【摘要】 位于黄河流域上游干旱半干旱的青铜峡灌区是宁夏工农业生产和社会经济发展的重要地区。灌区耗水量占宁夏引黄灌区耗水量的绝大部分,因此,对该灌区耗水量研究非常必要。本论文在查阅大量国内外相关文献资料的基础上,针对灌区农业耗水量的影响因素做了分析,将引水量、排水量、降雨量、蒸发量和地下水位埋深作为最主要的影响因子,建立青铜峡河东灌区农业灌溉耗水量的BP神经网络预测模型。结果表明,网络函数的选取和结构设计较为合理,误差满足要求,精度较高,可应用于进行其它年份灌溉耗水量的预测。应用此网络对2003年河东灌区农田灌溉耗水量进行了预测,证明该模型适合于灌区农业耗水量的预测,具有一定的推广价值。更多还原

【Abstract】 Located in the upper reaches of the Yellow river in Ningxia Qing Tongxia industrial and agricultural production and irrigation is the essence of regional socio-economic development.Ningxia irrigation water accounts for the vast majority of irrigation water consumption,so the amount of irrigation water is necessary.In this thesis,access to a large number of domestic and foreign literature,based on water consumption for irrigation of agricultural factors were analyzed,the amount of water,discharge,rainfall,evaporation and groundwater depth as the most important factors affecting the establishment of Qing Tongxia east of the river irrigation water consumption BP neural network prediction model.The results show that the selection function and structure of the network design is more reasonable,error meets the requirements of high precision,and can be applied to other years of irrigation water consumption in the forecast.Application of this network,the irrigation area in 2003 east of the river irrigation water consumption is predicted.The model proves to be suitable for irrigated agriculture and water consumption forecast is worth promoting.更多还原