Injection Molding Process Quality Analysis and Prediction Based on Batch Dynamic Augmentation

In order to predict the quality of batch process accurately, this paper takes some work for the typical batch process, injection molding process. Due to the slow time varying nature of the injection molding process, the characteristics of different batches may vary, so it is not easy to predict product quality with a single model. Therefore, in this paper, a batch process quality prediction method based on batch dynamic analysis is proposed for the injection molding process. The sliding window model is constructed, and the region of support is redefined, based on which batch dynamic is included into the building of the PLS regression model. The offline quality analysis and online quality prediction of products in the injection molding process are completed, and compared with the traditional quality prediction method based on PLS, the advantages of the proposed method are presented.