Theory And Algorithm Research Of Injection Molding Simulation And Z-Mold Software Improvement

At present, injection molding is one of the most widely used and effective manufacturing processes in the plastics industry. It can directly realize the near-net shape of three dimensional complex parts, which is characterized by high production efficiency, mass production. Since the advent of injection molding CAE techonoly in the 1950s, it has been very widely applied and has become an indispensable for the plastics processing enterprises. CAE analysis can make mold design and process parameters setting more scientific, and has important guiding value for shortening the production cycle and improving product quality. However, the injection molding CAE technology still has many problems, such as the simulation accuracy need to be improved, the parameter design is still dependent on the experience, the mesh division and repair is very complex and time consuming, the calculation time is not efficient and so on.The above problems show that the theory model of CAE and the calculation method are not perfect enough, and many problems still need to be further studied. Under this background and based on the requirements of the injection molding simulation software Z-Mold development and upgrading, the dissertation carries out the theory and algorithm research of heat transfer, warpage prediction and algorithm efficiency increase improvement. The objective is to improve the simulation accuracy of Z-Mold, so that the software has a stronger guiding role for the actual production. The main research work includes the following aspects:1. The integral solving and the operator-splitting solving of the energy conservation equation are discussed in Hele-Shaw flow simulation. Based on the operator-splitting method, a new algorithm named sub time step with variable size is suggested to deal with the thermal convection in this paper, in which the sub time step using dichotomy is introduced that bounds the tracking path in a certain element, which effectively solve the problem that the tracking needs to pass through a few elements and the reverse tracking may fail when the elemental velocity is very high. Based on this method, the heat transfer simulation of Z-Mold is improved. The new algorithm makes the computation more simple and effective. The numerical examples show that the new method has same accuracy as one using uniform small time step and high solving stability, but calculating time is dramatically reduced.2. The finite element equation to calculate the pressure field is derived in Hele-Shaw flow simulation by means of Galerkin method. In solving the pressure field, the coefficient matrix of the equation repeatedly needs to be rebuilt in every time step, which is very time consuming. To solve the problem, the pre-conditioned conjugate gradient method (PCG) is suggested to solve the equation, in which the approximate triangular decomposition method is selected to deal with the pre-condition matrix according to the characteristics of the pressure field approximation in adjacent time step and the bandwidth distribution of the equation coefficient matrix. Based on this method, the PCG solver is developed with Visual C++ and is integrated into Z-Mold. These examples show that the solution time of PCG is 20% less than that of Gauss-Seidel.3. A special algorithm to predict the deformation of the injection-molded plastic strip-like part is suggested, namely the two-step finite element method (FEM). The strip-like plastic part is regarded as a little-curved beam macroscopically, and is divided into a few one-dimensional elements. On the section of each elemental node location, two-dimensional thermal finite element analysis is made to obtain the non-uniform thermal stress caused by the time difference of the solidification of the plastic melt in the mold. The stress relaxation, or equivalently, strain creep is dealt with by using a special computing model. On the bases of in-mold elastic stress, the final bending moment to the beam is obtained and the warpage is predicted. The CAE software StripMolding to predict the warpage of the injection-molded strip-like plastic parts is developed with Visual C++and is integrated into Z-Mold. The numerical examples showed good agreement with real practice.4. For the injection-molded strip-like plastic parts, they are more easily deformed and have the larger deflection than the products with conventional size, the traditional method to control part warpage is not very effective for them. To solve the problems mentioned above, taking a typical strip-like plastic part, chafing strip on the automobile door as example, the pre-deformed reverse compensating method is adopted to control the plastic part warpage, and CAE technology is repeatedly employed to guide the pre-deformed design and verify the pre-deformed design for obtaining the final pre-deformation value and curve of the plastic part. Based on the final pre-deformed design, the mould is designed and made and the real product meeting the accuracy requirements is produced. The practice shows that the pre-deformed design expands the injection molding process window, effectively improves the success rate of once mould tryout and production efficiency and is the powerful technique for controlling strip-like plastic part warpage, moreover, CAE technology is an important tool for determining the pre-deformation direction and curve of strip-like plastic part.According to the current situation of CAE technology, combined with the requirements of Z-Mold development and upgrading, the dissertation carries out the key technology research, which develops the theory and algorithm of injection molding. These works optimize and improve the Z-Mold software, so that the software has important application value for the real production.