| Monitoring point refers to the specific spatial location of monitoring station or monitoring equipment used to obtain sampling data in the monitoring process.For ecological environment monitoring,reasonable distribution of monitoring sites can achieve maximum sampling efficiency under the premise of saving economic cost.Therefore,the quantitative and accurate evaluation of the rationality of spatial monitoring sites layout is an important basic research issue to realize the transformation and upgrading of ecological environment improvement from "quantitative change" to "qualitative change" under the background of the new normal of ecological environment monitoring in China since the14 th Five-Year Plan.Its scientific and credible assessment report is of great scientific significance for optimizing the layout of monitoring sites,improving the ability of disaster prevention and resistance,and maintaining the ecological environment.Since most of the existing methods for evaluating the rationality of spatial distribution of monitoring points are described from the perspective of heterogeneity,this paper chooses to start from the heterogeneity of spatial distribution characteristics of monitoring points and background attribute characteristics.However,there are some problems in the current methods :(1)different methods are biased to varying degrees in the level of perception data distribution characteristics,and a single model or index cannot robustly evaluate the data of monitoring points with different prior distributions,resulting in low credibility and availability of the evaluation results;(2)The existing methods are susceptible to the influence of data and model parameters,which makes it impossible to achieve horizontal comparison of reasonableness degree of data from multiple monitoring points in the same dimension.In view of the shortcomings of the existing research,this paper firstly proposes an evaluation model of the rationality of the layout of monitoring points from the perspective of spatial distribution by borrowing the idea of integrated learning,so as to solve the problems of strong bias and low credibility of the traditional evaluation methods.Furthermore,coupled with the attribute distribution characteristics of monitoring targets,an evaluation method for the rationality of monitoring point layout oriented to ecological environment attribute sampling is proposed to achieve quantitative evaluation of the rationality of monitoring point layout from the comprehensive perspective of "spatial-attribute".Specifically,(1)Evaluation model of rationality of monitoring point layout from the perspective of spatial distribution.Under the perspective of data driven,draw lessons from the thought of integrated framework,using multiple indicators to by a variety of point distribution model of integrated training data sets heterogeneity measure,extracting information parameter,and carries on the information fusion based on voting mechanism,put forward a kind of intelligent learning space monitoring layout rationality assessment model.With the actual surveillance camera distribution in a sensitive area of a border city in China as experimental data,the proposed method is proved to be more accurate and robust than the traditional method through experimental analysis.(2)Evaluation method of spatial layout rationality of monitoring points based on coupling attribute characteristics.Under the space constraint considering properties of space monitoring characteristics of heterogeneous measurement method based on the monitoring background attributes to evaluate its layout reasonable,and compared with the result of a variety of measures,the coupling space in the distribution and layout of the background attribute of the rationality of evaluation methods under the guidance of monitoring stations layout optimization.With the background attribute information of air quality monitoring points in Tianjin,Fujian and Guizhou as experimental data,the experimental analysis proves that the research method has high practical application value.Finally,a review of the whole paper is made,and some problems to be further studied are prospected. |
