Mountainous area meteorological rainfall supplementation observation station site selection evaluation method based on high-influence weather influence path

Abstract

The invention discloses a mountainous area meteorological rainfall supplementation observation station site selection evaluation method based on a high-influence weather influence path in the technical field of meteorological observation, and the method comprises the steps: obtaining meteorological grid point data in an observation region, screening the meteorological grid point data to obtain a rainstorm process, generating a path sequence based on the rainstorm data in the rainstorm process, and carrying out the site selection evaluation of a mountainous area meteorological rainfall supplementation observation station. A dynamic time warping (DTW) method is adopted to generate an influence area with a widened rainstorm path, and grid point rainfall forecast error data is generated based on multi-source forecast data and station meteorological data. According to the invention, by extracting the evolution path of the rainstorm process in time and space and combining the multi-source mode forecast error, the regional rainfall distribution, the existing station network density, the topographic feature and the settlement distance element, the evaluation method can be dynamically quantified and the monitoring demand and cost constraint can be balanced; the problems of data missing in the initial stage of a rainstorm system, insufficient upstream observation of a weather system path, insufficient station network coverage rate and weak data representativeness are solved.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of meteorological observation, and particularly relates to a mountainous area meteorological precipitation supplementary observation station site selection evaluation method based on a high-impact weather influence path. BACKGROUND

[0002] The mountainous area is a high-incidence area of natural disasters such as rainstorm, mountain flood and mud-rock flow due to complex terrain and changeable climate, and meteorological precipitation observation stations are often set up in the mountainous area to observe the meteorological conditions of the mountainous area. At present, the layout of the meteorological precipitation observation station mainly depends on geographical experience judgment and local meteorological statistics, and the microclimate difference, the moving speed and the direction change rate of the rainstorm path are easily ignored, so that it is difficult to quantize the transmission influence of the upstream of the rainstorm path on the downstream disaster, and the layout of the observation station is difficult to capture the early warning signal in the early stage of the rainstorm. For the disaster prevention and early warning demand, it is particularly important to monitor in advance in the aspect of meteorological observation.

[0003] According to the patent document with the publication number CN118780092A, a ground rain station network optimization layout technical method, system and device are disclosed, which comprises the following steps: determining the tributary basin distribution in the study basin according to the basin area and elevation data of the study basin, determining each tributary basin as a target basin; determining the contour belt and the equal rainfall belt in the study basin according to the elevation data and rainfall data of the study basin, overlapping the target basin and the contour belt and the equal rainfall belt, and determining each region formed after overlapping as a configuration unit; determining the river network density of the configuration unit according to the river network length and the configuration unit area of the configuration unit; determining the multi-year runoff contribution proportion of the configuration unit according to the multi-year runoff average value of the configuration unit and the multi-year runoff average value of the study basin; determining the priority order of the configuration unit according to the river network density and the multi-year runoff contribution proportion of the configuration unit; determining the potential configuration unit in each configuration unit according to the rain measurement covered range of each configuration unit and the set threshold; constructing a multi-objective optimization function of the potential configuration unit according to the spatial continuity constraint function and the joint information entropy constraint function of the potential configuration unit determined in advance; determining the multi-objective optimization index of each proposed construction plan of the potential configuration unit through the multi-objective optimization function, wherein the proposed construction plan comprises the point position and the number of the proposed rain station; determining the target plan in the proposed construction plan according to the multi-objective optimization index and the proposed construction plan; and performing rain station construction layout according to the priority order of the configuration unit and the target plan of the potential configuration unit.

[0004] However, the current meteorological precipitation observation station layout mainly relies on geographical experience and local meteorological statistics, which easily ignores the microclimate differences. The spatial and temporal dynamic evolution characteristics of the storm path, such as the moving speed and the direction change rate, are ignored, the transmission influence of the upstream on the downstream disaster cannot be quantified, and the observation station layout cannot capture the early warning signal of the initial stage of the storm system. The data spatial representation and application mechanism need to be improved, and the dynamic quantitative evaluation is difficult. It is difficult to scientifically balance the monitoring demand and cost constraint, and technical upgrading is urgently needed. SUMMARY

[0005] The purpose of the present application is to provide a mountainous area meteorological precipitation supplementary observation station site evaluation method based on high-impact weather influence path, which focuses on the influence path of regional high-impact storm, comprehensively considers the factors of multi-source model forecast precipitation error, regional precipitation distribution, existing observation network density, slope, slope direction, distance from populated area, and river network density, further constructs a research area precipitation supplementary observation station layout suitability evaluation system based on the information entropy of each factor, completes the preliminary site selection of the precipitation supplementary observation station, and adjusts the site selection of the supplementary observation station by combining with the field investigation and evaluation of the traffic accessibility, communication stability and energy supply stability of the site selection, optimizes the site selection scheme, and can quantize the transmission influence of the upstream on the downstream disaster, which is beneficial to capture the early warning signal of the initial stage of the storm system.

[0006] To achieve the above purpose, the present application provides the following technical scheme:

[0007] A mountainous area meteorological precipitation supplementary observation station site evaluation method based on high-impact weather influence path, comprising: obtaining meteorological grid point data in an observation area, screening stormy days from the meteorological grid point data, regarding consecutive stormy days as a storm process, extracting the central longitude and latitude and the precipitation intensity based on the hourly precipitation data in the storm process, generating a path sequence, and using the dynamic time warping (DTW) method to generate the influence area data of the storm path widening, wherein the meteorological grid point data includes daily precipitation and hourly precipitation in a certain period.

[0008] Based on the precipitation observation data of the meteorological station in the observation area, the inverse distance weighted interpolation method is used to generate regional grid precipitation data, the spatial resolution of the regional grid precipitation data is unified, the root mean square error is calculated to obtain the model prediction error, and the minimum error of the model prediction error is selected as the prediction error factor based on the root mean square error;

[0009] The coordinates of the meteorological station are obtained, the grid station network density of the position coordinates is counted, the regional elevation data of the meteorological station is obtained, the slope and slope direction of the meteorological station are calculated, and the populated area adjacent relationship of the meteorological station is generated;

[0010] Based on multi-index data, entropy weight method is used to calculate information entropy and weight of each index, so as to calculate site selection score of the grid according to the index data and the corresponding weight, and the grid point with a score reaching a preset score is selected as a preliminary site selection area in combination with the influence area data of the site selection corresponding to the storm path.

[0011] The terrain environment data of the grid point is collected to determine the target site selection area, wherein the terrain environment data includes terrain features, soil types, geological conditions, surface coverage, traffic accessibility, communication stability and energy supply stability.

[0012] As a further scheme of the present application: the meteorological grid point data of the meteorological station in the observation area is obtained, and storm days are screened from the meteorological grid point data, and consecutive storm days are regarded as a storm process, and based on the hourly precipitation data in the storm process, the central longitude and latitude and precipitation intensity are extracted to generate a path sequence, including:

[0013] Based on a pre-set daily precipitation threshold, storm days are screened from the meteorological grid point data;

[0014] A plurality of consecutive days of the storm days are regarded as a storm process, and the number of times of the storm process in the meteorological grid point data is obtained;

[0015] Based on the hourly precipitation data of the storm process, central longitude and latitude data and precipitation intensity data of each storm process are obtained to generate a path sequence of the storm path.

[0016] As a further scheme of the present application: the method of dynamic time warping (DTW) is used to generate the influence area data of the widening of the storm path, including:

[0017] The data transformation method is used to pre-process the path sequence to remove noise data in the path sequence to obtain a noise reduction sequence;

[0018] The resampling method is used to resample the noise reduction sequence into an equal-length sequence, and the average moving speed formula is used:

[0019] The average moving speed of the storm path is calculated, wherein Δd is the path distance, and Δt is the time interval;

[0020] The direction change rate formula is used:

[0021] θ=△α / △d is used to calculate the direction change rate of the storm path, wherein △α is the direction angle change;

[0022] The dynamic time warping (DTW) method is used to perform time sequence clustering on the average moving speed and the direction change rate to obtain a clustering result;

[0023] Determine the optimal cluster number (K) of the cluster result based on the elbow rule and the contour coefficient, and calculate the average trajectory and precipitation intensity distribution data of the rainstorm path to generate the influence area data of the rainstorm path widening.

[0024] As a further scheme of the present application: the calculation of the root mean square error obtains the mode prediction error, comprising:

[0025] Adopt the root mean square error formula:

[0026] Wherein, P obes,i is the observed precipitation, P mod,i is the precipitation, and M is the number of grid points.

[0027] As a further scheme of the present application: obtaining the coordinates of the weather station, and counting the grid point station network density of the position coordinates, comprising:

[0028] Statistically assign the number of stations in the grid point.

[0029] As a further scheme of the present application: the acquisition of the regional elevation data of the weather station, the calculation of the slope and the slope direction of the weather station, comprising:

[0030] Based on the regional elevation data, the slope calculation formula is adopted:

[0031] The slope is calculated, wherein △z is the height difference of the regional elevation data, and △h is the horizontal distance of the regional elevation data.

[0032] The slope direction calculation formula is adopted:

[0033] A = arctan2(△y,△x) is adopted to calculate the slope direction, wherein A is the slope direction, △y is the height difference component in the vertical direction, and △x is the distance component in the horizontal direction.

[0034] As a further scheme of the present application: the generation of the weather station's settlement neighborhood relationship, comprising:

[0035] Based on the settlement position of the weather station, the nearest neighbor analysis method and the nearest neighbor distance formula are adopted:

[0036] The neighborhood relationship is calculated, wherein (x p ,y p ) is the settlement coordinate.

[0037] As a further scheme of the present application: the multi-index data includes the historical average precipitation of the grid point, the precipitation prediction error, the slope, the slope direction, the weather station density and the settlement neighborhood relationship.

[0038] As a further scheme of the present application: the multi-index data is used to calculate the information entropy and weight of each index by entropy weight method, including:

[0039] The entropy weight method formula is used:

[0040] The information entropy of the index is calculated, wherein, is the feature proportion of the jth grid point under the ith index, M is the total number of the grid points, E i is the information entropy, when p ij = 0, p ij lnp ij = 0 is defined.

[0041] The weight formula is used:

[0042] The weight w i of the ith index is calculated.

[0043] As a further scheme of the present application: the site selection score of the grid point is calculated according to the index data and the corresponding weight, the influence area data of the rainstorm path corresponding to the site selection of the grid point is combined, and the grid point with a score reaching a preset score is selected as a preliminary site selection area, including:

[0044] The grid point site selection score formula is used:

[0045] The site selection score of the grid point is calculated.

[0046] Compared with the prior art, the present application has the following advantages:

[0047] 1. In the present application, the evolution path of the rainstorm process in time and space is extracted, the dynamic quantitative evaluation method is balanced, the monitoring demand and the cost constraint are balanced, the multi-source mode prediction error, the regional precipitation distribution, the existing station network density, the terrain characteristics and the distance elements of the settlement are combined, and the problems of insufficient observation on the upstream of the weather system path, insufficient station network coverage and weak data representativeness are solved.

[0048] 2. In the present application, the scientific evaluation system is constructed by using the entropy weight method, the site selection scheme is optimized according to the score of each grid point, the data representativeness and the disaster prevention and reduction ability are improved, the influence path and the range are obtained by providing the rainstorm influence path analysis method, the multi-dimensional index objective weighting is further combined, the transmission influence of the upstream on the downstream disaster is quantified, the early warning signal of the initial stage of the rainstorm system is captured in a targeted manner, and the disaster prevention and reduction ability is improved.

[0049] 3、In the present application, by introducing the analysis method of the moving speed and direction change rate of the rainstorm process influence path, and based on the DTW time series clustering method, the moving track and influence area of the rainstorm path center can be quantified, and the problem of insufficient observation on the upstream of the weather system path can be solved.

[0050] 4、In the present application, by comprehensively considering the precipitation distribution, forecast error, station network density, topographic features, distance to populated areas, and multiple indexes of rainstorm influence path, and by using the entropy weight method to construct an objective site selection evaluation system and combining with field investigation to optimize site selection, the monitoring demand after site selection and the construction feasibility of the meteorological station at the grid point can be balanced, the dependence of site selection on experience judgment can be reduced, the upstream observation capability can be improved, the data loss of the existing station network at the initial stage of the rainstorm system can be made up, and the forecast lead time and accuracy can be improved. BRIEF DESCRIPTION OF DRAWINGS

[0051] Figure 1 The method flowchart of the present application is shown in the figure.

[0052] Figure 2 The method step diagram of the present application is shown in the figure.

[0053] Figure 3 The embodiment result diagram of the present application is shown in the figure. DETAILED DESCRIPTION

[0054] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0055] Embodiment:

[0056] The layout of the mountainous precipitation observation station needs to consider regional representativeness, site selection, site, equipment, observation and maintenance, etc. The observation should have regional representativeness, be able to represent the average meteorological conditions within a certain range around the observation point, ensure that the observation data meet the requirements of representativeness and accuracy, and keep the station site stable and unchanged, so the regional precipitation distribution, geological stability, etc. need to be considered when selecting the site.

[0057] In this embodiment, the site selection requirements include topography and environment, and the relatively flat terrain is preferred, and the steep slope, canyon and wind gap are avoided. When it is impossible to completely avoid obstacles, the distance between the rain gauge and the obstacle is not less than 2 times the height of the obstacle (or the height difference between the obstacle and the instrument port). In addition, the environment requirements are to be away from smoke and dust, industrial pollution sources and strong electromagnetic interference.

[0058] In this embodiment, the installation equipment needs regular maintenance, and the maintenance includes site maintenance and equipment maintenance, which need to be carried out regularly, so the accessibility of the site and the proximity to the populated area should be considered when selecting the site.

[0059] To ensure the smooth transmission of observation records, the stability of communication should also be considered when selecting the site. The use data includes regional range information, regional elevation, regional meteorological station multi-year precipitation observation data, regional meteorological station location information, regional populated area location, regional water system information, etc.

[0060] Please refer to Figures 1-2 In the embodiment of the application, a mountainous area meteorological precipitation supplementary observation station site selection evaluation method based on high-impact weather influence path is provided, which includes the following steps:

[0061] S1: Obtain meteorological grid point data in the observation area, select stormy days from the meteorological grid point data, regard consecutive stormy days as a storm process, extract the central longitude and latitude and the precipitation intensity based on the hourly precipitation data in the storm process, generate a path sequence, and generate the influence area data of the storm path widening by using the DTW method, wherein the meteorological grid point data includes daily precipitation and hourly precipitation in a certain period;

[0062] S2: Based on the precipitation observation data of the meteorological station in the observation area, generate regional grid point precipitation data by using the inverse distance weighted interpolation method, unify the spatial resolution of the regional grid point precipitation data, calculate the root mean square error to obtain the model prediction error, and select the minimum error of the model prediction error as the prediction error factor based on the root mean square error;

[0063] S3: Obtain the coordinates of the meteorological station, count the grid point station network density of the location coordinates, obtain the regional elevation data of the meteorological station, calculate the slope and slope direction of the meteorological station, and generate the populated area adjacent relationship of the meteorological station;

[0064] S4: Based on the multi-index data, calculate the information entropy and weight of each index by using the entropy weight method, calculate the site selection score of the grid point according to the index data and the corresponding weight, and select the grid point with a score reaching a preset score as the preliminary site selection area in combination with the influence area data of the storm path corresponding to the site selection of the grid point.

[0065] S5: Collect the terrain environment data of the grid point, and determine the target site selection area, wherein the terrain environment data includes terrain features, soil types, geological conditions, surface coverage, traffic accessibility, communication stability and energy supply stability.

[0066] Preferably, step S1 includes: selecting rainstorm days from meteorological grid data based on a pre-set daily precipitation threshold; taking several consecutive days of rainstorms as a rainstorm process to obtain the number of rainstorm processes in the meteorological grid data; and obtaining the center latitude and longitude data and precipitation intensity data of each rainstorm process based on the hourly precipitation data of the rainstorm process to generate a path sequence of rainstorm paths.

[0067] Preferably, step S1 further includes:

[0068] A data transformation method is used to preprocess the path sequence to remove noise data, resulting in a denoised sequence. A resampling method is then used to resample the denoised sequence into a sequence of equal length, using the average moving speed formula:

[0069] Calculate the average moving speed of the rainstorm path, where Δd is the path distance and Δt is the time interval; use the formula for the rate of change of direction:

[0070] θ = △α / △d is used to calculate the rate of change of direction of the rainstorm path, where △α is the change of direction angle. The DTW method is used to perform time series clustering on the average moving speed and the rate of change of direction to obtain the clustering results. The optimal number of clusters is determined based on the elbow rule and the profile coefficient. The average trajectory and precipitation intensity distribution data of the rainstorm path are calculated to generate the impact area data of the widening rainstorm path.

[0071] Preferably, step S2 includes:

[0072] The root mean square error formula is used:

[0073] Among them, P obes,i To observe precipitation, P mod,i For precipitation, M is the number of grid points, and the number of stations within each grid point is statistically assigned.

[0074] Preferably, step S3 includes:

[0075] Based on regional elevation data, the slope calculation formula is used:

[0076] The slope is calculated, where △z is the height difference of the regional elevation data and △h is the horizontal distance of the regional elevation data;

[0077] The slope aspect calculation formula is used as follows:

[0078] The slope aspect is calculated by A = arctan2(Δy,Δx), where A is the slope aspect, Δy is the vertical height difference component, and Δx is the horizontal distance component.

[0079] Based on the location of the settlement of the weather station, the nearest neighbor analysis method and the nearest neighbor distance formula are adopted:

[0080] The neighboring relationship is calculated, wherein (x p ,y p ) are the coordinates of the settlement.

[0081] Preferably, the multi-index data includes the historical average precipitation of the grid, the precipitation forecast error, the slope, the slope direction, the weather station density and the settlement neighboring relationship.

[0082] In this embodiment, the specific method for calculating the grid station density of the location coordinates is to count the number of stations in each grid.

[0083] Preferably, step S4 comprises:

[0084] The entropy weight method formula is adopted:

[0085] The information entropy of the index is calculated, wherein, is the feature proportion of the jth grid under the ith index, M is the total number of grids, and E i is the information entropy, when p ij =0, it is defined that p ij lnp ij =0;

[0086] The weight formula is adopted:

[0087] The weight w i of the ith index is calculated.

[0088] The grid site selection score formula is adopted:

[0089] The site selection score of the grid is calculated.

[0090] In this embodiment, the following Table 1 provides the terms and annotations:

[0091]

[0092] Table 1

[0093] Taking a certain river basin as an example, according to the historical five-year data, three types of influence paths can be clustered as follows:

[0094] The rainstorm path type 1 is the most important rainstorm type in the three types, accounting for 57.9% of the total samples. From Figure 3It can be observed from the figure that the type 1 influence area is mainly located in the southwest of the study area (around 28 °N, 97 °E), showing a local feature. The average duration of this type is 2.9 days, and the average maximum daily rainfall is 111.8 mm, showing a clear feature of heavy rainfall. Its average moving speed is relatively slow (1.5 km / d), and the average direction change rate is 1.418 rad / km, indicating that this type of rainstorm system moves slowly and is relatively stable.

[0095] From the influence area, type 1 rainstorm system is far from the two river mouths of Yalongjiang River, and has limited direct impact on the two river mouths. This type of system may be related to shear line weather system. When the cold air from Qinghai and Gansu meets the warm and humid air in the south of Sichuan and the north of Yunnan, a wind shear is formed at 700 hPa height. However, due to its slow moving speed, the influence range is limited, mainly affecting the southwest region of the study area, and the rainfall in the two river mouths is relatively small.

[0096] Type 2 rainstorm path accounts for 25% of the total samples, from Figure 3 It can be seen from the figure that its influence area is distributed in a northeast-southwest direction, across the middle of the study area (about 29 °N, 97 °E to 101 °E). The average duration of this type is 3.7 days, which is the longest among the three types, and the average maximum daily rainfall is 118.1 mm, which is the largest among the three types. Its significant feature is fast average moving speed (9.5 km / d) and drastic direction change (average direction change rate is as high as 4.991 rad / km).

[0097] From the trajectory, type 2 rainstorm system moves from southwest to northeast, passing through the southern edge of the two river mouths of Yalongjiang River. This type of system may be closely related to the southwest low vortex weather type. When the southwest low vortex develops and moves eastward at 700 hPa height, it will bring significant rainfall. Although the rainstorm center may not directly affect the two river mouths, due to its fast moving speed and wide coverage, the two river mouths may be affected by the peripheral cloud system, bringing moderate to heavy rain, and may appear as a rainstorm when the moving path of the rainstorm system is close.

[0098] Type 3 rainstorm path accounts for 17.1% of the total samples, which is the smallest among the three types. From Figure 3 It can be seen from the figure that its influence area is mainly concentrated in the east of the study area (about 29.5 °N, 100 °E to 102 °E), showing a relatively narrow east-west band distribution. The average duration of this type is 2.9 days, and the average maximum daily rainfall is 105.9 mm, which is relatively small among the three types. Its average moving speed is 6 km / d, and the average direction change rate is 1.762 rad / km, showing a medium moving speed and a relatively stable moving direction.

[0099] From the geographical position, the influence area of the type 3 rainstorm system just covers the two estuary regions of the Yalong River. The system moves from east to west and directly crosses the two estuary regions, so the precipitation influence on the two estuary regions is the most direct and significant. This kind of system may be related to the plateau trough weather type. When the plateau trough on the 500 hPa isobaric surface moves eastward to the western edge of the subtropical high, different frontogenesis factors will produce frontogenesis effect at the same time, causing severe weather in the Yalong River basin, especially in the two estuary regions. When this kind of system affects, the two estuary regions may appear sustained heavy rain, which needs to be highly concerned about its hydrological response. When the extraction of the evolution path of the rainstorm process is completed, the scoring of the grid is calculated according to the index data and the corresponding weight, combined with the influence area data of the rainstorm path corresponding to the grid selection of the grid point, the grid point whose score reaches the preset score is selected as the preliminary site selection area, and the target site selection area is determined combined with the terrain environment data.

[0100] The above is only the preferred specific embodiment of the present application, but the protection scope of the present application is not limited to this. Any person skilled in the art can make equivalent replacement or change according to the technical solution and the inventive concept of the present application within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A mountainous weather precipitation supplementary observation station site evaluation method based on a high-impact weather influence path, characterized in that, The method comprises the following steps: Obtaining meteorological grid point data in an observation area, screening rainstorm days from the meteorological grid point data, regarding consecutive rainstorm days as a rainstorm process, extracting the central longitude and latitude and the precipitation intensity based on the hourly precipitation data in the rainstorm process, and generating a path sequence, and using a dynamic time warping method to generate the influence area data of the widening of the rainstorm path, wherein the meteorological grid point data comprises daily precipitation and hourly precipitation in a certain period; Based on the precipitation observation data of the meteorological stations in the observation area, using the inverse distance weighted interpolation method to generate regional grid precipitation data, unifying the spatial resolution of the regional grid precipitation data, calculating the root mean square error to obtain the model prediction error, and selecting the minimum prediction error as the prediction error factor based on the root mean square error; Obtaining the coordinates of the meteorological stations, counting the grid station network density of the position coordinates, obtaining the regional elevation data of the meteorological stations, calculating the slope and slope direction of the meteorological stations, and generating the neighborhood relationship of the meteorological stations in the populated areas; Based on multi-index data, using the entropy weight method to calculate the information entropy and weight of each index, calculating the site selection score of the grid point according to the index data and the corresponding weight, and selecting the grid point with a score reaching a preset score as the preliminary site selection area in combination with the influence area data of the rainstorm path corresponding to the site selection of the grid point; Collecting the terrain environment data of the grid point, and determining the target site selection area, wherein the terrain environment data comprises terrain features, soil types, geological conditions, surface coverage, traffic accessibility, communication stability and energy supply stability.

2. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 1, characterized in that: The method for obtaining meteorological grid point data of meteorological stations in an observation area, screening rainstorm days from the meteorological grid point data, regarding consecutive rainstorm days as a rainstorm process, extracting the central longitude and latitude and the precipitation intensity based on the hourly precipitation data in the rainstorm process, and generating a path sequence, comprises the following steps: Screening rainstorm days from the meteorological grid point data based on a pre-set daily precipitation threshold; Regarding consecutive rainstorm days as a rainstorm process to obtain the number of rainstorm processes in the meteorological grid point data; Based on the hourly precipitation data of the rainstorm process, obtaining the central longitude and latitude data and the precipitation intensity data of each rainstorm process to generate a path sequence of the rainstorm path.

3. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 2, characterized in that: The method for generating the influence area data of the widening of the rainstorm path using a dynamic time warping method comprises the following steps: Using a data transformation method to preprocess the path sequence to remove noise data in the path sequence to obtain a denoising sequence; Using a resampling method to resample the denoising sequence into an equal-length sequence, using an average moving speed formula: calculating an average moving speed of the storm path, where Δd is the path distance and Δt is the time interval; Using a direction change rate formula: θ = △α / △d to calculate the direction change rate of the rainstorm path, wherein △α is the direction angle change; Using a dynamic time warping method to perform time sequence clustering on the average moving speed and the direction change rate to obtain a clustering result; Based on the elbow rule and the contour coefficient, determining the best clustering number of the clustering result, and calculating the average trajectory and the precipitation intensity distribution data of the rainstorm path to generate the influence area data of the widening of the rainstorm path.

4. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 3, characterized in that: The method for calculating the root mean square error to obtain the model prediction error comprises the following steps: Using a root mean square error formula: where P obes,i is the observed precipitation, P mod,i is the precipitation, and M is the number of grid points.

5. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 4, characterized in that: The method comprises the following steps: The number of stations in the grid is counted and assigned.

6. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 5, characterized in that: The method comprises the following steps: Based on the regional elevation data, the slope calculation formula is used: The slope is calculated, where Δz is the height difference of the area elevation data, and Δh is the horizontal distance of the area elevation data. The slope calculation formula is used: A = arctan2 (△y,△x) is used to calculate the slope direction, wherein A is the slope direction,△y is the vertical height difference component, and△x is the horizontal distance component.

7. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 6, characterized in that: The method comprises the following steps: Based on the location of the meteorological station, the nearest neighbor analysis method and the nearest neighbor distance formula are used: The proximity relationship is calculated, where (x p ,y p ) are the coordinates of the populated place.

8. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 7, characterized in that: The multi-index data includes the historical average precipitation, precipitation forecast error, slope, slope direction, meteorological station density and the proximity of the meteorological station.

9. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 8, characterized in that: The method comprises the following steps: The entropy weight method formula is used: calculating information entropy of the index, wherein, is the feature proportion of the jth grid point under the ith index, M is the total number of the grid points, E i is the information entropy, when p ij = 0, it is defined that p ij lnp ij = 0; The weight formula is used: calculating the weight w of the i-th indicator i .

10. The mountain weather precipitation supplementary observation station site evaluation method based on high-impact weather influence path according to claim 9, characterized in that: The method comprises the following steps: The grid site selection score is calculated according to the index data and the corresponding weight, and the influence area data of the corresponding rainstorm path of the grid site selection is combined to select the grid points with a score reaching a preset score as the preliminary site selection area. The site selection score of the grid point is calculated.

Images