区域水资源可持续利用评价的神经网络方法

在论述区域水资源可持续利用与水资源承载能力关系的前提下,提出了区域水资源可持续利用评价的神经网络方法。建立的神经网络模型对汉中盆地水资源可持续利用程度进行了评价。应用实例表明:建立的模型与其他模糊综合评价和属性识别方法相比,评价结果更合理可行和可靠。对汉中盆地来说,城固的水资源可持续利用程度为低级,其他5个区县的水资源可持续利用程度为中级偏低,其中又以南郑为最高,勉县为最低。在研究的7项评价指标中,水资源利用率对可持续利用程度的影响最大,灌溉率的影响次之,生态环境用水率的影响最小,从而为提高区域水资源可持续利用程度提供了理论依据。表3,参9。

On assessment of sustainable development level of regional water resource using artificial neural networks

Based on the discussion of relationship between sustainable developmental level and bearing capacity of regional water resource, an assessment model of sustainable development level of regional water resource (SDLRWR) using artificial neural networks (ANN) was put forward and applied to assess the SDLRWR of Hanzhong Basin.The case study shown that the presented model was more reliable and the assessed results were more reasonable and practicable than that of other methods such as fuzzy and comprehensive mathematics, attribute recognition method etc.The SDLRWR of Chenggu county was low-level and the SDLRWR of other five counties was medium-level.The SDLRWR of Nanzheng county was best in the five counties, and that of Mianxian worst.In the seven evaluation indexes, investigated in this paper, influencing on the SDLRWR, the utilization ratio of water resource was the most important index, then the irrigation ratio of irrigated area to land area, and the eco-environmental water-consumption ratio of the amount of water used in eco-environment to total water was the lest important index.