An emergency evacuation path planning method and device, a terminal and a storage medium

Abstract

This invention provides an emergency evacuation route planning method, device, terminal, and storage medium. The method includes: constructing an undirected graph of an emergency evacuation network based on the evacuation route; acquiring risk factors that affect the risk level of the evacuation route, and classifying each risk factor into a risk level based on the potential harm to evacuees; wherein different risk levels are assigned corresponding risk indices; constructing a single-objective route planning model (Model I) considering the risk level of the evacuation route based on the risk indices and the undirected graph of the emergency evacuation network, and constructing a single-objective route planning model (Model II) considering the length of the evacuation route based on the undirected graph of the emergency evacuation network. This invention solves the route planning problem with two optimization objectives—route risk and route length—by transforming the two established single-objective models into a single bi-objective route planning model.

Description

Technical Field

[0001] This invention belongs to the field of emergency evacuation planning technology, and in particular relates to an emergency evacuation route planning method, device, terminal and storage medium. Background Technology

[0002] The optimal path problem is essentially the shortest path algorithm problem based on road network topology. Traditional shortest path problems only consider the shortest path length, with weights in the road network representing static physical lengths. However, in practical applications, other factors affecting path quality must be considered. To explore the impact of fire byproducts on evacuation, many scholars have conducted research on quantifying risk factors such as fire smoke and temperature. Most of these studies integrate different calculation formulas to characterize risk. For example, Wang Yanfu cited Jin's research from Japan on quantifying the effect of temperature and Halim's research from the United States on quantifying the effect of carbon monoxide to characterize the risk level of evacuation routes. While this approach can calculate risk numerically, the inconsistency in the methods used to quantify various risk factors limits the accuracy of the results. Summary of the Invention

[0003] In view of this, the present invention aims to provide an emergency evacuation route planning method, device, terminal and storage medium to solve the problem of poor accuracy in emergency evacuation route planning.

[0004] To achieve the above objectives, the technical solution of the present invention is implemented as follows:

[0005] In a first aspect, the present invention provides an emergency evacuation route planning method, comprising:

[0006] An undirected graph of the emergency evacuation network is constructed based on the evacuation routes; wherein, the undirected graph of the emergency evacuation network includes the nodes within the evacuation routes and the edges between the nodes;

[0007] The risk factors that affect the risk level of the evacuation routes are identified, and each risk factor is classified into a risk level according to the potential harm to the evacuees; wherein, different risk levels are assigned corresponding risk indices.

[0008] Based on the risk index and the undirected graph of the emergency evacuation network, a single-objective path planning model Model I considering the risk level of the evacuation path is constructed, and a single-objective path planning model Model II considering the length of the evacuation path is constructed based on the undirected graph of the emergency evacuation network.

[0009] Based on the single-objective path planning model Model I and the single-objective path planning model Model II, a dual-objective path planning model Model III is constructed to find the Pareto optimal solution of the two models;

[0010] A dynamic adjustment algorithm DABD based on Dijkstra's algorithm is constructed to dynamically adjust the weights of the objective functions of path risk level and path length in the dual-objective path planning model ModelⅢ, so as to find the path with the minimum weight F for evacuees; wherein, the path with the minimum weight F is the optimal evacuation path.

[0011] The risk index corresponding to each risk factor within the evacuation route is obtained in real time. Based on the risk index corresponding to each risk factor within the evacuation route in real time, the weights of the two objective functions in the dual-objective path planning model ModelⅢ are adjusted using the dynamic adjustment algorithm DABD to obtain the optimal evacuation route.

[0012] Furthermore, the step of acquiring risk factors that affect the risk level of the evacuation route, and classifying each risk factor into a risk level based on the potential harm to evacuees, includes:

[0013] Five risk factors that affect the risk level of the evacuation routes were identified: CO concentration, HCN concentration, ambient temperature, visibility, and congestion.

[0014] Based on the potential harm of each risk factor to the escapees, each risk factor is divided into four risk levels: Level I, Level II, Level III, and Level IV.

[0015] Risk indices were set for the four risk levels.

[0016] Furthermore, the step of constructing a single-objective path planning model Model I considering the risk level of evacuation paths based on the risk index and the undirected graph of the emergency evacuation network, and constructing a single-objective path planning model Model II considering the length of evacuation paths based on the undirected graph of the emergency evacuation network, includes:

[0017] For each risk factor in the evacuation path, a risk factor weight correction factor is set, and based on the risk factor weight correction factor, the weighted average of the risk index corresponding to the risk factor in each edge of the undirected graph of the emergency evacuation network is calculated to obtain the risk level of each edge in the undirected graph of the emergency evacuation network; wherein, the sum of the risk factor weight correction factors corresponding to each risk factor is 1.

[0018] Based on the risk level of each edge in the undirected graph of the emergency evacuation network, a single-objective path planning model Model I considering the risk level of the evacuation path is constructed.

[0019] Based on the undirected graph of the emergency evacuation network, a single-objective path planning model, Model II, is constructed, which considers the length of the evacuation path.

[0020] Furthermore, the construction of a bi-objective path planning model Model III, based on the single-objective path planning model Model I and the single-objective path planning model Model II, for finding the Pareto optimal solution of the two models, includes:

[0021] A bi-objective path planning model, Model III, is constructed to find the Pareto optimal solution for the two models, as shown in the following formula:

[0022]

[0023] In the formula:

[0024] γ1 represents the weighting factor for risk level, γ2 represents the weighting factor for path length, and γ1 and γ2 represent the relative importance of risk level and path length in Model III, respectively, where γ1≥0, γ2≥0, and γ1+γ2=1; f1 represents the risk level in a certain evacuation route; f1 * f1 represents the risk level in the ideal evacuation route, and f2 represents the minimum risk level from the source point to the destination point, calculated by the single-objective path planning model Model I; f2 represents the actual length of a certain evacuation route; f2 * The ideal evacuation path length represents the shortest distance from the source point to the destination, calculated by the single-objective path planning model Model II.

[0025] Furthermore, the construction of the dynamic adjustment algorithm DABD based on Dijkstra's algorithm, to dynamically adjust the weights of the path risk level objective function and the path length objective function in the dual-objective path planning model Model III, to find the path with the minimum weight F for the escapees, includes:

[0026] When only considering the risk level objective problem, set γ2 = 0 and call algorithm b to obtain the optimal evacuation path;

[0027] When only considering the path length objective, set γ1 = 0 and call algorithm b to obtain the optimal evacuation path;

[0028] When considering a bi-objective problem involving risk level and path length, initial values ​​for γ1 and γ2 are set, and the optimal evacuation path is constructed by searching the weight vector space at the end of the call to algorithm b; where γ1≥0, γ2≥0, and γ1+γ2=1.

[0029] The goal of calling algorithm b is that when Dijkstra's algorithm is implemented, each subsequent node will update the weights of the network based on the found evacuation paths and output the evacuation path with the smallest weight F and its information.

[0030] Furthermore, after constructing the Dijkstra-based dynamic adjustment algorithm DABD to dynamically adjust the weights of the path risk level objective function and the path length objective function in the dual-objective path planning model Model III, and finding the path with the minimum weight F for the escapees, the method further includes:

[0031] Based on Dijkstra's algorithm, a virtual endpoint connecting all exits is set to change the original network structure of the undirected graph of the emergency evacuation network. The weights of the edges connecting each exit to the virtual endpoint are all set to zero, and the virtual endpoint is transformed into the source node, while each node is transformed into the endpoint.

[0032] Furthermore, after constructing the Dijkstra-based dynamic adjustment algorithm DABD to dynamically adjust the weights of the path risk level objective function and the path length objective function in the dual-objective path planning model Model III, and finding the path with the minimum weight F for the escapees, the method further includes:

[0033] Append a no-entry threshold for the risk factor to each edge of the undirected graph of the emergency evacuation network;

[0034] When using the DABD algorithm to search for the path with the smallest weight F, remove the edges where a certain risk factor exceeds the forbidden threshold, and use the remaining network structure to search for the optimal evacuation path.

[0035] When a node has no feasible path, the DABD algorithm is used to sequentially remove the prohibition thresholds of each risk factor attached to the undirected graph of the emergency evacuation network according to the probability of survival.

[0036] Secondly, the present invention also provides an emergency evacuation route planning device, comprising:

[0037] A module is established to construct an undirected graph of the emergency evacuation network based on the evacuation routes; wherein the undirected graph of the emergency evacuation network includes nodes within the evacuation routes and edges between nodes;

[0038] The acquisition module is used to acquire risk factors that affect the risk level of the evacuation route, and classify each risk factor into a risk level according to the potential harm of each risk factor to the evacuees; wherein, different risk levels are set with corresponding risk indices.

[0039] The planning module is used to construct a single-objective path planning model Model I that considers the risk level of the evacuation path based on the risk index and the undirected graph of the emergency evacuation network, and to construct a single-objective path planning model Model II that considers the length of the evacuation path based on the undirected graph of the emergency evacuation network.

[0040] The construction module is used to construct a dual-objective path planning model Model III for finding the Pareto optimal solution of the two models, based on the single-objective path planning model Model I and the single-objective path planning model Model II.

[0041] The adjustment module is used to construct a dynamic adjustment algorithm DABD based on Dijkstra's algorithm, so as to realize the dynamic adjustment of the weights of the objective function of path risk level and objective function of path length in the dual-objective path planning model ModelⅢ, and find the path with the minimum weight F for evacuees; wherein, the path with the minimum weight F is the optimal evacuation path that takes into account both safety and efficiency.

[0042] The processing module is used to obtain the real-time risk index corresponding to each risk factor within the evacuation route, and adjust the weights of the two objective functions in the dual-objective path planning model ModelⅢ using the dynamic adjustment algorithm DABD based on the real-time risk index corresponding to each risk factor within the evacuation route, so as to obtain the optimal evacuation route.

[0043] Thirdly, the present invention also provides a terminal, comprising:

[0044] One or more processors;

[0045] Storage device for storing one or more programs;

[0046] A camera is used to capture images;

[0047] Sensors are used to collect environmental parameters;

[0048] A display is used to indicate evacuation routes; the display includes a main body and a group of guide cursors arranged around the main body. Each group of guide cursors includes a no-entry guide cursor arranged on the main body and a compliant guide cursor arranged around the no-entry guide cursor. The compliant guide cursors are set to correspond one-to-one with the evacuation routes.

[0049] When the one or more programs are executed by the one or more processors, the one or more processors implement the emergency evacuation route planning method provided in the above embodiments.

[0050] Fourthly, the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the emergency evacuation route planning method provided in the above embodiments.

[0051] Compared with existing technologies, the emergency evacuation route planning method, device, terminal, and storage medium described in this invention have the following advantages:

[0052] This invention discloses an emergency evacuation route planning method, device, terminal, and storage medium. First, it establishes two single-objective route planning models considering both route risk and route length, identifying five risk factors: CO concentration, HCN concentration, ambient temperature, visibility, and congestion. Then, it employs a semi-quantitative calculation method to characterize the risk values ​​of these five factors using a risk index normalization method. Finally, it transforms the two single-objective models into a single bi-objective route planning model, solving the route planning problem with both route risk and route length optimization objectives.

[0053] The present invention discloses an emergency evacuation route planning method, device, terminal, and storage medium. Based on the traditional Dijkstra algorithm, the present invention proposes a dynamic adjustment algorithm DABD. By introducing a virtual endpoint, the efficiency of the traditional Dijkstra algorithm in solving multi-exit problems is optimized. Furthermore, mechanisms such as dynamic adjustment and no-entry thresholds are added, enabling the DABD algorithm to serve as an optimization algorithm for a dual-objective path planning model, finding the optimal evacuation route that balances safety and efficiency for evacuees. Attached Figure Description

[0054] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an undue limitation of the invention. In the drawings:

[0055] Figure 1 Here is a flowchart of an emergency evacuation route planning method provided in Embodiment 1 of the present invention;

[0056] Figure 2 The flowchart of the DABD algorithm provided in Embodiment 1 of the present invention;

[0057] Figure 3 This is a flowchart of an emergency evacuation route planning method provided in Embodiment 2 of the present invention;

[0058] Figure 4 This is a schematic diagram of the structure of an emergency evacuation route planning device provided in Embodiment 3 of the present invention;

[0059] Figure 5This is a schematic diagram of the structure of a terminal provided in Embodiment 4 of the present invention;

[0060] Figure 6 This is a schematic diagram of the structure of a display provided in Embodiment 4 of the present invention. Detailed Implementation

[0061] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0062] Example 1

[0063] Figure 1 This is a flowchart of an emergency evacuation route planning method provided in Embodiment 1 of the present invention. This embodiment is applicable to the emergency evacuation of people escaping from a fire-stricken building. This method can plan the optimal evacuation route for evacuees, improving their survival probability. The method specifically includes the following steps:

[0064] Step 101: Construct an undirected graph of the emergency evacuation network based on the evacuation routes; wherein the undirected graph of the emergency evacuation network includes nodes within the evacuation routes and edges between nodes.

[0065] Because the optimal evacuation route planning for sudden fires is subject to various influencing factors, such as smoke spread, route length, and stampede, simply measuring the merits of an evacuation route by its actual length cannot guarantee the safety of evacuees in practice. The risk level within the evacuation route is more important than its length. Furthermore, the optimal evacuation route should not be fixed but dynamically adjustable based on the development of the fire. This embodiment designs a route planning model for building fires based on risk factors with significant impact. In the graph network structure, the edge weights are set as a time-varying function of route length and fire risk. The path with the minimum time-varying weight at a given moment is the optimal evacuation route, ensuring both the safety of evacuees and evacuation efficiency. This eliminates the limitations of the traditional shortest path problem and better represents emergency evacuation networks with time-varying and stochastic characteristics.

[0066] During an escape, path selection is a multi-objective optimization problem that requires weighing many factors. From a graph theory perspective, the intricate network of passageways within a building forms a connected network graph, where passageways are considered edges between nodes in the network, and intersections between passageways are considered nodes in the network.

[0067] For example, the emergency evacuation network is defined as an undirected graph G(V,E), where V is the set of nodes and E is the set of edges; v1, v2, ... v n For nodes in the network, i.e., V = {v1, v2, ... v} n};(v i ,v j ) represents node v i With node v j The edge between; v s v represents the source point (i.e., the starting point) of the selected evacuation route. n Represents the destination (i.e., the exit).

[0068] Step 102: Obtain the risk factors that affect the risk level of the evacuation route, and classify each risk factor into a risk level according to the potential harm to the evacuees; wherein, different risk levels are assigned corresponding risk indices.

[0069] Smoke, high temperatures, low visibility, and overcrowding in building fires pose significant threats to evacuees. Any fire-related risk that harms the physical and psychological well-being of evacuees increases the likelihood of death. Asphyxiating gases such as CO and HCN, along with high temperatures, are major causes of death during and after a fire. Smoke produced during a fire spreads along the evacuation route, reducing visibility and delaying escape. Furthermore, overcrowding increases the time evacuees spend at the fire scene and significantly increases the probability of stampedes. An experiment using Social Force Networks (SNA) to analyze 78 typical stampede incidents from 2010 to 2019 showed that overcrowding is one of the leading causes of stampedes. Therefore, this embodiment considers five risk factors that evacuees may encounter along evacuation routes: CO concentration, HCN concentration, ambient temperature, visibility, and crowding level, to characterize the risk level along the route.

[0070] In practical application, this embodiment selects AEGLs standard values ​​as the risk assessment index for gases in evacuation routes. AEGLs represent the exposure limit thresholds for the general public, applicable to emergency exposure times from 10 minutes to 8 hours. AEGL levels are distinguished by different degrees of toxicity, and the toxicity exposure levels are applicable to the general population, including infants, children, and other potentially sensitive or susceptible individuals. The three-level standard classification of AEGLs—AEGL-1, AEGL-2, and AEGL-3—can serve as the basis for determining the risk level of hazardous gases. Given the urgency of evacuation in fire accidents, AEGLs standard values ​​at exposure times of 10 minutes and 30 minutes are selected to determine the risk level of CO concentration.

[0071] When the concentration of hazardous gas is below the AEGL-2 standard value for an exposure time of 30 minutes, it will not cause the escapee to lose mobility or cause serious long-term effects; therefore, the concentration of hazardous gas within this range is relatively safe. When the concentration of hazardous gas is between the AEGL-2 standard values ​​for an exposure time of 10 minutes and 30 minutes, passage within 30 minutes is very likely to cause the escapee to lose mobility; at this point, the concentration of hazardous gas is in a relatively dangerous state. When the concentration of hazardous gas is between the AEGL-2 standard value for an exposure time of 10 minutes and the AEGL-3 standard value for 30 minutes, the escapee will lose mobility within ten minutes and face life-threatening danger after about half an hour; therefore, the concentration of hazardous gas at this point is very dangerous. When the concentration of hazardous gas is higher than the AEGL-3 standard value for an exposure time of 30 minutes, the escapee will quickly lose mobility and face the threat of death within half an hour; at this point, the concentration of hazardous gas is unacceptable.

[0072] Specifically, the risk index of CO concentration is determined by the AEGLs standard values ​​for exposure times of 10 min and 30 min, thus deriving the CO concentration risk index. The values ​​are shown in Table 1.1:

[0073] Table 1.1 Risk index values ​​for CO

[0074]

[0075] The risk index for HCN concentration is also determined by the AEGLs standard values ​​for exposure times of 10 min and 30 min, converting the HCN concentration level range into its risk index. As shown in Table 1.2:

[0076] Table 1.2 Risk index values ​​for HCN

[0077]

[0078] The normal human body temperature ranges from 36.3℃ to 37.2℃. When the ambient temperature exceeds 40℃, the human body can experience significant discomfort, and some people cannot even inhale air above 65℃. The maximum tolerance time of the human body in high-temperature environments is limited. In this embodiment, the risk index of ambient temperature is determined based on the maximum tolerance time of the human body at different temperatures. According to the relationship between the maximum tolerance time τ(T) and temperature: τ(T) = 1812e -0.046T(See Gao Rui, Jiang Zhong'an, Dong Feng, et al. Mathematical Model and Algorithm for Dynamic Optimal Rescue Route in Mine Fire Based on MapObject [J]. Journal of Beijing University of Science and Technology, 2008, (07): 705-709+755). Data from different time periods shows that the higher the temperature, the shorter the maximum tolerance time for the human body. At 120℃, the maximum tolerance time is only 7 minutes. Therefore, 120℃ is defined as the highest risk level, meaning the passage is potentially fatal. When the ambient temperature is below 40℃, the human body will not feel significant discomfort, so the risk level at this time can be considered the lowest, meaning the passage is safe. Therefore, based on the maximum tolerance time data of the human body at different temperatures, different ranges of ambient temperature are converted into their risk indices. As shown in Table 1.3:

[0079] Table 1.3 Risk Index Values ​​for Ambient Temperature

[0080] Table 1.3Risk index values ​​for environment temperature

[0081]

[0082] The risk index for visibility distance (i.e., visibility) is calculated based on data from a simulated evacuation experiment conducted abroad (Casey C. Grant PE, John R. Halljr. PD, Robert E. Solomon PE. NFPA-Fire Protection Handbook-2008-20th [M]. National Fire Protection Association, 2008.). The experimental results show that when the visibility distance in the passage is less than 0.6 meters, 100% of the participants choose to change direction. Therefore, in this embodiment, a visibility distance of 0.6 meters is defined as the highest risk level, indicating a very dangerous passage. When the visibility distance in the passage is greater than 3.7 meters, only 9% of the participants choose to change direction. Therefore, a visibility distance of 3.7 meters can be considered an acceptable risk, indicating a relatively safe passage. Thus, based on the proportion of evacuees choosing to change direction at different visibility distances, different ranges of visibility distance are converted into their risk indices. As shown in Table 1.4:

[0083] Table 1.4 Risk Index Values ​​for Visibility Distance

[0084] Table 1.4Table of risk index values ​​for visibility

[0085]

[0086] Crowding is a decisive factor in the walking speed of evacuees and has a significant impact on evacuation time. A crowd density of 4 people / m² is ideal. 2 When people walk together, they will come into contact and stop (see Ibrahim AM, Venkat I, Subramanian KG, et al. Intelligent Evacuation Management Systems[J]. ACM Transactions on Intelligent Systems and Technology, 2016, 7(3): 1-27.). Therefore, this embodiment will have 4 people / m 2 The highest risk level is defined as a crowd density of 1.8 people / m², indicating an extremely dangerous passage. Entering the passage in this state could cause a stampede, resulting in widespread casualties. 2 In the following situations, survivors can move freely and will not come into contact, therefore the density is 1.8 people / m². 2 Defined as acceptable risk, i.e., the lowest risk level and a relatively safe passage condition; the risk index of congestion is determined based on the impact of crowd density on crowd movement speed. Based on the risk level assessment, the range of crowd density is converted into a risk index of congestion. As shown in Table 1.5:

[0087] Table 1.5 Risk Index Values ​​for Crowding

[0088] Table 3.10Table of risk index values ​​for congestion

[0089]

[0090] In practical applications, the risk index can be defined as the value converted from the real-time changes of risk factors in the path, and the risk level in a certain path is characterized by the risk index of five risk factors. The specific steps are as follows:

[0091] Step 1021: Obtain five risk factors that affect the risk level of the evacuation route, namely CO concentration, HCN concentration, ambient temperature, visibility distance, and congestion.

[0092] Step 1022: Based on the potential harm of each risk factor to the escaped personnel, each risk factor is divided into four risk levels: Level I, Level II, Level III, and Level IV.

[0093] Step 1023: Set the risk indices for the four risk levels to 0, 0.5, 0.7 and 1 respectively.

[0094] Step 103: Construct a single-objective path planning model Model I that considers the risk level of evacuation paths based on the risk index and the undirected graph of the emergency evacuation network, and construct a single-objective path planning model Model II that considers the length of evacuation paths based on the undirected graph of the emergency evacuation network.

[0095] This embodiment sets two overall optimization objectives: path risk level and path length. The optimal evacuation path is the path with the lowest possible risk level and the shortest possible evacuation distance. The dual-objective optimization can be expressed as follows:

[0096] F1 = minf1;

[0097] F2 = minf2;

[0098] In the formula:

[0099] f1 — the risk level in a certain evacuation route;

[0100] F1 – The lowest risk level among all evacuation routes;

[0101] f2 — The actual length of a certain evacuation route;

[0102] F2 – The shortest actual length of all evacuation routes.

[0103] For example, step 103 includes the following specific steps:

[0104] Step 1031: Set risk factor weight correction factors for each risk factor in the evacuation path, and calculate the weighted average of the risk indices corresponding to the risk factors in each edge of the undirected graph of the emergency evacuation network according to the risk factor weight correction factors, so as to obtain the risk level of each edge in the undirected graph of the emergency evacuation network; wherein, the sum of the risk factor weight correction factors corresponding to each risk factor is 1.

[0105] Since asphyxiating gases such as CO and HCN, along with high temperatures, are the leading causes of death during and after fires, asphyxiating gases, especially CO, are the most lethal risk factor. Although inhaling HCN can lead to rapid incapacitation, its effects are often limited, for example, by the amount of nitrogen-containing materials in the building. Furthermore, all fires that produce cyanide also produce CO, and combustion conditions that produce high HCN also produce high CO. Therefore, compared to HCN, CO is often the most lethal risk factor in building fires. This paper assigns a higher weight to asphyxiating gases, especially CO, with high temperature having the second-highest weight after CO and HCN. The risks of visibility and crowding primarily affect the speed of escape and do not directly harm escapees; therefore, their weights are relatively lower. Considering the characteristics of the building in the case study, the risk factor weighting correction factors for CO concentration, HCN concentration, ambient temperature, visibility, and crowding can be set as α1 = 0.4, α2 = 0.3, α3 = 0.2, α4 = 0.05, and α5 = 0.05, respectively.

[0106] Step 1032: Based on the risk level of each edge in the undirected graph of the emergency evacuation network, construct a single-objective path planning model Model I that considers the risk level of the evacuation path.

[0107] In the event of a fire, emergency evacuation routes with lower risk should be prioritized to minimize the possibility of injury to people during evacuation. Therefore, minimizing the risk level is one of the main optimization objectives in the model. The single-objective path planning model (Model I) considering fire risk is expressed as follows:

[0108] Model I:

[0109]

[0110] St

[0111]

[0112] x ij =0,1,i=1,2,…,n,j=1,2,…,n;

[0113] In the formula:

[0114] s represents the node number of the evacuation path source point;

[0115] n represents the node number of the end point of the evacuation route;

[0116] x ij Denotes the decision variables in the model, when edge (v i ,v j When x is included in the evacuation route, ij=1, when edge (v i ,v j When x is not included in the evacuation route, ij =0.

[0117] F1 is the overall objective of Model I: minimizing the risk level along the road. The constraint St is determined by limiting x. ij The value of v is used to guarantee the source point v s To the destination v n The feasible paths are determined; considering the rationality of the evacuation plan and the urgency of the evacuation time, constraints are used to ensure that there are no loop paths.

[0118] Step 1033: Construct a single-objective path planning model Model II that considers the evacuation path length based on the undirected graph of the emergency evacuation network.

[0119] Besides considering the risk level of emergency evacuation routes, the length of these routes is also a key factor to consider. In some cases, even if the chosen evacuation route has a lower risk, it may require a lot of detours, and congestion and stampedes often occur in areas with complex road conditions. On the other hand, the longer evacuees spend on the road, the greater their chance of being harmed by the fire. Therefore, minimizing the length of the evacuation route is another major optimization objective of the model.

[0120] Therefore, based on the single-objective model that considers fire risk proposed above, this embodiment also establishes a single-objective path planning model that considers path length, with the goal of minimizing the length of evacuation paths.

[0121] The expression function of the single-objective path planning model (Model II) considering path length is as follows: Model II:

[0122]

[0123] St

[0124]

[0125]

[0126] x ij =0,1,i=1,2,…,n,j=1,2,…,n;

[0127] In the formula:

[0128] e ij Represents an edge (v) i ,v j The actual path length.

[0129] Step 104: Based on the single-objective path planning model Model I and the single-objective path planning model Model II, construct a dual-objective path planning model Model III for finding the Pareto optimal solution of the two models.

[0130] Since Model I and Model II can plan the evacuation routes with the lowest risk level and the shortest length, respectively, in order to obtain the comprehensive optimal evacuation route considering both fire risk and route length, this embodiment also establishes a dual-objective route planning model based on Model I and Model II to find the Pareto optimal solution of the two objective functions.

[0131] For example, the expression function of the bi-objective path planning model (Model III) considering path risk and length is as follows:

[0132] Model III:

[0133]

[0134] St

[0135]

[0136]

[0137]

[0138] x ij =0,1,i=1,2,…,n,j=1,2,…,n.

[0139] In the formula:

[0140] γ1 and γ2 are weighting correction factors for fire risk and path length, respectively, representing the relative importance of fire risk and path length in Model III. γ1≥0, γ2≥0, γ1+γ2=1. Changes in γ1 and γ2 will result in corresponding increases or decreases in f1 and f2.

[0141] —The risk level in an ideal evacuation route represents the risk level at the source point v. s To the destination v n The minimum risk level is calculated using Model I.

[0142] —The ideal evacuation path length represents the length of the source point v. s To the destination v n The shortest distance is calculated using Model II.

[0143] Step 105: Construct a dynamic adjustment algorithm DABD based on Dijkstra's algorithm to dynamically adjust the weights of the objective functions of path risk level and path length in the dual-objective path planning model ModelⅢ, so as to find the path with the minimum weight F for evacuees; wherein, the path with the minimum weight F is the optimal evacuation path that takes into account both safety and efficiency.

[0144] Because bi-objective optimization problems involve multiple objective functions and constraints, when one objective reaches its optimum, the other may deteriorate. Given the mutual constraints and influences among the objective functions, it is difficult to obtain optimal solutions for all objectives, and a specific Pareto optimal solution may not meet practical needs. In such cases, multiple Pareto solutions can be sought based on the optimization algorithm, and the results can be dynamically adjusted as the fire progresses.

[0145] Dijkstra's algorithm is a widely recognized and excellent algorithm for solving the single-source shortest path problem. Proposed by Dutch computer scientist EWDijkstra in 1959, it is also known as Dijkstra's algorithm. Dijkstra's algorithm uses a breadth-first search-like approach to determine the path with the minimum weight from a given point to other nodes in a weighted graph. It's important to note that Dijkstra's algorithm cannot handle graphs with negative weight edges. As a general algorithm for solving the shortest path problem, Dijkstra's algorithm has wide applications in many fields. By modifying it according to the characteristics and constraints of the actual network topology, it can adapt to computational conditions under different circumstances.

[0146] Figure 2 The flowchart of the DABD algorithm provided in Embodiment 1 of the present invention is shown below. Figure 2 The specific process of the DABD algorithm can be summarized in the following steps:

[0147] Step 1051: Initialize the data, introduce constants Q=0, M=1, and the number of algorithm iterations i=1; initially, the fire risk weight correction factor γ1 is set to 0.7, and the path length weight correction factor γ2 is set to 0.3.

[0148] Step 1052: Determine if the algorithm has not reached the maximum number of iterations N. max If yes, proceed to step 1053; otherwise, output the path directly.

[0149] Step 1053: Use algorithm b to perform path search and obtain p(v) j ), B j f1(δ) and f2(δ), proceed to step 1054.

[0150] Step 1054: Determine whether the risk level of the optimal evacuation route is lower than r. *If yes, output the path directly; otherwise, proceed to step 1055.

[0151] Step 1055: Increase the value of the fire risk weight factor using constants Q and M, i = i + 1, and proceed to step 1056.

[0152] Step 1056: Call algorithm b to determine if there is a feasible path from the node to the safe exit. If there is, proceed to step 1052; otherwise, proceed to step 1057.

[0153] Step 1057, Adjust r * Let r be the value of r. * =r * +0.05, γ1=0.7, γ2=1-γ1, proceed to step 1052.

[0154] The implementation steps of the above-mentioned algorithm b are as follows:

[0155] (1) First, introduce two sets (S, U). S contains nodes whose minimum weight has been found (initially an empty set), and U contains nodes whose minimum weight has not been found. V is the set of all nodes. Introduce an auxiliary array D, where each element D(i) represents the currently found node v from the source node. s to each other node v i The weights, relative to the source point v s Non-adjacent nodes have a weight of ∞. Let S = v s ,enter γ1 and γ2;

[0156] (2) If S = V, then the algorithm stops operating and outputs p(v). j ), B j f1(δ) and f2(δ), otherwise proceed to step (3);

[0157] (3) Calculate and save the weights (F, f1, f2) of adjacent nodes, and select the node v with the smallest weight F from U. k Such that D(k) = Min{D(k)|v k ∈US}, S=S∪v k U = US, proceed to step (4);

[0158] (4) Traverse v k Find the neighboring node v with the smallest weight F among the neighboring nodes. k+1 Update all nodes in U to the source node v s If there is a node with a weight less than the current path, the optimal path is updated; otherwise, it is not updated and proceeds to step (5).

[0159] (5) Repeat steps (2), (3), and (4) until all nodes have been traversed. Find the path with the minimum weight from the source to the destination and output p(v). j ), B j , f1(δ) and f2(δ).

[0160] For example, the main body of the DABD algorithm is as follows: Figure 2 As shown, it is mainly responsible for dynamically adjusting the weights of the fire risk objective function and the path length objective function in Model III, and finding multiple Pareto optimal solutions as the fire develops. Algorithm b is the main part of the calling algorithm. When only a single objective problem is considered, γ1 = 0 or γ2 = 0 is set, and only algorithm b needs to be called; if a dual objective problem is considered, the optimal evacuation path will be reached at the end of algorithm b. Figure 2 The search space of the weight vectors in the algorithm is used to construct the algorithm. Initially, γ1 is set to 0.7 and γ2 is set to 0.3, indicating a higher degree of emphasis on fire risk. The basic idea behind algorithm b is that when implementing Dijkstra's algorithm, each subsequent node updates the network weights based on the found path and outputs the evacuation path with the smallest weight F and its information. Here, the information refers to the p(v) output by algorithm b. j ), B j , f1(δ) and f2(δ).

[0161] Specifically, let p(v) j B is the minimum value of weight F among all evacuation routes. j Represents F = p(v) j The evacuation path; δ represents an optimal point, and f1(δ) and f2(δ) represent F = p(v) respectively. j The values ​​of f1 and f2 in the path are given; K represents the number of edges in the selected path; i is the record of the algorithm iteration count, which increases by 1 for each iteration; N max This represents the maximum number of iterations of the algorithm; if the number of iterations exceeds N... max It will directly output the optimal evacuation path.

[0162] Where, r * This represents a pre-set standard, signifying the upper limit of the ideal risk level; exceeding this level could result in a potential threat from fire. In this embodiment, r... * The initial value is set to 0.7, corresponding to the third level of risk. Specifically, the mechanism works as follows: the evacuation path with the smallest weight F (the optimal evacuation path) found by algorithm b will be assessed to determine whether its overall risk level exceeds r. *The risk factor f1(δ) is the ratio of the sum of risks of all segments along the path to the number of edges (segments) in the path. If the value is less than or equal to 0.7, it will be output as the optimal path. If it exceeds 0.7, the algorithm will use constants Q and M to increase the value of the fire risk weight factor to plan the optimal evacuation path. However, as the fire progresses, each path may reach a high risk level, and in extreme cases, there may be no usable path in the network graph. To ensure that there is a feasible path from the node to the safe exit, r... * The value can be increased in units of 0.05.

[0163] Step 106: Obtain the real-time risk index corresponding to each risk factor within the evacuation route, and adjust the weights of the two objective functions in the dual-objective path planning model ModelⅢ using the dynamic adjustment algorithm DABD based on the real-time risk index corresponding to each risk factor within the evacuation route, so as to obtain the optimal evacuation route.

[0164] First, by obtaining the real-time risk index corresponding to each risk factor within the evacuation route, and then based on the real-time risk index corresponding to each risk factor within the evacuation route, the subsequent algorithm can obtain the optimal evacuation route based on these real-time risk indices when planning the route, thereby increasing the survival probability of evacuees.

[0165] This embodiment first establishes two single-objective path planning models considering both path risk and path length, identifying five risk factors: CO concentration, HCN concentration, ambient temperature, visibility, and congestion. Then, a semi-quantitative calculation method is used to characterize the risk values ​​of these five factors through risk index normalization. The risk index for hazardous gases is determined based on AGGLs standard values, the temperature risk index is determined based on the maximum tolerance time of the human body at high temperatures, and the risk indices for visibility and congestion are determined based on experimental data from existing literature. Finally, the two single-objective models are transformed into a single bi-objective path planning model to solve the path planning problem with optimization objectives for both path risk and path length.

[0166] Example 2

[0167] Figure 3This is a flowchart of an emergency evacuation route planning method provided in Embodiment 2 of the present invention. This embodiment is based on the above embodiment and optimized. In this embodiment, after constructing the dynamic adjustment algorithm DABD based on Dijkstra's algorithm to dynamically adjust the weights of the objective functions of path risk level and path length in the dual-objective path planning model Model III, and finding the path with the minimum weight F for evacuees, the following steps are added: Based on Dijkstra's algorithm, a virtual endpoint connecting all exits is set to change the original network structure of the undirected graph of the emergency evacuation network. The weights of the edges connecting each exit to the virtual endpoint are set to zero, and the virtual endpoint is transformed into a source node, and each node is transformed into an endpoint. A prohibition threshold for risk factors is attached to each edge in the undirected graph of the emergency evacuation network. When searching for the path with the minimum weight F using the DABD algorithm, edges where a certain risk factor exceeds the prohibition threshold are removed, and the optimal evacuation path is searched using the remaining network structure. When a node has no feasible path, the prohibition thresholds for each risk factor attached to the undirected graph of the emergency evacuation network are removed sequentially according to the survival probability using the DABD algorithm.

[0168] Accordingly, the emergency evacuation route planning method provided in this embodiment specifically includes:

[0169] Step 201: Construct an undirected graph of the emergency evacuation network based on the evacuation routes; wherein the undirected graph of the emergency evacuation network includes nodes within the evacuation routes and edges between nodes.

[0170] Step 202: Obtain the risk factors that affect the risk level of the evacuation route, and classify each risk factor into a risk level according to the potential harm of each risk factor to the evacuees; wherein, different risk levels are set with corresponding risk indices.

[0171] Step 203: Construct a single-objective path planning model Model I that considers the risk level of the evacuation path based on the risk index and the undirected graph of the emergency evacuation network, and construct a single-objective path planning model Model II that considers the length of the evacuation path based on the undirected graph of the emergency evacuation network.

[0172] Step 204: Based on the single-objective path planning model Model I and the single-objective path planning model Model II, construct a dual-objective path planning model Model III for finding the Pareto optimal solution of the two models.

[0173] Step 205: Construct a dynamic adjustment algorithm DABD based on Dijkstra's algorithm to dynamically adjust the weights of the objective function of path risk level and the objective function of path length in the dual-objective path planning model ModelⅢ, so as to find the path with the minimum weight F for evacuees; wherein, the path with the minimum weight F is the optimal evacuation path that takes into account both safety and efficiency.

[0174] Step 206: Based on Dijkstra's algorithm, set a virtual endpoint connecting all exits to change the original network structure of the undirected graph of the emergency evacuation network. Set the weight of the edge connecting each exit to the virtual endpoint to zero, and transform the virtual endpoint into the source node and each node into the endpoint.

[0175] Since most modern buildings have multiple exits, to ensure the safety of evacuees and the efficiency of evacuation, people in different locations need to follow corresponding evacuation routes to different safe exits. Fire evacuation route planning in buildings with multiple safe exits can be abstracted as a multi-source, multi-sink problem, where people at different nodes in the emergency evacuation network need to safely and quickly reach different safe exits within a certain time. However, the traditional Dijkstra's algorithm, as a single-source shortest path algorithm, cannot be directly used to solve multi-source, multi-sink problems.

[0176] For example, suppose a network has m nodes (excluding exit nodes) and n exit nodes, with a total number of nodes of m+n. The following scheme can be used to solve the multi-source multi-sink problem using the traditional Dijkstra algorithm:

[0177] (1) Scheme 1: Using the exit as the destination and other nodes in the network as the source. This requires repeatedly calling Dijkstra's algorithm m times. Each time Dijkstra's algorithm is executed with a certain node as the source, the shortest path from that node to all other nodes is found. Finally, the shortest paths from m nodes to the exit node are obtained. The time complexity of this scheme is O(m(m+n)). 2 ).

[0178] (2) Scheme 2: Using the exit node as the source and other nodes in the network as destinations. This requires repeatedly calling Dijkstra's algorithm n times. Each time Dijkstra's algorithm is called with a specific exit node as the source, m shortest paths from that exit node to the destination can be found. Similarly, the shortest paths from other source nodes can be calculated, ultimately obtaining n shortest paths from each source node to any destination in the set of destinations. The time complexity of this scheme is O(n(m+n)). 2 ).

[0179] Both of the above schemes suffer from inefficiency due to the need for multiple calls to Dijkstra's algorithm, resulting in repeated node expansion. To improve computational efficiency, the DABD algorithm modifies the original network structure by introducing a virtual endpoint connecting all exits to Dijkstra's algorithm. The weights of the edges connecting each exit to the virtual endpoint are set to zero, and the virtual endpoint is transformed into a source node, while each node becomes an endpoint. This allows the calculation of the optimal path from the virtual endpoint to each node to be solved with only one call to Dijkstra's algorithm. This transforms the original multi-source, multi-sink problem into a single-source shortest path problem, significantly improving the solution speed to O(m+n). 2 The actual destination is transformed into a node in the path, which does not affect the final result.

[0180] Specifically, after adding a virtual endpoint to the undirected graph of the emergency evacuation network, the weight of the edge between the exit and the virtual endpoint can be set to 0. Dijkstra's algorithm is then executed with the virtual endpoint as the source to plan the optimal path for the modified emergency evacuation network. In the optimal path planning problem of this embodiment, the virtual endpoint is first placed into the set S of nodes whose minimum weight has been determined (S is initially an empty set), and other nodes are placed into the set U of nodes whose minimum weight has not been determined. At this point, the network expands outward layer by layer with the virtual endpoint as the source. Each time the minimum weight path of a node is determined, the algorithm adds that node to set U. Finally, the minimum weight path from all nodes to the virtual endpoint is obtained, which is the optimal evacuation path from all nodes to multiple safe exits.

[0181] Step 207: Attach the prohibition threshold of the risk factor to each edge in the undirected graph of the emergency evacuation network; when using the DABD algorithm to search for the path with the smallest weight F, remove the edge of a risk factor that exceeds the prohibition threshold, and use the remaining network structure to search for the optimal evacuation path; when a node has no feasible path, use the DABD algorithm to remove the prohibition thresholds of each risk factor attached to the undirected graph of the emergency evacuation network in order of the probability of survival.

[0182] Because there may be situations where a single risk index is too high but the overall risk level does not exceed the limit, the algorithm output may not be the optimal path. Therefore, special processing is needed for paths with excessively high single risk indices. In this embodiment, prohibition thresholds for five risk factors are attached to each edge in the network. When the algorithm searches for the path with the smallest weight F, if the risk factor of a path corresponding to a node reaches the set prohibition threshold, the algorithm will not consider that node, but will instead search for the path with the smallest weight F other than that node. The prohibition thresholds can be determined based on the research on risk level quantification in Chapter 3, setting a single risk index of 1 as the prohibition threshold. Therefore, the prohibition threshold for CO is set to 600 ppm, the prohibition threshold for HCN is set to 21 ppm, the prohibition threshold for ambient temperature is set to 120℃, the prohibition threshold for visibility distance is set to 0.6 m, and the prohibition threshold for congestion is set to 4 people / m. 2 .

[0183] In graph theory, the prohibition threshold mechanism involves removing edges that exceed the prohibition threshold for a certain risk factor and then using the remaining network structure to search for the optimal evacuation path. As the fire progresses, some nodes may become unusable. In this case, the prohibition threshold mechanism is removed, and the algorithm is used again to traverse the emergency evacuation network to search for the optimal evacuation path. When a node has no feasible paths, the algorithm removes the prohibition thresholds for each risk factor in the network in order of survival probability (prioritizing the restoration of paths with higher survival probability). The priority order for removing prohibition thresholds is: congestion > visibility > ambient temperature > HCN concentration > CO concentration. If, after restoring all paths for a risk factor that exceeds its prohibition threshold, the node still has no feasible paths, the algorithm will restore all paths for the next risk factor that exceeds its prohibition threshold in sequence.

[0184] The algorithm first recovers paths where congestion reaches the prohibited threshold, followed by paths where visibility reaches the prohibited threshold. This is because the risks of visibility and congestion primarily affect the speed of escape, not the direct harm to those escaping. If no feasible path remains at a node, paths where ambient temperature reaches the prohibited threshold will be recovered, followed by HCN concentration, and finally CO concentration. This is because asphyxiating gases such as CO and HCN, along with high temperatures, are the main causes of death during and after a fire. Among these, asphyxiating gases, especially CO, are the most lethal risk factor; at high CO concentrations, the chances of survival are extremely low.

[0185] Step 208: Obtain the real-time risk index corresponding to each risk factor within the evacuation route, and adjust the weights of the two objective functions in the dual-objective path planning model ModelⅢ using the dynamic adjustment algorithm DABD based on the real-time risk index corresponding to each risk factor within the evacuation route, so as to obtain the optimal evacuation route.

[0186] This embodiment proposes a dynamic adjustment algorithm based on Dijkstra's algorithm, DABD (Dynamic Adjustment Algorithm Based on Dijkstra's Algorithm). Building upon the traditional Dijkstra algorithm, it incorporates mechanisms such as dynamic adjustment, no-entry thresholds, and virtual endpoints. This algorithm adjusts the weights of the two objective functions in Model III based on fire development, improving its efficiency in solving multi-source, multi-sink shortest path problems and overcoming the limitation of Dijkstra's algorithm being unsuitable for computation in dynamic networks. DABD can be used as an optimization algorithm in Model III to find the path with the minimum weight F for evacuees, i.e., the optimal evacuation path that balances safety and efficiency.

[0187] Example 3

[0188] Figure 4 This is a schematic diagram of the structure of an emergency evacuation route planning device provided in Embodiment 3 of the present invention, as shown below. Figure 4 As shown, the device includes:

[0189] A module 301 is established to construct an undirected graph of the emergency evacuation network based on the evacuation routes; wherein the undirected graph of the emergency evacuation network includes nodes within the evacuation routes and edges between nodes.

[0190] The acquisition module 302 is used to acquire risk factors that affect the risk level of the evacuation route, and classify each risk factor into a risk level according to the potential harm of each risk factor to the evacuees; wherein, different risk levels are set with corresponding risk indices.

[0191] The planning module 303 is used to construct a single-objective path planning model Model I that considers the risk level of the evacuation path based on the risk index and the undirected graph of the emergency evacuation network, and to construct a single-objective path planning model Model II that considers the length of the evacuation path based on the undirected graph of the emergency evacuation network.

[0192] Module 304 is used to construct a bi-objective path planning model Model III for finding the Pareto optimal solution of the two models, based on the single-objective path planning model Model I and the single-objective path planning model Model II.

[0193] The adjustment module 305 is used to construct a dynamic adjustment algorithm DABD based on Dijkstra's algorithm to dynamically adjust the weights of the objective function of path risk level and the objective function of path length in the dual-objective path planning model ModelⅢ, so as to find the path with the minimum weight F for evacuees; wherein, the path with the minimum weight F is the optimal evacuation path that takes into account both safety and efficiency.

[0194] The processing module 306 is used to obtain the real-time risk index corresponding to each risk factor within the evacuation path, and adjust the weights of the two objective functions in the dual-objective path planning model ModelⅢ using the dynamic adjustment algorithm DABD based on the real-time risk index corresponding to each risk factor within the evacuation path, so as to obtain the optimal evacuation path.

[0195] The planning device provided in this embodiment improves the survival probability of evacuees by acquiring the real-time risk index of each risk factor within the evacuation route and using an algorithm to obtain the optimal evacuation route based on these real-time risk indices.

[0196] The planning device provided in the embodiments of the present invention can execute the emergency evacuation route planning method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0197] Example 4

[0198] Figure 5 This is a schematic diagram of the structure of a terminal provided in Embodiment 4 of the present invention. Figure 5 A block diagram is shown of an exemplary terminal 12 suitable for implementing embodiments of the present invention. Figure 5 The terminal 12 shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0199] like Figure 5 As shown, terminal 12 is presented in the form of a general-purpose computing device. The components of terminal 12 may include, but are not limited to: one or more processors or processing units 16, system memory 28, and bus 18 connecting different system components (including system memory 28 and processing unit 16).

[0200] Bus 18 represents one or more of several bus architectures, including a memory bus or memory controller, a peripheral bus, a graphics acceleration port, a processor, or a local bus using any of the various bus architectures. For example, these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MAC) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect (PCI) bus.

[0201] Terminal 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by terminal 12, including volatile and non-volatile media, removable and non-removable media.

[0202] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Terminal 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be used to read and write non-removable, non-volatile magnetic media (… Figure 5 Not shown; usually referred to as a "hard drive"). Although Figure 5 Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk") and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of the present invention.

[0203] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data. Each or some combination of these examples may include an implementation of a network environment. Program modules 42 typically perform the functions and / or methods described in the embodiments of the present invention.

[0204] Terminal 12 can also communicate with one or more external devices 14 (e.g., keyboard, pointing device, sensor, camera, and display 24, etc.), and with one or more devices that enable a user to interact with the terminal 12, and / or with any device that enables the terminal 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). This communication can be performed via input / output (I / O) interface 22. Furthermore, terminal 12 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 20. As shown, network adapter 20 communicates with other modules of terminal 12 via bus 18. It should be understood that, although not shown in the figures, other hardware and / or software modules can be used in conjunction with terminal 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems, etc.

[0205] The processing unit 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the emergency evacuation route planning method provided in the embodiments of the present invention.

[0206] In practical applications, the detection data of path risk coefficients play a decisive role in the calculation results of models and algorithms. When the dual-objective model and algorithm are applied to multi-exit buildings, CO concentration, HCN concentration, and visibility distance can be detected by fire smoke sensors; ambient temperature along the path can be detected by temperature sensors; and crowd density along the path can be detected by IR infrared cameras. Through real-time calculation of path risk and length, the optimal evacuation path can be displayed on a monitor. The monitor can change the direction of indication in real time, unlike static traditional evacuation signs, which may lead evacuees to dangerous paths. As mentioned earlier, modern buildings have complex internal structures, high population density, and numerous, intersecting evacuation routes, forming many intersections. Traditional emergency evacuation signs are insufficient to meet the needs of evacuation guidance. Therefore, this embodiment uses a monitor to indicate the optimal evacuation path.

[0207] However, existing intelligent evacuation signs mainly adopt a two-way adjustable method, which can provide visual directional guidance for evacuees. However, in large complexes and underground spaces, evacuation routes are crisscrossed with many intersections, making it difficult for two-way adjustable signs to provide effective directional guidance, resulting in poor evacuation guidance. Therefore, this embodiment provides a novel display.

[0208] Figure 6 This is a schematic diagram of the structure of a display provided in Embodiment 4 of the present invention, as shown below. Figure 6 As shown, the display includes a main body and a group of guide cursors arranged around the main body. Each group of guide cursors includes a no-entry guide cursor arranged on the main body and a compliant guide cursor arranged around the no-entry guide cursor. The compliant guide cursors are set to correspond one-to-one with the evacuation path.

[0209] For example, when the display is applied to an intersection, four sets of directional cursors can be set around the display body. Each directional cursor set includes one prohibition cursor and three permitted cursors, with the three permitted cursors facing the left, front, and right directions of the intersection, respectively. The prohibition cursor can be an existing red no-entry indicator light, and the permitted cursors can be existing green permitted cursors. Both the prohibition and permitted cursors can use existing power supply and control methods to achieve controlled illumination of the indicator lights. Those skilled in the art can also select other directional cursors and corresponding power supply and control methods according to actual needs to achieve passage indication of evacuation routes, which will not be elaborated here.

[0210] For buildings, especially road intersections, intelligent evacuation signs should first and foremost be simple and easy to identify. Whether people are in a passageway or at a road intersection, they should be able to identify the indicated path through the indicator lights on the signs. When people face danger, they will instinctively exert their subjective initiative and obtain information through various channels, of which about 90% is obtained through vision (see Liu Shengpeng. Design of Guidance Signs for Evacuation Passage Intersections [J]. Fire Science and Technology, 2017, 36(11): 1552-1554). Therefore, vision should be the first choice as the information transmission channel for evacuation signs.

[0211] The intelligent evacuation display designed in this embodiment can be suspended and placed above the intersection of passageways and roads. It has evacuation indicator units on all four sides, each displaying directional indicators for left, forward, and right, as well as a "no entry" indicator. When no fire is occurring, the indicators are off. In the event of a fire, the intelligent evacuation display illuminates the corresponding directional indicator based on the optimal evacuation route calculated using a dual-objective path planning model and the DABD algorithm. When all three directions are unavailable at a certain location (in a dangerous state), the directional indicators are off, and the red "no entry" indicator (shaped like a cross) illuminates. By displaying the optimal evacuation route through the intelligent evacuation display, the traditional concept of "escape to the nearest exit" is transformed into "safe escape."

[0212] Furthermore, the information transmission of the intelligent evacuation display provided in this embodiment can also be implemented based on ZigBee wireless communication technology. ZigBee is a short-range, low-complexity, two-way wireless communication technology characterized by high data transmission rate, low power consumption, energy saving, reliability, short latency, large network capacity, and low cost of security equipment. For example, fire and crowd density information can be collected in real time by several ZigBee evacuation terminals located in different areas. Each ZigBee terminal consists of an intelligent evacuation display, smoke detector, temperature sensor, light sensor, and crowd density detection device. The information from the ZigBee terminals is aggregated by a regional router and then transmitted to the ZigBee coordinator in this system. When the emergency evacuation server receives the information from the ZigBee coordinator, the system identifies and encodes the locations of the ZigBee evacuation terminals in the fire area, generates the optimal evacuation route information in real time using a dual-objective path planning model and the DABD algorithm, and sends it back to all ZigBee evacuation terminals in the system network. The evacuation terminals then drive their intelligent evacuation displays to provide intelligent guidance based on the received evacuation routes. The management terminal can view the status around the ZigBee evacuation terminal in real time, and can also send control commands in unicast or broadcast form to achieve rapid evacuation and rescue.

[0213] Example 5

[0214] Embodiment 5 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform any of the emergency evacuation route planning methods provided in the above embodiments.

[0215] The computer storage medium of this invention can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be—but is not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this embodiment, the computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0216] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0217] The program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0218] Computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as "C" or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0219] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. An emergency evacuation path planning method characterized by, include: An undirected graph of the emergency evacuation network is constructed based on the evacuation routes; wherein, the undirected graph of the emergency evacuation network includes the nodes within the evacuation routes and the edges between the nodes; The risk factors that affect the risk level of the evacuation routes are identified, and each risk factor is classified into a risk level according to the potential harm to the evacuees; wherein, different risk levels are assigned corresponding risk indices. Based on the risk index and the undirected graph of the emergency evacuation network, a single-objective path planning model Model I considering the risk level of the evacuation path is constructed, and a single-objective path planning model Model II considering the length of the evacuation path is constructed based on the undirected graph of the emergency evacuation network. Based on the single-objective path planning model Model I and the single-objective path planning model Model II, a dual-objective path planning model Model III is constructed to find the Pareto optimal solution of the two models; A dynamic adjustment algorithm DABD based on Dijkstra's algorithm is constructed to dynamically adjust the weights of the objective functions of path risk level and path length in the dual-objective path planning model ModelⅢ, so as to find the path with the minimum weight F for evacuees; wherein, the path with the minimum weight F is the optimal evacuation path. The risk index corresponding to each risk factor within the evacuation route is obtained in real time. Based on the risk index corresponding to each risk factor within the evacuation route in real time, the weights of the two objective functions in the dual-objective path planning model ModelⅢ are adjusted using the dynamic adjustment algorithm DABD to obtain the optimal evacuation route.

2. The method of claim 1, wherein, The process of identifying risk factors that affect the risk level of the evacuation routes, and classifying each risk factor into a risk level based on the potential harm to evacuees, includes: Five risk factors that affect the risk level of the evacuation routes were identified: CO concentration, HCN concentration, ambient temperature, visibility, and congestion. Based on the potential harm of each risk factor to the escapees, each risk factor is divided into four risk levels: Level I, Level II, Level III, and Level IV. Risk indices were set for the four risk levels.

3. The method of claim 1, wherein, The construction of a single-objective path planning model (Model I) considering the risk level of evacuation paths based on the risk index and the undirected graph of the emergency evacuation network, and the construction of a single-objective path planning model (Model II) considering the length of evacuation paths based on the undirected graph of the emergency evacuation network, includes: For each risk factor in the evacuation path, a risk factor weight correction factor is set, and based on the risk factor weight correction factor, the weighted average of the risk index corresponding to the risk factor in each edge of the undirected graph of the emergency evacuation network is calculated to obtain the risk level of each edge in the undirected graph of the emergency evacuation network; wherein, the sum of the risk factor weight correction factors corresponding to each risk factor is 1. Based on the risk level of each edge in the undirected graph of the emergency evacuation network, a single-objective path planning model Model I considering the risk level of the evacuation path is constructed. Based on the undirected graph of the emergency evacuation network, a single-objective path planning model, Model II, is constructed, which considers the length of the evacuation path.

4. The method of claim 1, wherein, The construction of a dual-objective path planning model, Model III, based on the single-objective path planning model Model I and Model II, for finding the Pareto optimal solution of the two models includes: A bi-objective path planning model, Model III, is constructed to find the Pareto optimal solution for the two models, as shown in the following formula: In the formula: γ1 represents the weighting factor for risk level, γ2 represents the weighting factor for path length, and γ1 and γ2 represent the relative importance of risk level and path length in Model III, respectively, where γ1≥0, γ2≥0, and γ1+γ2=1; f1 represents the risk level in a certain evacuation route; f1 * f1 represents the risk level in the ideal evacuation route, f2 represents the minimum risk level from the source point to the destination point, and is calculated by the single-objective path planning model ModelⅠ; f2 represents the actual length of a certain evacuation route. The ideal evacuation path length represents the shortest distance from the source point to the destination, calculated by the single-objective path planning model Model II.

5. The method of claim 4, wherein, The construction of the dynamic adjustment algorithm DABD based on Dijkstra's algorithm is used to dynamically adjust the weights of the objective functions of path risk level and path length in the dual-objective path planning model Model III, in order to find the path with the minimum weight F for escapees. This includes: When only considering the risk level objective problem, set γ2 = 0 and call algorithm b to obtain the optimal evacuation path; When only considering the path length objective, set γ1 = 0 and call algorithm b to obtain the optimal evacuation path; When considering a bi-objective problem involving risk level and path length, initial values ​​for γ1 and γ2 are set, and the optimal evacuation path is constructed by searching the weight vector space at the end of the call to algorithm b; where γ1≥0, γ2≥0, and γ1+γ2=1. The goal of calling algorithm b is that when Dijkstra's algorithm is implemented, each subsequent node will update the weights of the network based on the found evacuation paths and output the evacuation path with the smallest weight F and its information.

6. The method of claim 1, wherein, After constructing the Dijkstra algorithm-based dynamic adjustment algorithm DABD to dynamically adjust the weights of the path risk level objective function and the path length objective function in the dual-objective path planning model Model III, and finding the path with the minimum weight F for the escapees, the method further includes: Based on Dijkstra's algorithm, a virtual endpoint connecting all exits is set to change the original network structure of the undirected graph of the emergency evacuation network. The weights of the edges connecting each exit to the virtual endpoint are all set to zero, and the virtual endpoint is transformed into the source node, while each node is transformed into the endpoint.

7. The method according to claim 1 or 6, characterized in that, After constructing the Dijkstra algorithm-based dynamic adjustment algorithm DABD to dynamically adjust the weights of the path risk level objective function and the path length objective function in the dual-objective path planning model Model III, and finding the path with the minimum weight F for the escapees, the method further includes: Append a no-entry threshold for the risk factor to each edge of the undirected graph of the emergency evacuation network; When using the DABD algorithm to search for the path with the smallest weight F, remove the edges where a certain risk factor exceeds the forbidden threshold, and use the remaining network structure to search for the optimal evacuation path. When a node has no feasible path, the DABD algorithm is used to sequentially remove the prohibition thresholds of each risk factor attached to the undirected graph of the emergency evacuation network according to the probability of survival.

8. An emergency evacuation route planning device characterized by comprising: include: A module is established to construct an undirected graph of the emergency evacuation network based on the evacuation routes; wherein the undirected graph of the emergency evacuation network includes nodes within the evacuation routes and edges between nodes; The acquisition module is used to acquire risk factors that affect the risk level of the evacuation route, and classify each risk factor into a risk level according to the potential harm of each risk factor to the evacuees; wherein, different risk levels are set with corresponding risk indices. The planning module is used to construct a single-objective path planning model Model I that considers the risk level of the evacuation path based on the risk index and the undirected graph of the emergency evacuation network, and to construct a single-objective path planning model Model II that considers the length of the evacuation path based on the undirected graph of the emergency evacuation network. The construction module is used to construct a dual-objective path planning model Model III for finding the Pareto optimal solution of the two models, based on the single-objective path planning model Model I and the single-objective path planning model Model II. The adjustment module is used to construct a dynamic adjustment algorithm DABD based on Dijkstra's algorithm, so as to realize the dynamic adjustment of the weights of the objective function of path risk level and objective function of path length in the dual-objective path planning model ModelⅢ, and find the path with the minimum weight F for evacuees; wherein, the path with the minimum weight F is the optimal evacuation path that takes into account both safety and efficiency. The processing module is used to obtain the real-time risk index corresponding to each risk factor within the evacuation route, and adjust the weights of the two objective functions in the dual-objective path planning model ModelⅢ using the dynamic adjustment algorithm DABD based on the real-time risk index corresponding to each risk factor within the evacuation route, so as to obtain the optimal evacuation route.

9. A terminal, characterized by comprising: include: One or more processors; Storage device for storing one or more programs; A camera is used to capture images; Sensors are used to collect environmental parameters; A display is used to indicate evacuation routes; the display includes a main body and a group of guide cursors arranged around the main body. Each group of guide cursors includes a no-entry guide cursor arranged on the main body and a compliant guide cursor arranged around the no-entry guide cursor. The compliant guide cursors are set to correspond one-to-one with the evacuation routes. When the one or more programs are executed by the one or more processors, the one or more processors implement the emergency evacuation route planning method as described in any one of claims 1-7.

10. A storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform the emergency evacuation route planning method as described in any one of claims 1-7.

Images