Study On The Prediction Of Return Water And The Temporal And Spatial Variation Characteristics Of Water Quality In Large-scale Yellow River Diversion Irrigation Area

The composition and distribution of return water in large-scale irrigation areas of the Yellow River are more complex,and the residue of nitrogen and phosphorus and other chemical fertilizers carried in the return water pollutes the water quality of the river and causes the pollution of the water environment.By consulting a large number of relevant literature at home and abroad,combining theoretical analysis with empirical research,taking Jingdian first stage irrigation area as an example,the composition and variation law of surface return water in irrigation area are studied,and the relationship between the return water quantity and its influencing factors,using support vector machine and grey system model to establish the forecasting model.On the basis of studying the temporal and spatial variation of return water quality in irrigation area,the grey clustering method was used to evaluate the water quality comprehensively.The evaluation index system of farmland drainage and irrigation reuse suitability was established,and the fuzzy identification model was used to evaluate the reuse suitability of return water in the first phase of Jingdian irrigation area.The main achievements are as follows:(1)The composition of return water in Jingdian first stage irrigation area was studied,and the annual and interannual variation rules of the return water were revealed.The gray correlation method was used to determine the main influencing factors of the return water as the depth of groundwater,the amount of water diversion,and the amount of evaporation.(2)The support vector machine(SVM)model was established for the monthly return water in the irrigation area,and the prediction was carried out.By comparing the predicted value with the measured value,it was found that the prediction accuracy of the SVM model was higher.Established a gray GM(1,1)model to predict the amount of return water in Jingdian irrigation area in November and December.The results show that both models can be used to predict the monthly return water in Jingdian irrigation area.(3)According to the monitoring data of surface return water quality in Jingdian first stage irrigation area,the spatial and temporal distribution characteristics of different water quality indexes were analyzed.The results showed that the indexes of total phosphorus,ammonia nitrogen,pH,and dissolved oxygen changed relatively little with time,while indexes such as total nitrogen,nitrate nitrogen and chemical oxygen demand changed greatly with time.The main water quality index was selected as the evaluation factor,and the gray clustering method was used to evaluate the surface return water quality of Jingdian first stage irrigation area.The results showed that the return water quality of the irrigation area was better in the irrigation season and slightly worse in the non-irrigation season.(4)The suitability evaluation index system of irrigation and reuse of farmland drainage resources was established,based on the fuzzy recognition model,the suitability of the reutilization of the return water in Jingdian first stage irrigation area was evaluated.The results show that the return water at the pumping station is the most suitable for irrigation use,and the return water in each period can be reused for irrigation.The return water from other monitoring sites has a good degree of suitability for reuse from June to September.During this period,the return water from the irrigation area can be used to replace part of the water from the Yellow River for irrigation,while the suitable degree of return water from October to December is worse.