With the growing complexity of the human living environment, the environment-related industries have also been flourishing. Clustering of environmental data is a key task of environment research. The environmental data are characterized by diversification, boundary fuzzification, incompleteness, etc. This paper conducts a cluster analysis of environmental data based on fuzzy theory. To start with, the basic principle of fuzzy theory is analyzed, and the focus is on studying the membership function and the $\mathrm{\lambda }$-cut-set knowledge. Following that, the fuzzy clustering method and its process are studied. Finally, fuzzy evaluation is used to build the membership function after experiment on the MATLAB platform to evaluate the environmental quality. The fuzzy C-means clustering algorithm is used to realize the target identification of environmental data. In the process of fuzzy clustering, fuzzy evaluation of the seawater quality is realized, and the redundant data of the monitoring station are removed. Through experiment and analysis, experimental results are in line with the practical situations and show a high consistency with the data characteristics. Compared with the traditional clustering algorithm, fuzzy clustering is more suitable for environmental data processing in environmental data research and analysis.

随着人类生活环境的日益复杂，与环境有关的产业也蓬勃发展。环境数据的聚类是环境研究的关键任务。环境数据具有多样性，边界模糊，不完备等特点。本文基于模糊理论对环境数据进行了聚类分析。首先，分析了模糊理论的基本原理，重点是研究隶属函数和$\mathrm{\lambda }$切定知识。然后，研究了模糊聚类方法及其过程。最后，在MATLAB平台上进行实验后，通过模糊评估建立隶属函数，以评估环境质量。采用模糊C-均值聚类算法实现环境数据的目标识别。在模糊聚类的过程中，实现了对海水水质的模糊评价，去除了监测站的冗余数据。通过实验和分析，实验结果符合实际情况，与数据特征具有较高的一致性。与传统的聚类算法相比，模糊聚类更适合环境数据研究与分析中的环境数据处理。