新安江模型和人工神经网络的耦合应用 - 水土保持通报.pdf

30 6 Vol. 30, No . 6 2010 12 Bulletin of Soil and Water Conservation Dec. , 2010 宋晓猛, 孔凡哲 ( , 221116) : , , , , , GIS DEM ; , , ; , , , , : ; ; ; ; : B : 1000-288X(2010) 06-013 -04 : P338. 9 Application of Xinanjiang Model Coupling with Artificial Neural Networks SONG Xiao-meng, KONG Fan-zhe (School of Resources and Earth Science, China University of Mining and Technology, X uzhou, Jiangsu 221008, China) Abstract: A hybrid form of rainfal-l runoff models integrating artificial neural networks ( ANNs) w ith conceptual models is proposed. T he integrated model is a sem-i distributed form of conceptual rainfal-l runoff models, in consider- ation of the spatial variation of rainfall, the heterogeneity of w atershed characteristics and their impacts on runoff. Genetic algorithm is used to optimize the parameters of the conceptual model and GIS software and DEM data are used to divide the w hole catchment into sub-catchments based on the spatial distribution of rain-gage sta- tions. As a result, in each sub-catchment, runoff generation is simulated in consideration of the spatially dis- tributed model parameters and rainfall inputs. In runoff routing, instead of a linear superposition of routed runoff from all sub-catchments as traditionally performed in a sem-i distributed form of conceptual models, ar- tificial neural networks as an effective tool in nonlinear mapping are employed to estimate runoff. T he feasi- bility of the new approach is demonstrated in Dapoling w atershed, the upper tributary of Huaihe River basin, and the results of coupling model are com