水质的评价和预测模型

An Evaluation and Prediction Model of Water Quality

本文首先考虑到水质类别的差异和相同类别水质在数量上的差异对综合评价的影响,构造"S"形的变权函数,对属于不同水质类别的同种污染指标进行"动态加权",建立基于逼近理想点排序法的评价模型和利用灰色关联度的分析方法,对长江水质状况做出了综合评价;其次,根据7个观测站的位置将干流分成8段,把每段河道内所有污染源都等效为一个段中央的连续稳定源,分别利用稳态条件下的一维水质模型及质量守恒定律,得出中间6段每个月的排污量,综合比较各河段一年多来的总排污量得到主要污染源的分布区域;然后,用每年不可饮用类水的百分比之和刻画水质状况,综合利用灰色GM(1,1)模型和时间序列分折方法,对变化趋势进行了预测;最后,建立不可饮用类水的百分比与长江水总流量和废水排放量的线性回归模型,计算在满足约束条件下排污量的极限值,用排污量的预测值减去极限值,得到未来10年的污水处理量。

The difference of water quality and quantity have an effect on the comprehensive evaluation of water quality. At first an ＂S＂ -like variable-value function dynamically weights the pollution evaluation standards of identical kinds for different water quality categories. Then, by making an evaluation model based on TOPSIS and using the analysis method of Grey Correlation, the water quality of Changjiang River is evaluated. Secondly, the main stream is divided into eight parts according to the positions of the seven observation points. All the polluting sources in each part are equal to a continuously stable source in the center of the part. By using the one-dimensional water quality model in the stable state and Quality Conservation Law to get the monthly pollutant amount in the middle six parts, and by comprehensively comparing the pollutant amount of different parts over one year, the distribution area of the main polluting sources can be got. Thirdly, the description of the state of water quality by the yearly undrinkable water percentage is made, and by using of the Grey GM（1,1） model and time sequence analysis method, the variation tendency is estimated. Last, in light of developing a linear regression model of the undrinkable water percentage in relation to the overall flow and waste water flow in Changjiang River, the limit of pollutant amount under restrictions is resolved. By subtracting the limit from the estimation value of pollutant amount, the waste water disposal amount can be 10redicted in ten years.