Anlaysis Of Driving Forces And Forecast Of Water Utilization Structure In Jiangxi Province

Jiangxi province is a large agricultural province in China.Sufficient water resources play a decisive role in the development of agriculture and economy in Jiangxi province.Therefore,the rational determination and prediction of water structure is of great significance for the sustainable development of water resources in Jiangxi province.Jiangxi province water resources is abundant,but because of its geography,climate and other reasons caused by the uneven distribution of water resources in Jiangxi Province,and the negative influence of the human factors in Jiangxi Province,frequent drought and flood extreme meteorological and hydrological events.In this paper,based on a large number of data from population,the proportion of the urban population,animal husbandry and fishery output,industrial output,GDP,fixed assets investment,cultivated area,effective irrigation area,grain yield,precipitation,temperature,evaporation and city green coverage area as water structure index,reference principle of information entropy and equilibrium degree.And principal component analysis methods are analyzed and researched on the water structure of Jiangxi province in 1999-2015,and the BP neural network prediction model,PLS-SVM regression model and grey system GM(1,1)model respectively to predict the water structure in Jiangxi Province in the next 15 years,and compared,obtained the following conclusions.Water data of 1999-2015 in Jiangxi Province Based on the information entropy and equilibrium degree method in analyzing and researching the water consumption structure of Jiangxi province in 1999-2015,the results showed that: Jiangxi Province agricultural water water structure occupies a large proportion,by contrast,industrial water,living water,ecological water ratio is smaller however,the proportion of agricultural water,have a slow decline,a gradual increase in the proportion of other water.At the same time,the water balance of different cities in Jiangxi province is uneven,and there is a clear gap between strength and weakness.The results of principal component analysis showed that the principal component analysis was selected by GDP,industrial production,animal husbandry and fishery output value,fixed asset investment,population,city green coverage area,grain output,city population,cultivated area and irrigation area as the representative of social economic factors.Climate factors represented by evaporation,rainfall and temperature.Using grey system GM(1,1)model,PCR-BP model and PLS-SVM model of water were predicted,the results show that the gray system GM(1,1)model of the prediction error interval between [0.01,17.40],PCR-BP prediction model error interval between[0.19,13.58],PLS-SVM model error in the range of [0.09,4.92],PCR-BP consistent with PLS-SVM results.According to the PCR-BP and PLS-SVM model to predict the water consumption,the prediction results show that the agricultural water use in 2018 before the slow growth after2008,water consumption decreased slowly,the prediction of industrial water results very significant upward trend;the decline in living water has a certain extent,and then showed a slow upward trend;the ecological water use has been increasing slowly.