A method and system for predicting rainstorm disaster risks in mountainous scenic areas

Abstract

This invention provides a method and system for predicting rainstorm disaster risks in mountainous scenic areas. The current method includes: dividing the tourist route topology network into several continuous tourist route segment units, and matching corresponding risk-prone terrain parameters for each tourist route segment unit; deriving the temporal transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units, generating disaster risk time-series curves for each tourist route segment unit; collecting real-time personnel location data within the scenic area, extracting personnel movement characteristic parameters, and deriving the spatiotemporal distribution trajectory sequence of personnel in each tourist route segment unit based on the tourist route topology network; obtaining the quantitative value of personnel disaster risk in each tourist route segment unit under the corresponding time series; and generating a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated periodically based on the correlation between the quantitative value of personnel disaster risk and the time series of all tourist route segment units. This invention can significantly improve the efficiency of rainstorm disaster risk prediction in scenic areas.

Description

Technical Field

[0001] This invention relates to the field of meteorological forecasting technology, and in particular to a method and system for predicting the risk of rainstorm disasters in mountainous scenic areas. Background Technology

[0002] Mountainous scenic areas, characterized by dramatic terrain and frequent localized convective weather, are high-risk regions for torrential rains and the resulting chain reactions of disasters such as flash floods, landslides, and mudslides. Furthermore, the high concentration of tourists and the high mobility of people in these areas significantly alter the exposure characteristics of disaster-bearing bodies compared to ordinary mountainous watersheds. Therefore, accurate prediction of torrential rain disaster risks is a core technological support for ensuring the safe operation of these scenic areas. Currently, torrential rain disaster risk prediction for mountainous scenic areas generally adopts a technical system applicable to general mountainous watersheds, using natural sub-watersheds or 100-meter-level grids as the basic calculation unit. It primarily focuses on risk assessment of natural disaster-causing factors such as rainfall, topography, soil, and vegetation. This overall technical approach is designed for general scenarios such as regional flood control and geological disaster prevention, without being adapted and optimized for the specific management needs of mountainous scenic areas.

[0003] Existing technical solutions exhibit a significant disconnect between their spatial prediction granularity and the actual control scale of scenic areas. Their spatial resolution is mostly in the range of hundreds of meters to kilometers, failing to cover core control units within the scenic area, such as trail segments, viewing platforms, gully entrances and exits, and cliff-side gathering points, which are typically within the ten-meter range. Furthermore, general terrain parameter extraction algorithms are insufficient in characterizing steep and micro-topography, easily overlooking localized disaster-causing hazards such as small gullies and steep slopes. This makes it difficult for the location results of high-risk areas to correspond with the actual control targets within the scenic area, thus failing to support refined risk control at the point-to-point level.

[0004] Furthermore, existing risk prediction models mostly consider only the static hazards of natural disaster-causing events, failing to incorporate dynamic exposure factors such as real-time tourist distribution and population density into the core calculation logic of risk quantification. They can only output a uniform risk level for the entire region or watershed, unable to quantify the actual disaster risk to people in different spatial locations. Consequently, the generated warning signals lack specificity and struggle to match the differentiated emergency response needs of scenic areas, directly hindering the efficiency of tourist evacuation and the accuracy of emergency response during rainstorm disasters, indicating significant room for technological improvement. Summary of the Invention

[0005] Based on this, the purpose of this invention is to provide a method and system for predicting rainstorm disaster risks in mountainous scenic areas, so as to solve the problem that the existing technology can only output a uniform risk level at the whole area or watershed level, and cannot quantify the actual disaster risk of people in different spatial locations, resulting in insufficient targeting of the generated early warning signals.

[0006] The first aspect of the present invention proposes: A method for predicting rainstorm disaster risk in mountainous scenic areas, wherein the method includes: Based on the distribution of trails and the orientation of terrain corridors in mountainous scenic areas, a topological network of tourist routes is constructed. The topological network of tourist routes is divided into several continuous segment units, and corresponding risky terrain parameters are matched for each segment unit. Based on real-time rainfall monitoring data, disaster triggering points within the scenic area are identified, and the time-series transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units is deduced, generating disaster risk time-series curves for each tourist route segment unit. Collect real-time personnel location data within the scenic area, extract personnel movement characteristic parameters, and deduce the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network; By performing spatiotemporal registration calculations on the disaster risk time series curve and the spatiotemporal distribution trajectory sequence of personnel under the same time series, the quantitative value of the disaster risk of personnel in each walkway segment unit under the corresponding time series is obtained; Based on the quantitative value of disaster risk to personnel in all tourist route segments and the temporal correlation, a dynamic risk distribution map of rainstorm disasters in the scenic area is generated and updated time by time.

[0007] The beneficial effects of this invention are as follows: This solution addresses the shortcomings of existing technologies, such as the inability to quantify disaster risks to people in different spatial locations and insufficient targeted early warning. By constructing a topological network of hiking trails and terrain corridors that conforms to the orientation of mountainous scenic areas, and dividing continuous hiking trails into segmented units and matching them with corresponding risk-prone terrain parameters, this solution breaks away from the extensive model of uniform risk rating across the entire area, achieving precise binding between risk assessment and actual hiking routes. It generates segmented disaster risk time-series curves by simulating the temporal transmission process of disaster-causing factors along terrain corridors, reconstructing the spatiotemporal patterns of the dynamic evolution of rainstorm disasters. Combining real-time personnel positioning data to extract movement characteristics and simulating the spatiotemporal distribution trajectory of personnel, it calculates the quantified disaster risk value for each segment through spatiotemporal registration, achieving precise quantification of the actual disaster risk at different spatial locations. Finally, it generates a dynamic risk distribution map updated periodically, providing refined basis for rainstorm disaster early warning in scenic areas by road segment and time sequence, significantly improving the targetedness and timeliness of early warning signals, and supporting scenic areas in accurately carrying out personnel evacuation and risk management.

[0008] Furthermore, the steps of identifying disaster triggering points within the scenic area based on real-time rainfall monitoring data and deducing the temporal transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units include: Real-time rainfall data is matched with the risky terrain parameters of each travel segment unit using threshold matching to identify locations that meet the disaster triggering conditions as disaster triggering source points and determine the corresponding disaster type. Based on topographic slope, confluence direction and surface roughness parameters, the transmission velocity of disaster-causing factors along the topographic corridor and the intensity attenuation coefficient along the way are calculated. Based on the spatial intersection relationship between the travel route segment units and the terrain corridor, the time nodes and corresponding disaster intensity of the disaster-causing factors reaching each travel route segment unit are deduced, and the disaster risk time series curves corresponding to each travel route segment unit are generated.

[0009] Furthermore, the step of generating the disaster risk time series curve corresponding to each walkway segment unit includes: When the same travel segment unit corresponds to multiple disaster triggering sources, the time node and disaster intensity of the disaster-causing factor corresponding to each disaster triggering source unit are extracted respectively; Based on the disaster type, the superposition coupling coefficient between different disaster-causing factors is determined, and the multi-source disaster intensity at the same time point is coupled and calculated. Arrange the coupled disaster intensity values ​​in chronological order to generate the disaster risk time series curve corresponding to the current traveler segment unit.

[0010] Furthermore, the steps of collecting real-time personnel location data within the scenic area, extracting personnel movement feature parameters, and deducing the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network include: The real-time personnel location data is matched with the travel route to determine the current travel route segment unit and the direction of travel for each person; Historical movement speed, duration of stay at attractions, and turning probability at tourist route nodes are extracted as movement feature parameters. Based on the connectivity of the walkway topology network, the distribution of the number of people in each walkway segment unit in future time periods is inferred according to the movement characteristic parameters and the Markov chain model, and a sequence of spatiotemporal distribution trajectories of people is generated.

[0011] Furthermore, the step of performing line matching on the real-time collected personnel positioning data includes: Acquire ticket gate data, video crowd flow statistics, and mobile location data, and perform spatial coordinate correction and timestamp alignment respectively; The walk-line buffer matching algorithm is used to map various personnel data to the corresponding walk-line segment units and eliminate abnormal positioning points that exceed the walk-line buffer range; Weighted verification is performed on multi-source personnel statistics within the same travel segment unit to obtain the real-time personnel base number for each travel segment unit.

[0012] Furthermore, the step of performing spatiotemporal registration calculation between the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence under the same time series includes: Based on a preset time step, the disaster intensity value and the number of people at the corresponding time step are extracted from the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence, respectively. Extract the risk avoidance space capacity parameters of the corresponding travel line segment unit, and calculate the personnel retention risk coefficient of the current segment in combination with the personnel number value; The disaster intensity value and the personnel stranded risk coefficient are nonlinearly coupled to calculate the quantitative value of personnel disaster risk in the corresponding walk-line segment unit at the current time step.

[0013] Furthermore, the step of generating a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated periodically based on the quantitative value of the disaster risk of personnel in all tourist route segment units and the temporal correlation includes: The risk level of personnel in each segment of the travel route is classified according to the preset risk level threshold, and the corresponding risk level is marked. Map the tourist route segments with risk level labels to the scenic area's geographical base map, and mark the expected duration and peak arrival time of each segment's risk level. The disaster risk quantification value of personnel in all travel line segment units is iteratively calculated according to the preset update frequency to generate a dynamic risk distribution map that is updated in time period.

[0014] The second aspect of the present invention proposes: A rainstorm disaster risk prediction system for mountainous scenic areas, wherein the system includes: The module is used to construct a tour route topology network based on the distribution of trails and the orientation of terrain corridors in mountainous scenic areas. The tour route topology network is divided into several continuous tour route segment units, and corresponding risky terrain parameters are matched for each tour route segment unit. The simulation module is used to identify disaster triggering points in the scenic area based on real-time rainfall monitoring data, and to simulate the time-series transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units, generating disaster risk time-series curves for each tourist route segment unit. The data acquisition module is used to collect real-time personnel location data within the scenic area, extract personnel movement characteristic parameters, and deduce the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network. The registration module is used to perform spatiotemporal registration calculations on the disaster risk time series curve and the spatiotemporal distribution trajectory sequence of personnel under the same time series, so as to obtain the quantitative value of the disaster risk of personnel in each walkway segment unit under the corresponding time series. The generation module is used to generate a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated time by time based on the quantitative value of the disaster risk of personnel in all tourist route segment units and the temporal correlation.

[0015] Furthermore, the steps of identifying disaster triggering points within the scenic area based on real-time rainfall monitoring data and deducing the temporal transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units include: Real-time rainfall data is matched with the risky terrain parameters of each travel segment unit using threshold matching to identify locations that meet the disaster triggering conditions as disaster triggering source points and determine the corresponding disaster type. Based on topographic slope, confluence direction and surface roughness parameters, the transmission velocity of disaster-causing factors along the topographic corridor and the intensity attenuation coefficient along the way are calculated. Based on the spatial intersection relationship between the travel route segment units and the terrain corridor, the time nodes and corresponding disaster intensity of the disaster-causing factors reaching each travel route segment unit are deduced, and the disaster risk time series curves corresponding to each travel route segment unit are generated.

[0016] Furthermore, the step of generating the disaster risk time series curve corresponding to each walkway segment unit includes: When the same travel segment unit corresponds to multiple disaster triggering sources, the time node and disaster intensity of the disaster-causing factor corresponding to each disaster triggering source unit are extracted respectively; Based on the disaster type, the superposition coupling coefficient between different disaster-causing factors is determined, and the multi-source disaster intensity at the same time point is coupled and calculated. Arrange the coupled disaster intensity values ​​in chronological order to generate the disaster risk time series curve corresponding to the current traveler segment unit.

[0017] Furthermore, the steps of collecting real-time personnel location data within the scenic area, extracting personnel movement feature parameters, and deducing the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network include: The real-time personnel location data is matched with the travel route to determine the current travel route segment unit and the direction of travel for each person; Historical movement speed, duration of stay at attractions, and turning probability at tourist route nodes are extracted as movement feature parameters. Based on the connectivity of the walkway topology network, the distribution of the number of people in each walkway segment unit in future time periods is inferred according to the movement characteristic parameters and the Markov chain model, and a sequence of spatiotemporal distribution trajectories of people is generated.

[0018] Furthermore, the step of performing line matching on the real-time collected personnel positioning data includes: Acquire ticket gate data, video crowd flow statistics, and mobile location data, and perform spatial coordinate correction and timestamp alignment respectively; The walk-line buffer matching algorithm is used to map various personnel data to the corresponding walk-line segment units and eliminate abnormal positioning points that exceed the walk-line buffer range; Weighted verification is performed on multi-source personnel statistics within the same travel segment unit to obtain the real-time personnel base number for each travel segment unit.

[0019] Furthermore, the step of performing spatiotemporal registration calculation between the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence under the same time series includes: Based on a preset time step, the disaster intensity value and the number of people at the corresponding time step are extracted from the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence, respectively. Extract the risk avoidance space capacity parameters of the corresponding travel line segment unit, and calculate the personnel retention risk coefficient of the current segment in combination with the personnel number value; The disaster intensity value and the personnel stranded risk coefficient are nonlinearly coupled to calculate the quantitative value of personnel disaster risk in the corresponding walk-line segment unit at the current time step.

[0020] Furthermore, the step of generating a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated periodically based on the quantitative value of the disaster risk of personnel in all tourist route segment units and the temporal correlation includes: The risk level of personnel in each segment of the travel route is classified according to the preset risk level threshold, and the corresponding risk level is marked. Map the tourist route segments with risk level labels to the scenic area's geographical base map, and mark the expected duration and peak arrival time of each segment's risk level. The disaster risk quantification value of personnel in all travel line segment units is iteratively calculated according to the preset update frequency to generate a dynamic risk distribution map that is updated in time period.

[0021] The third aspect of the present invention proposes: A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the rainstorm disaster risk prediction method for mountainous scenic areas as described above.

[0022] The fourth aspect of the present invention proposes: A readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, implements the rainstorm disaster risk prediction method for mountainous scenic areas as described above.

[0023] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0024] Figure 1 A flowchart of a method for predicting rainstorm disaster risks in mountainous scenic areas provided in the first embodiment of the present invention; Figure 2 The structural block diagram of the rainstorm disaster risk prediction system for mountainous scenic areas provided in the third embodiment of the present invention.

[0025] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0026] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0027] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0029] Please see Figure 1 The image shows a method for predicting rainstorm disaster risks in mountainous scenic areas according to the first embodiment of the present invention. The method for predicting rainstorm disaster risks in mountainous scenic areas provided in this embodiment can ultimately generate a dynamic risk distribution map that is updated every time period, providing a refined basis for rainstorm disaster early warning in scenic areas by road segment and time sequence, significantly improving the pertinence and timeliness of early warning signals, and supporting scenic areas to accurately carry out personnel evacuation and risk management.

[0030] Specifically, this embodiment provides: A method for predicting rainstorm disaster risk in mountainous scenic areas, wherein the method includes: Step S10: Construct a tour route topology network based on the distribution of trails and the orientation of terrain corridors in the mountainous scenic area, divide the tour route topology network into several continuous tour route segment units, and match corresponding risky terrain parameters for each tour route segment unit. It should be noted that, based on high-precision surveying and mapping of the scenic area's trail vector data, and combined with the natural orientation of the terrain corridors, a linear trail topology network was constructed to replace the traditional planar grid as the basic unit for risk assessment. According to characteristic points such as terrain slope change points, catchment area boundaries, scenic spot nodes, and intersections, the continuous trail network was divided into 92 homogeneous trail segment units, with each segment's length controlled between 100-300 meters to ensure relatively consistent terrain conditions and disaster exposure characteristics within each segment. Simultaneously, corresponding risk-prone terrain parameters were matched to each segment, covering eight core indicators: elevation, slope, aspect, runoff accumulation, topographic humidity index, vertical distance from valleys, slope height, and surface vegetation coverage, comprehensively characterizing the disaster-prone conditions of each trail segment. The disaster risk of mountainous scenic areas is highly dependent on the distribution of trails, with all tourist activities concentrated along these trails. Using trail segments as assessment units directly anchors the spatial carriers of personnel exposure, avoiding the problem of planar assessments being disconnected from actual personnel distribution, and also providing a linear spatial reference for subsequent disaster transmission calculations.

[0031] Step S20: Identify disaster triggering points within the scenic area based on real-time rainfall monitoring data, and deduce the temporal transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units, generating disaster risk time series curves for each tourist route segment unit; It should be noted that real-time minute-level rainfall monitoring data from automatic weather stations and rain gauges in and around the scenic area, combined with the risk-prone terrain parameters of each tourist route segment, identifies locations that meet the triggering conditions for disasters such as flash floods, landslides, and debris flows as disaster triggering sources. Based on the natural conduction characteristics of terrain corridors such as valleys and gullies, the temporal transmission process of disaster-causing factors generated by each triggering source along the terrain corridor and along the segmented units of the tourist route is deduced, and the change law of disaster intensity over time for each segment of the tourist route is calculated, generating disaster risk time-series curves for each segment of the tourist route. Secondary disasters caused by rainstorms in mountainous scenic areas do not occur simultaneously across the entire area, but rather dynamically transmit downstream along terrain corridors from source points such as gully heads and water catchment points. Time-series deduction can reconstruct the dynamic evolution process of the disaster, clarify the arrival time and peak intensity of the disaster for each segment of the tourist route, and reserve a decision-making window for personnel evacuation.

[0032] Step S30: Collect real-time personnel location data within the scenic area, extract personnel movement characteristic parameters, and deduce the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network; It should be noted that the system collects real-time personnel data from three sources: ticket gates, video surveillance, and mobile location data within the scenic area. From this data, it extracts movement characteristic parameters such as visitor speed, preference for stopping at attractions, and turning patterns at intersections. Based on the connectivity of the tourist route topology network, it projects changes in the number of people in each segment of the tourist route over the next 60 minutes, generating a sequence of spatiotemporal distribution trajectories. Tourists move along fixed routes within the mountain scenic area, and their distribution dynamically changes over time. The number of people on the same route can vary several times at different times. Dynamic trajectory projection can reconstruct the scale of personnel exposure in future periods, avoiding the lag in risk calculations based solely on current personnel numbers.

[0033] Step S40: Perform spatiotemporal registration calculation on the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence under the same time series to obtain the quantitative value of personnel disaster risk for each walkway segment unit under the corresponding time series; It should be noted that the disaster risk time-series curve values ​​at the same time point are spatially and temporally registered with the corresponding values ​​of the personnel spatiotemporal distribution trajectory sequence, and the quantitative value of personnel disaster risk at the corresponding time point is calculated segment by segment. The registration process couples disaster intensity and personnel exposure at the same time and spatial unit, taking into account both the severity of the disaster itself and the number of people and evacuation conditions in the area. The output quantitative value directly reflects the comprehensive risk level of that segment of the trajectory at the corresponding time.

[0034] Step S50: Based on the quantitative value of personnel disaster risk and the temporal correlation of all tourist route segment units, generate a dynamic risk distribution map of rainstorm disaster in the scenic area that is updated time by time.

[0035] It should be noted that the system integrates the quantified risk values ​​of personnel affected by disasters from all segments of the tourist route, and combines the temporal correlations and spatial adjacencies between segments to generate a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated periodically. Using the tourist routes as a visual medium, the map intuitively presents the risk levels and evolution trends of different road sections, directly supporting the scenic area management's emergency decision-making for tiered early warning, personnel evacuation, and temporary control measures.

[0036] Second Embodiment Furthermore, the steps of identifying disaster triggering points within the scenic area based on real-time rainfall monitoring data and deducing the temporal transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units include: Real-time rainfall data is matched with the risky terrain parameters of each travel segment unit using threshold matching to identify locations that meet the disaster triggering conditions as disaster triggering source points and determine the corresponding disaster type. Based on topographic slope, confluence direction and surface roughness parameters, the transmission velocity of disaster-causing factors along the topographic corridor and the intensity attenuation coefficient along the way are calculated. Based on the spatial intersection relationship between the travel route segment units and the terrain corridor, the time nodes and corresponding disaster intensity of the disaster-causing factors reaching each travel route segment unit are deduced, and the disaster risk time series curves corresponding to each travel route segment unit are generated.

[0037] It should be noted that real-time rainfall data, including minute-level rainfall intensity, hourly cumulative rainfall, and daily cumulative rainfall, is matched with risk-prone terrain parameters for each segment of the tourist route using a multi-threshold matching process. When both rainfall intensity and terrain conditions simultaneously meet the trigger thresholds for the corresponding disaster, the location is determined as the disaster trigger point, and the corresponding disaster type is simultaneously identified. The threshold system is based on historical disaster cases in the scenic area and regional geological disaster risk zoning. For example, the trigger conditions for flash floods are: 1-hour rainfall ≥ 30 mm, with a corresponding catchment area ≥ 0.8 km² and a valley slope ≥ 15°; the trigger conditions for shallow landslides are: 24-hour cumulative rainfall ≥ 100 mm, with a slope slope ≥ 35° and surface soil weathering reaching a strong weathering level. Taking the upstream catchment area of ​​Baiyun Creek as an example, when 1-hour rainfall reaches 32 mm, corresponding to a catchment area of ​​1.2 km² and an average slope of 18°, the flash flood trigger conditions are met, and the location is determined as the flash flood trigger point. Threshold matching can accurately pinpoint the spatial location where a disaster first occurs, serving as the starting point for the entire transmission simulation.

[0038] Based on three core parameters of the terrain corridor—slope, confluence direction, and surface roughness—the transmission velocity and intensity attenuation coefficient of the disaster-causing factor along the corridor were calculated. Slope is the core factor affecting transmission velocity, which can be calculated using empirical formulas: the transmission velocity is approximately 3.5 m / s at a slope of 15°, and can reach 6 m / s at a slope of 25°. The smaller the slope and the wider the valley, the slower the flow velocity. Surface roughness corresponds to hindering factors such as vegetation cover, boulder obstruction, and bend curvature. The higher the roughness, the greater the energy loss of the disaster-causing factor and the faster the intensity attenuation. For example, the attenuation coefficient in dense shrubland sections is approximately 1.5 times that in open, bare rock valleys, with an intensity attenuation of about 8% per 100 meters. Taking the Baiyunxi flash flood source as an example, the average valley slope is 18°, and the surface is mainly covered with shrubs and grasses. The calculated transmission velocity is approximately 4.2 m / s, and the intensity attenuation coefficient per 100 meters is 0.93.

[0039] Based on the spatial intersection relationship between the tourist route segments and the terrain corridors, the intersection types are divided into three categories: traversing, parallel, and crossing. Traversing tourist routes cross valleys and are directly impacted by disasters, resulting in the highest degree of influence. Parallel tourist routes are laid out along one side of the valley and are affected by overflow and lateral erosion, resulting in a lower degree of influence. Crossing tourist routes cross valleys as bridges and have the least impact. The specific time points at which each disaster-causing factor arrives at each tourist route segment are calculated, along with the corresponding disaster intensity at arrival. The intensity change process of each segment over time is recorded to generate disaster risk time series curves for each tourist route segment. Taking the Baiyunxi flash flood as an example, the calculated time for the flash flood to reach the valley floor of the first loop tourist route is 14 minutes after the rainfall triggers, with a peak intensity of 0.68. The disaster process lasts for approximately 42 minutes, and the curve generally shows a typical pattern of "rapid rise - maintaining peak value - slow decline".

[0040] Furthermore, the step of generating the disaster risk time series curve corresponding to each walkway segment unit includes: When the same travel segment unit corresponds to multiple disaster triggering sources, the time node and disaster intensity of the disaster-causing factor corresponding to each disaster triggering source unit are extracted respectively; Based on the disaster type, the superposition coupling coefficient between different disaster-causing factors is determined, and the multi-source disaster intensity at the same time point is coupled and calculated. Arrange the coupled disaster intensity values ​​in chronological order to generate the disaster risk time series curve corresponding to the current traveler segment unit.

[0041] It should be noted that when the same travel segment unit corresponds to multiple disaster triggering sources, such as the Erhuan Valley bottom section being simultaneously affected by flash floods from both the main ditch of Baiyunxi and the tributary on the north side, and facing the risk of shallow landslides on the left slope, three core parameters are extracted for each disaster triggering source point: arrival time, peak intensity, and duration of the causative factor reaching the current travel segment unit. The extraction process is entirely based on independent calculations of single-source transmission, preserving the complete temporal characteristics of each disaster source under independent action, providing lossless input data for subsequent coupled calculations. For the example of the Erhuan Valley bottom section, the arrival time of the flash flood in the main ditch is 16 minutes, with a peak intensity of 0.62; the arrival time of the flash flood in the tributary is 19 minutes, with a peak intensity of 0.28; and the arrival time of the landslide on the left is 22 minutes, with a peak intensity of 0.25.

[0042] Based on the differences in the combination of disaster types, the superposition coupling coefficient between different disaster-causing factors is determined, and then the multi-source disaster intensity at the same time point is coupled and calculated. The coupling coefficient is determined based on the physical mechanism of the disaster superposition effect: for the superposition of similar disasters, cumulative coupling is used. For example, if flash floods from two valleys flow into the same downstream section at the same time, the coupling coefficient is taken as 1.0, and the intensity is directly superimposed; for the superposition of dissimilar disasters, amplification coupling is used. For example, when flash floods and landslides occur simultaneously, the water flow carrying landslide soil is prone to forming small debris flows, and the destructive force is significantly higher than that of a single disaster, so the coupling coefficient is taken as 1.2-1.5; if there is a suppressive effect between disasters, reduction coupling is used. Taking the example of the second ring valley bottom section, the two flash floods belong to the same superposition. The superposition intensity of the flash flood is first calculated as 0.62 + 0.28 = 0.90; then, it is coupled with the landslide in a dissimilar way, and the coupling coefficient is taken as 1.25. The final comprehensive disaster intensity is 0.90 × 1.25 = 1.125. After normalization in the 0-1 interval, the corresponding standardized intensity is 0.78, which is significantly higher than the risk level of a single flash flood. By using differentiated coupling coefficients, the combined effects of different disaster combinations can be accurately reflected, avoiding the risk calculation bias caused by simple summation.

[0043] The disaster intensity values ​​after coupling at all time points are arranged in chronological order, and the intensity values ​​within the time intervals are completed using cubic spline interpolation to form a continuous and smooth disaster risk time series curve. The curve after multi-source coupling may show different forms such as single peak, multiple peaks, and stepped rise, corresponding to scenarios where different disaster sources arrive successively or simultaneously, fully presenting the risk change process of the traveler segment throughout the entire disaster process.

[0044] Furthermore, the steps of collecting real-time personnel location data within the scenic area, extracting personnel movement feature parameters, and deducing the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network include: The real-time personnel location data is matched with the travel route to determine the current travel route segment unit and the direction of travel for each person; Historical movement speed, duration of stay at attractions, and turning probability at tourist route nodes are extracted as movement feature parameters. Based on the connectivity of the walkway topology network, the distribution of the number of people in each walkway segment unit in future time periods is inferred according to the movement characteristic parameters and the Markov chain model, and a sequence of spatiotemporal distribution trajectories of people is generated.

[0045] It should be noted that spatial matching is performed on the real-time collected personnel positioning data, mapping each discrete positioning point to a corresponding walkway segment unit. Simultaneously, the direction of movement is determined, categorized as uphill, downhill, or stationary. Within the mountain scenic area, tourists' activities are strictly confined to the walking trails; they will not arbitrarily enter unmarked mountain forest areas. Therefore, matching discrete positioning points to linear walkways eliminates positioning drift errors caused by GPS signal obstruction in valleys, accurately determining the specific route and movement status of each tourist, and providing an accurate initial position for subsequent trajectory calculation.

[0046] From the historical location big data of the scenic area over the past six months, three core movement characteristic parameters were extracted: First, terrain-differentiated movement speed, distinguishing the average walking speed under three terrain types: uphill, downhill, and flat roads. In the Xihai Grand Canyon area, the average speed uphill is about 2.2 km / h, downhill is about 3.8 km / h, and flat roads are about 3.2 km / h. Second, the dwell time at different levels of scenic spots, with core scenic spots such as the valley bottom viewing platform averaging about 25 minutes, ordinary viewing platforms about 5 minutes, and rest platforms about 3 minutes. Third, the probability of turning at intersections, statistically analyzing the proportion of tourists choosing different branch routes at each intersection. For example, at the Paiyun Pavilion intersection, 85% of tourists chose to go downhill into the Second Ring Road, while 15% chose to go towards Beihai. These three characteristics, from the dimensions of speed, dwell time, and route selection, depict the movement behavior patterns of tourists and form the behavioral basis for trajectory extrapolation.

[0047] Based on the connectivity of the tourist route topology network, and taking the current number of people in each segment as the initial state, a Markov chain model is used to predict the distribution of people in each segment unit of the tourist route in future time periods, combined with extracted movement feature parameters. The core characteristic of the Markov chain is that the distribution of people in the next time moment is only related to the current state and transition probability, which is very suitable for the scenario of tourist flow prediction along a fixed tourist route in mountain scenic areas. With a time step of 5 minutes, the inflow, outflow and dwell time of people in each segment are calculated in time periods to generate a continuous spatiotemporal distribution trajectory sequence of people. Taking the Paiyunting entrance segment as an example, there are currently 320 tourists, of which 85% choose to tour down the second ring road and 15% choose to stay and rest. Based on the downhill speed, tourists can advance about 317 meters every 5 minutes, covering 2 tourist route segments. It can be predicted that the number of people in the upper segment of the second ring road will reach a peak of 280 people after 10 minutes, and the number of people in the valley segment will rise to 210 people after 25 minutes. This sequence can output the changes in the number of people in each segment of the tour route over the next 60 minutes, fully presenting the dynamic process of passenger flow moving along the tour route.

[0048] Furthermore, the step of performing line matching on the real-time collected personnel positioning data includes: Acquire ticket gate data, video crowd flow statistics, and mobile location data, and perform spatial coordinate correction and timestamp alignment respectively; The walk-line buffer matching algorithm is used to map various personnel data to the corresponding walk-line segment units and eliminate abnormal positioning points that exceed the walk-line buffer range; Weighted verification is performed on multi-source personnel statistics within the same travel segment unit to obtain the real-time personnel base number for each travel segment unit.

[0049] It should be noted that three types of data sources were acquired simultaneously: ticket gate data, recording the number of tourists entering and exiting each entrance and exit; this data is highly accurate but only covers node locations; video crowd flow statistics, deployed at key tourist route nodes and attractions, counts the number of people passing by, but is subject to occlusion errors; and mobile location data, from tourists' mobile app and operator signaling, covering the entire tourist route but showing location drift in valley areas. Spatial coordinate correction was performed on each of these three types of data, unifying them to the scenic area's independent coordinate system, and high-precision DEM was used to correct elevation deviations. Simultaneously, timestamps were unified to the BeiDou standard clock, achieving dual spatial and temporal alignment and establishing a consistent benchmark system for subsequent fusion processing.

[0050] A buffer zone matching algorithm is employed to set differentiated buffer zones for different types of trails: 7.5 meters on each side for trails at the valley floor, with a total width of 15 meters; and 5 meters on each side for trails on the ridge, with a total width of 10 meters. Positioning data for each type of person is mapped to the corresponding trail segment unit. Positioning points that fall completely outside the buffer zone are identified as abnormal drift points and are discarded. In mountainous scenic valley environments, GPS signals are easily blocked by mountains, often resulting in positioning points drifting to hillsides or valleys. Buffer zone matching effectively filters out these abnormal positioning points, ensuring the accurate binding relationship between personnel location and trail.

[0051] Weighted fusion verification was performed on multi-source personnel statistics within the same travel segment unit, with weights assigned based on the reliability of different data sources: ticket gate data had the highest accuracy, with a weight of 0.4; video statistics had high local accuracy, with a weight of 0.35; and mobile location data had wide coverage but slightly lower accuracy, with a weight of 0.25. For the example of the core valley section, ticket gate backtracking yielded approximately 240 people in the area, video flow statistics converted to a real-time number of 220 people, and mobile location matching identified 190 people. The weighted calculation yielded a real-time personnel base of 240 × 0.4 + 220 × 0.35 + 190 × 0.25 = 96 + 77 + 47.5 = 220.5, rounded down to 221 people. Multi-source weighted verification effectively balanced the errors of different data sources, improving personnel statistics accuracy by more than 30% compared to a single data source.

[0052] Furthermore, the step of performing spatiotemporal registration calculation between the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence under the same time series includes: Based on a preset time step, the disaster intensity value and the number of people at the corresponding time step are extracted from the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence, respectively. Extract the risk avoidance space capacity parameters of the corresponding travel line segment unit, and calculate the personnel retention risk coefficient of the current segment in combination with the personnel number value; The disaster intensity value and the personnel stranded risk coefficient are nonlinearly coupled to calculate the quantitative value of personnel disaster risk in the corresponding walk-line segment unit at the current time step.

[0053] It should be noted that, based on a preset 5-minute time step, the disaster intensity value for the corresponding time step is extracted from the disaster risk time series curve, and the number of people for the corresponding time step is extracted from the spatiotemporal distribution trajectory sequence of personnel. A unified time benchmark ensures strict alignment of disaster data and personnel data in the time dimension, avoiding risk calculation errors caused by time series misalignment; the 5-minute step setting balances the timeliness of early warning with computational efficiency, aligning with the rapid evolution of disasters in mountainous scenic areas.

[0054] Extract the evacuation space capacity parameter for the corresponding hiking trail segment, which is the rated capacity of high platforms, sturdy caves, and non-slip steps within a 50-meter radius of that segment for temporary evacuation. Combine this with the current number of people in the segment to calculate the personnel retention risk coefficient, which is equal to the ratio of the number of people in the segment to the evacuation space capacity. When the coefficient is greater than 1, it indicates that the number of people exceeds the evacuation capacity; the larger the value, the higher the evacuation difficulty and the greater the risk of personnel exposure. For example, in a 200-meter-long section at the bottom of a valley, there is only one small evacuation platform with a rated capacity of 60 people. If the projected number of people at the current time is 180, then the personnel retention risk coefficient is 180 / 60 = 3.0, indicating that the number of people far exceeds the evacuation capacity, making evacuation extremely difficult. This coefficient is a key intermediate variable connecting disaster intensity and personnel casualty risk, compensating for the inadequacy of simply using the number of people to measure exposure risk.

[0055] The disaster intensity value and the personnel retention risk coefficient are nonlinearly coupled to calculate the quantitative value of personnel disaster risk for the corresponding travel segment unit at the current time step. The coupling adopts an exponential growth model, with the formula: Risk Value = 1 - exp(-α × Disaster Intensity × Retention Risk Coefficient), where α is a calibration coefficient, approximately 0.35. This model conforms to the evolution law of disaster risk: when the disaster intensity is low or the number of people is small, the risk increases slowly; when the disaster intensity exceeds the critical threshold or the number of people far exceeds the evacuation capacity, the risk shows an accelerated upward trend. Taking the aforementioned valley segment as an example, with a disaster intensity of 0.7 and a retention risk coefficient of 3.0, the calculated risk value is 1 - exp(-0.35 × 0.7 × 3.0) = 1 - exp(-0.735) ≈ 0.52, corresponding to a high-risk level. The final output quantitative value is a normalized value between 0 and 1; the higher the value, the higher the comprehensive disaster risk of that road segment at that time.

[0056] Furthermore, the step of generating a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated periodically based on the quantitative value of the disaster risk of personnel in all tourist route segment units and the temporal correlation includes: The risk level of personnel in each segment of the travel route is classified according to the preset risk level threshold, and the corresponding risk level is marked. Map the tourist route segments with risk level labels to the scenic area's geographical base map, and mark the expected duration and peak arrival time of each segment's risk level. The disaster risk quantification value of personnel in all travel line segment units is iteratively calculated according to the preset update frequency to generate a dynamic risk distribution map that is updated in time period.

[0057] It should be noted that, according to the preset four-level risk level thresholds, the risk of personnel casualties in each segment of the tourist route is quantified and classified into four levels: 0-0.25 is low risk (blue), 0.25-0.4 is medium risk (yellow), 0.4-0.6 is high risk (orange), and 0.6-1 is extremely high risk (red). Each segment is also marked with a corresponding risk level identifier. This four-level classification standard corresponds one-to-one with the national emergency warning levels, facilitating the scenic area management to directly match appropriate emergency response measures: Blue (low risk) maintains normal tours and issues rainfall warnings; Yellow (medium risk) closes high-risk sections and guides tourists to safe areas; Orange (high risk) activates the evacuation plan and dispatches security personnel to guide evacuation; Red (extremely high risk) issues an emergency evacuation order and organizes people to take shelter nearby.

[0058] The risk level-labeled sections of the hiking trail are mapped onto a high-resolution geographic base map of the scenic area. Colors corresponding to the risk level are rendered along the actual path of the trail, creating a linear risk visualization effect. Two key temporal information items are also labeled next to each section: the estimated duration of the risk level for that section and the estimated time of the risk peak. High-risk sections also include the location of the nearest evacuation point and suggested evacuation routes. This rich temporal and guidance information helps management accurately grasp the pace of disaster and rationally plan evacuation routes and control measures.

[0059] Following a preset update frequency of once every 5 minutes, the system reconnects to the latest rainfall monitoring data and personnel location data, iteratively calculates the quantified disaster risk value for personnel in all tourist route segments, and synchronously updates information such as risk level, peak time, and duration on the risk distribution map, generating a dynamic risk distribution map of rainstorm disasters for the scenic area that is updated on a rolling basis over time. For example: the initial forecast map released at 10:00 showed that the valley floor segment reached a high-risk level from 10:20, with the peak occurring at 10:35 and lasting until 11:10; at 10:05, the latest rainfall data was received, increasing the hourly rainfall intensity from 30mm to 45mm. After iterative calculation, the map was updated to show that the valley floor reached an extremely high risk from 10:15, with the peak occurring earlier at 10:30 and lasting until 11:20, while the north slope segment was added as a high-risk area. This dynamic update mechanism ensures that the risk prediction results always align with the latest rainfall trends and changes in personnel distribution. As the rainfall process develops, the prediction results are continuously revised, and the accuracy of the early warning gradually improves over time, providing continuous and accurate decision-making basis for emergency response in the scenic area.

[0060] Please see Figure 2 The third embodiment of the present invention provides: A rainstorm disaster risk prediction system for mountainous scenic areas, wherein the system includes: The module is used to construct a tour route topology network based on the distribution of trails and the orientation of terrain corridors in mountainous scenic areas. The tour route topology network is divided into several continuous tour route segment units, and corresponding risky terrain parameters are matched for each tour route segment unit. The simulation module is used to identify disaster triggering points in the scenic area based on real-time rainfall monitoring data, and to simulate the time-series transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units, generating disaster risk time-series curves for each tourist route segment unit. The data acquisition module is used to collect real-time personnel location data within the scenic area, extract personnel movement characteristic parameters, and deduce the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network. The registration module is used to perform spatiotemporal registration calculations on the disaster risk time series curve and the spatiotemporal distribution trajectory sequence of personnel under the same time series, so as to obtain the quantitative value of the disaster risk of personnel in each walkway segment unit under the corresponding time series. The generation module is used to generate a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated time by time based on the quantitative value of the disaster risk of personnel in all tourist route segment units and the temporal correlation.

[0061] Furthermore, the steps of identifying disaster triggering points within the scenic area based on real-time rainfall monitoring data and deducing the temporal transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units include: Real-time rainfall data is matched with the risky terrain parameters of each travel segment unit using threshold matching to identify locations that meet the disaster triggering conditions as disaster triggering source points and determine the corresponding disaster type. Based on topographic slope, confluence direction and surface roughness parameters, the transmission velocity of disaster-causing factors along the topographic corridor and the intensity attenuation coefficient along the way are calculated. Based on the spatial intersection relationship between the travel route segment units and the terrain corridor, the time nodes and corresponding disaster intensity of the disaster-causing factors reaching each travel route segment unit are deduced, and the disaster risk time series curves corresponding to each travel route segment unit are generated.

[0062] Furthermore, the step of generating the disaster risk time series curve corresponding to each walkway segment unit includes: When the same travel segment unit corresponds to multiple disaster triggering sources, the time node and disaster intensity of the disaster-causing factor corresponding to each disaster triggering source unit are extracted respectively; Based on the disaster type, the superposition coupling coefficient between different disaster-causing factors is determined, and the multi-source disaster intensity at the same time point is coupled and calculated. Arrange the coupled disaster intensity values ​​in chronological order to generate the disaster risk time series curve corresponding to the current traveler segment unit.

[0063] Furthermore, the steps of collecting real-time personnel location data within the scenic area, extracting personnel movement feature parameters, and deducing the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network include: The real-time personnel location data is matched with the travel route to determine the current travel route segment unit and the direction of travel for each person; Historical movement speed, duration of stay at attractions, and turning probability at tourist route nodes are extracted as movement feature parameters. Based on the connectivity of the walkway topology network, the distribution of the number of people in each walkway segment unit in future time periods is inferred according to the movement characteristic parameters and the Markov chain model, and a sequence of spatiotemporal distribution trajectories of people is generated.

[0064] Furthermore, the step of performing line matching on the real-time collected personnel positioning data includes: Acquire ticket gate data, video crowd flow statistics, and mobile location data, and perform spatial coordinate correction and timestamp alignment respectively; The walk-line buffer matching algorithm is used to map various personnel data to the corresponding walk-line segment units and eliminate abnormal positioning points that exceed the walk-line buffer range; Weighted verification is performed on multi-source personnel statistics within the same travel segment unit to obtain the real-time personnel base number for each travel segment unit.

[0065] Furthermore, the step of performing spatiotemporal registration calculation between the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence under the same time series includes: Based on a preset time step, the disaster intensity value and the number of people at the corresponding time step are extracted from the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence, respectively. Extract the risk avoidance space capacity parameters of the corresponding travel line segment unit, and calculate the personnel retention risk coefficient of the current segment in combination with the personnel number value; The disaster intensity value and the personnel stranded risk coefficient are nonlinearly coupled to calculate the quantitative value of personnel disaster risk in the corresponding walk-line segment unit at the current time step.

[0066] Furthermore, the step of generating a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated periodically based on the quantitative value of the disaster risk of personnel in all tourist route segment units and the temporal correlation includes: The risk level of personnel in each segment of the travel route is classified according to the preset risk level threshold, and the corresponding risk level is marked. Map the tourist route segments with risk level labels to the scenic area's geographical base map, and mark the expected duration and peak arrival time of each segment's risk level. The disaster risk quantification value of personnel in all travel line segment units is iteratively calculated according to the preset update frequency to generate a dynamic risk distribution map that is updated in time period.

[0067] The fourth embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the rainstorm disaster risk prediction method for mountainous scenic areas as described above.

[0068] The fifth embodiment of the present invention provides a readable storage medium on which a computer program is stored, wherein when the current program is executed by a processor, it implements the rainstorm disaster risk prediction method for mountainous scenic areas as described above.

[0069] In summary, the rainstorm disaster risk prediction method and system for mountainous scenic areas provided by the above embodiments of the present invention can ultimately generate a dynamic risk distribution map that is updated on a time-by-time basis, providing a refined basis for rainstorm disaster early warning in scenic areas by road segment and time sequence, significantly improving the pertinence and timeliness of early warning signals, and supporting scenic areas to accurately carry out personnel evacuation and risk management.

[0070] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0071] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0072] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0073] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0074] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with the present embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0075] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the appended claims.

Claims

1. A method for predicting the risk of rainstorm disasters in mountainous scenic areas, characterized in that, The method includes: Based on the distribution of trails and the orientation of terrain corridors in mountainous scenic areas, a topological network of tourist routes is constructed. The topological network of tourist routes is divided into several continuous segment units, and corresponding risky terrain parameters are matched for each segment unit. Based on real-time rainfall monitoring data, disaster triggering points within the scenic area are identified, and the time-series transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units is deduced, generating disaster risk time-series curves for each tourist route segment unit. Collect real-time personnel location data within the scenic area, extract personnel movement characteristic parameters, and deduce the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network; By performing spatiotemporal registration calculations on the disaster risk time series curve and the spatiotemporal distribution trajectory sequence of personnel under the same time series, the quantitative value of the disaster risk of personnel in each walkway segment unit under the corresponding time series is obtained; Based on the quantitative value of disaster risk to personnel in all tourist route segments and the temporal correlation, a dynamic risk distribution map of rainstorm disasters in the scenic area is generated and updated time by time.

2. The method for predicting rainstorm disaster risks in mountainous scenic areas according to claim 1, characterized in that, The steps of identifying disaster triggering points within the scenic area based on real-time rainfall monitoring data and deducing the temporal transmission process of disaster-causing factors generated by each triggering point along the terrain corridor and tourist route segment units include: Real-time rainfall data is matched with the risky terrain parameters of each travel segment unit using threshold matching to identify locations that meet the disaster triggering conditions as disaster triggering source points and determine the corresponding disaster type. Based on topographic slope, confluence direction and surface roughness parameters, the transmission velocity of disaster-causing factors along the topographic corridor and the intensity attenuation coefficient along the way are calculated. Based on the spatial intersection relationship between the travel route segment units and the terrain corridor, the time nodes and corresponding disaster intensity of the disaster-causing factors reaching each travel route segment unit are deduced, and the disaster risk time series curves corresponding to each travel route segment unit are generated.

3. The method for predicting rainstorm disaster risks in mountainous scenic areas according to claim 2, characterized in that, The step of generating the disaster risk time series curve corresponding to each walkway segment unit includes: When the same travel segment unit corresponds to multiple disaster triggering sources, the time node and disaster intensity of the disaster-causing factor corresponding to each disaster triggering source unit are extracted respectively; Based on the disaster type, the superposition coupling coefficient between different disaster-causing factors is determined, and the multi-source disaster intensity at the same time point is coupled and calculated. Arrange the coupled disaster intensity values ​​in chronological order to generate the disaster risk time series curve corresponding to the current traveler segment unit.

4. The method for predicting rainstorm disaster risks in mountainous scenic areas according to claim 1, characterized in that, The steps of collecting real-time personnel location data within the scenic area, extracting personnel movement feature parameters, and deducing the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network include: The real-time personnel location data is matched with the travel route to determine the current travel route segment unit and the direction of travel for each person; Historical movement speed, duration of stay at attractions, and turning probability at tourist route nodes are extracted as movement feature parameters. Based on the connectivity of the walkway topology network, the distribution of the number of people in each walkway segment unit in future time periods is inferred according to the movement characteristic parameters and the Markov chain model, and a sequence of spatiotemporal distribution trajectories of people is generated.

5. The method for predicting rainstorm disaster risks in mountainous scenic areas according to claim 4, characterized in that, The step of performing line matching on the real-time collected personnel positioning data includes: Acquire ticket gate data, video crowd flow statistics, and mobile location data, and perform spatial coordinate correction and timestamp alignment respectively; The walk-line buffer matching algorithm is used to map various personnel data to the corresponding walk-line segment units and eliminate abnormal positioning points that exceed the walk-line buffer range; Weighted verification is performed on multi-source personnel statistics within the same travel segment unit to obtain the real-time personnel base number for each travel segment unit.

6. The method for predicting rainstorm disaster risks in mountainous scenic areas according to claim 1, characterized in that, The steps for performing spatiotemporal registration calculations between disaster risk time series curves and personnel spatiotemporal distribution trajectory sequences under the same time series include: Based on a preset time step, the disaster intensity value and the number of people at the corresponding time step are extracted from the disaster risk time series curve and the personnel spatiotemporal distribution trajectory sequence, respectively. Extract the risk avoidance space capacity parameters of the corresponding travel line segment unit, and calculate the personnel retention risk coefficient of the current segment in combination with the personnel number value; The disaster intensity value and the personnel stranded risk coefficient are nonlinearly coupled to calculate the quantitative value of personnel disaster risk in the corresponding walk-line segment unit at the current time step.

7. The method for predicting rainstorm disaster risks in mountainous scenic areas according to claim 6, characterized in that, The steps for generating a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated time-by-time based on the quantitative value of personnel disaster risk and the temporal correlation of all tourist route segment units include: The risk level of personnel in each segment of the travel route is classified according to the preset risk level threshold, and the corresponding risk level is marked. Map the tourist route segments with risk level labels to the scenic area's geographical base map, and mark the expected duration and peak arrival time of each segment's risk level. The disaster risk quantification value of personnel in all travel line segment units is iteratively calculated according to the preset update frequency to generate a dynamic risk distribution map that is updated in time period.

8. A rainstorm disaster risk prediction system for mountainous scenic areas, characterized in that, The system includes: The module is used to construct a tour route topology network based on the distribution of trails and the orientation of terrain corridors in mountainous scenic areas. The tour route topology network is divided into several continuous tour route segment units, and corresponding risky terrain parameters are matched for each tour route segment unit. The simulation module is used to identify disaster triggering points in the scenic area based on real-time rainfall monitoring data, and to simulate the time-series transmission process of disaster-causing factors generated by each disaster triggering point along the terrain corridor and tourist route segment units, generating disaster risk time-series curves for each tourist route segment unit. The data acquisition module is used to collect real-time personnel location data within the scenic area, extract personnel movement characteristic parameters, and deduce the spatiotemporal distribution trajectory sequence of personnel in each segment unit of the tourist route based on the tourist route topology network. The registration module is used to perform spatiotemporal registration calculations on the disaster risk time series curve and the spatiotemporal distribution trajectory sequence of personnel under the same time series, so as to obtain the quantitative value of the disaster risk of personnel in each walkway segment unit under the corresponding time series. The generation module is used to generate a dynamic risk distribution map of rainstorm disasters in the scenic area that is updated time by time based on the quantitative value of the disaster risk of personnel in all tourist route segment units and the temporal correlation.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the rainstorm disaster risk prediction method for mountainous scenic areas as described in any one of claims 1 to 7.

10. A readable storage medium having a computer program stored thereon, characterized in that, When the current program is executed by the processor, it implements the rainstorm disaster risk prediction method for mountainous scenic areas as described in any one of claims 1 to 7.

Images