MODELING OF HOT EXTRUSION PROCESS USING ARTIFICIAL NEURAL NETWORKS IMPLANTED WITH GENETIC ALGORITHM
Keywords:
Finite Element Simulation, Process modeling, Extrusion Load, Artificial Neural Network, Genetic AlgorithmAbstract
Hot extrusion is a complex metal-forming process and requires careful selection of parameters, and control and inspection through a precise simulation and analysis. This paper proposes modeling of hot extrusion using multi-layered perceptron trained by Genetic Algorithm (GA). The data obtained from Finite Element Method simulations of a typical hot extrusion process are modeled in a multi layered Artificial Neural Networks (ANN) with four inputs to get an output of extrusion load. The proposed method also uses a Genetic Algorithm procedure to find the optimal weights, which makes the model efficient and accurate. The final trained network model will predict the requisite forces for given parameters combinations in real time with out any extensive and expensive computations.
Downloads
References
David S, J., Darrell Whitely, Larry J.Eshehman. Combinations of Genetic Algorithms and Neural Networks: A survey of the state of the Art, Proceedings of the IEEE International workshop on combinations of genetic algorithms and neural networks, Baltimore 1-37.
Djajasaputra S. R. Application of Artificial Neural Networks in Industrial Engineering. Proceedings of National conference Institute of Industrial Engineering, 1998.
Hansraj K., R. S. Sharma S. Srivastava and C. Patvardhan, Modelling of manufacturing Process with ANNs for intelligent manufacturing, International Journal of Machine tools and Manufacture 2000, V40, 851 – 868.
Hansraj K., S. Srivastava, S. Pal, S. Verma, R. S. Sharma and C. Patvardhan, Cutting force optimisation using multi objective neuro stochastic search technique in intelligent manufacturing, Proceedings of International Conference on Intelligent Flexible Autonomous Manufacturing Systems 2000, 535 – 541,India
Hou T, H., and L.Lin, Manufacturing Process Monitoring using Neural Networks, Computers Electrical Engineering 1993, V19, 129 – 141.
Seiffert U. Multiple layer perceptron training using genetic algorithm, Proceedings of European symposium on Artificial Neural Networks Bruges, 2001,159-164.
Zhang H.C and S. H. Huang, Application of neural networks in manufacturing: a state of the art survey, International Journal of Production research, 1995,V33, 705-728.
