基于改进贴近度法的水质监测点优化布设

贴近度法作为一种系统优化方法已被广泛应用于水质监测点优化中,然而此方法存在几方面的问题:一是仅考虑空间因素而未考虑时间因素对区域监测重要性的影响;二是用各指标监测数据平均值进行计算,从而导致聚类效果较差;三是未考虑指标权重对测点优化布设的影响。为此,将层次分析法与贴近度法相结合,提出了基于改进贴近度法的水质监测点优化布设方案。应用此方法对某引水明渠水质监测点进行优化布设,结果表明,改进方法有效提高了聚类效果,易于进行水质监测点的排序与分类。

On the optimized monitoring project layout for the river water quality based on the analytic hierarchy process

As a scientific method for optimizing the layout in the river water quality monitoring practice,the similarity method had been widely used for years.However,there exist some inevitable defects with the method.First of all,it only pays attention to the importance of the temporal factors,whereas it ignores the impact of the time on the rational monitoring.Secondly,since it is not possible to guarantee the clustering effects when the average value of monitoring data concerning the various water quality indexes is used for calculation.Thirdly,it tends to neglect the importance of the index weights in the optimized layout.Seeing the above said problems,we have in this paper managed to improve the method by joining the analytic hierarchy process and the similarity method comprehensively.What is more,we have applied the long series of monitoring data to the decision matrix in the said method,too.And,finally,we have classified the monitoring points of the water quality according to the degrees of similarity.With the new method introduced,the first step is to build up the decision matrix for each monitoring index.And,next,attempts to be made are to obtain the scheme of optimized layout for all the monitoring indexes,with the changing rate of each index being used to build up the index weight vectors.Thus,we have managed to join each scheme of the optimized layout and the corresponding index weight vectors together to work out the final scheme through analytic hierarchy process.By making the method optimized for its manual monitoring points,we can not only make full use of the flexibility of hand-operating monitoring,but can also reflect the changes of significance of monitoring regions.For example,as compared with the original similarity method,the new renovated method has improved the clustering effect remarkably.As a consequence,the water quality monitoring points can be sorted and classified by the decision makers much more easily.Besides,with the new method,it has made us possible to optimize the layout schemes of each season by choosing different schemes in different seasons.Thus,it can not only reflect the changes of the significance of monitoring regions,but can also reduce the manpower resources.