| Half feed combine harvester has been widely used in the existing grain combine harvester,and research,along with the development of the combine harvester for threshing performance requirements more and more high,as the core device of the combine harvester threshing device,determine the performance of the combine harvester threshing,study of motion in the grain on the threshing device and process,It plays an important role in the optimization of threshing device.Based on the semi-feeding combine harvester Kubota PRO 588,this paper carried out theoretical analysis and research on threshing device,simulation test of structure and performance,and structural optimization of key parts.The specific work is as follows:Firstly,the threshing separation device size of the existing combine harvester was measured and the data were recorded,and the three-dimensional modeling was carried out in SolidWorks.In the finite element ANSYS Workbench,the statics analysis,modal analysis,transient analysis and ADAMS intermediate motion balance analysis of the core component threshing drum were carried out successively.The validity and rationality of the structure are tested.The discrete element simulation model of threshing and separation device was established,and the threshing process simulation test was carried out in EDEM to simulate the reliability of actual working conditions.Secondly,based on the simulation model of the discrete element threshing device,three influencing factors including drum speed,feeding amount and vibration frequency of the concave screen were selected from EDEM and their influence rules on loss rate and impurity rate were analyzed.Single factor tests were carried out successively,and the influence rules of three factors were analyzed.Multi-factor orthogonal experiment was designed,and the optimal combination of factor levels was obtained by range analysis,and the significant degree of factor effects was obtained by variance analysis.Finally,the quadratic regression center combination test was designed,and the experimental data were analyzed and processed by polynomial fitting.The mathematical regression equation and the optimal combination of the level range of factors of the corresponding surface graph were obtained,as well as the optimal combination of the influencing factors parameters.Next,according to the optimal parameter combination and combined with relevant design theory,the threshing separation device(threshing drum,threshing element,screw head)and the corresponding structure size optimization design,use SolidWorks to carry out three-dimensional modeling of the optimized design of threshing device parts,after preliminary verification of the structure and assembly rationality of each part,Finite element structure simulation analysis and discrete element performance simulation analysis were conducted successively,and compared with the original basic threshing device,and the superiority of the optimized and improved threshing device was verified.Finally,on the basis of three factors three levels orthogonal experiment,comprehensive experiment design,use the BP neural network nonlinear function and strong generalization ability,roller speed,feed rate and concave plate screen vibration frequency as the network input layer,loss rate and impurity rate as the network output layer,set up with two layers of hidden layer three layer BP neural network model,realize the threshing performance prediction,the results show that the prediction model of threshing performance by BP neural network has good prediction accuracy and generalization ability. |
