A big data-based urban lifeline facility flood disaster chain simulation method

Abstract

The application provides a city lifeline facility flood disaster chain simulation method based on big data, and the method steps comprise: city flood model simulation and verification based on big data, city lifeline facility flood disaster chain construction based on big data, lifeline facility flood disaster chain dynamic simulation coupled with numerical simulation and cascade failure method, and influence evaluation of extreme storm flood on lifeline facility disaster chain evolution. The method provided by the application is based on big data, adopts a technical scheme of "numerical simulation+big data+cascade failure", and creatively realizes dynamic simulation of the lifeline facility disaster chain, innovates the technical method in the field, and has significant technical progress.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of urban public safety and disaster prevention and mitigation, and is a kind of urban lifeline facility flood disaster chain simulation method based on big data, which is the intersection of public safety guarantee technology and new generation information technology (big data application), and particularly relates to a kind of urban lifeline facility flood disaster chain simulation method based on big data. BACKGROUND

[0002] In extreme situations, floods not only cause serious casualties and property losses, but also easily lead to the linkage failure between key infrastructures, inducing "disaster chain". In general intensity rainstorm, such chain reaction is rare, but in extreme rainstorm, urban lifeline facilities such as water supply, power supply, road, communication, gas supply and the like have the characteristics of spatial high concentration and functional high coupling, once a key node fails, it can be rapidly propagated and expanded through the network structure, leading to systematic collapse, which seriously threatens the safe operation of city and social stability.

[0003] In extreme flood events, conventional monitoring equipment often fails due to power failure, immersion or communication interruption, etc., resulting in the problem of missing observation data in the calibration and verification stage of traditional numerical model, which affects the accuracy and practicability of simulation results. Social media and other big data resources can provide supplementary support in two aspects: on the one hand, the data source of flood evolution process can be indirectly supplemented through the public uploaded text and image information; on the other hand, through the crawling and analysis of the location information in the public platform, the spatial distribution of urban lifeline facilities and their disaster bodies can also be identified, which provides a basis for constructing the structure of facilities network and simulating the propagation path of disaster chain.

[0004] The previous disaster chain modeling usually relies on preset and static facility failure path, which has strong subjectivity and insufficient dynamics, and is difficult to truly reflect the dynamic evolution characteristics of urban lifeline facility disaster chain in flood process. SUMMARY

[0005] In order to overcome the problems of the prior art, the present application proposes a kind of urban lifeline facility flood disaster chain simulation method based on big data by fusing "numerical simulation + big data + cascade failure". The method fuses numerical simulation and cascade effect through big data based technical means, dynamically simulates the propagation process of urban lifeline facility disaster chain, innovates the technical method in the field, and provides scientific support for urban disaster prevention and mitigation decision and resilience improvement under extreme rainstorm.

[0006] The purpose of the present application is achieved as follows:

[0007] The present application provides a kind of urban lifeline facility flood disaster chain simulation method based on big data, which includes the following steps:

[0008] Step 1, simulation and verification of urban flood model based on big data

[0009] Existing methods including two-dimensional water dynamic model are used to construct urban flood model, and the constructed model is verified by field measurement or network public data;

[0010] Among them, the network public data is obtained by parsing the social media webpage HTML document, the parsing process includes extracting the label content containing flood related keywords, identifying the geographic location information, and taking the spatial coincidence rate of the identified water accumulation point and the survey water accumulation point as the basis for verifying the accuracy of the model.

[0011] Step 2, construction of urban lifeline facility flood disaster chain based on big data

[0012] S21, extraction of lifeline facility disaster body point

[0013] Open map API is called by using a request set containing spatial grid and point of interest keywords to extract lifeline facility data set and disaster body data set;

[0014] The request set is constructed by the following formula:

[0015] R={r ij =(q j ,g i )|i=1,...M;j=1,...,C}

[0016] D=D (L) ∪D (T)

[0017] In the formula, R is the set of all data requests; r ij is a data request composed of spatial grid g i and POI keyword q j ; q j is the jth POI keyword; g i is any spatial grid; M is the total number of spatial grids; C is the number of POI keywords; D is the data set extracted according to the request; D (L) is the lifeline facility data set; D (T) is the disaster body data set;

[0018] S22, screening of lifeline facility disaster body point

[0019] According to the population correlation degree of spatial point, the location advantage factor, the distance from the city center and the distance and quantity constraints of spatial point, the facility and disaster body point are screened;

[0020] The importance level of spatial point is calculated by the following formula:

[0021]

[0022] In the formula, represents the spatial point importance level; is the population quantity associated with the spatial point; represents the location advantage factor of the spatial point; is the distance of the spatial point from the city center; α, β, ξ, ε are the weights of the respective indexes; λ is the penalty factor;

[0023] S23, construction of a lifeline facility flood disaster chain

[0024] Based on the functional correlation of lifeline facilities, the lifeline facility disaster chain is constructed according to a weighted directed network, wherein the edge weight is determined by the in-degree, out-degree, spatial coordinates and related parameters of the nodes in the disaster chain;

[0025] The edge weight is determined by the following formula:

[0026]

[0027] In the formula, w ij is the edge weight; θ is an adjustable parameter; k i and k j are the in-degree and out-degree of the two nodes connected by the edge in the disaster chain; (x i ,y i ) and (x j ,y j ) are the spatial point coordinates; d i and are functional adjustable parameters.

[0028] Step 3, dynamic simulation of lifeline facility flood disaster chain by coupling numerical simulation and cascading failure method

[0029] Set the functional failure threshold of lifeline facilities, input different rainfall scenarios into the city flood model verified in step 1, obtain the city flood numerical simulation results under different scenarios, and realize the dynamic simulation of lifeline facility disaster chain failure at different times by spatial superposition of simulated water depth and lifeline facility disaster chain;

[0030] The spatial superposition relationship of the simulated water depth and the lifeline facility disaster chain is represented as:

[0031] L(p i ,t)=H(x i ,y i ,t)·f(H(x i ,y i ,t))

[0032]

[0033] where L(p i ,t) is the influence degree of facility p i at time t, H(x i ,y i ,t) is the water depth at location (x i ,y i ) at time t, f(H) is the influence function, H is the water depth simulated by the urban flood model, and H safe is the water depth at which the facility function can normally work.

[0034] Step 4: Influence of extreme rainstorm flood on evolution of lifeline facility disaster chain

[0035] Based on network redundancy, path entropy, number of disaster-bearing bodies connected to lifeline facilities, and service function of lifeline facilities, a lifeline facility disaster chain loss index is constructed to quantitatively evaluate the evolution trend of the disaster chain under extreme rainstorm.

[0036] Further, in step 1, when parsing the HTML document of the social media webpage, the bs4 library is used to filter the target object according to the keyword of the tag, and the keywords include wash away, waterlogging, water, waterlogging, drainage, water immersion, waterlogging, waterlogging, waterlogging, waterlogging, waterlogging, waterlogging, and waterlogging.

[0037] The HTML object is parsed by the following formula:

[0038] T(H)={(N i ,A i ,T(C i ))|i=1,2...,n}

[0039] where T(H) is the number of HTML objects, N i is the name of the i-th HTML tag, A i is the attribute set of the tag, C i is the child content of the tag, including the microblog content published by the user, T(C i ) is the number of objects calculated recursively for the child nodes, and n is the total number of tags in HTML.

[0040] Further, in step 2, S22, the location advantage factor is calculated by the following formula:

[0041]

[0042] where u is the average of the maximum and minimum distances of the spatial point, and σ is the control attenuation amplitude.

[0043] Further, in step 2, S22, the distance and number of spatial points satisfy the constraint:

[0044]

[0045] N≤N max

[0046] In the formula, T min is the minimum distance between any two selected spatial points; T max is the maximum distance between any position in the study area Omega and the spatial points in the study area; dist(x i , x j ) is the distance between any spatial point and between; dist(x, x j ) is the distance between any position and the spatial point; I is the set of spatial points; N and N max are the current and maximum number of spatial points, respectively.

[0047] Further, in step 4, the formula for calculating the lifeline facility disaster chain loss index is:

[0048]

[0049] In the formula, DCNDI is the lifeline facility disaster chain loss index; RD ori is the initial network redundancy; RD t is the network redundancy at time t; PE ori is the initial path entropy; PE t is the path entropy at time t; N ori is the number of hazard-affected bodies connected to the lifeline facility at the initial time; N t is the number of hazard-affected bodies connected to the lifeline facility at time t; S ori is the service function of the lifeline facility at the initial time; S t is the service function of the lifeline facility at time t; w1, w2, w3 and w4 are the weights of each index.

[0050] Further, in step 1, the lifeline facilities include water supply, power supply, road, communication, gas supply facilities, and the hazard-affected bodies are people-intensive or key function areas including schools, residential areas, hospitals, subway stations, bus stations, and parking lots.

[0051] The present application fuses the technical scheme of "numerical simulation + big data + cascading failure", and realizes multi-dimensional technical breakthroughs for the technical pain points of urban lifeline facility flood disaster chain simulation, and has the following beneficial effects:

[0052] 1. Break through the traditional data bottleneck, improve the accuracy and practicality of the model

[0053] In view of the problem that conventional monitoring equipment is prone to failure and observation data is missing in existing extreme flood events, the application innovatively introduces social media and public platform big data resources, extracts flood-related text information and geographic location data by analyzing web HTML documents, and supplements the key data source of the flood evolution process. At the same time, the spatial distribution of lifeline facilities and disaster-bearing bodies is identified with the aid of big data, providing multiple support for model calibration and verification, further solving the simulation deviation problem caused by insufficient data in traditional numerical models, and significantly improving the accuracy and practicality of the model.

[0054] 2. Realize dynamic simulation of disaster chain, break through the limitation of static modeling

[0055] Traditional disaster chain modeling relies on preset static failure paths, which has the defects of strong subjectivity and inability to reflect dynamic evolution. The application realizes real-time dynamic simulation of the propagation process of the disaster chain by coupling numerical simulation and cascading failure method, based on the dynamic water depth simulation results under different rainfall scenarios, and combining with the function failure threshold of lifeline facilities. The simulation results can clearly show the diffusion law of facility failure at different times (such as t = 12h, t = 18h and the maximum water depth moment), and truly reflect the dynamic characteristics of "a node failure-network chain reaction-system collapse" in extreme rainstorm, breaking through the simplification and distortion of the evolution process of disaster chain in static modeling.

[0056] 3. Precisely construct disaster chain network to improve the scientificity of simulation

[0057] The application innovatively proposes a POI big data extraction method based on grid public data and semantic keywords, and realizes the precise screening of lifeline facilities and disaster-bearing body points by combining point importance evaluation (considering population correlation, location advantage, spatial distance and other factors). At the same time, the disaster chain is constructed by a weighted directed network, and the edge weight is determined based on the node in-degree, out-degree, spatial coordinates and function attributes, so that the correlation between facilities is more in line with the actual service logic (such as the coupling relationship between power supply facilities and water supply, gas supply facilities). This structured network construction method solves the problem of strong subjectivity of facility correlation and disconnection between network structure and actual function in traditional models, and lays a scientific foundation data support for disaster chain simulation.

[0058] 4. Establish a quantitative evaluation system to support disaster prevention and mitigation decision-making

[0059] The present application innovatively constructs a lifeline facility disaster chain loss index (DCNDI), integrates network redundancy, path entropy, number of connections of disaster-affected bodies and service function four key indicators, and realizes quantitative evaluation of the influence of extreme rainstorm on disaster chain evolution. Through analyzing the nonlinear change trend of the indicators (such as the three stages of 'less disaster influence-increased rate-stable'), the key nodes of disaster chain evolution (such as the mutation moments of t=14h and t=17h) can be accurately identified, which provides a scientific basis for the city to develop phased disaster prevention strategies and optimize resource allocation, and effectively improves the resilience and risk resistance of urban lifeline system under the scenario of extreme rainstorm. BRIEF DESCRIPTION OF DRAWINGS

[0060] The present application will be further described below in conjunction with the drawings and examples.

[0061] Figure 1 is a flow chart of the urban lifeline facility flood disaster chain simulation method based on big data according to the embodiment of the present application;

[0062] Figure 2 shows the spatial distribution of the water accumulation points identified by field measurement investigation and network public data;

[0063] Figure 3 shows the simulation results of the typical rainfall flood numerical model;

[0064] Figure 4 shows the comparison results before and after the screening of lifeline facilities and disaster-affected bodies;

[0065] Figure 5 is a schematic diagram of the failure of lifeline facility disaster chain;

[0066] Figure 6 shows the simulation results of the urban flood model;

[0067] Figure 7 shows the dynamic simulation results of the lifeline facility disaster chain;

[0068] Figure 8 shows the time variation process of the key indicators;

[0069] Figure 9 shows the time variation process of the DCNDI. DETAILED DESCRIPTION

[0070] In order to make the purpose, technical scheme and advantages of the present application clearer, the following will further describe the disclosed embodiments of the present application in conjunction with the drawings. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0071] Embodiment:

[0072] As Figure 1As shown, the embodiment provides a big data-based city lifeline facility flood disaster chain simulation method, including the following steps:

[0073] Step 1, big data-based city flood model simulation and verification

[0074] TELEMAC-2D and other numerical methods are used to construct a city flood model, and the constructed model is verified through field measurement or network public data, to supplement city flood monitoring.

[0075] Network public data is obtained by parsing social media web page HTML documents, and the parsing process includes extracting label content containing flood-related keywords, identifying geographic location information, and taking the spatial coincidence rate of the identified water accumulation points and the survey water accumulation points as the basis for model accuracy verification.

[0076] Specifically, the following formula is used to parse the HTML object:

[0077] T(H)={(N i ,A i ,T(C i ))|i=1,2...,n} (1)

[0078] In the formula: T(H) is the number of HTML objects; N i is the name of the i-th HTML tag; A i is the attribute set of the tag; C i is the child content of the tag, including user-posted microblog content such as location, time, keywords, and other flood-related descriptions; T(C i ) is the number of objects calculated recursively for the child nodes; and n is the total number of tags in HTML.

[0079] O i =(N i ,A i ,V i ,ID(N i )) (2)

[0080] In the formula: O i is the instantiated HTML object; V i is the text content of the object; and ID(N i ) is a unique identifier used to distinguish different nodes.

[0081] bs4 can extract certain specific tags based on the keywords in the tag content, which is considered as selecting specific nodes from the parsed tree.

[0082]

[0083] In the formula: S(T,Vtarget V represents the set of target objects selected from the parse tree; target This refers to the specific tag content that needs to be extracted.

[0084] Typically, bs4 parsing HTML can be likened to the mapping of functions onto a collection.

[0085] For example, when a specific tag is specified, the actual operation is to map this set of tags to a subset that meets the criteria:

[0086] F:H→S,F(N)={O i |O i ∈H,P(O i )=Ture)} (4)

[0087] In the formula: H is the set of all objects in the HTML file; S is the subset of objects that meet the filtering criteria; F is the mapping function; P(O i The filtering criteria are as follows. After parsing, the content is filtered based on specific tags, including keywords related to floods such as washed away, water accumulation, water ingress, waterlogging, drainage, flooding, flood disaster, rising water, wading, flooding, and rising water.

[0088] The selected content is further identified to obtain its geographical location information, and the geographic information points are plotted on the GIS platform. The accuracy of the model is verified by the spatial overlap rate between the identified waterlogging points and the surveyed waterlogging points. Step 2: Construction of Urban Lifeline Facility Flood Disaster Chain Based on Big Data

[0089] S21, Extraction of disaster-bearing points for lifeline facilities

[0090] This application proposes a big data extraction method for Points of Interest (POIs) based on publicly available grid data and semantic keywords, as shown in the following formulas (5-6). Using a request set of spatial grids and POI keywords, the method automatically obtains POI data by calling the Open Maps API, and extracts the lifeline facility dataset D. (L) With disaster-bearing body dataset D (T) Furthermore, it is refined into multiple functional subclasses to form a POI semantic network with nested structural features, providing basic data support for subsequent modeling in this method.

[0091] R = {r ij =(q j ,g i )|i=1,...M; j=1,...,C} (5)

[0092] D = D (L) ∪D (T) (6)

[0093] Wherein, R is a set of all data requests; r ij is a spatial grid g i combined with POI keywords q j ; q j is the jth POI keyword; g i is any spatial grid; M is the total number of spatial grids; C is the number of POI keywords; D is a set of data extracted by request; D (L) is a lifeline facility data set, including water, electricity, and gas networks; D (T) is a hazard-affected body data set, including densely populated buildings such as residential areas.

[0094] S22, lifeline facility and hazard-affected body point screening

[0095] According to the population correlation degree of the spatial point, the location advantage factor, the distance from the city center, and the distance and quantity constraints of the spatial point, the facility and hazard-affected body point are screened.

[0096] The application innovatively proposes a lifeline facility and hazard-affected body point screening method, important points are selected according to the score, and the spatial point importance level is calculated by the following formula:

[0097]

[0098] Wherein, represents the spatial point importance level; is the number of populations associated with the spatial point; represents the location advantage factor of the spatial point; is the distance from the spatial point to the city center; α, β, ξ, and ε are the weights corresponding to each index respectively; and λ is a penalty factor.

[0099] Each index needs to be normalized before use. Among them, the location advantage factor is used to simulate the spatial distribution trend from each spatial point to the city center, and is calculated by the following formula:

[0100]

[0101] Wherein, u is the average value of the maximum and minimum distances of the spatial point; and σ is a control attenuation amplitude.

[0102] The distance and quantity constraints of the spatial point satisfy:

[0103]

[0104] N≤N max (11)

[0105] Wherein, T minT is the minimum distance between any two selected spatial points; max dist(x) represents the maximum distance between any location within the study area Ω and a spatial point within the study area. i ,x j ) represents the distance between any point in space and ; dist(x, x j ) represents the distance between any location and a point in space; I is the set of points in space; N and N max These represent the current and maximum number of spatial points, respectively.

[0106] S23, Construction of lifeline facilities for flood disaster relief

[0107] Based on the functional relationships of lifeline facilities, a disaster chain for lifeline facilities is constructed according to a weighted directed network. The edge weights are determined by calculating the in-degree, out-degree, spatial coordinates, and related parameters of the nodes in the disaster chain.

[0108] Specifically, Python and GIS tools are used to identify the locations of lifeline facilities and disaster-bearing entities, which are then plotted on the GIS platform. Lifeline facilities include water, electricity, roads, communications, and gas (water supply, power supply, roads, communications, and gas supply facilities), while disaster-bearing entities include schools, tourist attractions, supermarkets, subway stations, residential areas, parking lots, nursing homes, hotels, hospitals, companies, and bus stops. Based on the functions of lifeline facilities, the disaster chain connections are as follows: power supply facilities can connect to water supply, gas supply, and communications facilities, as well as schools, tourist attractions, supermarkets, residential areas, nursing homes, hotels, hospitals, and companies; water supply, gas supply, and communications facilities can connect to schools, tourist attractions, supermarkets, residential areas, nursing homes, hotels, hospitals, and companies; roads can connect to subway stations and bus stops; and gas stations can connect to parking lots.

[0109] Based on the weighted directed network, a disaster chain for lifeline facilities is constructed. Considering the location of the disaster-bearing body of the lifeline facility and the service function of the lifeline facility, the edge weights are determined as follows:

[0110]

[0111] In the formula, w ij θ is the edge weight; θ is an adjustable parameter, the larger the value of θ, the greater the difference in edge weights in the disaster chain, and the value ranges from 0 to 1; k i and k j Let x be the in-degree and out-degree of the edge connecting two nodes in the disaster chain, respectively; i ,y i ) and (x j ,y j ) represents the coordinates of a point in space; d i and It is an adjustable parameter with a value ranging from 0 to 1.

[0112] Step 3, Dynamic simulation of lifeline facility flood disaster chain based on coupling of numerical simulation and cascade failure method

[0113] The threshold of lifeline facility failure is set, and the lifeline facility is considered to fail when it exceeds the threshold, at which time the lifeline facility can only be affected by other lifeline facilities and cannot affect other lifeline facilities or disaster-bearing bodies. According to the urban flood numerical model verified in step 1, different rainfall scenarios are input to obtain the results of urban flood numerical simulation under different scenarios, and the spatial superposition of water depth and lifeline facility disaster chain is realized to achieve dynamic simulation of lifeline facility disaster chain failure at different times. The superposition relationship can be expressed as:

[0114] L(p i ,t)=H(x i ,y i ,t)·f(H(x i ,y i ,t)) (13)

[0115]

[0116] In the formula, L(p i ,t) is the influence degree of facility p i at time t, H(x i ,y i ,t) is the water depth at position (x i ,y i ) at time t. The influence function f(H) can set different influence degrees according to the water depth threshold; H is the simulated water depth of the urban flood model; H safe is the water depth at which the facility function can normally work, which can be set according to the actual situation. When there is no flood, the lifeline facility disaster chain is not affected. When the flood occurs, the lifeline facility disaster chain is affected differently at different times under different rainfall scenarios, and the disaster chain presented is different.

[0117] Step 4, Influence evaluation of extreme rainstorm flood on lifeline facility disaster chain evolution

[0118] The lifeline facility disaster chain loss index (DCNDI) is established to evaluate the influence of extreme rainstorm on the lifeline facility disaster chain.

[0119] Based on the network redundancy, path entropy, number of disaster-bearing bodies connected by lifeline facilities, and service function of lifeline facilities, the lifeline facility disaster chain loss index is constructed to quantitatively evaluate the evolution trend of the disaster chain under extreme rainstorm.

[0120]

[0121] In the formula, RD is network redundancy; |E| is the total number of edges; |V| is the number of points; PE is path entropy; n is the total number of effective paths of all lifeline facilities to the disaster body; p i is the important probability of the path, which is calculated as the proportion of the total weight of the path to the total weight of all paths; N is the number of disaster bodies effectively connected by the lifeline facility; L is the set of all lifeline facility types; L t is the tth type of lifeline facility; l is a certain facility of the tth type of lifeline facility; h is a disaster body adjacent to the lifeline facility; x lh is the adjacent edge; S is the service function of the lifeline facility; |L t | is the number of tth type of lifeline facilities; w lh is the weight of the adjacent edge; RD ori is the network redundancy at the initial time; RD t is the network redundancy at t time; PE ori is the path entropy at the initial time; PE t is the path entropy at t time; N ori is the number of disaster bodies effectively connected by the lifeline facility at the initial time; N t is the number of disaster bodies effectively connected by the lifeline facility at t time; S ori is the service function of the lifeline facility at the initial time; S t is the service function of the lifeline facility at t time; w1, w2, w3 and w4 are the weights of each index.

[0122] Application Example:

[0123] In this application example, a specific case area is selected to simulate the urban lifeline facility flood disaster chain by the method described in Embodiment 1, and the specific results are as follows:

[0124] Step 1, Simulation and verification of urban flood model based on big data

[0125] As shown in Figure 2 , compared with the investigation of water accumulation points and the identification of water accumulation points, the number of identified water accumulation points in the case area is 24, of which 17 are spatially coincident with the investigation of water accumulation points, and the spatial coincidence rate is 71%, indicating that network public data can be used as a supplement to urban flood monitoring.

[0126] As shown in Figure 3 , the investigation of water accumulation points, the identification of water accumulation points and the superposition of the results of typical rainfall flood numerical simulation can obtain the spatial distribution of the investigation of water accumulation points and the identification of water accumulation points, which is basically consistent with the results of numerical simulation, and most of them are distributed in the areas with larger water depth in the simulation results, verifying the accuracy of the model.

[0127] Step 2, Construction of urban lifeline facility flood disaster chain based on big data

[0128] Identify the lifeline facilities and disaster body points in the case area, screen and build the lifeline disaster chain. The comparison results before and after screening of lifeline facilities and disaster bodies are shown in Figure 4 .

[0129] Figure 5 The failure schematic diagram of lifeline disaster chain.

[0130] Step 3, dynamic simulation of lifeline flood disaster chain by coupling numerical simulation and cascade failure method

[0131] According to the simulation results of the urban flood model Figure 6 ), t = 12h, the water depth is small and the distribution is sparse at this time, the rainfall accumulation is small, and the comparison shows that the lifeline disaster chain is almost not affected; t = 18h, the rainfall accumulation is large, the water depth distribution is large and the distribution is dense, and the large water depth affects part of the infrastructure points; at the moment of maximum water depth, most of the facilities fail, and the normally operating facilities cannot cover most of the disaster bodies.

[0132] Figure 7 The dynamic simulation results of the lifeline disaster chain are shown.

[0133] Step 4, influence evaluation of extreme rainstorm flood on the evolution of lifeline disaster chain

[0134] As shown in Figure 8 , Figure 9 , according to the time evolution process of the four key indicators and DCNDI, it can be seen that the time variation trend of the key indicators and DCNDI can be divided into three stages of "small disaster impact - disaster impact rate increases - disaster impact stabilizes", showing a nonlinear trend, and there are obvious changes at t = 14h and t = 17h. Using this index, the influence of extreme rainstorm flood on the evolution of lifeline disaster chain can be quantitatively evaluated.

[0135] Finally, it should be pointed out that the above is only used to illustrate the technical solutions of the present application and not to limit, although the present application has been described in detail with reference to the preferred arrangement, those skilled in the art should understand that the technical solutions of the present application (such as the use of various formulas, the order of steps, etc.) can be modified or replaced equivalently without departing from the spirit and scope of the technical solutions of the present application.

Claims

1. A method for simulating flood disaster chains in urban lifeline facilities based on big data, characterized in that, The method includes the following steps: Step 1: Simulation and validation of urban flooding models based on big data: An urban flood model was constructed using existing methods, including numerical methods for two-dimensional hydrodynamic models, and the constructed model was validated using field measurements or publicly available online data. Among them, publicly available online data was obtained by parsing HTML documents of social media web pages. The parsing process included extracting tag content containing flood-related keywords, identifying geographical location information, and using the spatial overlap rate between the identified water accumulation points and the surveyed water accumulation points as the basis for verifying the accuracy of the model. Step 2, Construction of Urban Lifeline Infrastructure Flood Disaster Chain Based on Big Data: S21, Location extraction of disaster-bearing points for lifeline facilities: The Open Maps API was invoked using a request set containing spatial grids and points of interest keywords to extract lifeline facility datasets and disaster-bearing body datasets. The request set is constructed using the following formula: ; In the formula, R is the set of all data requests; For spatial grid Keywords related to POI A combined data request; For the j-th type of POI keyword; For any spatial grid; M is the total number of spatial grids; C is the number of POI keywords; D is the dataset extracted according to the request; Data set for lifeline facilities; Data set for disaster-bearing entities; S22, Screening of disaster-bearing sites for lifeline facilities: Based on the population correlation of spatial points, location advantage factors, distance from the city center, and spatial point distance and quantity constraints, the locations of facilities and disaster-bearing bodies are screened. The importance level of a spatial point is calculated using the following formula: ; In the formula, Indicates the importance level of spatial points; The population associated with a spatial point; Locational advantage factors representing spatial points; denoted as , where is the distance between the spatial point and the city center; α, β, ξ, and ε represent the weights of each indicator; and λ is the penalty factor. S23, Lifeline Facility Flood Disaster Chain Construction: Based on the functional relationships of lifeline facilities, a disaster chain for lifeline facilities is constructed according to a weighted directed network. The edge weights are determined by calculating the in-degree, out-degree, spatial coordinates, and related parameters of the nodes in the disaster chain. The edge weight is determined using the following formula: ; In the formula, The edge weight; These are adjustable parameters; and These represent the in-degree and out-degree of the edges connecting two nodes in the disaster chain; , )and( , () represents the coordinates of a point in space; and It is a function with adjustable parameters; Step 3: Dynamic simulation of the lifeline facility flood disaster chain coupled with numerical simulation and cascading failure method: Set a failure threshold for lifeline facilities, input different rainfall scenarios into the urban flood model verified in step 1, and obtain numerical simulation results of urban flood under different scenarios. By simulating the spatial superposition of water depth and lifeline facility disaster chain, the dynamic simulation of lifeline facility disaster chain failure at different times is realized. The spatial superposition relationship between the simulated water depth and the disaster chain of lifeline facilities is expressed as follows: ; ; In the formula, L( , t) is the facility The degree of influence at time t, H( , (, t) represents time t and position ( , The water depth is f(H); f(H) is the influence function; H is the simulated water depth in the urban flooding model. The water depth at which the facilities can function normally; Step 4, Impact Assessment of Extreme Rainstorms and Floods on the Evolution of Disaster Chains for Lifeline Facilities: Based on four indicators—network redundancy, path entropy, the number of disaster-bearing bodies connected to lifeline facilities, and the service functions of lifeline facilities—a lifeline facility disaster chain loss index is constructed to quantitatively assess the evolution trend of disaster chains under extreme rainstorms.

2. The flood disaster chain simulation method according to claim 1, characterized in that, In step 1, when parsing the HTML document of the social media webpage, the bs4 library is used to filter target objects based on tag keywords. The keywords include: washed away, water accumulation, water inflow, waterlogging, drainage, water immersion, flooding, flood disaster, rising water, wading, flooding, rising water. Use the following formula to parse HTML objects: ; In the formula: T(H) is the number of HTML objects; Let i be the name of the i-th HTML tag; This is the set of attributes for the tag; This is a sub-content of the tag, containing the microblog content posted by the user; This is used to recursively calculate the number of child nodes; n is the total number of tags in the HTML.

3. The flood disaster chain simulation method according to claim 1, characterized in that, In step 2, S22, the locational advantage factor Calculated using the following formula: ; In the formula, u is the average of the maximum and minimum distances between spatial points; To control the attenuation rate.

4. The flood disaster chain simulation method according to claim 1, characterized in that, In step 2, S22, the distance and number constraints of the spatial points satisfy: ; ; ; In the formula, The minimum distance between any two selected spatial points; For the study area The maximum distance between any location within the study area and a spatial point within the study area; The distance between any points in space; Let N be the distance between any location and a point in space; let I be the set of points in space; and N and ... These represent the current and maximum number of spatial points, respectively.

5. The flood disaster chain simulation method according to claim 1, characterized in that, In step 4, the formula for calculating the lifeline facility disaster chain loss index is as follows: ; In the formula, DCNDI is the disaster chain loss index for lifeline facilities; The initial network redundancy; Let t be the network redundancy at time t; The path entropy at the initial moment; Let be the path entropy at time t; The number of disaster-bearing entities that are effectively connected to lifeline facilities at the initial moment; S represents the number of disaster-bearing bodies effectively connected to the lifeline facility at time t; ori The service functions of the lifeline facility at the initial moment; The service functions of the lifeline facility at time t; , , and The weights of each indicator.

6. The flood disaster chain simulation method according to claim 1, characterized in that, In step 1, the lifeline facilities include water supply, power supply, roads, communication and gas supply facilities, and the disaster-bearing bodies are densely populated or critical functional areas including schools, residential areas, hospitals, subway stations, bus stations and parking lots.

Images