%0 Journal Article
%T DBSCAN算法在优化农业灌区划分中的应用研究
The Application Research of DBSCAN Algorithm in Optimizing the Division of Agricultural Irrigation Areas
%A 毛新宇
%A 刘泓辰
%A 郜婷婷
%A 董欣月
%A 彭亮
%J Geomatics Science and Technology
%P 109-128
%@ 2329-7239
%D 2025
%I Hans Publishing
%R 10.12677/gst.2025.132014
%X 随着农业现代化进程的不断推进，农业灌区划分的优化工作显得尤为重要。本研究运用一种基于密度分类的DBSCAN算法作为农业灌区划分的新方法，该方法结合当地灌溉井、水系、耕地土壤类型、土地利用现状等影响因素，并基于Visual Studio 2019软件与C#编程语言，编写DBSCAN算法代码及聚类评估代码，以灌溉井坐标信息为例进行聚类分析，得出当地农业灌区的划分结果。结果表明：① 对比传统划分方法需要考虑的八大方面，本研究采用的基于DBSCAN算法的灌区划分方法可以在保证准确率(92.86%)的前提下，仅考虑四个方面，即通过更少的工作量完成灌区划分的前期准备工作。② 基于C#语言的DBSCAN算法架构及聚类评估体系，可以有效反映不同要素对灌区划分结果的影响，即可以通过改变坐标点的含义达到侧重性划分的目的。③ 算法中的参数Eps和MinPts可以代表目标点的半径及密度限制，这为以后的新建、续建和维护灌区提供了广泛的适用性。研究成果对新增灌区建设及续建工程的规划有一定借鉴意义。
As the process of agricultural modernization continues to advance, the optimization of agricultural irrigation area division has become particularly important. This study employs a density-based clustering algorithm known as DBSCAN as a new method for dividing agricultural irrigation areas. The method integrates local factors such as irrigation wells, water systems, soil types of cultivated land, and current land use status, and is based on the Visual Studio 2019 software and the C# programming language to write DBSCAN algorithm code and clustering evaluation code. Taking the coordinates of irrigation wells as an example, cluster analysis is conducted to obtain the division results of local agricultural irrigation areas. The results show: ① Compared to the eight aspects that need to be considered in traditional division methods, the DBSCAN algorithm-based irrigation area division method used in this study can, while ensuring accuracy (92.86%), only consider four minor aspects, that is, to complete the preliminary preparation work of irrigation area division with less workload. ② The DBSCAN algorithm architecture and clustering evaluation system based on the C# language can effectively reflect the impact of different factors on the division results of irrigation areas, that is, it is possible to achieve a focused division by changing the meaning of coordinate points. ③ The parameters Eps and MinPts in the algorithm can represent the radius and density limit of the target point, which provides broad applicability for the future construction, continuation, and maintenance of irrigation areas. The research findings have certain reference significance for the planning of new irrigation area construction and continuation projects.
%K 农业灌区，
%K 聚类分析，
%K DBSCAN算法，
%K SC-DB-V-Measure
Agricultural Irrigation Areas
%K Cluster Analysis
%K DBSCAN Algorithm
%K SC-DB-V-Measure
%U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=113239