基于模糊综合评价和BP神经网络的景观河道水质评价研究

发布时间：2018-05-10 22:15

  本文选题：模糊综合评价法 + BP神经网络　； 参考：《河北工业大学》2015年硕士论文

【摘要】：水是生命之源,是一切生命生存和发展不可或缺的重要自然资源,但是由于人们对水资源保护意识不强,加之污染物种类随社会发展日趋复杂,水资源短缺和水质污染逐渐成为全球性的重大问题,水环境的相关研究也随之发展起来。水质评价是水环境研究的一个重要组成部分,在水资源的管理和维护中占有重要地位。本文在对天津市中心城区景观河道长时间、高密度地采样监测的基础上,结合模糊综合模型和BP神经网络理论,依据天津市中心城区景观河道的实际监测数据分析了其主要污染特征,采用优化的模糊综合评价法以及基于BP神经网络的水质评价方法,对一级河道、二级河道和排水河道的现状水质进行了综合评价。并针对造成水体水质现状的原因进行了简要的分析。结果表明:模糊综合评价模型宜采取迭加隶属度原则;考虑TN指标超标严重的情况,宜对含TN因子和不含TN因子的情况分别进行评价;一级河道、二级河道7-9月份各指标均较差,两种评价结果均为Ⅴ类,其余月份除TN指标外各指标情况较好,不含TN因子评价结果均优于Ⅳ类;一级河道、二级河道水体评价结果为Ⅴ类的比例分别为84.06%和71.74%,优于Ⅴ类的比例分别为15.94%和28.26%;排水河道由于其河道的功能原因,水质整体超标严重。将模糊综合评价法和BP神经网络水质评价法进行比较。结果表明:模糊综合评价法和BP神经网络水质评价法得到的水质评价结果基本一致,二者都考虑了各个水质指标,都为综合的水质评价方法,其中BP神经网络由于其网络的自学习性,评价结果更加客观、准确,成为水质评价方法的首选。本研究可以为天津市中心城区景观河道的水质分析提供理论依据,也可为今后的水体治理和水质研究提供一定的数据支持以及为其他地区水体的水质分析提供有益的借鉴作用。
[Abstract]:Water is the source of life and an indispensable and important natural resource for the survival and development of all life. However, due to the lack of awareness of water resources protection and the increasing complexity of pollutant types with the development of society, water is the source of life. Water shortage and water pollution have gradually become a major problem in the world, and the research on water environment has also developed. Water quality assessment is an important part of water environment research and plays an important role in the management and maintenance of water resources. On the basis of long time and high density sampling and monitoring of landscape watercourses in downtown Tianjin, this paper combines fuzzy synthesis model and BP neural network theory. Based on the actual monitoring data of landscape river in downtown Tianjin, the main pollution characteristics are analyzed, and the water quality evaluation method based on BP neural network is used to evaluate the water quality of the first class river. The current water quality of the secondary channel and the drainage channel was evaluated comprehensively. The causes of water quality are analyzed briefly. The results show that the fuzzy comprehensive evaluation model should adopt the principle of superposition degree of membership, consider the serious situation of the TN index exceeding the standard, and evaluate the situation with and without TN factor respectively. In July and September, the indexes of secondary river were all worse, and the two kinds of evaluation results were classified into category 鈪，

本文编号：1871149