| The state and variation history of polymer in cavity have direct influence on final part performance and quality in injection molding. It is an important method in polymer processing field to simulate the polymer processing, predict the state and variation of the melt and aid to design the mold and setup the processing parameters. However, simulation of molding can only replace the trial-and-error and the correct also depend on the design's experience. It is difficult to get the best process condition, and the influence of disturbance make it more difficult to get consistent good quality part. The investigation of on-line quality control for injection molding to make the molding keep on the desirable level and produce high quality part consistently is very valuable practically. This dissertation focus on the simulation of filling/post-filling stage, analysis and modeling of the relationship between quality indexes and processing parameters, optimization of processing parameters and control of process and part quality in injection molding. The theories and algorithms are investigated and developed for optimization of process and control of part quality in injection molding, and a novel part quality control system is proposed and implemented. The achievements included:1. Based on the compressible flow of viscous fluid under non-isothermal conditions, a unified theoretical model of filling/post-filling stage for 3D thin parts is developed based on some assumes and simplifications. Compressibility and phase change are considered in the unified model. The numerical model is developed by hybrid finite-element/finite-difference /control-volume method with un-uniform time step and variable difference mesh across the half-gap thickness, and the numerical simulation of unified filling/post-filling model is implemented. The simulation can predicted filling pattern, location of weld line and air trap, and the variations of the temperature, pressure, density, shear rate etc with time and space. The part quality indexes such as volumetric shrinkage, sink mark index and weight are further predicted based on the simulation. The simulation results are satisfied with experimental results.2. Orthogonal array experiments based on Taguchi DOE are conducted using material ABS, HDPE and PP to investigate relationships of quality indexes (such as sink mark index, volumetric shrinkage variation and weight) with processing parameters based on simulation results of filling/post-filling stage, and significance of processing parameters are determined and the optimal factor level combinations are obtained.3. The control variables are selected based on significance of processing parameters, and the process and part quality characteristics are extracted using simulation results based on baseline condition of near optimized condition. The relationship models of quality indexes with processing parameters, which is the basis of part quality control, are built by ANN technique.4. The theories and algorithms are developed to optimize the processing parameters of single-objective and multi-objective problem for injectionmolding by hybrid genetic-algorithms/grads-descend method based on ANN model, and the constrains of genetic-algorithms are deal with non-stationary penalty function. Grads-descend operator improves the convergence performance, and the use of non-stationary penalty function allows the GA to explore more of the space before confining to feasible region to ensure that the global optimal solution is obtained in process optimization. The global optimal solutions provide process setups and reference quality indexes for part quality control. The optimization of process parameters in numerical examples improves the distribution of volumetric shrinkage and sink mark of the part obviously.5. Closed-loop part quality control strategy with series connection of process control loop and part quality control loop is proposed based on the mechanism of injection molding, the relationships of process setups...
|
PDF Full Text Request