摘要
为了深入探求地下水动态变化规律,以陕西洛惠渠灌区多年实测数据为例,首先引用3层前馈型BP网络建模方法,对灌区自然-人工-生物条件下地下水动态进行了研究,其次采用改进的灰色斜率关联度方法分析了各影响因子与地下水埋深的敏感程度,根据斜率关联度进行了综合排序.结果表明,该人工神经网络模型具有较高的精度,可以真实地定量描述地下水动态与各影响因子之间的响应关系;改进的灰色斜率关联度法能够很好的分析各因子对地下水动态的影响程度;蒸发量是影响该灌区地下水动态的主要因子,各因子之间相互作用,相互影响,形成了复杂条件下的耦合关系.将这两种方法结合运用到灌区地下水动态评价中是切实可行的,是对传统地下水动态研究方法的补充与完善.
In order to deeply seek the variation of groundwater dynamic, groundwater dynamic under natural-artificial-biological conditions was studied with measured data of Luohuiqu irrigation district in Shaanxi as an example based on the application of BP network of three layers. Secondly, the improved grey slope coefficient correlation degree analysis method was applied to analyze indetail the influence degree between various factors and groundwater depth. The factors are comprehensively sequenced on the basis of grey slope coefficient correlation degree. The results show that the artificial neural network model well expresses quantitatively the responsive relationship between groundwater dynamic and various factors with sufficient high accuracy. The improved grey slope coefficient correlation degree analysis method can well analyze the influence degree amongst various factors on groundwater dynamic. The evaporation is the main factor affecting groundwater dynamic in this irrigation district, and the interaction and interrelationship amongst various factors form coupling relationship under complicated conditions. The a application of the combination with two methods to the appraisal of groundwater dynamic in irrigation district,is feasible and practical and it is a complement and perfection for the traditional research methods of groundwater dynamic.
出处
《西安建筑科技大学学报（自然科学版）》
CSCD
北大核心
2009年第4期566-570,共5页
Journal of Xi'an University of Architecture & Technology(Natural Science Edition)
基金
国家科技支撑项目(2006BAD09B02)
国家科学自然基金资助项目(40771124)
中日合作项目(SBS-379)
关键词
人工神经网络
灰色斜率关联度
地下水动态
洛惠渠灌区
artificial neural network
grey slope coefficient correlation degree
groundwater dynamic
Luohuiqu irrigation district
