The river health evaluation is typically complex non-linear system with characteristics of fuzziness and randomness. However, conventional gray clustering method has difficult to effectively describe fuzzy and random information simultaneously. For this purpose, the cloud model and fuzzy entropy theory are introduced to establish 2D gray cloud clustering-fuzzy entropy comprehensive evaluation model. Different with health level models, it reflects river health situation from aspects of health level and corresponding water body complexity simultaneously. The health level is obtained by gray cloud whitened weight function (first sub-system) and fuzzy entropy represents complexity and fuzziness of river health situation (second sub-system). Moreover, multi-level river health evaluation indicator system is constructed with dividing indicators into common and distinct sections according to differences on river characteristics. Meanwhile, indicator weights are determined by renewed combined weighting method based on minimum deviation principle. Finally, we conduct health evaluation work for rivers in the Taihu basin. The evaluation health levels and fuzzy entropy for river A–G are H3 (0.4888, relatively significant); H2 (0.5476, relatively fuzzy); H2 (0.7526, fuzzy); H2 (0.4731, relatively significant); H2 (05138, relatively fuzzy); H3 (0.5822, relatively fuzzy), and H2 (0.4064, relatively significant), respectively. Results are consistent with current river health situation and more intuitive than compared models. Furthermore, evaluation results with four different weighting methods are compared to further demonstrate rationality of the weighting method and evaluation model. Hence, the model proposed is demonstrated to provide new insight for solving river health assessment problem effectively.

河流健康评估通常是具有模糊性和随机性特征的复杂非线性系统。然而，传统的灰色聚类方法难以同时有效地描述模糊和随机信息。为此，引入云模型和模糊熵理论，建立了二维灰色云聚类-模糊熵综合评价模型。与健康水平模型不同，它同时从健康水平和相应的水体复杂性方面反映河流的健康状况。通过灰云白化权重函数（第一子系统）获得健康水平，模糊熵表示河流健康状况（第二子系统）的复杂性和模糊性。此外，建立了多级河流健康评价指标体系，根据河流特征的不同，将指标分为共同部分和不同部分。同时，指标权重根据最小偏差原理通过更新的组合加权法确定。最后，我们对太湖流域的河流进行健康评估。A–G河的评估健康水平和模糊熵为H3（0.4888，相对显着）；H2（0.5476，相对模糊）；H2（0.7526，模糊）；H2（0.4731，相对显着）；H2（05138，相对模糊）；H3（0.5822，相对模糊）和H2（0.4064，相对显着）。结果与当前河流健康状况一致，并且比对比模型更直观。此外，比较了四种不同加权方法的评估结果，以进一步证明加权方法和评估模型的合理性。因此，所提出的模型被证明为有效解决河流健康评估问题提供了新的见识。