Improved Method And Its Application For River Water Quality Assessment And Prediction

To evaluate the water quality and forecast the tendency of the water quality in the near future are the basic tasks for water environment planning and management.Based on the characteristic of fuzziness,random,seasonal variation,and uncertainty of water environmental system,matter element analytical method and set pair analysis method are used to build water quality evaluation model,and grey GM(1,1)theory,wavelet analysis,and generalized autoregressive conditional heteroskedasticity(GARCH)model are used to build water quality forecasting model.The application of both the water quality evaluation model and the water quality forecasting model in Songhua river and Xiaoqing river,and the results show that the advantages the two model have as follow:For water quality assessment,face the deficiency that occurs in the traditional water quality assessment methods which can't take the complexities and the certainty and uncertainty of the water quality elements into account,the given article is to employ set pair theory in combination with the matter-element connection number,and propose a entropy weight connection number matter-element comprehensive assessment and variance tendency prediction model of water quality,for the traditional matter-element model of water quality can be successfully used to describe the complex system with the uncertain characteristics.The combination of the above said two methods can be joined to establish a comprehensive assessment and trend projection model based on the six-element connection numbers of the matter-element and partial connection number based on the qualitative analysis for the ambiguous characteristics of the water quality elements.For water quality forecasting,a short time series hybrid forecasting method based on wavelet transform,grey model and autoregressive conditional heteroscedasticity(GARCH)is proposed.Firstly,the actual measured data ratio has been worked out,and consider the ratio as the condition to decide whether use the GM(1,1)as the initial forecast model or not,if the final forecasting has less error than actual measured data and the class of accuracy arrive at 1 level(posterior error ratio C?0.35),then take the model as the forecast model,else take the grey-wavelet-GARCH as the final forecast model,in this model,the time series is decomposed and reconstructed into approximate series and detailed series.Secondly,when using the grey-wavelet-GARCH model,try to use DBi to decompose and reconstruct into approximate series and detailed series,and analyze the approximate series data and test its ration in condition of grey model until its ratio meet required precision(in this paper,the precision is the class of accuracy arrive at 1 level),then use GM(1,1)to forecast the approximate series so as to find out its future values;while the detailed series future values are forecasted by GARCH model,the reason use GARCH model to forecastdetailed series is the large fluctuation the detailed series has,and the GARCH model has advantage in time series forecast with large fluctuation.Finally,the sum of the approximate and detailed series future values is used as the final forecast values.In the end,the typical Chinese river in different area has been chosen as case study,and the results show that the river quality forecast model and the river quality evaluation model are efficient.Except for water,it can be used widely in air,soil,sea,stock,the risk evaluation in project management and population.