城市生活污水排放量的影响因子分析及关联性研究

Factor Analysis of Domestic Wastewater Discharge and Correlation Study

生活污水排放系统复杂、影响因素多种多样,从其内涵出发,分析主要影响因子,共选择3大类14个影响因子,它们之间存在着严重相关性问题,为了解决多重相关性问题,引入偏最小二乘回归方法,该方法可以有效克服多重相关性,并能够实现多种数据分析方法的综合应用;而人工神经网络具有学习和记忆能力,将二者相关联,可以较好地解决非线性问题。为检验影响因子选择的合理性和方法的适用性,以郑州市为例,对生活污水排放量和影响因子进行定量分析。结果表明,主要影响因子的选择合理,拟合和预测精度均较好。

Major affecting factors in domestic wastewater discharge system were analyzed. Partial least squares regression （PLS） capable of multi-data analysis combined with neural network which has ability of self adaptive learning and remembering were used to analyze nonlinearity of domestic wastewater discharge system. To check the rationality of the selected factors and applicability of the method, domestic wastewater discharge and affecting factors were quantitatively analyzed with a case study in Zhengzhou, Henan Province. Results indicated that selection of main factors was reasonable with preferable prediction and forecasting precision.