Search

CN-122020196-A - Method, system, equipment and storage medium for distinguishing and clustering similarity of small-stream-area middle-stream-area and small-stream-area of yellow river

CN122020196ACN 122020196 ACN122020196 ACN 122020196ACN-122020196-A

Abstract

The invention discloses a method, a system, equipment and a storage medium for distinguishing and clustering similarity of small watershed in yellow river, which are used for obtaining static topography index data and dynamic hydrologic index data of a target small watershed, carrying out standardized processing on the static topography index data and the dynamic hydrologic index data to obtain standardized data, calculating variation coefficients of each static topography index and each dynamic hydrologic index, determining weights of each static topography index and each dynamic hydrologic index according to the variation coefficients, calculating feature space distance between the small watershed by adopting a weighted Euclidean distance algorithm based on the weights and the standardized data, determining similarity coefficients of the small watershed according to the feature space distance, grouping the small watershed by utilizing a K-means clustering algorithm, calculating the optimal clustering number of the contour coefficients, and obtaining a small watershed similarity clustering result. The similarity between the watersheds can be quantified comprehensively and accurately.

Inventors

  • Fei Gengjie

Assignees

  • 中国科学院地球环境研究所

Dates

Publication Date
20260512
Application Date
20260123

Claims (10)

  1. 1. A method for discriminating and clustering similarity of small-stream-area middle-stream-area and small-stream-area of yellow river is characterized by comprising the following steps: Acquiring static topography index data and dynamic hydrologic index data of a small target drainage basin, and carrying out standardization processing on the static topography index data and the dynamic hydrologic index data to obtain standardized data; calculating variation coefficients of each static topography index and each dynamic hydrologic index, and determining weights of each static topography index and each dynamic hydrologic index according to the variation coefficients; Based on the weight and the standardized data, calculating the feature space distance between the small waterbasins by adopting a weighted Euclidean distance algorithm, and determining the similarity coefficient of the small waterbasins according to the feature space distance; Grouping the small watersheds by using a K-means clustering algorithm, calculating a contour coefficient to determine the optimal clustering number, and obtaining a small watershed similarity clustering result.
  2. 2. The method for distinguishing and clustering the similarity of small and medium-stream vamps in a yellow river according to claim 1, wherein the static topography indexes comprise a river area, a river basin shape factor, an average gradient, a topography humidity index, a river length, a river specific drop, a cultivated land rate, a forest land rate, a grassland rate, a terrace ratio and a silting dam quantity, and the dynamic hydrologic indexes comprise a rainfall, a river runoff and a river sand content.
  3. 3. The method for distinguishing and clustering similarity of small and medium-stream river basins according to claim 1, wherein the standardization process adopts a Z-Score standardization method to eliminate dimension differences of static topography indexes and dynamic hydrologic indexes.
  4. 4. The method for distinguishing and clustering similarity of small and medium-stream river basins according to claim 1, wherein the step of determining each static topography index and each dynamic hydrologic index weight comprises the steps of: Wherein E i is the coefficient of variation of index i, sigma i and mu i are the standard deviation and the mean of the index, omega i is the weight of index i, and n is the total number of indexes.
  5. 5. The method for distinguishing and clustering similarity of small river basins in the middle of yellow river according to claim 1, wherein the process of calculating the characteristic space distance between the small river basins comprises the following steps: Where D ij is the weighted Euclidean distance between basin i and basin j, ω k is the weight of the kth index, and x ik and x jk are the normalized values of basin i and basin j on the kth index, respectively.
  6. 6. The method for discriminating and clustering similarity of small river basins in yellow river according to claim 5 wherein determining the similarity coefficient of small river basin comprises: Wherein: As a coefficient of similarity(s) to be used, In order for the euclidean distance to be the same, Is the maximum distance.
  7. 7. The method for distinguishing and clustering similarity of small-stream-area and medium-stream-area in yellow-river according to claim 1, wherein the process of calculating the contour coefficient to determine the optimal clustering number comprises the following steps: Wherein a i is the average distance from the sample i to all other samples in the same cluster to reflect the degree of compactness in the cluster, b i is the average distance from the sample i to all samples in the nearest neighbor cluster to reflect the degree of separation between clusters, S i has the value range of [ -1,1], the larger the value is, the more reasonable the sample cluster is, and for a complete clustering result, the whole quality is measured by the average value of all the sample contour coefficients, namely the average contour coefficient S k : In the formula, n is the total number of samples, S k corresponding to different preset k values is calculated, and the k value which maximizes S k is selected as the final cluster number.
  8. 8. A yellow river basin middle and small river basin similarity discrimination and clustering system is characterized by comprising: The index data acquisition module is used for acquiring static topography index data and dynamic hydrologic index data of the small-basin target, and carrying out standardization processing on the static topography index data and the dynamic hydrologic index data to obtain standardized data; The index weight determining module is used for calculating variation coefficients of each static topography index and each dynamic hydrologic index and determining the weights of each static topography index and each dynamic hydrologic index according to the variation coefficients; The similarity calculation module is used for calculating the characteristic space distance between the small watersheds by adopting a weighted Euclidean distance algorithm based on the weight and the standardized data, and determining the similarity coefficient of the small watersheds according to the characteristic space distance; and the clustering module is used for grouping the small watersheds by using a K-means clustering algorithm, calculating the contour coefficient to determine the optimal clustering number, and obtaining a small watershed similarity clustering result.
  9. 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 the processor, when executing the computer program, carries out the steps of the method for discriminating and clustering similarity of small-stream vaults in yellow river vaults according to any one of claims 1 to 7.
  10. 10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the steps of the method for discriminating and clustering similarity of small-stream-area middle-stream-area and small-stream-area according to any one of claims 1 to 7.

Description

Method, system, equipment and storage medium for distinguishing and clustering similarity of small-stream-area middle-stream-area and small-stream-area of yellow river Technical Field The invention belongs to the field of hydrologic prediction, and relates to a method, a system, equipment and a storage medium for distinguishing and clustering similarity of small-stream-area and middle-stream-area of a yellow river. Background The loess plateau area in the yellow river basin is a continuous loess accumulation area with the largest global standard, and the soil erosion problem of the area is prominent, so that the loess plateau area is one of the areas with the most serious global water and soil loss. The yellow river has huge sand production in the section, and the confluence mechanism is complex. In recent years, vegetation coverage in most areas of loess plateau has been significantly improved (zhangetal, 2000). However, the negative balance of soil moisture is also caused at the same time of vegetation restoration, the problems of soil desiccation and dry soil layer thickening are increasingly prominent, and challenges are posed to vegetation sustainable restoration and ecological system stability (Shao Mingan, et al, 2016). The Huang Hezhong water-sand-travelling process and vegetation dynamic state have complex coupling relation (Shang Qiuhong and the like, 2023), and the complexity of a yield and confluence mechanism directly influences the sediment transport rule of the yellow river and has important influences on the hydrologic forecast of the river, flood control, drought resistance, water resource regulation and the like. Therefore, the development of water and sand prediction research in loess plateau areas has important significance for optimizing the water resource allocation of yellow river basin, implementing water and sand regulation, perfecting a flood control and disaster reduction system and promoting comprehensive management of river channels (Li Linqi and the like, 2025). The reliable data support is the basis of hydrologic forecasting (Vasilotal.2021), but the condition space heterogeneity of the underlying area is strong, the influence of human activities (such as town, ecological restoration, silt dam construction and the like) is remarkable, and moreover, hydrologic sites are sparse and unevenly distributed, so that a plurality of small waterbasins face the dilemma of data shortage, and the deep development of water and sand forecasting work is seriously restricted. The watershed feature similarity method based on hydrologic similarity theory (Liu Jintao, et al, 2014) becomes an effective way for solving the hydrologic prediction problem in the non-data or non-data areas. According to the method, the existing mature hydrologic model is transplanted to a data-free watershed with similar water-sand characteristics by quantifying the similarity degree of attribute characteristics among watersheds, so that scientific prediction of the hydrologic process is realized. Although the watershed feature similarity method based on the hydrologic similarity theory is an effective way for solving the prediction of the non-data area, the similarity degree between watersheds is difficult to comprehensively and accurately quantify aiming at the areas with high sand content such as the middle stream of the yellow river in the prior art, and the problems of the hydrologic model transplanting and the accurate classification treatment of the non-data watershed cannot be effectively solved. Disclosure of Invention The invention aims to overcome the defects of the prior art, and provides a method, a system, a device and a storage medium for distinguishing and clustering similarity of small and medium-stream river basins, which can comprehensively and accurately quantify the similarity between the river basins and effectively solve the problems of hydrological model transplanting and accurate classification treatment of the non-data river basins. In order to achieve the purpose, the invention is realized by adopting the following technical scheme: A method for discriminating and clustering similarity of small river basins in yellow river basin includes: Acquiring static topography index data and dynamic hydrologic index data of a small target drainage basin, and carrying out standardization processing on the static topography index data and the dynamic hydrologic index data to obtain standardized data; calculating variation coefficients of each static topography index and each dynamic hydrologic index, and determining weights of each static topography index and each dynamic hydrologic index according to the variation coefficients; Based on the weight and the standardized data, calculating the feature space distance between the small waterbasins by adopting a weighted Euclidean distance algorithm, and determining the similarity coefficient of the small waterbasins according to the feature space distance; Gro