Neural network feature and architecture optimization for injection molding surface defect prediction of model polypropylene

Predictive Methodology for Quality Assessment in Injection Molding Comparing Linear Regression and Neural Networks

The use of recycled polypropylene in industry to reduce environmental impact is increasing. Design for manufacturing and process simulation is a key stage in the development of plastic parts. Traditionally, a trial-and-error methodology is followed to eliminate uncertainties regarding geometry and process. A new proposal is presented, combining simulation with the design of experiments and creating prediction models for seven different process and part quality output features. These models are used to optimize the design without developing additional time-consuming simulations. The study aims to compare the precision and correlation of these models. The methods used are linear regression and artificial neural network (ANN) fitting. A wide range of eight injection parameters and geometry variations are used as inputs. The predictability of nonlinear behavior and compensatory effects due to the complex relationships between this wide set of parameter combinations is analyzed further in the state of the art. Results show that only Back Propagation Neural Networks (BPNN) are suitable for correlating all quality features in a single formula. The use of prediction models accelerates the optimization of part design, applying multiple criteria to support decision-making. The methodology is applied to the design of a plastic support for induction hobs. Furthermore, this methodology has demonstrated that a weight reduction of 27% is feasible. However, it is necessary to combine process parameters that differ from the standard ones with a non-uniform thickness distribution so that the remaining injection parameters, material properties, and dimensions fall within tolerances.

Artificial Neural Network Training Using Structural Learning with Forgetting for Parameter Analysis of Injection Molding Quality Prediction

The analysis of influential machine parameters can be useful to plan and design a plastic injection molding process. However, current research in parameter analysis is mostly based on computer-aided engineering (CAE) or simulation which have been demonstrated to be inadequate for analyzing complex behavioral changes in the real injection molding process. More advanced approaches using machine learning technology specifically with artificial neural networks (ANNs) brought promising results in terms of prediction accuracy. Nevertheless, the black box and distributed representation of ANN prevent humans from gaining an insight into which process parameters give a significant influence on the final prediction output. Therefore, in this paper, we develop a simpler ANN model by using structural learning with forgetting (SLF) as the algorithm for the training process. Instead of typical backpropagation which generated a fully connected layer of the ANN model, SLF only reveals the important neurons and connections. Hence, the training process of SLF leaves only influential connections and neurons. Since each of the neurons specifically on the input layer represent each of the injection molding parameters, the ANN-SLF model can be further investigated to determine the influential process parameters. By applying SLF to the ANN training process, this experiment has successfully extracted a set of significant injection molding process parameters.

Natural Rubber Blend Optimization via Data-Driven Modeling: The Implementation for Reverse Engineering

Natural rubber formulation methodologies implemented within industry primarily implicate a high dependence on the formulator’s experience as it involves an educated guess-and-check process. The formulator must leverage their experience to ensure that the number of iterations to the final blend composition is minimized. The study presented in this paper includes the implementation of blend formulation methodology that targets material properties relevant to the application in which the product will be used by incorporating predictive models, including linear regression, response surface method (RSM), artificial neural networks (ANNs), and Gaussian process regression (GPR). Training of such models requires data, which is equal to financial resources in industry. To ensure minimum experimental effort, the dataset is kept small, and the model complexity is kept simple, and as a proof of concept, the predictive models are used to reverse engineer a current material used in the footwear industry based on target viscoelastic properties (relaxation behavior, tanδ, and hardness), which all depend on the amount of crosslinker, plasticizer, and the quantity of voids used to create the lightweight high-performance material. RSM, ANN, and GPR result in prediction accuracy of 90%, 97%, and 100%, respectively. It is evident that the testing accuracy increases with algorithm complexity; therefore, these methodologies provide a wide range of tools capable of predicting compound formulation based on specified target properties, and with a wide range of complexity.