Optimization Of Defict Irrigation Schedule In Irrigation District Based On Groundwater Level Depth Prediction

The total amount of water resources is insufficient in China,and the per capita occupancy volume is much lower than the world average.Especially,this situation is aggravated by unreasonable exploitation and utilization of water resources,which even destroys the local ecology and environment.In order to alleviate the present situation and realize the sustainable utilization of water resources,we should vigorously develop water-saving agriculture and improve the efficiency of water resources utilization.Hetao irrigation area is located in arid and semi-arid areas of Northwest China,and it is imperative to develop water-saving agriculture for its water shortage.However,with the development of water-saving irrigation,its groundwater recharge will decrease and groundwater level depth will decline accordingly,which will not only destroy the balance of groundwater,but also reduce the contribution of groundwater to crop water consumption.Therefore,it is of practical and guiding significance to carry out the research on prediction of groundwater level depth and optimization of deficit irrigation schedule under different water-saving conditions for selecting water-saving measures and protecting ecological environment.According to the relative shortage of water resources,soil salinization and desertification in the future,this paper selects Jiefangzha irrigation district as a research case,and constructs the prediction of groundwater level depth model and the optimization model of deficit irrigation schedule in irrigation district respectively,then corresponding solution methods are given.The main research contents and conclusions are as the follows:(1)Research on groundwater level depth prediction.According to the statistical analysis of influencing factors of groundwater level depth,this paper takes the water diversion from the Yellow River,precipitation,evaporation,drained water amounts,groundwater level depth as input variables,then constructs the controlled auto-regressive(CAR)model,the back propagation(BP)neural network model and the improved GABP model(a BP neural network model based on improved genetic algorithm)respectively,which predict the monthly groundwater level depth in Jiefangzha irrigation district in 2012.The Nash-Sutcliffe efficiency coefficients are 0.887,0.883 and 0.911 respectively,and the percent biases are-0.2%,5.5%and 4.7%respectively.(2)Improved particle swarm optimization(PSO).The PSO algorithm is improved respectively by using LDIW(linear decreasing inertia weight),CF(constricition factor),MF(mutation factor),CF-MF(combination of constricition factor with mutation factor)and LDIW-MF(combination of linear decreasing inertia weight with mutation factor).The optimal values of three test functions,including Sphere,Griewank and Rastrigrin,are calculated respectively by the above improved PSOs.The results show that the improved CF-MPSO and LDIW-MPSO have higher accuracy.(3)Research on optimization of defict irrigation schedule in irrigation district.This paper constructs an optimization model of defict irrigation schedule in irrigation district based on the prediction of groundwater level depth.The paper takes the maximal sum of relative yields of main crops in irrigation district as the objective function,which is constrained by water diversion,crop planting area,groundwater level depth,etc.Then dynamic programming algorithm and improved PSO algorithm are used to calculate this model.Under the conditions of different WEIs(water efficiencys of irrigation),RWCs(ratios of well irrigation area to canal irrigation area)and water diversion,this paper puts forward optimal crop irrigating quota in different growth stages,monthly water diversion and groundwater level depth,which provides reference for water-saving planning and water management in Jiefangzha irrigation district in the future.(4)Considering the ecological requirement of groundwater level in crop growth period,some simulated results are given under different water-saving conditions.Under the condition of current RWC and increasing WEI,the maximum water-saving potential is 340 million m3 when developing deficit irrigation,and corresponding WEI is 0.55,the water diversion is 950 million m3,the sum of main crops relative yields is 2.859 and the groundwater level depth is 2.50 m.Under the condition of current WEI and increasing RWC,the maximum water-saving potential is 290 million m3 when developing deficit irrigation,and the corresponding water diversion is 1.0 billion m3.When the RWC is increased from 1:25 to 1:5,the sum of main crops relative yields is increased from 2.232 to 2.619,and the maximum groundwater level depth is 2.50 m in crop growth period.