Prediction of cutting process parameters in boring operations using artificial neural networks

2013 ◽  
Vol 21 (6) ◽  
pp. 1043-1054 ◽  
Author(s):  
K Ramesh ◽  
T Alwarsamy ◽  
S Jayabal
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Process Parameters ◽  
Cutting Process ◽  
Artificial Neural

scholarly journals Development of turning process digital twin based on machine learning

2021 ◽  
pp. 32-41
Author(s):  
D. A. Rastorguev ◽  
◽  
A. A. Sevastyanov ◽  
Keyword(s):  
Neural Networks ◽  
Mathematical Model ◽  
Artificial Neural Networks ◽  
Information Technologies ◽  
Cutting Process ◽  
Turning Process ◽  
Force Load ◽  
Cnc Machine ◽  
Digital Twin ◽  
Artificial Neural

Today, manufacturing technologies are developing within the Industry 4.0 concept, which is the information technologies introduction in manufacturing. One of the most promising digital technologies finding more and more application in manufacturing is a digital twin. A digital twin is an ensemble of mathematical models of technological process, which exchanges information with its physical prototype in real-time. The paper considers an example of the formation of several interconnected predictive modules, which are a part of the structure of the turning process digital twin and designed to predict the quality of processing, the chip formation nature, and the cutting force. The authors carried out a three-factor experiment on the hard turning of 105WCr6 steel hardened to 55 HRC. Used an example of the conducted experiment, the authors described the process of development of the digital twin diagnostic module based on artificial neural networks. When developing a mathematical model for predicting and diagnosing the cutting process, the authors revealed higher accuracy, adaptability, and versatility of artificial neural networks. The developed mathematical model of online diagnostics of the cutting process for determining the surface quality and chip type during processing uses the actual value of the cutting depth determined indirectly by the force load on the drive. In this case, the model uses only the signals of the sensors included in the diagnostic subsystem on the CNC machine. As an informative feature reflecting the force load on the machine’s main motion drive, the authors selected the value of the energy of the current signal of the spindle drive motor. The study identified that the development of a digital twin is possible due to the development of additional modules predicting the accuracy of dimensions, geometric profile, tool wear.

Modeling & Analysis of End Milling Process Parameters Using Artificial Neural Networks

2014 ◽  
Vol 592-594 ◽  
pp. 2733-2737 ◽  
Author(s):  
G. Harinath Gowd ◽  
K. Divya Theja ◽  
Peyyala Rayudu ◽  
M. Venugopal Goud ◽  
M .Subba Roa
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Process Parameters ◽  
End Milling ◽  
Milling Process ◽  
Machining Process ◽  
Depth Of Cut ◽  
Machining Parameters ◽  
Artificial Neural ◽  
End Milling Process

For modeling and optimizing the process parameters of manufacturing problems in the present days, numerical and Artificial Neural Networks (ANN) methods are widely using. In manufacturing environments, main focus is given to the finding of Optimum machining parameters. Therefore the present research is aimed at finding the optimal process parameters for End milling process. The End milling process is a widely used machining process because it is used for the rough and finish machining of many features such as slots, pockets, peripheries and faces of components. The present work involves the estimation of optimal values of the process variables like, speed, feed and depth of cut, whereas the metal removal rate (MRR) and tool wear resistance were taken as the output .Experimental design is planned using DOE. Optimum machining parameters for End milling process were found out using ANN and compared to the experimental results. The obtained results provβed the ability of ANN method for End milling process modeling and optimization.

Prediction of Fracture Force of Weld Line Based on Artificial Neural Networks

2011 ◽  
Vol 314-316 ◽  
pp. 547-553
Author(s):  
Peng Fei Zhu ◽  
Xiao Fang Sun ◽  
Ying Jun Lu ◽  
Hai Tian Pan
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Process Parameters ◽  
Weld Line ◽  
Percentage Error ◽  
Maximum Relative Error ◽  
Melt Temperature ◽  
Data Set ◽  
Fracture Force ◽  
Artificial Neural

A feed-forward three-layer neural network was proposed to predict the fracture force of injection-molded parts’ weld line. Firstly, the most significant process parameters which affect the fracture force of weld line were analyzed. Secondly, melt temperature, injection pressure, holding pressure and holding time were chosen as import variables and the fracture force of weld line was chosen as output variable to construct artificial neural networks. Furthermore, the performance of ANN was evaluated and tested by its application to verification tests with process parameters randomly selected which all of them were not used in the network training. Results showed that the ANN predictions yield mean absolute percentage error (MAPE) in the range of 0.86％,and maximum relative error (MRE) in the range of 1.84％ for the test data set, and which can comparatively accurately reflect the influence relation of the injection process parameters on part’s quality index under the circumstance of data deficiencies.

Prediction of Powder Injection Molding Process Parameters Using Artificial Neural Networks

2012 ◽  
Vol 59 (2) ◽  
Author(s):  
Javad Rajabi ◽  
Norhamidi Muhamad ◽  
Maryam Rajabi ◽  
Jamal Rajabi
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Injection Molding ◽  
Process Parameters ◽  
Powder Injection Molding ◽  
Powder Injection ◽  
Injection Molding Process ◽  
Ann Model ◽  
Back Propagation Algorithm ◽  
Artificial Neural

The parameters of Powder Injection Molding (PIM) process were modeled by artificial neural networks (ANNs). The feed-forward multilayer perceptron was utilized and trained by back-propagation algorithm. Particle size, particle morphology, debinding time, and sintering temperature were taken into account and regarded as inputs of the ANN model. The outputs included relative density, wax loss, shrinkage, and hardness. The results obtained using the ANN model were in good agreement with the experimental data. In fact, they displayed an average R-value of 0.95 versus the experimental values. The optimum architecture of ANN was 7-4-1, in which the network was trained with Levenberg–Marquardt training algorithm. Thus, the ANN model can be used to evaluate, calculate, and forecast PIM process parameters.

Classification of disc damage status by discovering knowledge from experimental data in marble cutting process

2016 ◽  
Vol 231 (13) ◽  
pp. 2407-2416 ◽  
Author(s):  
Emre Akarslan ◽  
Fatih O Hocaoğlu ◽  
Ismail Ucun
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Cutting Process ◽  
Mean Values ◽  
Current Voltage ◽  
The Status ◽  
Artificial Neural ◽  
Natural Stones ◽  
Industrial Injuries

In marble industry, it is of vital importance to determine the damaged discs on time to prevent possible industrial injuries. Therefore, in this study, it is proposed to classify the status of the cutting discs that are used while cutting the natural stones. To classify the deflections of the discs, 673 different experiments are performed. Cutting discs corresponding to four different damage classes (undamaged disc, less damaged disc, much damaged disc, and broken disc) are employed in the tests. Eight different parameters (cutting forces (Fx, Fy, Fz), noise, peripheral speed of the disc, current, voltage, power consumption) are measured and recorded in the experiments. For each experiment, mean values of different measured data are studied. Artificial neural networks are employed as classifiers. In the first stage, all of these mean values corresponding to eight parameters are selected as the input vectors of the artificial neural networks, whereas in the second stage, the dimension of input vector is decreased by leaving out the parameters one by one. In this stage, it is aimed to determine the most important parameter that caries much more information about the cutting process.

Relative density prognosis for directed energy deposition with the help of artificial neural networks

2021 ◽  
Vol 63 (1) ◽  
pp. 41-47
Author(s):  
Angelina Marko ◽  
Andreas Schafner ◽  
Julius Raute ◽  
Michael Rethmeier
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Artificial Neural Network ◽  
Artificial Neural Networks ◽  
Relative Density ◽  
Process Parameters ◽  
Energy Deposition ◽  
Data Sets ◽  
Artificial Neural ◽  
Directed Energy

Abstract Additive manufacturing, and therefore directed energy deposition, is gaining more and more interest from industrial users. However, quality assurance for the components produced is still a challenge. Machine learning, especially using artificial neuronal networks, is a potential method for ensuring a high-quality standard. Based on process parameters and monitoring data, part quality can be predicted. A further advantage is the ability to constantly learn and adopt to slight process changes. First tests using artificial neural networks focus on the prediction of track geometry. The results show that even a small data set is enough to provide high accuracy in the predictions. In this work, an artificial neural network for the predictive analysis of relative density in laser powder cladding has been developed. A central composite experimental design is used to generate 19 data sets. Input variables are laser power, feed rate and powder mass flow. Cubes are built up where density is considered as a target value. Several neural networks are trained and evaluated with these data sets. Different topologies and initial weights are considered. The best network reaches a confidence level of around 90 % for the prediction of relative density based on the process parameters. Finally, the optimization of the generalization performance is investigated. To this purpose, methods of variation in error limit as well as cross-validation are applied. In this way, density is predictable by an artificial neural network with an accuracy of about 95 %.

scholarly journals A Strategy for Problem Solving of Filling Imbalance in Geometrically Balanced Injection Molds

Polymers ◽  
2020 ◽  
Vol 12 (4) ◽  
pp. 805
Author(s):  
Krzysztof Wilczyński ◽  
Przemysław Narowski
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Problem Solving ◽  
Process Parameters ◽  
Experimental Studies ◽  
Injection Pressure ◽  
Optimization Procedure ◽  
Operating Conditions ◽  
Melt Temperature ◽  
Artificial Neural

Simulation and experimental studies were performed on filling imbalance in geometrically balanced injection molds. An original strategy for problem solving was developed to optimize the imbalance phenomenon. The phenomenon was studied both by simulation and experimentation using several different runner systems at various thermo-rheological material parameters and process operating conditions. Three optimization procedures were applied, Response Surface Methodology (RSM), Taguchi method, and Artificial Neural Networks (ANN). Operating process parameters: the injection rate, melt temperature, and mold temperature, as well as the geometry of the runner system were optimized. The imbalance of mold filling as well as the process parameters: the injection pressure, injection time, and molding temperature were optimization criteria. It was concluded that all the optimization procedures improved filling imbalance. However, the Artificial Neural Networks approach seems to be the most efficient optimization procedure, and the Brain Construction Algorithm (BSM) is proposed for problem solving of the imbalance phenomenon.

Sign in / Sign up

Export Citation Format

Share Document