| Thin-walled parts are characterized by light weight,high specific strength,and saving materials.They are widely applied in aerospace and precision machine tools,such as thin-walled frames and tilting pad bearings.Such parts will be affected by cutting force,vibration and other factors in the traditional mechanical processing process,resulting in great deformation.Wire-cut Electric Discharge Machining is an important non-traditional processing technology.There is no macroscopic force on the workpiece during the processing,so it is an ideal process for processing thin-walled parts.However,due to the poor rigidity of the thin-walled parts,they are easily affected by thermal stress which will lead to large thermal-elastic-plastic deformation.Moreover,the electrical machining erosion process changes the wall thickness of the thin-walled parts and the volume of rmaterial undergoing thermal deformation.Therefore,it is difficult to control the thermal deformation of thin-walled parts,which limits the application of wire-cut EDM in the processing of thin-walled parts.Therefore,this paper focuses on the thermal deformation behavior mechanism of WEDM,mainly from the erosion process and its effect on the thermal deformation and the WEDM heat source model,and finally realizes the purpose of suppressing the thermal deformation.The main research contents are as follows:Aiming at quantitatively describing the effects of different electrical erosion methods on the equivalent erosion temperature,a molecular dynamics simulation platform was established to calculate different erosion methods by simulating the dynamic behavior of atoms during processing under different energies.Equivalent erosion temperature was calculated and the effect of the erosion mode on the equivalent erosion temperature was studied.Simulation results show that under low energy conditions,the material was mainly eroded in the form of a single atom or small atom clusters.The equivalent erosion temperature under this erosion mode was higher.At high energy,the material is mainly eroded in the form of larger particles and the equivalent etching temperature was lower in this erosion mode.The research in this chapter visualizes the difficult-to-observe electrical machining process and studies the effect of input energy on the erosion volume at the microscopic scale.Aiming at the problem that the simulation time of the molecular dynamics model is too short and the simulation system is too small,based on the theory of specific discharge energy,this paper proposes an equivalent erosion method that can predict the erosion volume of thin-walled components in actual processing scale and time.First,the concept of specific discharge energy is modified,and a numerical prediction method based on specific discharge energy is proposed.This method is validated by the calculation and experimentation of aluminum,Inconel 718 and SUS 304.To solve the problem that the model only validated when the material was mainly melted,this paper combines the input energy and specific discharge energy.Then this model was further applied to vaporization-based erosion methods,and finally a dynamic equivalent erosion temperature model was proposed.In this chapter,a dynamic equivalent erosion temperature model was established the for different erosion modes under different processing conditions and realizes the calculation of the erosion amount during the electric machining process of thin-walled parts.Aiming at the complex influence of electrode wire vibration on the surface thermal stress distribution of thin-walled components,a vibration-Gaussian continuous pulse heat source model for wire-cut machining was established to calculate the temperature distribution of the workpiece surface.The vibration of the electrode wire was studied first,and the influence of the vibration of the electrode wire on the distribution of the discharge point was studied in different directions.Then the influence of the vibration of the electrode wire on the distribution of the discharge point was introduced into the single pulse heat source model.A continuous pulse heat source model for electric discharge wire cutting was used to solve the heat source distribution under different vibration conditions using parallel calculation.The model was further optimized and a simplified vibration-Gaussian heat source model was obtained,which significantly improved the calculation efficiency.The effectiveness of the model was verified by comparing the recast layer thickness with the experimental results.Based on the research on the effects of thermal erosion and the model of wire-cutting vibration-Gaussian heat source,a thermo-elastic-plastic finite element model of WEDM under the effect of erosion was established to simulate the erosion and deformationprocess.This model adopted the Python-Fortran hybrid programming method to realize the function of automatic heat source movement,melting unit automatic identification and automatic deletion.This method avoids participitation of the eroded units in the calculation of elastic-plastic deformation.This model proves that the thermal deflection will decrease with the increase of erosion volume,and explains that the thermal deformation will decrease with the increase of the input energy when the input energy is large.In order to solve the problem that the thermal deformation of thin-walled parts is difficult to control,this paper proposes a compensation method based on image compensation and response surface design for WEDM.This method effectively suppressed the thermal deformation of thin-walled beams,and reduced the straightness error by 95%.In addition,this method was further extended to other shapes of thin-walled parts of the WEDM and verified by the square thin-walled frame.The results of this study were applied to the processing of ultra-thin electrode arrays with a wall thickness of only0.1 mm.The straightness error is within 1 μm.In addition to the study of thermal deformation suppression,this paper also took advantage of the thermal deformation to process thin-walled parts with complex surfaces by using BPNN-MEA algorithm. |