城市供水管网水质检验监测模型研究

发布时间：2018-05-18 05:25

本文选题：输水管网 + 水质模型　；参考：《苏州科技学院》2013年硕士论文

【摘要】：经自来水厂处理后的出厂水,已经满足生活饮用水的卫生标准。但经过复杂的城市输水管网输送到用户时,水质可能发生明显的变化。为了改善管网的水质,加强对管网水质的运行管理,有必要研究水质在管网中的变化情况。通过建立供水管网的水质模型,可以对管网水质进行预测,为水质的运行管理提供可靠的理论基础,对保障饮用水的水质安全具有重要的意义。本文以苏州市某管网末梢水质指标监测中的9个监测点为研究对象,结合实际情况,对监测得的实验数据采用4D检验模型、Dixon检验模型、Q检验模型、Grubbs检验模型这4种统计学检验模型进行离群数据数据的检验,并对检验结果进行分析,做出科学的取舍判断,为水质模型的建立和水质的预测奠定基础。以浑浊度为模型的控制指标,并选用多元统计回归和人工神经网络方法来建立管网水质统计模型。其中,运用SPSS软件建立多元回归模型并进行了求解,运用MATLAB来进行人工神经网络的训练检验。对研究范围内的对1号和9号监测点建立了水质模型,然后对这两个监测点管网水的浊度进行了预测。结果表明神经网络的模拟效果要比多元统计回归理想,模型的准确性较高,能应用于实际,而且进行离异数据筛选后的数据应用于模型后预测效果更佳。最后,结合研究范围内的实际情况,提出了水质统计模型的建立必须以大量的数据为基础,这是保证模型准确性和精度的前提；另外还提出了在建立水质模型之前,有必要进行离群数据的检验。在分析引起预测误差原因的同时,对水质统计模型的研究发展做出了一定的展望。
[Abstract]:After the treatment of water plant, the factory water has met the sanitary standards of drinking water. However, the water quality may change obviously when the complex urban pipeline network is transported to the user. In order to improve the water quality of the pipe network and strengthen the operation and management of the water quality of the pipe network, it is necessary to study the change of the water quality in the pipe network. By establishing the water quality model of the water supply network, the water quality of the pipe network can be predicted, which provides a reliable theoretical basis for the operation and management of the water quality, and is of great significance to the safety of the drinking water quality. In this paper, 9 monitoring points in the monitoring of the water quality index of the end of a pipe network in Suzhou City are taken as the research object, and combined with the actual situation, The 4 kinds of statistical test models such as 4D test model, Dixon test model, Q test model and Grubbs test model were used to test the outlier data, and the test results were analyzed to make a scientific decision. It lays a foundation for the establishment of water quality model and the prediction of water quality. Taking turbidity as the control index, multivariate statistical regression and artificial neural network were used to establish the statistical model of pipe network water quality. The multivariate regression model is established by SPSS software and solved. MATLAB is used to test the training of artificial neural network. A water quality model was established for monitoring sites 1 and 9 in the scope of study, and the turbidity of pipe network water at these two monitoring points was predicted. The results show that the simulation effect of neural network is better than that of multivariate statistical regression, the accuracy of the model is higher, and the model can be applied to practice. Finally, according to the actual situation in the scope of the study, it is proposed that the establishment of water quality statistical model must be based on a large number of data, which is the premise to ensure the accuracy and accuracy of the model, and it is also proposed that before the establishment of the water quality model, It is necessary to test outlier data. At the same time, the research and development of water quality statistical model are prospected.
【学位授予单位】：苏州科技学院
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TU991.33;TU991.2

【相似文献】