贵州典型岩溶流域日降雨径流过程模拟研究

以蓄满产流方式描述岩溶地区的产流过程,采用稳定入渗法划分径流,将非线性水箱模型用于模拟单元流域的调蓄过程,以遗传算法为基础率定模型参数,建立了基于遗传算法率定参数的概念性岩溶水文模型,并以贵州普定后寨河流域为例,采用流域内老黑潭、六谷、后寨测站的水文资料对所建模型进行检验,以相对误差、互相关系数和确定性系数来评定模型.结果表明,本模型预报结果的相对误差均小于士10%,互相关系数均大于0.80,确定性系数均大于或等于0.70,说明所构建的模型能够模拟及预报岩溶地区的降雨径流过程. 

Simulation to daily rainfall-runoff process in typical karst basin in Guizhou

In this study, the full-stored runoff pattern is used to describe runoff process in karst area, and the stream flow process is separated based on the method of stable infiltration. The non-linear Tank Model is applied to simulate regulation and storage of catchments. According to the method above, conceptual karst hydrological model is established, whose parameters are calibrated and determined by genetic algorithm. The Houzhaihe karst catchment in Puding county, Guizhou province is set an example to analysis of the rainfallrunoff process in karst area. Hydrological data from Laoheitan, Liugu and Houzhai monitor station is applied to test the models. Comparative analysis is made with relative error, cross-correlation coefficient and deterministic coefficient to evaluate model. Relative error of the predicted flow is less than 10%, and cross-correlation coefficient is greater than 0.80, and deterministic coefficient is greater than or equal to 0.70. The analyed results show that the hydrological processes could be simulated and forecasted with the presented model.