Melt temperature learning control of injection molding process based on CMAC neural network

2011 ◽  
Vol 31 (1) ◽  
Author(s):  
Yonggang Peng ◽  
Wei Wei
Keyword(s):  
Neural Network ◽  
Injection Molding ◽  
Control Method ◽  
System Modeling ◽  
Learning Control ◽  
Injection Molding Process ◽  
Cerebellar Model Articulation Controller ◽  
Molding Process ◽  
Melt Temperature ◽  
Cmac Neural Network

Abstract Accurately monitoring and controlling melt temperature in the injection molding process can be a challenge. A barrel temperature model was achieved with the use of a system modeling method. Because of time variance, uncertainty and non-linearity of injection molding barrel temperature, a learning control method based on the use of a cerebellar model articulation controller (CMAC) neural network was proposed. Simulations and experimental results have demonstrated that this method elicits high response speed and excellent control accuracy of barrel melt temperature.

scholarly journals Induced network-based transfer learning in injection molding for process modelling and optimization with artificial neural networks

2021 ◽  
Vol 112 (11-12) ◽  
pp. 3501-3513
Author(s):  
Yannik Lockner ◽  
Christian Hopmann
Keyword(s):  
Injection Molding ◽  
Transfer Learning ◽  
Injection Molding Process ◽  
Process Data ◽  
Molding Process ◽  
Melt Temperature ◽  
Processing Industry ◽  
Source Models ◽  
Plastics Processing ◽  
Artificial Neural

AbstractThe necessity of an abundance of training data commonly hinders the broad use of machine learning in the plastics processing industry. Induced network-based transfer learning is used to reduce the necessary amount of injection molding process data for the training of an artificial neural network in order to conduct a data-driven machine parameter optimization for injection molding processes. As base learners, source models for the injection molding process of 59 different parts are fitted to process data. A different process for another part is chosen as the target process on which transfer learning is applied. The models learn the relationship between 6 machine setting parameters and the part weight as quality parameter. The considered machine parameters are the injection flow rate, holding pressure time, holding pressure, cooling time, melt temperature, and cavity wall temperature. For the right source domain, only 4 sample points of the new process need to be generated to train a model of the injection molding process with a degree of determination R2 of 0.9 or and higher. Significant differences in the transferability of the source models can be seen between different part geometries: The source models of injection molding processes for similar parts to the part of the target process achieve the best results. The transfer learning technique has the potential to raise the relevance of AI methods for process optimization in the plastics processing industry significantly.

Determination of Optimum Process Parameters in Gas-Assisted Injection Molding

2011 ◽  
Vol 143-144 ◽  
pp. 494-498
Author(s):  
Ke Ming Zi ◽  
Li Heng Chen
Keyword(s):  
Injection Molding ◽  
Delay Time ◽  
Optimization Design ◽  
Gas Pressure ◽  
Optimum Process ◽  
Injection Molding Process ◽  
Technological Parameters ◽  
Molding Process ◽  
Element Analysis ◽  
Melt Temperature

With finite element analysis software Moldflow, numerical simulation and studies about FM truck roof handle were conducted on gas-assisted injection molding process. The influences of melt pre-injection shot, gas pressure, delay time and melt temperature were observed by using multi-factor orthogonal experimental method. According to the analysis of the factors' impact on evaluation index, the optimized parameter combination is obtained. Therefore the optimization design of technological parameters is done. The results show that during the gas-assisted injection molding, optimum pre-injection shot is 94%,gas pressure is 15MPa,delay time is 0.5s,melt temperature is 240 oC. This study provided a more practical approach for the gas-assisted injection molding process optimization.

Optimization of Injection Molding Process Parameters with Material Properties Based on GA and BP

2013 ◽  
Vol 345 ◽  
pp. 586-590 ◽  
Author(s):  
Xiao Hong Tan ◽  
Lei Gang Wang ◽  
Wen Shen Wang
Keyword(s):  
Injection Molding ◽  
Process Parameters ◽  
Mold Temperature ◽  
Injection Molding Process ◽  
Axial Deformation ◽  
Molding Process ◽  
Melt Temperature ◽  
Bp Network ◽  
Injection Parameters ◽  
Moldflow Simulation

To obtain optimal injection process parameters, GA was used to optimize BP network structure based on Moldflow simulation results. The BP network was set up which considering the relationship between volume shrinkage of plastic parts and injection parameters, such as mold temperature, melt temperature, holding pressure and holding time etc. And the optimal process parameters are obtained, which is agreed with actual results. Using BP network to predict injection parameters impact on parts quality can effectively reduce the difficulty and workload of other modeling methods. This method can be extended to other quality prediction in the process of plastic parts.Keyword: Genetic algorithm (GA);Neural network algorithm (BP);Injection molding process optimization;The axial deformation

Simulation of Ceramic Injection Molding for Zirconia Optical Ferrule

2007 ◽  
Vol 336-338 ◽  
pp. 997-1000 ◽  
Author(s):  
Mei Min Zhang ◽  
Bin Lin
Keyword(s):  
Injection Molding ◽  
Injection Molding Process ◽  
Molding Process ◽  
Melt Temperature ◽  
Ceramic Injection Molding ◽  
Rectangular Distribution ◽  
Simulation Results ◽  
Die Temperature ◽  
Ram Speed ◽  
Short Shot

Zirconia Ferrule is a key part for manufacturing fiber connectors. The ceramic injection molding (CIM) process of the optical ferrule was simulated with the commercial CAE software moldflow. In order to obtain the optimum results, the orthogonal method was introduced to discuss the influence of each parameter such as die temperature, melt temperature, ram speed and gate dimension with the two kinds of distribution layout system respectively. The simulation results show that the curved distribution runner system is more suitable than the rectangular distribution one in the optical ferrule molding. Moreover, the effect of gravity on the ceramic injection molding process was discussed for determining a more reasonable balanced runner system of the special designed two-plate mold with six die cavities. It was found that short shot occurred at the top of the die cavity while other five cavities were filled well in the original designed mold. And when the top die cavity’s round runner with section diameter of 4.0mm was increased to 4.17 mm, each cavity was balanced filled without short shot.

The simulation of the warpage rule of the thin-walled part of polypropylene composite based on the coupling effect of mold deformation and injection molding process

2018 ◽  
Vol 25 (3) ◽  
pp. 593-601 ◽  
Author(s):  
Jixiang Zhang ◽  
Xiaoyi Yin ◽  
Fengzhi Liu ◽  
Pan Yang
Keyword(s):  
Injection Molding ◽  
Coupling Effect ◽  
Mold Temperature ◽  
Injection Molding Process ◽  
Plastic Part ◽  
Molding Process ◽  
Melt Temperature ◽  
Packing Pressure ◽  
Thin Walled ◽  
Packing Time

Abstract Aiming at the problem that a thin-walled plastic part easily produces warpage, an orthogonal experimental method was used for multiparameter coupling analysis, with mold structure parameters and injection molding process parameters considered synthetically. The plastic part deformation under different experiment schemes was comparatively studied, and the key factors affecting the plastic part warpage were analyzed. Then the injection molding process was optimized. The results showed that the important order of the influence factors for the plastic part warpage was packing pressure, packing time, cooling plan, mold temperature, and melt temperature. Among them, packing pressure was the most significant factor. The optimal injection molding process schemes reducing the plastic part warpage were melt temperature (260°C), mold temperature (60°C), packing pressure (150 MPa), packing time (2 s), and cooling plan 3. In this situation, the forming plate flatness was better.

The Effect of Injection Molding Process Parameters on the Buckling Properties of PBT/Short Glass Fiber Composites

2011 ◽  
Vol 284-286 ◽  
pp. 550-556 ◽  
Author(s):  
Ming Hsiung Ho ◽  
Pin Ning Wang ◽  
Chin Ping Fung
Keyword(s):  
Injection Molding ◽  
Glass Fiber ◽  
Process Parameters ◽  
Injection Molding Process ◽  
Skin Area ◽  
Molding Process ◽  
Melt Temperature ◽  
Short Glass Fiber ◽  
Filling Time ◽  
Short Glass

This study investigates the effect of various injection molding process parameters and fiber amount on buckling properties of Polybutylene Terephthalate (PBT)/short glass fiber composite. The buckling specimens were prepared under injection molding process. These forming parameters about filling time, melt temperature and mold temperature that govern injection molding process are discussed. The buckling properties of neat PBT, 15 wt%, and 30 wt% are obtained using two ends fixed fixture and computerized closed-loop server-hydraulic material testing system. The fracture surfaces are observed by scanning electron microscopy (SEM). The global buckling forces are raised when increased the fiber weight percentage of PBT. Also, the fracture mechanisms in PBT and short glass fiber matrix are fiber pullout in skin area and fiber broken at core area. It is found that the addition of short glass fiber can significantly strengthen neat PBT.

scholarly journals Fuzzy PD Control of Networked Control Systems Based on CMAC Neural Network

2012 ◽  
Vol 2012 ◽  
pp. 1-10 ◽  
Author(s):  
Li-lian Huang ◽  
Jin Chen
Keyword(s):  
Neural Network ◽  
Control Systems ◽  
Control Method ◽  
Networked Control Systems ◽  
Networked Control ◽  
Cerebellar Model Articulation Controller ◽  
Pd Controller ◽  
Control Effect ◽  
Cmac Neural Network ◽  
Reasoning Mechanism

The network and plant can be regarded as a controlled time-varying system because of the random induced delay in the networked control systems. The cerebellar model articulation controller (CMAC) neural network and a PD controller are combined to achieve the forward feedback control. The PD controller parameters are adjusted adaptively by fuzzy reasoning mechanism, which can optimize the control effect by reducing the uncertainty caused by the network-induced delay. Finally, the simulations show that the control method proposed can improve the performance effectively.

Sign in / Sign up

Export Citation Format

Share Document