| Large-scale and deep cavities injection dies, usually characterized by complex multi-surface, high requirement of machining accuracy, high requirement of surface quality and a long manufacturing cycle, etc., are the key manufacturing equipment to produce the plastic products of the shell of Large-screen color TV sets, computer monitors, car dashboards, interior of vehicle door, etc. At present, the process of manufacturing of such die depends too much on EDM, manual polishing and grinding, personal skills and experience of manufacturer, which make it difficult to guarantee the machining quality, extend the manufacturing cycle greatly and is hard to meet the growing market competition. The matching of lengthened shrink-fit holder (LSFH) and cutter, which has the characteristics of high clamping intensity, high rotary balance, easily close to deep cavity, can meet the diversity requirement of cutter in machining complex multi-surface, etc., become an important means to achieve a high-speed, high-efficiency and high-quality milling of this above die. Nowdays, the advantage of the matching of LSFH-cutter do not give full play due to the lack of in deep study of the registration property, processing parameter optimization and the machining precision control of which when LSFH is used.This paper focused on the contact characteristics, radial clamping rigidity, dynamics, processing parameter optimization and deformation error compensation of the matching of LSFH and cutter according to the urgent demand of suitable cutter system when high speed milling of large-scale and deep cavity injection. Meanwhile, the analysis model established in the course of study, the effectiveness and practicality of processing parameter optimization and deformation error compensation are verified by experiments.Firstly, basing on Back-Propagation Neural Network (BPNN), a predictive transient milling force mode of the matching of LSFH-cutter is set up since the transient milling force is the basis of the study of dynamics and deformation error compensation. Before the milling force model is set up, the experiment data are processed properly and the time-parameter is introduced into input vector, then the limitation of the past neural network model only can forecast the average cutting force is overcome. Usually simple gradient descent method to network training easily converges to a local minimum. In order to overcome this shortcoming, the momentum gradient descent learning function and momentum gradient descent BP algorithm training function are used in this forecast model. Test results show that the prediction accuracy is very higher than that of the usual analytical model.On contact characteristics and radial clamping rigidity of the matching of LSFH and cutter, a parametric finite element model is established and a new method to determine the contact stiffness coefficient of model is given too. The influence of initial basic interference, fit length, fit diameter and centrifugal force caused by high rotation speed on equivalent stress, contact deformation and contact pressure are studied. On this basis, variation coefficient and its calculation method is first introduced to control the reasonable contact state between LSFH and cutter. Subsequently, the influence of interference fit, clamping length and spindle speed on radial clamping rigidity are studied, and at the same time, the critical bending moment under different interference fit is defined theoretically and an effective method to evaluate the static radial clamping rigidity is established too. The results show that: not only the accuracy of analysis results but also the convergence of analysis process can be guaranteed by using the new iterative algorithm presented in this chapter to rapidly determine the stiffness coefficient of the matching of LSFH-cutter; the reasonable contact state should be controlled according to the diameter of cutter; the method to control the reasonable contact given in this paper not can only guarantee the stable and reliable clamping of cutter but also improve the durability of LSFH; in order to have a secure and reliable cutter clamping on condition of high rotation speed, the weaken role of centrifugal force to radial clamping rigidity must be taken into account; the finite element simulation model established in this paper is proved to be correct by the evaluation results of static radial clamping rigidity experiment and the analysis results can provide the reference to the design and selection of the matching of LSFH and cutter.On dynamics of the matching of LSFH and cutter, the first time, FEA (Finite Element Analysis) technology and EMA (Experiment Modal Analysis) technology are combined to set up the comprehensive finite element model verification and correction process. The influence of overhang length of cutter and spindle speed on dynamic characteristics (include natural frequency and mode shape) are studied. Based on above, transient dynamic of the matching of LSFH and cutter are analyzed in a given processing conditions, and the actual transient displacement response was evaluated. The results show that: the overhang length of cutter affects the natural frequency more significantly; the influence of high rotation speed on dynamic characteristics should be taken into account when the rotation speed exceeds a certain limit; in order to obtain the accurate deformation and force of the matching of LSFH and cutter in an actual high-speed processing condition the influence of transient dynamic must be considered so that the obtained results agree with the actual milling conditions. The results of numerical simulation are verified by experiments. The transient dynamic model can meet the engineering needs and can be used to analyze the actual milling process.On machining parameters optimization when LSFH is used in high speed milling, a fuzzy optimization model of high speed milling is established on the basis of taking fully into account the fuzziness of all restrictive conditions and takes the mechanics and dynamics of the matching of LSFH and cutter as physical restrictive conditions. The optimization results show that the high speed milling time reduces 5.95%, 8.54% and the surface roughness decrease 5.42, 6.85% compared with the results of the conventional optimization without considering the fuzziness of restrictive conditions and the milling results using the parameters recommended by the cutter manual.On machining deformation error compensation when LSFH is used in high speed milling, a new compensation method is presented based on an algorithm of iterative balance between milling force and cutter deformation, and the previous finite element model, milling force model and optimization results, then the high speed milling error caused by the deformation of cutter is compensated offline. The milling example shows that the maximum milling error is decreased from 42μm to 9.5μm after compensation, and the milling error reduce about 77.4%. When machining large-scale and deep cavity injection die, high dimensional accuracy of surface finish can be obtained using this compensation method.
|
PDF Full Text Request