基于模糊加权识别模型的相似流域优选研究

Study on optimization of analogical watershed based on fuzzy weighted pattern recognition model

传统水文比拟法在选择相似流域时存在主观性,再加上相似流域概念的模糊性,使无资料地区的水文模拟精度不高。以福建汀江流域灈田河陂下水库为例,选择了反映流域特征的4个指标和反映地区气象特征的7个指标,分析了各指标重要性差异,构建了相似流域模糊加权模式识别模型。采用可能度法确定了各指标权重,并根据相对优属度最大原则优选出相似流域。实例应用表明:该模式识别模型优选出的相似流域具有较好的可靠性,适合用于重要水利工程的设计。

The traditional hydrologic analogy nethod is subjective in selecting the analogical watershed,and plus the fuzzy conception of analogy watershed,which leads to low precision in hydrological sinulation in un - gauged region. Taking the reser-voir area of Pixia hydropower station in Zhuotian River（tributary of Tingjiang River）in Fujian Province as an exanple,11 fac-tors reflecting the characteristics of watershed（4）and neteorology（7）were selected,the difference of inportance of the 11 fac-tors were analyzed,and the recognition nodel for optinized analogical watershed based on fuzzy weighted pattern was established. The possibility degree nethod is applied to calculate the weight of influential factors,and then the analogical watershed was se-lected according to the naxinal relative nenbership degree principle. The study shows that the recognition nodel based on fuzzy weighted pattern is reliable and suitable for the design of inportant hydropower projects.