A method for selecting locations for disaster relief material storage stations in cities

Abstract

This invention relates to a method for selecting disaster relief material reserve stations in a city. The method includes: S1, inputting relevant data of the city to be selected as a disaster relief material reserve station; S2, obtaining data features based on the relevant data of the city to be selected as a disaster relief material reserve station; the data features include: population distribution features, urban road network structure features, spatial location information features of urban infrastructure, geographic information features, and urban POI data features; S3, selecting the optimal area in the city as the city's disaster relief material reserve station based on the data features, a pre-set site selection model, and pre-set rules.

Description

Technical Field

[0001] This invention relates to the field of urban site selection technology, and in particular to a method for selecting a site for a disaster relief material storage station in a city. Background Technology

[0002] Earthquakes, snowstorms, and other sudden natural disasters have become increasingly frequent and complex in recent years, and their suddenness and destructiveness can have a significant negative impact on society and the environment. To reduce the losses caused by sudden natural disasters, the speed and effectiveness of emergency response measures are crucial. Since disaster relief material storage stations are essential facilities in post-disaster emergency response operations, their location is paramount. A well-chosen disaster relief material storage station not only ensures timely emergency response after a disaster but also plays a significant role in calming the emotions of people in the disaster area. Therefore, selecting appropriate disaster relief material storage stations to improve the speed of post-disaster emergency response is extremely important.

[0003] Currently, solutions to location selection problems are mainly divided into two categories: heuristic location selection methods and clustering-based location selection methods. Heuristic location selection methods are commonly used in traditional location selection problems, mainly employing heuristic algorithms to effectively search the solution space. Clustering-based location selection methods provide a multi-level balanced location selection algorithm, dividing all demand points into non-overlapping delivery ranges, and then using the centroid method to select a single delivery center within each delivery range. In other words, a large location selection problem is divided into multiple smaller location selection problems for solution.

[0004] Chinese patent "CN202110138907.4" proposes a method for selecting sites for public service facilities. This patent obtains basic data and screening criteria, selects candidate sites that meet the criteria, encodes these candidate sites in binary, and uses a multi-objective particle swarm optimization algorithm to find the optimal site. Chinese patent "CN202110554771.5" proposes a method and system for selecting shopping mall sites based on artificial intelligence and big data. It obtains several candidate sites for the proposed shopping mall based on the similarity and competitive relationship between the proposed shopping mall and existing shopping malls; obtains the trade zones of existing shopping malls and candidate sites, adjusts the trade zones, and calculates the trade zone feature scores of existing shopping malls and candidate sites based on the trade zones before or after adjustment; calculates the trade zone feature weights; calculates the repulsion force of existing shopping malls on the proposed shopping mall based on the distance between candidate sites and existing shopping malls, the weights and scores of trade zone features; obtains several preferred sites for the proposed shopping mall based on the repulsion force, and selects the optimal site from these preferred sites. The public service facility site selection method proposed in Chinese patent "CN202110138907.4" and the shopping mall site selection method based on artificial intelligence and big data proposed in Chinese patent "CN202110554771.5" share the same problem: like most logistics center site selection methods, they both use economic cost as the primary criterion for site selection. However, for facilities like disaster relief material reserve stations, which are oriented towards population coverage and the timeliness of post-disaster emergency response, economic cost cannot be used as the primary criterion for site selection. Furthermore, different cities have different geographical environments and infrastructure compositions, and corresponding site selection schemes should be provided for different urban structures. However, current research methods have not yet taken into account the impact of geographical information factors and spatial location information of urban infrastructure on the model. Summary of the Invention

[0005] (a) Technical problems to be solved

[0006] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method for selecting disaster relief material storage stations in cities, which solves the technical problem that the prior art only considers economic costs in the process of selecting disaster relief material storage stations, resulting in unsuitable disaster relief material storage stations.

[0007] (II) Technical Solution

[0008] To achieve the above objectives, the main technical solutions adopted by the present invention include:

[0009] This invention provides a method for selecting a site for a disaster relief material storage station in a city, the method comprising:

[0010] S1. Enter the relevant data of the cities to be selected as disaster relief material reserve stations;

[0011] S2. Based on the relevant data of the cities to be selected as disaster relief material reserve stations, obtain data characteristics; the data characteristics include: population distribution characteristics, urban road network structure characteristics, spatial location information characteristics of urban infrastructure, geographic information characteristics, and urban POI data characteristics.

[0012] S3. Based on the data characteristics, the pre-set site selection model and the pre-set rules, select the optimal area in the city as the city's disaster relief material reserve station;

[0013] The pre-defined location selection model is as follows:

[0014]

[0015] P represents the probability of location selection for m areas in the city; P = (p1, p2, ..., pn) i ...、p m ) T ;P m This represents the probability that the i-th region out of m regions will be selected as a disaster relief material storage station.

[0016] X is a matrix of data features, where X = (X1, X2, X3, X4, X5);

[0017] X1 represents population size characteristics; X2 represents urban road network structure characteristics; X3 represents spatial location information of urban infrastructure; X4 represents geographic information characteristics; X5 represents urban POI data characteristics.

[0018] w is the first parameter vector that needs to be trained in the pre-defined location selection model;

[0019] b is the second parameter vector that needs to be trained in the pre-defined location selection model;

[0020] b = (b1, b2, b3, b4, b5) T ;w=(w1,w2,w3,w4,w5) T ;

[0021] Wherein, b1 and w1 are the first and second parameters corresponding to the population quantity feature X1, respectively; b2 and w2 are the first and second parameters corresponding to the urban road network structure feature X2, respectively; b3 and w3 are the first and second parameters corresponding to the spatial location information feature X3 of urban infrastructure, respectively; b4 and w4 are the first and second parameters corresponding to the geographic information feature X4, respectively; and b5 and w5 are the first and second parameters corresponding to the urban POI data feature X5, respectively.

[0022] Preferably,

[0023] The relevant data for the cities to be selected as disaster relief material reserve stations include: the administrative division structure within the cities to be selected as disaster relief material reserve stations, the population of each administrative region of the cities to be selected as disaster relief material reserve stations, the urban road network structure data of the cities to be selected as disaster relief material reserve stations, the location information of urban infrastructure of the cities to be selected as disaster relief material reserve stations, and the urban topography information of the cities to be selected as disaster relief material reserve stations.

[0024] Preferably,

[0025] The population distribution characteristics refer to the pre-defined levels to which the population density of each administrative region within the city belongs;

[0026] The structural characteristics of urban road networks are the shortest distances between nodes in the urban road network structure.

[0027] The spatial location information of urban infrastructure is characterized by a connectivity map composed of the geographical locations of urban infrastructure.

[0028] Geographic information features include the city's topography and landform information;

[0029] The characteristics of urban POI data are the division information of functional areas within the city.

[0030] Preferably, S3 specifically includes:

[0031] S31. Based on the data characteristics and the first initial value of the randomly generated first parameter vector w, the first initial value of the second parameter vector b, the second initial value of the first parameter vector w, and the second initial value of the second parameter vector b, the first initial value of the location probability and the second initial value of the location probability are obtained respectively using the pre-set location selection model.

[0032] S32. Based on the first initial value of the location probability and the second initial value of the location probability, a first value of the location parameter corresponding to the first initial value of the location probability and a second value of the location parameter corresponding to the second initial value of the location probability are obtained by using a pre-set calculation strategy.

[0033] S33. Compare the first value of the addressing parameter with the second value of the addressing parameter, and obtain the comparison result;

[0034] S34. If the comparison result is that the first value of the addressing parameter is greater than the second value of the addressing parameter, then adjust the second initial value of the first parameter vector w and the second initial value of the second parameter vector b according to the preset first adjustment rule to obtain the new value of the first parameter vector w and the new value of the second parameter vector b. Based on the data characteristics and the new values ​​of the first parameter vector w and the second parameter vector b, the preset addressing model is used to obtain the new value of the second initial value of the addressing probability corresponding to the new value of the first parameter vector w and the new value of the second parameter vector b. Further, based on the new value of the second initial value of the addressing probability, a preset calculation strategy is used to obtain the new value of the second value of the addressing parameter. Steps S33-S34 are repeated until the first condition occurs.

[0035] If the comparison result is that the first value of the addressing parameter is less than the second value of the addressing parameter, then the first initial value of the first parameter vector w and the first initial value of the second parameter vector b are adjusted according to the preset first adjustment rule to obtain the new value of the first parameter vector w and the new value of the second parameter vector b. Based on the data features and the new values ​​of the first parameter vector w and the second parameter vector b, the preset addressing model is used to obtain the new value of the first initial value of the addressing probability corresponding to the new values ​​of the first parameter vector w and the second parameter vector b. Further, based on the new value of the first initial value of the addressing probability, a preset calculation strategy is used to obtain the new value of the first value of the addressing parameter, and steps S33-S34 are repeated until the first condition occurs.

[0036] S35. If the first condition is met, obtain the new values ​​of the first parameter vector w and the second parameter vector b corresponding to the new values ​​of the first and second values ​​of the current location parameters, respectively. Based on the new values ​​of the first and second values ​​of the current location parameters, respectively, select the optimal area in the city as a disaster relief material reserve station.

[0037] Preferably, S32 specifically includes:

[0038] S32. Based on the first initial value of the location probability and the second initial value of the location probability, a first value of the location parameter corresponding to the first initial value of the location probability and a second value of the location parameter corresponding to the second initial value of the location probability are obtained by using a pre-set calculation strategy.

[0039] S321. Select the regions corresponding to the first s largest probability values ​​among the first initial values ​​of the location probability, and the regions corresponding to the first s largest probability values ​​among the second initial values ​​of the location probability, respectively.

[0040] S322. Use formula (1) to calculate the population coverage rate of the area corresponding to the largest first s probability values ​​in the first initial value of site selection probability and the population coverage rate of the area corresponding to the largest first s probability values ​​in the second initial value of site selection probability.

[0041]

[0042] η cover This represents the population coverage value corresponding to the area corresponding to the largest of the first s probability values ​​among the first or second initial site selection probabilities, when the area is selected as the site selection target.

[0043] in,

[0044]

[0045] cover i This represents the population covered by the i-th region when it is selected as the location;

[0046] t represents the number of adjacent regions of the i-th region among m regions in the city;

[0047] person i This represents the population of the i-th region in the city;

[0048] person j This represents the population of the j-th region among the regions adjacent to the i-th region in the city.

[0049] S322. Calculate the total material allocation cost corresponding to the first initial value and the second initial value of the location probability using formula (2):

[0050]

[0051] in,

[0052] This represents the cost of intra-city material allocation in the i-th region of the city;

[0053] This represents the average cost of material input in the i-th region of the city;

[0054] cost i Let $i$ be the cost of resource allocation in the i-th region of the city.

[0055] S323. Based on the population coverage rate and total cost of material allocation corresponding to the first initial value of the site selection probability and the second initial value of the site selection probability, the first value of the site selection parameter corresponding to the first initial value of the site selection probability and the second value of the site selection parameter corresponding to the second initial value of the site selection probability are calculated respectively using formula (3).

[0056]

[0057] Preferably,

[0058] The first condition is that the new value of the first value of the current addressing parameter is equal to the new value of the second value of the current addressing parameter, or steps S33-S34 are repeated a preset number of times.

[0059] Preferably,

[0060] The preset number of times is greater than or equal to 200.

[0061] Preferably,

[0062] The first pre-defined adjustment rule is the gradient descent rule.

[0063] Preferably, S35 includes:

[0064] S351. If the first condition occurs, then obtain the average value of the new value of the first parameter vector w and the average value of the new value of the second parameter vector b corresponding to the new value of the current addressing parameter first value and the new value of the current addressing parameter second value, respectively.

[0065] S352. Based on the data characteristics and the average value of the new values ​​of the first parameter vector w and the second parameter vector b corresponding to the new values ​​of the current first and second location parameters, the average value of the location probability PA corresponding to the average value of the new values ​​of the first and second location parameters is obtained by using the pre-set location model.

[0066] S353. Based on the average value PA of the location probability, the average value of the location parameters corresponding to the average value PA of the location probability is obtained by using a pre-set calculation strategy;

[0067] S354. Based on the current location parameter first value, the current location parameter second value, and the average value of the location parameters, select the optimal area in the city as a disaster relief material reserve station.

[0068] Preferably,

[0069] The optimal areas in the city are the top k areas with the highest probability value corresponding to the selection parameter with the largest value among the current location parameter first value, the current location parameter first value, the current location parameter first value, the current location parameter new value, and the average value of the location parameters.

[0070] Where 1≤k≤3.

[0071] (III) Beneficial Effects

[0072] The beneficial effects of this invention are as follows: The method for selecting disaster relief material reserve stations in cities according to this invention involves inputting relevant data of the city to be selected as the disaster relief material reserve station; obtaining data characteristics based on the relevant data of the city to be selected as the disaster relief material reserve station; and selecting the optimal area in the city as the disaster relief material reserve station based on the data characteristics, a pre-set site selection model, and pre-set rules. Compared with existing technologies, this method considers the city's population distribution characteristics, urban road network structure characteristics, spatial location information of urban infrastructure, geographic information characteristics, and urban POI data characteristics during site selection, making the selected location of the disaster relief material reserve station more suitable. Attached Figure Description

[0073] Figure 1 This is a flowchart illustrating a method for selecting a site for a disaster relief material storage station in a city, according to the present invention.

[0074] Figure 2 The administrative division map of Shenyang City is shown in the example.

[0075] Figure 3 This is a schematic diagram of step S3 in an embodiment of the present invention;

[0076] Figure 4 This is a diagram showing the effect of selecting a site for a disaster relief material storage station in a city according to a method of the present invention. Detailed Implementation

[0077] To better explain and facilitate understanding of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0078] To better understand the above technical solutions, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that the present invention can be understood more clearly and thoroughly, and that the scope of the present invention can be fully conveyed to those skilled in the art.

[0079] See Figure 1 A method for selecting a site for a disaster relief material storage station in a city, the method comprising:

[0080] S1. Enter the relevant data of the cities to be selected as disaster relief material reserve stations.

[0081] The relevant data for the cities to be selected as disaster relief material reserve stations include: the administrative division structure within the cities to be selected as disaster relief material reserve stations, the population of each administrative region of the cities to be selected as disaster relief material reserve stations, the urban road network structure data of the cities to be selected as disaster relief material reserve stations, the location information of urban infrastructure of the cities to be selected as disaster relief material reserve stations, and the urban topography information of the cities to be selected as disaster relief material reserve stations.

[0082] The administrative division structure within a city is unique to each city, and the natural division within a city is its administrative division. This method of division is considered the most reasonable way to divide a city. For example... Figure 2 The image shows a map illustrating the administrative divisions of Shenyang City.

[0083] The population density of each administrative region can be determined by comparing the population size with the area of ​​that region. Based on population density, the administrative regions of a city can be divided into two categories: "sparsely populated" and "densely populated."

[0084] Urban road network structure data reflects the connectivity between nodes within a city, abstracts the urban spatial structure, and facilitates the calculation of accessibility relationships between nodes within the city.

[0085] Information on the location of urban infrastructure: There are various types of infrastructure within a city, which to some extent facilitate the allocation of supplies at disaster relief material storage stations.

[0086] Urban topography information: The topography within a city can affect the transportation of supplies and have a significant impact on the response speed of disaster relief material storage stations.

[0087] S2. Based on the relevant data of the cities to be selected as disaster relief material reserve stations, obtain data characteristics; the data characteristics include: population distribution characteristics, urban road network structure characteristics, spatial location information of urban infrastructure, geographic information characteristics, and urban POI data characteristics.

[0088] The population distribution characteristics refer to the pre-defined hierarchical levels of population density in each administrative region within the city. In this embodiment, the population distribution characteristics are obtained by pre-clustering the population density of each administrative region to obtain different levels of population density for each administrative region, including sparse, regular, and dense levels.

[0089] The structural characteristics of urban road networks are the shortest distances between nodes in the urban road network.

[0090] The spatial location information of urban infrastructure is characterized by a connectivity map composed of the geographical locations of urban infrastructure. In this embodiment, urban infrastructure includes fire hydrants, indoor warehouses, etc.

[0091] Geographic information features include the city's topography and landforms.

[0092] The characteristics of city POI data are the functional area division information within the city. In this embodiment, the functional areas include industrial areas, commercial areas, and residential areas.

[0093] S3. Based on the data characteristics, the pre-set site selection model, and the pre-set rules, select the optimal area in the city as the city's disaster relief material reserve station.

[0094] The pre-defined location selection model is as follows:

[0095]

[0096] P represents the probability of location selection for m areas in the city; P = (p1, p2, ..., pn) i ...、p m ) T ;P m This represents the probability that the i-th region out of m regions will be selected as a disaster relief material storage station.

[0097] X is a matrix of data features, where X = (X1, X2, X3, X4, X5).

[0098] X1 represents population size characteristics; X2 represents urban road network structure characteristics; X3 represents spatial location information of urban infrastructure; X4 represents geographic information characteristics; and X5 represents urban POI data characteristics.

[0099] w is the first parameter vector that needs to be trained in the pre-defined location selection model.

[0100] b is the second parameter vector that needs to be trained in the pre-defined location selection model.

[0101] b = (b1, b2, b3, b4, b5) T ;w=(w1,w2,w3,w4,w5) T .

[0102] Wherein, b1 and w1 are the first and second parameters corresponding to the population quantity feature X1, respectively; b2 and w2 are the first and second parameters corresponding to the urban road network structure feature X2, respectively; b3 and w3 are the first and second parameters corresponding to the spatial location information feature X3 of urban infrastructure, respectively; b4 and w4 are the first and second parameters corresponding to the geographic information feature X4, respectively; and b5 and w5 are the first and second parameters corresponding to the urban POI data feature X5, respectively.

[0103] See Figure 3 S3 specifically includes:

[0104] S31. Based on the data characteristics and the first initial value of the randomly generated first parameter vector w, the first initial value of the second parameter vector b, the second initial value of the first parameter vector w, and the second initial value of the second parameter vector b, the first initial value of the location probability and the second initial value of the location probability are obtained respectively using the pre-set location selection model.

[0105] S32. Based on the first initial value of the location probability and the second initial value of the location probability, a first value of the location parameter corresponding to the first initial value of the location probability and a second value of the location parameter corresponding to the second initial value of the location probability are obtained by using a pre-set calculation strategy.

[0106] In practical applications of this embodiment, S32 specifically includes:

[0107] S321. Select the regions corresponding to the first s largest probability values ​​in the first initial value of the location probability and the regions corresponding to the first s largest probability values ​​in the second initial value of the location probability, respectively.

[0108] S322. Use formula (1) to calculate the population coverage rate of the area corresponding to the largest first s probability values ​​in the first initial value of the site selection probability and the population coverage rate of the area corresponding to the largest first s probability values ​​in the second initial value of the site selection probability.

[0109]

[0110] η cover This represents the population coverage value corresponding to the area corresponding to the largest of the first s probability values ​​among the first or second initial site selection probabilities, when the area is chosen as the site selection target.

[0111] in,

[0112]

[0113] cover i This represents the population covered by the i-th region when it is selected as the location.

[0114] t represents the number of adjacent regions of the i-th region among m regions in the city.

[0115] person i This represents the population of the i-th region in the city.

[0116] person j This represents the population of the j-th region among the regions adjacent to the i-th region in the city.

[0117] S322. Calculate the total material allocation cost corresponding to the first initial value and the second initial value of the location probability using formula (2):

[0118]

[0119] in,

[0120] This represents the cost of intra-city material allocation in the i-th region of the city.

[0121] This represents the average material input cost of the i-th region in the city.

[0122] cost i Let be the cost of resource allocation in the i-th region of the city.

[0123] S323. Based on the population coverage rate and total cost of material allocation corresponding to the first initial value of the site selection probability and the second initial value of the site selection probability, the first value of the site selection parameter corresponding to the first initial value of the site selection probability and the second value of the site selection parameter corresponding to the second initial value of the site selection probability are calculated respectively using formula (3).

[0124]

[0125] S33. Compare the first value of the addressing parameter with the second value of the addressing parameter to obtain the comparison result.

[0126] S34. If the comparison result is that the first value of the addressing parameter is greater than the second value of the addressing parameter, then adjust the second initial value of the first parameter vector w and the second initial value of the second parameter vector b according to the preset first adjustment rule to obtain the new value of the first parameter vector w and the new value of the second parameter vector b. Based on the data characteristics and the new values ​​of the first parameter vector w and the second parameter vector b, the preset addressing model is used to obtain the new value of the second initial value of the addressing probability corresponding to the new values ​​of the first parameter vector w and the second parameter vector b. Further, based on the new value of the second initial value of the addressing probability, a preset calculation strategy is used to obtain the new value of the second value of the addressing parameter. Steps S33-S34 are repeated until the first condition occurs.

[0127] In this embodiment, the first adjustment rule is a gradient descent rule.

[0128] If the comparison result is that the first value of the addressing parameter is less than the second value of the addressing parameter, then the first initial value of the first parameter vector w and the first initial value of the second parameter vector b are adjusted according to the preset first adjustment rule to obtain the new values ​​of the first parameter vector w and the second parameter vector b. Based on the data characteristics and the new values ​​of the first parameter vector w and the second parameter vector b, the preset addressing model is used to obtain the new value of the first initial value of the addressing probability corresponding to the new values ​​of the first parameter vector w and the second parameter vector b. Further, based on the new value of the first initial value of the addressing probability, a preset calculation strategy is used to obtain the new value of the first value of the addressing parameter, and steps S33-S34 are repeated until the first condition occurs.

[0129] In this embodiment, the first condition is that the new value of the first value of the current addressing parameter and the new value of the second value of the current addressing parameter are equal, or steps S33-S34 are repeated a predetermined number of times.

[0130] Specifically, the preset number of times is greater than or equal to 200.

[0131] S35. If the first condition is met, obtain the new values ​​of the first parameter vector w and the second parameter vector b corresponding to the new values ​​of the first and second values ​​of the current location parameters, respectively. Based on the new values ​​of the first and second values ​​of the current location parameters, respectively, select the optimal area in the city as a disaster relief material reserve station.

[0132] In practical applications of this embodiment, S35 includes:

[0133] S351. If the first condition occurs, then obtain the average value of the new value of the first parameter vector w and the average value of the new value of the second parameter vector b corresponding to the new value of the current addressing parameter first value and the new value of the current addressing parameter second value, respectively.

[0134] S352, and based on the data characteristics and the average value of the new values ​​of the first parameter vector w and the second parameter vector b corresponding to the new values ​​of the current first and second location parameters, the average value of the location probability PA corresponding to the average value of the new values ​​of the first and second parameter vectors is obtained by using the pre-set location model.

[0135] S353. Based on the average value PA of the location probability, the average value of the location parameters corresponding to the average value PA of the location probability is obtained by using a pre-set calculation strategy.

[0136] S354. Based on the current location parameter first value, the current location parameter second value, and the average value of the location parameters, select the optimal area in the city as a disaster relief material reserve station.

[0137] See Figure 4 The optimal area in the city is the top k areas with the highest probability value corresponding to the selection parameter that has the largest value among the current location parameter first value new value, the current location parameter first value new value, and the average value of the location parameters.

[0138] Where 1≤k≤3.

[0139] like Figure 4 As shown in the figure, the arrows indicate the three selected areas as disaster relief material storage stations.

[0140] This embodiment of a method for selecting disaster relief material reserve stations in a city involves inputting relevant data of the city to be selected as the disaster relief material reserve station; obtaining data features based on the relevant data of the city to be selected as the disaster relief material reserve station; and selecting the optimal area in the city as the disaster relief material reserve station based on the data features, a pre-set site selection model, and pre-set rules. Compared with the prior art, this method considers the city's population distribution characteristics, urban road network structure characteristics, spatial location information of urban infrastructure, geographic information characteristics, and urban POI data characteristics during site selection, making the selected location of the disaster relief material reserve station more suitable.

[0141] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0142] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, as well as combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions.

[0143] It should be noted that any reference numerals placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In claims that enumerate several means, several of these means may be embodied by the same hardware. The use of the terms first, second, third, etc., is merely for convenience of expression and does not indicate any order. These terms can be understood as part of the component names.

[0144] Furthermore, it should be noted that in the description of this specification, the terms "one embodiment," "some embodiments," "embodiment," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Furthermore, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0145] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the claims should be interpreted to include both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0146] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, then this invention should also include these modifications and variations.

Claims

1. A method for selecting sites for disaster relief material storage stations in cities, characterized in that, The method includes: S1. Enter the relevant data of the cities to be selected as disaster relief material reserve stations; S2. Based on the relevant data of the cities to be selected as disaster relief material reserve stations, obtain data characteristics; the data characteristics include: population distribution characteristics, urban road network structure characteristics, spatial location information characteristics of urban infrastructure, geographic information characteristics, and urban POI data characteristics. S3. Based on the data characteristics, the pre-set site selection model and the pre-set rules, select the optimal area in the city as the city's disaster relief material reserve station; The pre-defined location selection model is as follows: ； P represents the probability of location selection for m areas in the city; P = (p1, p2, ..., pn) i ...、p m ) T ;P m This represents the probability that the i-th region out of m regions will be selected as a disaster relief material storage station. X is a matrix of data features, where X = (X1, X2, X3, X4, X5). X1 represents population size characteristics; X2 represents urban road network structure characteristics; X3 represents spatial location information of urban infrastructure; X4 represents geographic information characteristics; X5 represents urban POI data characteristics. w is the first parameter vector that needs to be trained in the pre-defined location selection model; b is the second parameter vector that needs to be trained in the pre-defined location selection model; b = (b1, b2, b3, b4, b5) T w = (w1, w2, w3, w4, w5) T ; Wherein, b1 and w1 are the first and second parameters corresponding to the population size feature X1, respectively; b2 and w2 are the first and second parameters corresponding to the urban road network structure feature X2, respectively; b3 and w3 are the first and second parameters corresponding to the spatial location information feature X3 of urban infrastructure, respectively; b4 and w4 are the first and second parameters corresponding to the geographic information feature X4, respectively; and b5 and w5 are the first and second parameters corresponding to the urban POI data feature X5, respectively. S3 specifically includes: S31, based on the data features and the first initial value of the randomly generated first parameter vector w, the first initial value of the second parameter vector b, the second initial value of the first parameter vector w, and the second initial value of the second parameter vector b, using the pre-set addressing model, obtaining the first initial value of the addressing probability and the second initial value of the addressing probability respectively; S32. Based on the first initial value of the location probability and the second initial value of the location probability, a first value of the location parameter corresponding to the first initial value of the location probability and a second value of the location parameter corresponding to the second initial value of the location probability are obtained by using a pre-set calculation strategy. S33. Compare the first value of the addressing parameter with the second value of the addressing parameter, and obtain the comparison result; S34. If the comparison result is that the first value of the addressing parameter is greater than the second value of the addressing parameter, then adjust the second initial value of the first parameter vector w and the second initial value of the second parameter vector b according to the preset first adjustment rule to obtain the new value of the first parameter vector w and the new value of the second parameter vector b. Based on the data characteristics and the new values ​​of the first parameter vector w and the second parameter vector b, the preset addressing model is used to obtain the new value of the second initial value of the addressing probability corresponding to the new value of the first parameter vector w and the new value of the second parameter vector b. Further, based on the new value of the second initial value of the addressing probability, a preset calculation strategy is used to obtain the new value of the second value of the addressing parameter. Steps S33-S34 are repeated until the first condition occurs. If the comparison result is that the first value of the addressing parameter is less than the second value of the addressing parameter, then the first initial value of the first parameter vector w and the first initial value of the second parameter vector b are adjusted according to the preset first adjustment rule to obtain the new value of the first parameter vector w and the new value of the second parameter vector b. Based on the data features and the new values ​​of the first parameter vector w and the second parameter vector b, the preset addressing model is used to obtain the new value of the first initial value of the addressing probability corresponding to the new values ​​of the first parameter vector w and the second parameter vector b. Further, based on the new value of the first initial value of the addressing probability, a preset calculation strategy is used to obtain the new value of the first value of the addressing parameter, and steps S33-S34 are repeated until the first condition occurs. S35. If the first condition occurs, obtain the new values ​​of the first parameter vector w and the second parameter vector b corresponding to the new values ​​of the first and second values ​​of the current location parameters, respectively. Based on the new values ​​of the first and second values ​​of the current location parameters, respectively, select the optimal area in the city as a disaster relief material reserve station. S32 specifically includes: S321. Select the regions corresponding to the first s largest probability values ​​among the first initial values ​​of the location probability, and the regions corresponding to the first s largest probability values ​​among the second initial values ​​of the location probability, respectively. S322. Use formula (1) to calculate the population coverage rate of the area corresponding to the largest first s probability values ​​in the first initial value of site selection probability and the population coverage rate of the area corresponding to the largest first s probability values ​​in the second initial value of site selection probability. （1）； This represents the population coverage value corresponding to the area corresponding to the largest of the first s probability values ​​among the first or second initial site selection probabilities, when the area is selected as the site selection target. in, ； cover i This represents the population covered by the i-th region when it is selected as the location; t represents the number of adjacent regions of the i-th region among m regions in the city; person i This represents the population of the i-th region in the city; person j This represents the population of the j-th region among the regions adjacent to the i-th region in the city. S322. Calculate the total material allocation cost corresponding to the first initial value and the second initial value of the location probability using formula (2): （2）； in, ; This represents the cost of intra-city material allocation in the i-th region of the city; This represents the average cost of material input in the i-th region of the city; Let $i$ be the cost of resource allocation in the i-th region of the city. S323. Based on the population coverage rate and total cost of material allocation corresponding to the first initial value of the site selection probability and the second initial value of the site selection probability, the first value of the site selection parameter corresponding to the first initial value of the site selection probability and the second value of the site selection parameter corresponding to the second initial value of the site selection probability are calculated respectively using formula (3). （3）。 2. The method according to claim 1, characterized in that, The relevant data for the cities to be selected as disaster relief material reserve stations include: the administrative division structure within the cities to be selected as disaster relief material reserve stations, the population of each administrative region of the cities to be selected as disaster relief material reserve stations, the urban road network structure data of the cities to be selected as disaster relief material reserve stations, the location information of urban infrastructure of the cities to be selected as disaster relief material reserve stations, and the urban topography information of the cities to be selected as disaster relief material reserve stations.

3. The method according to claim 2, characterized in that, The population distribution characteristics refer to the pre-defined levels to which the population density of each administrative region within the city belongs; The structural characteristics of urban road networks are the shortest distances between nodes in the urban road network structure. The spatial location information of urban infrastructure is characterized by a connectivity map composed of the geographical locations of urban infrastructure. Geographic information features include the city's topography and landform information; The characteristic of urban POI data is the division information of functional areas within the city.

4. The method according to claim 3, characterized in that, The first condition is that the new value of the first value of the current addressing parameter is equal to the new value of the second value of the current addressing parameter, or steps S33-S34 are repeated a preset number of times.

5. The method according to claim 4, characterized in that, The preset number of times is greater than or equal to 200.

6. The method according to claim 5, characterized in that, The first pre-defined adjustment rule is the gradient descent rule.

7. The method according to claim 6, characterized in that, The S35 includes: S351. If the first condition occurs, then obtain the average value of the new value of the first parameter vector w and the average value of the new value of the second parameter vector b corresponding to the new value of the current addressing parameter first value and the new value of the current addressing parameter second value, respectively. S352. Based on the data characteristics and the average value of the new values ​​of the first parameter vector w and the second parameter vector b corresponding to the new values ​​of the current first and second location parameters, the average value of the location probability PA corresponding to the average value of the new values ​​of the first and second location parameters is obtained by using the pre-set location model. S353. Based on the average value PA of the location probability, the average value of the location parameters corresponding to the average value PA of the location probability is obtained by using a pre-set calculation strategy; S354. Based on the current location parameter first value, the current location parameter second value, and the average value of the location parameters, select the optimal area in the city as a disaster relief material reserve station.

8. The method according to claim 7, characterized in that, The optimal areas in the city are the top k areas with the highest probability value corresponding to the selection parameter with the largest value among the current location parameter first value, the current location parameter first value, the current location parameter first value, the current location parameter new value, and the average value of the location parameters. Where 1≤k≤3.

Images