| Automatic drilling and riveting equipment can increase the riveting efficiency and riveting quality,and its application is increasingly widespread.However,due to the limited rigidity of its own structure,the drilling riveting machine will be elastically deformed by the pressure riveting force during the process of riveting,resulting in the position and orientation errors of the riveting rod and the top rod,affecting the pressure riveting quality.This paper calls this error as the drilling riveting machine stiffness error.By introducing the research background of automatic drilling and riveting technology,the importance of ensuring riveting quality was pointed out.By analyzing the status quo of the development of automatic drilling and riveting equipment at home and abroad and research status of equipment stiffness,it is determined that the main research object of this paper is the horizontal dual-machine joint automatic drilling and riveting machine to fill the existing research gap.The main research content is the stiffness error of the drilling and riveting machine.The influence of the pressure riveting quality and the modeling of the working space rigidity error of the drilling and riveting machine are based on RBF neural network.The main research methods are finite element numerical simulation combined with experimental verification.The riveting process was modeled.Based on the requirements of riveting technology,determine the matching relationship between rivets and sidings and the theoretical boring head diameter.The calculation formula of pressure riveting load was selected.Based on the theoretical diameter of the boring head,the theoretical pressure rivet displacement and pressure riveting force were calculated.Rigid body rivets and ejector rods were used instead of drilling riveters to complete the finite element simulation of the riveting process and to verify the correctness of the calculation formula for the pressure riveting load,which was used as the basis for the simulation of elastic riveting.The suitability of the riveting load calculation formula was validated by several sets of rivet-wall riveting simulations.This formula can be used to calculate the riveting process parameters based on the rivet-wall specification.Realize automatic pressure riveting system modeling.Based on the horizontal dual-machine combined automatic drilling and riveting machine,the finite element model of the coupling riveting process of the closed-chain multibody system for "pressure-bearing side riveting machine-rivet-wall plate-outside riveting head side drilling and riveting machine" was established.By analyzing the simulation results and comparing with the riveting results of the rigid body equipment in Chapter 2,the difference of the riveting effect under the displacement load and the force load was pointed out,and the deformation laws of the automatic drilling riveting machine,the rivet and the wallboard during the riveting process was revealed.And then a pressure riveting displacement compensation method was proposed.Through the results of finite element simulation under multiple sets of rivet-wallboard specifications,the applicability of the pressure riveting displacement compensation method was verified.Realize stiffness error modeling.Radial Basis Function Neural Network(AGA-RBF)based on automatic genetic algorithm optimization is used to establish the working space stiffness error distribution model.The stiffness error sampling is performed by orthogonal experiments combined with a full array of postures of the drilling riveting machine.Sampling data is used to neural network.Perform parameter training.In addition,several groups of drilling and riveting postures were selected to verify the generalization ability of the neural network.Through the model,the influence laws of drill riveting posture on its stiffness performance are analyzed,and the working positions with better stiffness and poorer performance are pointed out.Experiment verification.The measurement characteristics and measurement methods during the experiment was selected.By comparing the experimental data with the numerical simulation results,the correctness of the numerical simulation model was verified.Several groups of drilling and riveting machines were selected and the riveting experiments were performed according to the predicted values of the stiffness error neural network model.This verified the feasibility of off-line riveting displacement compensation using the stiffness error neural network. |
PDF Full Text Request