Quality Prediction for Injection Molding by Using a Multilayer Perceptron Neural Network

Abstract

Injection molding has been widely used in the mass production of high-precision products. The finished products obtained through injection molding must have a high quality. Machine parameters do not accurately reflect the molding conditions of the polymer melt; thus, the use of machine parameters leads to erroneous quality judgments. Moreover, the cost of mass inspections of finished products has led to strict restrictions on comprehensive quality testing. Therefore, an automatic quality inspection that provides effective and accurate quality judgment for each injection-molded part is required. This study proposes a multilayer perceptron (MLP) neural network model combined with quality indices for performing fast and automatic prediction of the geometry of finished products. The pressure curves detected by the in-mold pressure sensor, which reflect the flow state of the melt, changes in various indicators and molding quality, were considered in this study. Furthermore, the quality indices extracted from pressure curves with a strong correlation with the part quality were input into the MLP model for learning and prediction. The results indicate that the training and testing of the first-stage holding pressure index, pressure integral index, residual pressure drop index and peak pressure index with respect to the geometric widths were accurate (accuracy rate exceeded 92%), which demonstrates the feasibility of the proposed method.

Quality Classification of Injection-Molded Components by Using Quality Indices, Grading, and Machine Learning

“…Quality indices that are highly correlated with part quality are used as input parameters for model training in order to increase prediction accuracy. By referring to the pressure information, the following quality indices were selected [ 9 ].…”

“…This index is related to the total pressure characteristics of the polymer melt during the injection molding process. The variation in may reflect the change in part quality, particularly the change in part weight [ 9 ].…”

“…Because of advancements in sensing technology, pressure and temperature sensors are available for detecting changes in the properties of polymer melts in molds and for monitoring and controlling the injection molding process [ 3 , 4 , 5 , 6 , 7 , 8 ]. Although the cost of installing cavity sensors in production systems is higher than the cost of using machine profiles, the use of cavity profiles helps to reduce prediction errors [ 9 , 10 ]. A high correlation exists between part quality and features extracted from cavity pressure curves [ 11 , 12 ].…”

“…Guo et al [ 18 ] used input and output data obtained from finite element simulations to train various neural networks to function as warpage prediction models; they concluded that a hybrid particle swarm optimization–backpropagation algorithm is the ideal model for predicting the warpage of microcellular foaming material. A feedforward MLP neural network model has been proven to be suitable for the regression and classification of data in applications such as quality inspection and has been used in many studies [ 9 , 19 ]. The main advantage of the MLP model is to solve complex nonlinear problems, such as classification; however, the training process is time-consuming.…”

“…In particular, most previous studies have focused on using process parameters as input values for model learning, which may not be appropriate. Instead of using process parameter settings as input values, quality indices that are highly related to part quality can provide sufficient information for model learning, which has attracted attention recently [ 9 , 10 , 12 ].…”