• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Hydrology and Earth System Sciences >Clustering of heterogeneous precipitation fields for the assessment and possible improvement of lumped neural network models for streamflow forecasts
【24h】

Clustering of heterogeneous precipitation fields for the assessment and possible improvement of lumped neural network models for streamflow forecasts

机译：聚类非均质降水场以评估和可能改进集总神经网络模型以进行流量预报

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This work addresses the issue of better considering the heterogeneity ofprecipitation fields within lumped rainfall-runoff models where only arealmean precipitation is usually used as an input. A method using a Kohonenneural network is proposed for the clustering of precipitation fields.The evaluation and improvement of the performance of a lumpedrainfall-runoff model for one-day ahead predictions is then establishedbased on this clustering. Multilayer perceptron neural networks areemployed as lumped rainfall-runoff models. The Bas-en-Basset watershed inFrance, which is equipped with 23 rain gauges with data for a 21-yearperiod, is employed as the application case. The results demonstrate therelevance of the proposed clustering method, which produces groups ofprecipitation fields that are in agreement with the global climatologicalfeatures affecting the region, as well as with the topographic constraintsof the watershed (i.e., orography). The strengths and weaknesses of therainfall-runoff models are highlighted by the analysis of their performancevis-à-vis the clustering of precipitation fields. The results alsoshow the capability of multilayer perceptron neural networks to account forthe heterogeneity of precipitation, even when built as lumpedrainfall-runoff models.
机译：这项工作旨在更好地考虑集总降雨径流模型中降水场的非均质性问题，在该模型中通常仅使用区域降水作为输入。提出了一种使用Kohonenneural网络的方法对降水场进行聚类的方法，然后基于该聚类方法建立了针对未来一日预报的集总径流模型的性能评估和改进方法。多层感知器神经网络被用作集总降雨-径流模型。应用案例是法国的Bas-en-Basset流域，该流域配备了23个雨量计以及21年的数据。结果证明了拟议的聚类方法的相关性，该方法产生与影响该地区的全球气候特征以及分水岭（即地形）的地形约束相一致的降水场组。通过对降雨场聚类的性能分析，可以突出显示降雨径流模型的优缺点。结果还显示了多层感知器神经网络解决降水异质性的能力，即使将其构建为集总降雨-径流模型也是如此。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号