Design And Research On Vibration Harvesting Mechanism And Canopy Shaker For Camellia Oleifera

With the continuous promotion of Camellia oleifera industry,mechanized harvesting has become a key problem that restricts the development of Camellia oleifera industry.The problem of mechanized harvesting is mainly due to the simultaneous growth of flower and fruit,the short picking period of Camellia oleifera fruit,and the unstandardized planting modes.Therefore,this paper studies the canopy shaker,explores the dynamic characteristics of Camellia oleifera,constructs the prediction model of Camellia oleifera natural frequency,develops the canopy shaker based on two-dimensional trajectory excitation,and conducts experimental research.The research contents are as follows:1.The basic data determination of Camellia oleifera was carried out.Through the hammering test,and the natural frequencies of each order of the Camellia oleifera trees were measured.The branches of Camellia oleifera tree were graded,and the dimensions such as diameter,length and angle were measured,and the canopy of each branch on Camellia oleifera trees was weighed and recorded.2.The dynamic characteristics of Camellia oleifera were analyzed.The Camellia oleifera tree was constructed as a three degree of freedom vibration model,and the differential equations were established.According to the dimensions of the Camellia oleifera tree,the 3D Camellia oleifera model was established.Through finite element simulation,the simulated natural frequency was compared with the measured natural frequency,and the average relative error was3.72%.Adding canopy weight to the finite element model could improve the simulation accuracy of the finite element model to a certain extent,but the improvement effect was not obvious.3.The prediction model of Camellia oleifera natural frequency based on BP neural network was established.Re LU function was selected as the activation function,MSE was selected as the loss function and evaluation index,and Adam algorithm was selected as the optimizer.By changing the size parameters of each branch in the 3D Camellia oleifera model,the sample size of Camellia oleifera model was expanded.When the number of iterations was 12000,the global error was only 0.0214.The prediction accuracy test was performed on the test set,and the mean square error of the natural frequency prediction was 0.0221.The natural frequency of the sample trees was predicted,and the mean square error was 0.0408,which indicated that the natural frequency prediction model of Camellia oleifera tree could make a good prediction.4.Based on the planar 5R parallel mechanism,a Camellia oleifera canopy shaker was designed.The epitrochoid trajectory was used as the output excitation trajectory.The rotation angles of the two motors were obtained through the inverse kinematics position solution.Through dynamic simulation,it was verified that the stability of the output response and the excitation trajectory meet the design requirements.The control system of the canopy shaker was designed,the control of the driving mechanism by single chip microcomputer was realized,and outputs the two-dimensional excitation trajectory.5.The prototype of the Camellia oleifera canopy shaker was tested in the field.The influence of the layout of the exciter rods on the excitation response was compared,and the layout mode 4 of 7×7 staggered distribution was determined.It could avoid reducing the excitation response on the basis of reducing the number of excitation rods.The results of the canopy excitation test showed that the planar excitation used resulted in the acceleration value in the depth direction of the canopy being smaller than the acceleration in the excitation plane direction.The experimental results showed that the acceleration transfer efficiency was 77.63%,which verified the effectiveness of the two-dimensional trajectory excitation.