4 The step size is set to 0.1. A comparison shows that as the implementation rate of the change increases, the total running time t gradually decreases, and the variation in the overall rescue time decreases accordingly, gradually stabilizing. When T... 17 ≥0.83 days / time, T 14 When the time is ≥1.33 days / time and T9≥1.53 days / time, the overall rescue time changes very little.

[0076] Furthermore, according to the emergency management phase, among the three key stages, the approval of the aviation rescue application (T9) belongs to the early information feedback stage of the disaster, and the bucket operation to control the fire (T) 14 It is in the mid-stage of fire rescue and the remaining fire has been cleared. 17 This falls under the category of later-stage fire scene management. To explore the impact of the rescue aircraft approval process on other key processes and the overall system runtime, this embodiment simulates three different states: constant T9 implementation rate, 0.75 times the frequency, and 0.5 times the frequency. 14 and T 17 Regarding the variation pattern of the overall system uptime, see Figure 5 and Figure 6 .

[0077] Simulation results show that when T 14 and T 17 At low implementation rates, changing T9 has little impact on system runtime. However, as the implementation rate increases, the difference in system runtime across the three scenarios gradually widens. The impact of changing the T9 implementation rate on the overall system time is relatively smaller compared to T... 14 and T 17 While the impact may not be immediately apparent, the increased efficiency leads to reduced time for mid-stage rescue and post-rescue operations, and the system uptime can be reduced from 13.65 hours to 11.73 hours, making its influence more significant. This provides decision-makers with insights into not only the impact of changes in the implementation rate of key changes themselves on the overall system uptime, but also the connections between different changes, comprehensively considering the interrelationships among the three changes to further improve system efficiency.

[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for modeling and timeliness analysis of a forest fire air-ground coordinated emergency rescue process, characterized in that, Includes the following steps: Step 1: Using the rescue stage as the reservoir node, the tasks completed in the rescue stage as the transitions, and the arcs with arrows to represent the sequence of implementation between the various rescue stages, construct a "reservoir-transition-arc" flowchart for the air-ground coordinated emergency rescue process for forest fires. Step 2: According to the "Place-Transition-Arc" flowchart, the place node is H. x The transition to T y Given that x and y are both positive integers, construct an SPN model for the air-ground coordinated emergency rescue process for forest fires; Step 3: Use PIPE simulation software to test the SPN model and obtain the reachability graph of the SPN model. The reachability graph contains the identifier M. i For all warehouses H x The set of token states contained therein, 0≤i≤n-1, where n is the total number of identifiers; Step 4: Based on the identifier M in the reachability graph i Construct an isomorphic Markov chain for the SPN model, and calculate each identifier M based on this Markov chain. i The steady-state probability; Step 5: Calculate H for each storage location node x Busy probability and various transitions T y The utilization rate is then calculated, and the total runtime t of the SPN model is determined. The transitions with higher utilization rates, T, are then selected. y As a key factor restricting the efficiency of rescue efforts; Step 6: By changing the transition T of key links y The implementation rate is determined, and its relationship with the total running time t of the SPN model is fitted to determine the transition T of key links. y The optimal implementation rate provides a relevant entry point for decision-makers to further improve the efficiency of emergency rescue.

2. The method for modeling and timeliness analysis of a forest fire air-ground coordinated emergency rescue process according to claim 1, characterized in that, In step 4, identify M. i The steady-state probability matrix equation is as follows: Where A = [P(M0), P(M1), ..., P(M... i ),…,P(M n-1 Let P(M0), P(M1), ..., P(M2) represent the set of all identifiers of steady-state probabilities. i ),…,P(M n-1 ) represents the steady-state probability of the corresponding rescue stage; Let ξ be the transition excitation rate matrix. ij For M i To M j The transfer rate, M i M represents the i-th identifier. j Let i represent the j-th identifier, 0 ≤ j ≤ n-1; when i ≠ j, and M in the Markov chain... i To M j When there is a directed arc, ξ ij Equal to the transition excitation rate λ on the arc y Transition excitation rate λ y For the change T y The average implementation rate (d / time) was determined by analyzing data from previous forest fire cases to identify various changes in T. y The average implementation rate; if there is no directed arc between them, then ξ ij The value is 0; when i = j, the value corresponding to the diagonal of the Q matrix is ​​0. The value is M i With M j The excitation rate λ of strain transfer on the directed arc between them y The negative of the sum.

3. The method for modeling and timeliness analysis of a forest fire air-ground coordinated emergency rescue process according to claim 2, characterized in that, In step 5, the H of each storage node is calculated according to formula (2). x The probability of being busy is given by formula (2) as follows: Among them, P[M(H x [) = 1] represents different identifiers of the internal storage node H x The probability of being busy in an excited state, H x Being in the activated state means having one token. Indicates the identifier M j Contains the activated library node H x The cumulative probability value.

4. The method for modeling and timeliness analysis of a forest fire air-ground coordinated emergency rescue process according to claim 3, characterized in that, In step 5, each transition T is calculated according to formula (3). y The utilization rate is given by formula (3) as follows: Among them, U(T) y ) is able to trigger transition T y All identifiers M i The sum of the steady-state probabilities is the transition T. y Utilization rate; E is for allowing changes T y All implementable identifiers M i A set of.

5. The modeling and timeliness analysis method for a forest fire air-ground coordinated emergency rescue process according to claim 4, characterized in that, In step 5, the total running time t of the SPN model is calculated according to formula (4), which is as follows: in, This represents the average number of tags in the library under stable system conditions. ∑U(T y For all changes T y The sum of utilization rates, where R is the marked flow rate flowing into the system; R(T) y H x )=W(T y H x )U(T1)λ1, where W(T y H x ) represents the transition T y Pointing to the storage node H x The directed arc weights are given by U(T1), where U(T1) is the utilization rate of transition T1, and λ1 is the average implementation rate of transition T1.

6. The method for modeling and timeliness analysis of a forest fire air-ground coordinated emergency rescue process according to claim 1, characterized in that, In step 1, following the logical framework and emergency response approach of "fire monitoring → information transmission → situation assessment → rescue deployment → extinguishing open flames → clearing remaining fires", and combining the operational mechanism of aviation rescue in forest fires, a "location-change-arc" flowchart for the air-ground coordinated emergency rescue process for forest fires is constructed, with the rescue stage as the storage node, the tasks completed in the rescue stage as the change, and the arc with arrows representing the sequence of implementation between each rescue stage.

Images