Geological disaster early warning model construction method and system, and storage medium

Abstract

The application discloses a kind of geological disaster early warning model construction method, system and storage medium, the method includes: the slope analysis of digital elevation model of research area, obtain the slope value of each digital grid;According to the slope value, administrative division data and susceptibility unit data of research area, the spatial reclassification and segmentation of multiple digital grids are carried out, and the disaster factor coefficient of each slope unit is calculated according to the disaster point information;According to meteorological data, the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit are calculated respectively by rainfall reduction algorithm;Based on slope factor coefficient, susceptibility factor coefficient, disaster factor coefficient, historical rainfall factor coefficient and forecast rainfall factor coefficient, the initial early warning model is iteratively trained, and the trained early warning model is used as geological disaster early warning model.The geological disaster early warning model constructed by the present application realizes real-time automatic analysis and prediction of the grade and range of geological disasters in the region.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geological disaster early warning, and particularly relates to a geological disaster early warning model construction method and system and a storage medium. BACKGROUND

[0002] Geological disasters seriously threaten people's lives and property safety, and the main causes of geological disasters are topography, geological structure and heavy rainfall. In the prior art, a monitoring area is manually divided, and a correlation relationship is constructed by monitoring equipment to obtain underground water level changes, so as to make geological disaster prediction and early warning. However, a large number of monitoring equipment needs to be constructed, resulting in high monitoring cost, and each sub-monitoring area needs to be independently designed, so the implementation process is complex. Therefore, how to reduce the cost of equipment while effectively monitoring and warning the geological disasters in the region has become a problem to be solved.

[0003] The above content is only used to assist in understanding the technical solutions of the present application and does not represent the acknowledgement of the above content as prior art. SUMMARY

[0004] The main purpose of the present application is to provide a geological disaster early warning model construction method, system and storage medium, which aims to solve the technical problem of how to reduce the cost of equipment while effectively monitoring and warning the geological disasters in the region.

[0005] To achieve the above purpose, the present application provides a geological disaster early warning model construction method, which comprises:

[0006] Performing slope analysis on a digital elevation model of a study area to obtain slope values of each digital grid in the digital elevation model;

[0007] According to the slope values, administrative division data and susceptibility unit data of the study area, spatial reclassification and segmentation are performed on a plurality of digital grids to obtain a plurality of slope units, each slope unit comprising a corresponding slope factor coefficient and a susceptibility factor coefficient;

[0008] Determine the disaster point information of each slope unit by spatial overlay analysis method, and calculate the disaster factor coefficient of each slope unit according to the disaster point information by weighted summation and average formula;

[0009] According to the meteorological data of the study area, the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit are calculated by rainfall reduction algorithm;

[0010] Iteratively train an initial early warning model based on the slope factor coefficient, the susceptibility factor coefficient, the disaster factor coefficient, the historical rainfall factor coefficient, and the forecast rainfall factor coefficient, and take the trained early warning model as a geological disaster early warning model.

[0011] Optionally, the slope analysis on the digital elevation model of the study area to obtain the slope value of each digital grid in the digital elevation model comprises:

[0012] Extracting the elevation value and the grid geographic span of each grid from the digital elevation model of the study area;

[0013] Performing slope analysis on each digital grid according to the elevation value and the grid geographic span respectively to obtain the slope value of each grid.

[0014] Optionally, the spatial reclassification and segmentation of a plurality of digital grids according to the slope value, the administrative division data of the study area, and the susceptibility unit data to obtain a plurality of slope units comprises:

[0015] Performing spatial reclassification on a plurality of digital grids according to the slope value to form a plurality of slope classification units;

[0016] Segmenting a plurality of slope classification units based on the administrative division data of the study area to obtain a plurality of fine-grained slope units;

[0017] Performing spatial segmentation on a plurality of fine-grained slope units according to the susceptibility unit data of the study area to obtain a plurality of slope units.

[0018] Optionally, the calculation of the disaster factor coefficient of each slope unit according to the disaster point information through a weighted summation and averaging formula comprises:

[0019] Extracting the number of disaster points, the grade value of each disaster point, and the disaster influence area of each slope unit from the disaster point information;

[0020] Calculating the disaster factor coefficient of each slope unit through a weighted summation and averaging formula according to the number of disaster points, the grade value of each disaster point, and the disaster influence area.

[0021] Optionally, the calculation of the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit according to the meteorological data of the study area through a rainfall reduction algorithm comprises:

[0022] Matching the rainfall grid corresponding to each slope unit through a spatial indexing method;

[0023] Determining the historical actual rainfall value and the forecast rainfall value of the rainfall grid corresponding to each slope unit according to the meteorological data of the study area;

[0024] The historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit are calculated respectively according to the historical real-time rainfall value and the forecast rainfall value of the rainfall grid corresponding to each slope unit by a rainfall reduction algorithm.

[0025] Optionally, the matching of the rainfall grid corresponding to each slope unit by the spatial index method comprises:

[0026] The start and end row and column index numbers of each slope unit and the start and end row and column index numbers of each rainfall grid under the standard index grid are calculated respectively according to the polygon position information of each slope unit and the rectangular position information of each rainfall grid;

[0027] The start and end row and column index numbers of each slope unit are associated with the index number of the corresponding slope unit;

[0028] The rainfall grid corresponding to each slope unit is matched by the spatial index method based on the index number of the slope unit and the start and end row and column index numbers of each rainfall grid.

[0029] Optionally, the calculation of the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit according to the historical real-time rainfall value and the forecast rainfall value of the rainfall grid corresponding to each slope unit by the rainfall reduction algorithm comprises:

[0030] The grid intersection area between each slope unit and the rainfall grid corresponding to each slope unit is calculated respectively;

[0031] The historical rainfall factor coefficient of each slope unit is calculated according to the historical real-time rainfall value of the rainfall grid, the polygon area corresponding to each slope unit, the grid intersection area and a rainfall attenuation coefficient by the rainfall reduction algorithm;

[0032] The forecast rainfall factor coefficient of each slope unit is calculated according to the forecast rainfall value of the rainfall grid, the polygon area corresponding to each slope unit, the grid intersection area and the rainfall attenuation coefficient by the rainfall reduction algorithm.

[0033] Optionally, the rainfall reduction algorithm is:

[0034]

[0035]

[0036]

[0037] In the formula, the total rainfall of the xth slope unit on the zth day is The area of the ith rainfall grid is The area of the xth slope unit is ​is a historical or predicted rainfall value of a rainfall grid on the zth day, is a rainfall attenuation coefficient, is a historical rainfall factor coefficient, is a predicted rainfall factor coefficient.

[0038] In addition, to achieve the above object, the present application also provides a geological disaster early warning model construction system, which comprises:

[0039] a terrain factor processing unit, configured to perform slope analysis on a digital elevation model of a study area to obtain slope values of digital grids in the digital elevation model;

[0040] a susceptibility factor processing unit, configured to perform spatial reclassification and segmentation on the digital grids according to the slope values, administrative division data of the study area and susceptibility unit data to obtain a plurality of slope units, each of which comprises a corresponding slope factor coefficient and a susceptibility factor coefficient;

[0041] a disaster factor processing unit, configured to determine disaster point information of each slope unit by a spatial overlay analysis method and calculate a disaster factor coefficient of each slope unit by a weighted summation and averaging formula according to the disaster point information;

[0042] a meteorological factor processing unit, configured to calculate historical and predicted rainfall factor coefficients of each slope unit by a rainfall reduction algorithm according to meteorological data of the study area;

[0043] a warning level calculation unit, configured to perform iterative training on an initial early warning model based on the slope factor coefficients, the susceptibility factor coefficients, the disaster factor coefficients, the historical rainfall factor coefficients and the predicted rainfall factor coefficients, and take the trained early warning model as a geological disaster early warning model.

[0044] In addition, to achieve the above object, the present application also provides a geological disaster early warning model construction device, which comprises a memory, a processor and a geological disaster early warning model construction program stored in the memory and executable on the processor, and the geological disaster early warning model construction program is configured to implement the steps of the geological disaster early warning model construction method as described above.

[0045] In addition, to achieve the above object, the present application also provides a storage medium having a geological disaster early warning model construction program stored thereon, and the geological disaster early warning model construction program is configured to implement the steps of the geological disaster early warning model construction method as described above when executed by a processor.

[0046] The application firstly performs slope analysis on a digital elevation model of a research area to obtain slope values of each digital grid in the digital elevation model, then performs spatial reclassification and segmentation on multiple digital grids according to the slope values, administrative division data of the research area and susceptibility unit data to obtain multiple slope units, each slope unit including a corresponding slope factor coefficient and a susceptibility factor coefficient, then determines disaster point information of each slope unit through spatial overlay analysis, calculates a disaster factor coefficient of each slope unit through a weighted summation and averaging formula according to the disaster point information, and respectively calculates a historical rainfall factor coefficient and a forecast rainfall factor coefficient of each slope unit through a rainfall reduction algorithm according to meteorological data of the research area, and finally iteratively trains an initial early warning model based on the slope factor coefficient, the susceptibility factor coefficient, the disaster factor coefficient, the historical rainfall factor coefficient and the forecast rainfall factor coefficient, and takes the trained early warning model as a geological disaster early warning model. Through the geological disaster early warning model constructed by the application, real-time data and forecast data pushed by a meteorological bureau can be automatically received, and automatic analysis and prediction of the grade and range of geological disasters in the region can be realized, thereby effectively monitoring and warning regional geological disasters. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 FIG. 1 is a flowchart of a geological disaster early warning model construction method according to an embodiment of the application.

[0048] Figure 2 FIG. 2 is a schematic diagram of spatial reclassification using a slope factor coefficient according to the geological disaster early warning model construction method of the first embodiment of the application.

[0049] Figure 3 FIG. 3 is a rainfall grid matching schematic diagram according to the geological disaster early warning model construction method of the first embodiment of the application.

[0050] Figure 4 FIG. 4 is a structural block diagram of a geological disaster early warning model construction system according to an embodiment of the application.

[0051] The implementation, functional features and advantages of the application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0052] It should be understood that the specific embodiments described herein are merely intended to explain the application and not to limit the application.

[0053] The application provides a geological disaster early warning model construction method, which is described with reference to Figure 1 , Figure 1 FIG. 1 is a flowchart of a geological disaster early warning model construction method according to an embodiment of the application.

[0054] In this embodiment, the geological disaster early warning model construction method includes the following steps:

[0055] Step S10: performing slope analysis on the digital elevation model of the study area to obtain slope values of each digital grid in the digital elevation model.

[0056] It is easy to understand that the execution subject of the embodiment can be a geological disaster early warning model construction system with functions of data processing, network communication and program running, or other computer devices with similar functions, and the embodiment is not limited thereto.

[0057] It should be noted that the study area can be a region selected by the user.

[0058] In a specific implementation, the digital elevation model of the study area can be directly called, and then the elevation values and grid geographic spans of each digital grid are extracted from the digital elevation model of the study area. The slope of each digital grid is analyzed according to the elevation value and the grid geographic span, and the slope value of each digital grid is obtained.

[0059] Further, the processing mode of analyzing the slope of each digital grid according to the elevation value and the grid geographic span to obtain the slope value of each digital grid is that the slope value of each digital grid is calculated according to the elevation value and the grid geographic span through a slope formula.

[0060] The slope formula is:

[0061]

[0062] In the formula, is the slope value of the jth digital grid, n is the number of adjacent digital grids of the current digital grid (generally 8), is the elevation value of the ith adjacent digital grid, is the elevation value of the current digital grid, is the geographic span of the grid.

[0063] Step S20: performing spatial reclassification and segmentation on a plurality of digital grids according to the slope value, administrative division data and susceptibility unit data of the study area to obtain a plurality of slope units, each slope unit including a corresponding slope factor coefficient and a susceptibility factor coefficient.

[0064] Further, the plurality of digital grids are spatially reclassified according to the slope value to form a plurality of slope classification units. The plurality of slope classification units are segmented based on the administrative division data of the study area to obtain a plurality of fine-grained slope units. The plurality of fine-grained slope units are spatially cut according to the susceptibility unit data of the study area to obtain a plurality of slope units.

[0065] It should be noted that the administrative division data of the study area can be village-level administrative region range data.

[0066] The susceptibility is generally related to the geological structure, and in an actual application scenario, spatial range data of a research area is investigated, and each susceptibility unit has a corresponding susceptibility coefficient value (i.e., susceptibility unit data).

[0067] Reference Figure 2 , Figure 2 FIG. 1 is a schematic diagram of a method for constructing a geological disaster early warning model according to an embodiment of the present application, which shows the spatial reclassification using a slope factor coefficient. According to the slope value of each digital grid, different levels can be divided according to the precision requirement set by a user, for example, any value within 1-90 can be divided into one level (1-10), two levels (11-20), and so on. The continuous digital grids in the same level are divided into slope classification units of the level, Figure 2 The three digital grids in the purple box are one level, the four digital grids in the green box are two levels, and the three digital grids in the red box are one level. The purple, green, and red boxes are divided into corresponding slope classification units, respectively.

[0068] In this embodiment, the reclassified slope coefficient range data (i.e., multiple slope classification units) is cut using village-level administrative area range data (i.e., administrative division data) to form multiple fine-grained slope units (polygons, non-rectangular grids).

[0069] The multiple fine-grained slope units are further spatially cut according to the susceptibility coefficient values corresponding to the susceptibility units of each village-level administrative area to form more fine-grained grid units (i.e., slope units) containing susceptibility factor coefficient attributes and slope factor coefficient attributes, is the susceptibility factor coefficient of the xth slope unit. is the slope factor coefficient of the xth slope unit.

[0070] Step S30: Determine the disaster point information of each slope unit by the spatial superposition analysis method, and calculate the disaster factor coefficient of each slope unit by a weighted summation and averaging formula according to the disaster point information.

[0071] In a specific implementation, the disaster factor is determined by local investigation of disaster hidden danger points. There are multiple geological disaster hidden danger points in a region, and each hidden danger point has a risk level. In this embodiment, the disaster factor coefficient needs to be distributed to the corresponding slope unit.

[0072] According to the spatial superposition analysis method, the disaster points are distributed to the slope units in which they are located according to the spatial relationship:

[0073]

[0074] In the formula, is the xth slope unit, and the spatial intersection of the disaster point and the xth slope unit, represents the point of non-empty intersection (indicating within the slope unit).

[0075] Further, the processing mode of calculating the disaster factor coefficient of each slope unit according to the disaster point information through the weighted summation average formula is as follows: the number of disaster points in each slope unit, the grade value of each disaster point, and the disaster influence area are extracted from the disaster point information; the disaster factor coefficient of each slope unit is calculated through the weighted summation average formula according to the number of disaster points, the grade value of each disaster point, and the disaster influence area.

[0076] The weighted summation average formula is:

[0077]

[0078] In the formula, is the disaster factor coefficient of the xth slope unit, n is the number of disaster points in the slope unit, is the disaster grade value of the ith disaster point, is the disaster influence area of the ith disaster point.

[0079] Step S40: Calculate the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit respectively through the rainfall reduction algorithm according to the meteorological data of the study area.

[0080] Further, the corresponding rainfall grid of each slope unit is matched through the spatial index method; the historical real-time rainfall value and the forecast rainfall value of the corresponding rainfall grid of each slope unit are determined according to the meteorological data of the study area; the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit are calculated respectively through the rainfall reduction algorithm based on the historical real-time rainfall value and the forecast rainfall value of the corresponding rainfall grid of each slope unit.

[0081] The operation process of the present embodiment can take 15-day historical real-time and 24-hour forecast rainfall data, distribute the rainfall data to the corresponding slope unit, and form two meteorological factor coefficients: the historical rainfall factor coefficient and the forecast rainfall factor coefficient.

[0082] It should be noted that the historical real-time rainfall value can be grid data with a spatial resolution of 1 km and a time resolution of 1 hour, and the forecast rainfall value can be grid data with a spatial resolution of 2.5 km and a time resolution of 3 hours, which is not limited by the present embodiment.

[0083] The processing mode of matching the rainfall grid corresponding to each slope unit by the spatial index method is as follows: the start and end row and column index numbers of each slope unit in the standard index grid and the start and end row and column index numbers of each rainfall grid are calculated according to the polygon position information of each slope unit and the rectangular position information of each rainfall grid; the start and end row and column index numbers of each slope unit are associated with the corresponding slope unit index number; and the corresponding rainfall grid of each slope unit is matched by the spatial index method based on the slope unit index number and the start and end row and column index numbers of each rainfall grid.

[0084] In a specific implementation, all slope units (polygons, non-standard matrix grids) are traversed to construct a fast spatial query grid.

[0085] According to the polygon position information of each slope unit, the start and end row and column index numbers of each slope unit in the standard index grid are calculated:

[0086]

[0087]

[0088]

[0089]

[0090] In the formula, , are the start and end column index numbers of the polygon of a certain slope unit, , are the start and end row index numbers of the polygon of a certain slope unit, , are the spans of the slope unit in the x and y directions respectively, , , , are the coordinate ranges of the outer rectangle of the slope unit polygon respectively.

[0091] An effective query grid is added to the cache, and the slope grid number in the index is recorded (only the number is recorded here, the purpose is to sacrifice part of the performance and reduce the memory usage), and the data is stored in a Map container:

[0092]

[0093] In the formula, the first layer key is the row index number, the second layer key is the column index number, and the third layer key is the slope unit index number, The object storage extension attribute value.

[0094] It should be noted that the number of the possible associated slope polygon is quickly queried from the Map container through the calculated row and column numbers, and the geometric object of the slope polygon can be obtained through the number, without the need of loop comparison, thereby improving the search efficiency.

[0095] Further, the rain grid data is traversed, each rain grid is a rectangle, four coordinate values of the rectangular frame are included, according to the four coordinate values, the query grid position where the grid data is located can be calculated by calling the calculation query grid index range function again:

[0096]

[0097]

[0098]

[0099]

[0100] In the formula, 、 is the start and end column index number of a certain rain grid, 、 is the start and end row index number of a certain rain grid, 、 is the span of the rain grid in the x and y directions respectively, 、 、 、 is the range coordinate value of the rain grid.

[0101] In the embodiment, reference is made to Figure 3 is a rain grid matching schematic diagram of the first embodiment of the construction method of the geological disaster early warning model, the geometric object (i.e. the black solid line of the polygon) of the slope polygon can be obtained through the number, and the start and end row and column index numbers of the slope unit polygon in the standard index grid (i.e. the background dotted grid) are determined, then the rain grid (i.e. the red solid line) can be matched according to the start and end row and column index numbers.

[0102] Further, the processing mode of calculating the historical rainfall factor coefficient and the predicted rainfall factor coefficient of each slope unit based on the historical actual rainfall value and the predicted rainfall value of the rainfall grid corresponding to each slope unit through the rainfall reduction algorithm is as follows: the grid intersection area between each slope unit and the rainfall grid corresponding to each slope unit is calculated; the historical rainfall factor coefficient of each slope unit is calculated through the rainfall reduction algorithm based on the historical actual rainfall value of the rainfall grid, the polygon area corresponding to each slope unit, the grid intersection area, and the rainfall attenuation coefficient; and the predicted rainfall factor coefficient of each slope unit is calculated through the rainfall reduction algorithm based on the predicted rainfall value of the rainfall grid, the polygon area corresponding to each slope unit, the grid intersection area, and the rainfall attenuation coefficient.

[0103] The rainfall reduction algorithm is as follows:

[0104]

[0105]

[0106]

[0107] In the formula, is the total rainfall of the xth slope unit on the zth day, is the area of the ith rainfall grid, is the area of the xth slope unit, is the historical actual rainfall or predicted rainfall value of the rainfall grid on the zth day, is the rainfall attenuation coefficient (empirical value), is the historical rainfall factor coefficient, is the predicted rainfall factor coefficient.

[0108] Step S50: iteratively training the initial warning model based on the slope factor coefficient, the susceptibility factor coefficient, the disaster factor coefficient, the historical rainfall factor coefficient, and the predicted rainfall factor coefficient, and taking the trained warning model as the geological disaster warning model.

[0109] In the present embodiment, after the values of the above terrain influence factors (i.e., the slope factor coefficient), the susceptibility influence factors (i.e., the susceptibility factor coefficient), the disaster influence factors (i.e., the disaster factor coefficient), and the meteorological influence factors (i.e., the historical rainfall factor coefficient and the predicted rainfall factor coefficient) are obtained, the final warning level can be calculated according to matrix conversion.

[0110] However, the method of determining the conversion matrix through investigation has a huge workload and extremely high cost, and therefore the present scheme provides a method of machine learning to further extract the conversion matrix of each slope unit, and the specific implementation is as follows:

[0111] First, the mapping relationship between the early warning level and each factor is constructed, which can be achieved by using a multi-layer matrix method, and the fitting effect can be improved by increasing the matrix parameters.

[0112] For each slope grid, five influence factor values can be calculated, and the mapping relationship is as follows:

[0113]

[0114] In the formula, , , , , are the influence factor values, is the nth parameter of the jth layer matrix.

[0115] Then, combined with historical meteorological data, samples are generated according to the historical disaster situation of each slope unit and the historical results adjusted by experts.

[0116] Third, the initial values of the multi-layer matrix are set, the factor values of the samples are input into the model (note that each layer contains an activation function), and the predicted values are obtained.

[0117] Fourth, according to the final results adjusted by experts or the actual geological disaster results , the loss value is calculated:

[0118]

[0119] In the formula, n is the number of samples, is the predicted value of the ith sample (i.e., the ith slope unit), is the true value of the ith sample.

[0120] Fifth, the model performs back propagation, and gradually adjusts the parameters of each layer according to the loss value and learning rate.

[0121] Sixth, multiple iterations are trained to obtain the optimal solution to obtain the geological disaster early warning model.

[0122] In this embodiment, the system automatically receives the daily meteorological data of the predicted area, calculates the corresponding influence factors (i.e., slope factor coefficient, susceptibility factor coefficient, disaster factor coefficient, historical rainfall factor coefficient, and predicted rainfall factor coefficient), and then inputs the influence factors into the geological disaster early warning model to output the early warning result level of the predicted area. After the early warning result analysis is completed, the early warning consultation module is started, and the system pushes the analysis result to the meteorological experts. The meteorological experts consult with the pre-judgment on the map, video monitoring, and field investigation to determine the result and send the final achievement.

[0123] In the embodiment, first, slope analysis is performed on a digital elevation model of a research area to obtain slope values of digital grids in the digital elevation model, then multiple digital grids are spatially reclassified and segmented according to the slope values, administrative division data of the research area and susceptibility unit data to obtain multiple slope units, each slope unit includes a corresponding slope factor coefficient and a susceptibility factor coefficient, thereafter, disaster point information of each slope unit is determined through a spatial overlay analysis method, a disaster factor coefficient of each slope unit is calculated according to the disaster point information through a weighted summation and averaging formula, and historical rainfall factor coefficients and forecast rainfall factor coefficients of each slope unit are respectively calculated according to meteorological data of the research area through a rainfall reduction algorithm, finally, an initial warning model is iteratively trained based on the slope factor coefficient, the susceptibility factor coefficient, the disaster factor coefficient, the historical rainfall factor coefficient and the forecast rainfall factor coefficient, and the trained warning model is taken as a geological disaster warning model. The geological disaster warning model constructed in the embodiment can automatically receive real-time data and forecast data pushed by a meteorological bureau, and then automatically analyze and predict the grade and range of geological disasters in the region, thereby effectively monitoring and warning regional geological disasters.

[0124] Reference Figure 4 , Figure 4 FIG. 1 is a structural block diagram of a construction system of a geological disaster warning model according to an embodiment of the present application.

[0125] As shown in FIG. 1, the construction system of the geological disaster warning model according to the embodiment of the present application includes: Figure 4 A terrain factor processing unit 5001 is configured to perform slope analysis on a digital elevation model of a research area to obtain slope values of digital grids in the digital elevation model.

[0126] A susceptibility factor processing unit 5002 is configured to spatially reclassify and segment multiple digital grids according to the slope values, administrative division data of the research area and susceptibility unit data to obtain multiple slope units, each slope unit including a corresponding slope factor coefficient and a susceptibility factor coefficient.

[0127] A disaster factor processing unit 5003 is configured to determine disaster point information of each slope unit through a spatial overlay analysis method, and calculate a disaster factor coefficient of each slope unit according to the disaster point information through a weighted summation and averaging formula.

[0128] A meteorological factor processing unit 5004 is configured to calculate historical rainfall factor coefficients and forecast rainfall factor coefficients of each slope unit according to meteorological data of the research area through a rainfall reduction algorithm.

[0129]

[0130] ​The early warning level calculation unit 5005 is configured to perform iterative training on an initial early warning model based on the slope factor coefficient, the susceptibility factor coefficient, the disaster factor coefficient, the historical rainfall factor coefficient and the predicted rainfall factor coefficient, and take the trained early warning model as the geological disaster early warning model.

[0131] Other embodiments or specific implementations of the construction system of the geological disaster early warning model can refer to the above-mentioned method embodiments, which will not be described here.

[0132] It should be noted that in this paper, the term "include", "contain" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or includes elements inherent to such process, method, article or system. Without more limitations, the element defined by the sentence "includes a" does not exclude the presence of other identical elements in the process, method, article or system including the element.

[0133] The above-mentioned embodiment number of the present application is only for description, not representing the advantages and disadvantages of the embodiments.

[0134] Through the above description of the embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment method can be realized by means of software and necessary general hardware platform, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment. Based on such understanding, the technical solutions of the present application or the part of the prior art which contributes essentially can be embodied in the form of software product, which is stored in a storage medium (such as read-only memory / random access memory, magnetic disk, optical disk), including a plurality of instructions for making a terminal device (which can be a mobile phone, computer, server or network device) execute the method described in each embodiment of the present application.

[0135] The above is only the preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation made by using the content of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for constructing a geological disaster early warning model, characterized in that: The method comprises the following steps: Performing slope analysis on the digital elevation model of the study area to obtain the slope value of each digital grid in the digital elevation model; Spatially reclassifying and segmenting the plurality of digital grids according to the slope value, the administrative division data of the study area, and the susceptibility unit data to obtain a plurality of slope units, each slope unit including a corresponding slope factor coefficient and a susceptibility factor coefficient; Determine the disaster point information of each slope unit by using a spatial superposition analysis method, and calculate the disaster factor coefficient of each slope unit by using a weighted summation and averaging formula based on the disaster point information; According to the meteorological data of the study area, the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit are calculated by using the rainfall reduction algorithm; Iteratively training the initial early warning model based on the slope factor coefficient, the susceptibility factor coefficient, the disaster factor coefficient, the historical rainfall factor coefficient and the forecast rainfall factor coefficient, and using the trained early warning model as a geological disaster early warning model; The method of calculating the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit by using a rainfall reduction algorithm based on the meteorological data of the study area includes: The rainfall grid corresponding to each slope unit is matched by spatial index method; Determine the historical actual rainfall value and the forecast rainfall value of the rainfall grid corresponding to each slope unit according to the meteorological data of the study area; Based on the rainfall grid corresponding to each slope unit, the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit are calculated respectively by a rainfall reduction algorithm according to the historical actual rainfall value and the forecast rainfall value of the rainfall grid; The rainfall reduction algorithm is: Where, is the total rainfall on the zth day of the xth slope unit, is the i-th rainfall grid corresponding to the x-th slope unit, is the xth slope unit, is the historical actual rainfall or forecast rainfall value of the rainfall grid on day z, is the rainfall attenuation coefficient, is the historical rainfall factor coefficient, is the forecast rainfall factor coefficient, area is the area function.

2. The method according to claim 1, wherein The slope analysis of the digital elevation model of the study area to obtain the slope value of each digital grid in the digital elevation model includes: Extract the elevation value and geographic span of each digital grid from the digital elevation model of the study area; Slope analysis is performed on each digital grid according to the elevation value and the grid geographical span to obtain the slope value of each digital grid.

3. The method according to claim 1, wherein The spatial reclassification and segmentation of multiple digital grids according to the slope value, the administrative division data of the study area, and the susceptibility unit data to obtain multiple slope units include: spatially reclassifying the plurality of digital grids according to the slope values ​​to form a plurality of slope classification units; Segmenting multiple slope classification units based on administrative division data of the study area to obtain multiple fine-grained slope units; According to the susceptibility unit data of the study area, multiple fine-grained slope units are spatially cut to obtain multiple slope units.

4. The method according to claim 1, wherein The hazard factor coefficient of each slope unit is calculated by a weighted summation and averaging formula according to the hazard point information, including: Extracting the number of disaster points in each slope unit, the level value of each disaster point and the disaster-affected area from the disaster point information; The disaster factor coefficient of each slope unit is calculated according to the number of disaster points, the level value of each disaster point and the disaster affected area through a weighted summation and averaging formula.

5. The method according to claim 1, wherein The matching of the rainfall grid corresponding to each slope unit by the spatial index method includes: Calculate the starting and ending row and column index numbers of each slope unit and each rainfall grid under the standard index grid according to the polygonal position information of each slope unit and the rectangular position information of each rainfall grid; Associating the starting and ending row and column index numbers of each slope unit with the corresponding slope unit index number; The rainfall grid corresponding to each slope unit is matched by a spatial index method based on the slope unit index number and the start and end row and column index numbers of each rainfall grid.

6. The method according to claim 1, wherein The method of calculating the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit respectively based on the rainfall grid corresponding to each slope unit according to the historical actual rainfall value and the forecast rainfall value of the rainfall grid by using a rainfall reduction algorithm includes: Calculate the grid intersection area between each slope unit and the rainfall grid corresponding to each slope unit; Calculate the historical rainfall factor coefficient of each slope unit by a rainfall reduction algorithm based on the historical actual rainfall of the rainfall grid, the polygon area corresponding to each slope unit, the grid intersection area and the rainfall attenuation coefficient; The forecast rainfall factor coefficient of each slope unit is calculated by the rainfall reduction algorithm according to the forecast rainfall value of the rainfall grid, the polygon area corresponding to each slope unit, the grid intersection area and the rainfall attenuation coefficient.

7. A system for constructing a geological disaster early warning model, characterized in that: The system comprises: A terrain factor processing unit is used to perform slope analysis on the digital elevation model of the study area to obtain the slope value of each digital grid in the digital elevation model; a susceptibility factor processing unit, configured to spatially reclassify and segment the plurality of digital grids according to the slope value, the administrative division data of the study area, and the susceptibility unit data, to obtain a plurality of slope units, each of which includes a corresponding slope factor coefficient and a susceptibility factor coefficient; A disaster factor processing unit is used to determine the disaster point information of each slope unit through a spatial superposition analysis method, and calculate the disaster factor coefficient of each slope unit through a weighted summation and averaging formula based on the disaster point information; A meteorological factor processing unit is used to calculate the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit respectively through a rainfall reduction algorithm according to the meteorological data of the study area; a warning level calculation unit, configured to iteratively train an initial warning model based on the slope factor coefficient, the susceptibility factor coefficient, the disaster factor coefficient, the historical rainfall factor coefficient, and the forecast rainfall factor coefficient, and use the trained warning model as a geological disaster warning model; The meteorological factor processing unit is further configured to match the rainfall grid corresponding to each slope unit using a spatial index method; determine the historical actual rainfall value and the forecast rainfall value of the rainfall grid corresponding to each slope unit based on the meteorological data of the study area; and calculate the historical rainfall factor coefficient and the forecast rainfall factor coefficient of each slope unit using a rainfall reduction algorithm based on the historical actual rainfall value and the forecast rainfall value of the rainfall grid corresponding to each slope unit. The rainfall reduction algorithm is: Where, is the total rainfall on the zth day of the xth slope unit, is the i-th rainfall grid corresponding to the x-th slope unit, is the xth slope unit, is the historical actual rainfall or forecast rainfall value of the rainfall grid on day z, is the rainfall attenuation coefficient, is the historical rainfall factor coefficient, is the forecast rainfall factor coefficient, area is the area function.

8. A storage medium, characterized in that: The storage medium stores a program for constructing a geological hazard warning model. When the program for constructing a geological hazard warning model is executed by a processor, the steps of the method for constructing a geological hazard warning model as described in any one of claims 1 to 6 are implemented.

Images