Modeling Of Crop Response To Water Based On Robust Regression And Optimization Of Irrigation Schedule

In the situation of the shortage of the water resource, irrigation water volume in agriculture can not be added,whereas,crop yield production need to be increased,how to optimally allocate the limited irrigating water at the different growth stage of crops and how to ririgate the crop accuracy to get maximal benefit and become a lot of research in the agricultural irrigation field. Aim at the above problems, the writer made the research of the Model Crop Response to Water and Optimizing of Irrigation Schedule based on studying the major crop in hetao area. the results are following:(1) Through analyzing the influence of soil water stress to yield, physiological status (plant height, LAI), Enabled the crops of degree which received in the different period water scarcity affects to have the direct-viewing response.(2) In building model method application,Through compare the method of ordinary regression with Robust regression to build model of crop response to water, it had determined the superiority of the robust regression. Use the Robustfit order in MATLAB, can build crop model Response to Water very conveniently. The writer proposed from now on when establishment crops Response to Water, first uses the robust regression method.(3)Through consideration rainfall amount, crops water scarcity underproduction and so on foundation, According to field plot deficient irrigation experiment material. The writer analysis and establishment corn output and entire period of duration moisture content and between birth stage moisture content quantification relations, calculating the senditivity index in several water production functions and chose the Minhas model of crop response to water to suit in the local corn's model of crop response to Water .(4) This article uses the current advanced global optimization searching algorithm - Differential Evolution Algorithms as well as the pattern search method carries on the irrigation system's optimization. It has overcome formerly DPSA and NLP cannot restrain to the true optimal solution flaw, caused this algorithm in this kind of uncertainty, in the misalignment model solution to make the good progress. The solution result indicated that the model has solved the limited water volume well in the different period of duration optimized assignment.