Intelligent Design And Optimization Of Injection Molding Process Parameters

As the important molding methods of plastic products,injection molding is more and more widely used in the production of plastic products in various industries.Improving production efficiency has also become an urgent problem to be solved in the current injection molding production.Because of the long production cycle and high cost,the traditional mold test method of setting process parameters based on experience can not meet the current needs of injection molding production.Aiming at the setting of injection molding process parameters under the traditional mold test method,an injection molding process window setting process based on scientific mold test is proposed,which is composed of five parts: Determination of injection speed,cavity balance test,injection pressure setting,pressure holding pressure setting and pressure holding time setting.Finally,a molding window that can stabilize the production state can be formed,and this mold test process is realized based on mold flow analysis software,Through the first mock exam of two cavity products,the injection process window was obtained,that is,the injection speed interval was70mm/s～100mm/s,the melt temperature range was 290 ～315,the injection pressure was set to 162 Mpa,the holding pressure at the temperature of 29040～70Mpa was 40～70Mpa,when the temperature was 315 degrees,the holding time was 7～10s,which verified the feasibility of the scientific test process.Then,a scientific mold trial process window setting system is designed and developed,which integrates the mold trial process and results into the software to facilitate the result call and data storage.Aiming at the problem that the product quality corresponding to the process parameters in the injection molding process window cannot be known,an artificial intelligence algorithm model for the prediction of injection molding process parameters is established,and the long-term and short-term memory neural network(LSTM)is used as the prediction model.After the model is trained,the product weight is predicted as the output value.After 50 groups of prediction data are obtained,it is tested and verified on the injection molding machine,The predicted data are compared with the actual production data.The results show that the difference between the actual injection molded product quality and the product quality predicted by artificial intelligence algorithm is small,the maximum weight difference is 2.206 g,and the error is less than 5%,which meets the product quality requirements.Moreover,for 50 groups of process parameter combinations,the product quality is consistent,so the process parameter combinations with qualified quality and consistent quality can be produced,It can be determined as the optimal combination of process parameters.The neural network can predict the quality of the product in advance,and the accuracy of the neural network can be verified in terms of the quality of the molding process and the reliability of the product.