Hybrid neural models for pressure control in injection molding

Petrova, Tatiana; Kazmer, David O.

doi:10.1002/(sici)1098-2329(199921)18:1<19::aid-adv3>3.0.co;2-u

Cited by 17 publications

(9 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another approach to model and control the dynamics of the injection process without utilizing first-principles models are neural networks (NNs). Petrova and Kazmer combined a NN with process knowledge for the training of the NN and derived a process model for the injection pressure [30]. Michaeli and Schreiber developed a NN for cavity pressure control enabling a model-based predictive control of process variables [31].…”

Section: Literature Surveymentioning

confidence: 99%

Control-oriented modeling of servo-pump driven injection molding machines in the filling and packing phase

Froehlich

Kemmetmüller

Kugi

2018

Mathematical and Computer Modelling of Dynamical Systems

View full text Add to dashboard Cite

Servo-valves or variable displacement pumps are typically used to control conventional hydraulic injection molding machines (IMMs). Recent developments in electrical drive technology allow to utilize servo-motor driven pumps instead, which is beneficial due to their higher energy efficiency. Their dynamic behavior, however, is significantly different compared to the conventional setup. Thus, currently used mathematical models and control concepts cannot be directly applied. This paper presents a computationally efficient and scalable mathematical model of the injection process for these servo-pump driven IMMs. A first-principles model of the injection machine is combined with a phenomenological model describing the injection process, i.e. the compression of the melt and the polymer flow into the mold. The proposed model is tailored to real-time applications and serves as an ideal basis for the design of model-based control strategies. The feasibility of the proposed model is demonstrated by a number of different experiments. They confirm a high model accuracy over the whole operating range for different mold geometries. ARTICLE HISTORY

show abstract

Section: Literature Surveymentioning

confidence: 99%

Control-oriented modeling of servo-pump driven injection molding machines in the filling and packing phase

Froehlich

Kemmetmüller

Kugi

2018

Mathematical and Computer Modelling of Dynamical Systems

View full text Add to dashboard Cite

show abstract

“…This directly results in a higher uniformity of the process. Furthermore, the adaptive control concept [18,[25][26][27][28] and artificial neural networks [29][30][31] can be applied, which is able to accommodate the switchover temperature to the mold temperature, thus a high stability of the process and the quality can be accomplished.…”

Section: Cavity Pressure and Temperature Profilementioning

confidence: 99%

Filling-To-Packing Switchover Mode Based on Cavity Temperature for Injection Molding

Jian

Peng

Yang

2011

Polymer-Plastics Technology and Engineering

View full text Add to dashboard Cite

A filling-to-packing switchover mode based on cavity temperature was presented and compared with three other switchover modes. The part weight and dimension served as the main measure to probe the process capabilities. It revealed that the switchover mode based on cavity temperature yielded better part quality, more consistent part weight and dimension. The switchover temperature and sensor position should be considered for achieving such high process capabilities. The switchover mode based on cavity temperature can be used to accommodate the product weight change due to the variation of mold temperature, which is something that the other switchover modes cannot handle. INTRODUCTIONDuring injection molding, the position at which the machine switches from filling phase to packing phase is referred to as the switchover point. Plastic processors have noted for years that the repeatability of molded part qualities is significantly affected by the ability to identify the proper switchover point. Two inadequate filling-to-packing switchover conditions are: switchover occurring too early and switchover occurring too late. If the switchover takes place prematurely, it may lead to insufficient packing, the injection part can be easily rejected due to reduced dimensions, being underweight, or showing sink marks. The switchover occurring too late will cause an over-packed cavity, the molding will flash or may stick in the mold. Over-packing further adds weight and stress to the part, and makes de-molding more difficult.There are many filling-to-packing switchover modes that machines are capable of using today. The more common

show abstract

“…Among the many models of NN, the back propagation neural network (BPNN) is a known model for fault detection and prediction application [8][9][10][11]. Each neuron always receives information from the neurons of the previous layer and after, performing computations on that information transfers it to the neurons of the next layer.…”

Section: Design Of Back Propagation Network and Estimation Of Weld Pementioning

confidence: 99%

Prediction of Weld Penetration in FCAW of HSLA steel using Artificial Neural Networks

Asl¹,

Mostafa²,

Seyedkashi³

2011

AIP Conference Proceedings

View full text Add to dashboard Cite

Flux-cored arc welding (FCAW) is a semiautomatic or automatic arc welding process that requires a continuously-fed consumable tubular electrode containing a flux. The main FCAW process parameters affecting the depth of penetration are welding current, arc voltage, nozzle-to-work distance, torch angle and welding speed. Shallow depth of penetration may contribute to failure of a welded structure since penetration determines the stress-carrying capacity of a welded joint. To avoid such occurrences; the welding process parameters influencing the weld penetration must be properly selected to obtain an acceptable weld penetration and hence a high quality joint. Artificial neural networks (ANN), also called neural networks (NN), are computational models used to express complex non-linear relationships between input and output data. In this paper, artificial neural network (ANN) method is used to predict the effects of welding current, arc voltage, nozzle-to-work distance, torch angle and welding speed on weld penetration depth in gas shielded FCAW of a grade of high strength low alloy steel. 32 experimental runs were carried out using the bead-on-plate welding technique. Weld penetrations were measured and on the basis of these 32 sets of experimental data, a feedforward back-propagation neural network was created. 28 sets of the experiments were used as the training data and the remaining 4 sets were used for the testing phase of the network. The ANN has one hidden layer with eight neurons and is trained after 840 iterations. The comparison between the experimental results and ANN results showed that the trained network could predict the effects of the FCAW process parameters on weld penetration adequately.

show abstract

Hybrid neural models for pressure control in injection molding

Cited by 17 publications

References 4 publications

Control-oriented modeling of servo-pump driven injection molding machines in the filling and packing phase

Control-oriented modeling of servo-pump driven injection molding machines in the filling and packing phase

Filling-To-Packing Switchover Mode Based on Cavity Temperature for Injection Molding

Prediction of Weld Penetration in FCAW of HSLA steel using Artificial Neural Networks

Contact Info

Product

Resources

About