Neural networks for the dimensional control of molded parts based on a reverse process model

This paper presents the application of neural networks in suggesting the change of molding parameters for improving the dimensional quality of molded parts based on the concept of reverse process modeling. Instead of using the molding condition parameters as input values and dimensional outcomes as output values, the reverse process model configures the dimensional outcomes as inputs and the molding condition parameters as outputs. With the mapping on input and output layers of neural networks based on this configuration, the trained neural networks learn the correlation between the dimensional outcome values and the corresponding molding parameters. This model, which serves to learn from sample data and induce the values for change of the operating molding conditions, has been implemented for the dimensional improvement of injection molding parts, the dimensions of which are primarily determined by the process parameters such as injection time and cooling temperature.