Multi-objective Optimization Design Of The Injection Molding Process Parameters

Injection molding is an efficient and economic molding technology for plastic products. There are many non-linear and uncertain factors between molding processing conditions and quality of products, makes the process parameters optimization and quality control very difficult. Traditional methods of regulating the processing is not only time-consuming and high-cost, but also overly excessively dependent on experience and past cases, and its accuracy is not high, cannot meet the requirements of new products's express Production. Although the numerical simulation can reduce the cost of test, and the simulation results can also be acceptable to make a guidance to the adjust of process parameters, but the guidance is lack of quantitative precision. To obtain good results, experience, and repeated tests are still necessary. Design of Experiment can reduce the blindness of trial and error to some extent, and obtion the good combination of processing variables within the testing scope. Because of the limited values of per variable studied in the experiments, it is difficult achieve the global optimization of the molding process parameters.According to the existing research methods, the author analyzes the rheological behavior of polymer melt flow and heat transfer, as well as thermoplastic injection molding numerical simulation theory, use intelligent optimization technology into the injection molding process parameters optimization, obtained samples data from DOE numerical simulation, establish an neural net ensemble,then use the genetic algorithm and fuzzy weighted score to optimize the net model, proposed multi-objective optimization methods. Study a shell-type product example, and the results verify the feasibility of this method.In this thesis, the following research carried out and the corresponding conclusions are as follows:1. Deeply discussed the rheological theory of injection molding processing and effect of processing parameters on parts qulity at first. And on the base of the current CAE software of injection molding, the simulation theory is studied. According to the structural design of products and the use of the request, determine the quality indicators of the products and the process conditions.2. Take the data samples which is obtained form DOE numerical simulation and CAE software in range analysis and fuzzy comprehensive quality weighted score analysis, results are as follows:1) for the changes amount of volumetric shrinkage, the melt temperature have a biggest influence and the packing pressure on the second, the next is mold temperature; 2) for the average volumetric shrinkage, packing pressure have the biggest influence, mold temperature, melt temperature and fill time have a relatively large impact too; 3) for sink index, the melt temperature, packing pressure all have a bigger influence, and filling time next; 4) for Warpage, the pressure have a biggest pressure, melt temperature, filling time and cooling time have a greater influence; 5) it will be useful to effectively resolve the many conflicts and integrated quality optimization problem with the method that transform the multi-objective problem into single-objective problem.3. Neural net approximate calculation agent model is created to predict the product's quality, get an non-linear mapping mathematical mold which is useful for next step optimization between the processing parameters and the product's quality index. Adoption the natural evolution of genetic algorithm ideas, use the fuzzy comprehensive evaluation of the product's quality function as fitness function, Set of neural net approximate calculation agent model for global optimization, got the best process parameters, and achieve a multi-objective optimization of product's quality. The results showed that:The multi-objective optimization method is reasonable and feasible.Use the CAE simulation software, as well as the DOE and application of intelligent optimization algorithm, can take a variety of possible defects of the products at the design stage, and comprehensive optimizate the process parameters for shorter the design cycle. The idea of intelligent optimization is not only fit for the processing optimization of injection molding, but also for optimization of other problems with multi-facter effect, especially multi-index restriction, non-linear and uncertain relationship.