On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description
In recent years, the study of the hot deformation behavior of various materials is significantly marked by an increasing utilization of artificial neural networks, which are frequently employed for a hot flow curve description. This specific kind of description is commonly solved via a Feed-Forward...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-09-01
|
| Series: | Journal of Materials Research and Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2238785421007638 |
| _version_ | 1818910044195913728 |
|---|---|
| author | Petr Opěla Ivo Schindler Petr Kawulok Rostislav Kawulok Stanislav Rusz Horymír Navrátil |
| author_facet | Petr Opěla Ivo Schindler Petr Kawulok Rostislav Kawulok Stanislav Rusz Horymír Navrátil |
| author_sort | Petr Opěla |
| collection | DOAJ |
| description | In recent years, the study of the hot deformation behavior of various materials is significantly marked by an increasing utilization of artificial neural networks, which are frequently employed for a hot flow curve description. This specific kind of description is commonly solved via a Feed-Forward Multi-Layer Perceptron architecture and rarely via a Radial Basis architecture. Both network architectures are compared to assess their suitability in the process of a hot flow curve description under a wide range of thermomechanical conditions. The performed survey is also aimed on the eventual utilization of corresponding modifications of both studied networks, namely on a Cascade-Forward Multi-Layer Perceptron and Generalized Regression network. The main results have shown that the Feed-Forward Multi-Layer Perceptron architecture represents a good choice if very high accuracy is a crucial goal. However, in the case of this architecture, finding the proper parameters can be time-consuming and the hardware burdensome. On the contrary, for the flow curve description the almost unused Radial Basis network offers a very easy training procedure and significantly shorter computing time under acceptable accuracy. The results of the submitted research should then serve as a background for the selection and following application of a suitable network architecture in the process of solving future flow curve description tasks. |
| first_indexed | 2024-12-19T22:36:32Z |
| format | Article |
| id | doaj.art-40fecdeaa05f49dfb522ba89ff5de722 |
| institution | Directory Open Access Journal |
| issn | 2238-7854 |
| language | English |
| last_indexed | 2024-12-19T22:36:32Z |
| publishDate | 2021-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Materials Research and Technology |
| spelling | doaj.art-40fecdeaa05f49dfb522ba89ff5de7222022-12-21T20:03:10ZengElsevierJournal of Materials Research and Technology2238-78542021-09-011418371847On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve descriptionPetr Opěla0Ivo Schindler1Petr Kawulok2Rostislav Kawulok3Stanislav Rusz4Horymír Navrátil5Corresponding author.; Faculty of Materials Science and Technology, VSB–Technical University of Ostrava, 17. Listopadu 2172/15, 70800 Ostrava–Poruba, Czech RepublicFaculty of Materials Science and Technology, VSB–Technical University of Ostrava, 17. Listopadu 2172/15, 70800 Ostrava–Poruba, Czech RepublicFaculty of Materials Science and Technology, VSB–Technical University of Ostrava, 17. Listopadu 2172/15, 70800 Ostrava–Poruba, Czech RepublicFaculty of Materials Science and Technology, VSB–Technical University of Ostrava, 17. Listopadu 2172/15, 70800 Ostrava–Poruba, Czech RepublicFaculty of Materials Science and Technology, VSB–Technical University of Ostrava, 17. Listopadu 2172/15, 70800 Ostrava–Poruba, Czech RepublicFaculty of Materials Science and Technology, VSB–Technical University of Ostrava, 17. Listopadu 2172/15, 70800 Ostrava–Poruba, Czech RepublicIn recent years, the study of the hot deformation behavior of various materials is significantly marked by an increasing utilization of artificial neural networks, which are frequently employed for a hot flow curve description. This specific kind of description is commonly solved via a Feed-Forward Multi-Layer Perceptron architecture and rarely via a Radial Basis architecture. Both network architectures are compared to assess their suitability in the process of a hot flow curve description under a wide range of thermomechanical conditions. The performed survey is also aimed on the eventual utilization of corresponding modifications of both studied networks, namely on a Cascade-Forward Multi-Layer Perceptron and Generalized Regression network. The main results have shown that the Feed-Forward Multi-Layer Perceptron architecture represents a good choice if very high accuracy is a crucial goal. However, in the case of this architecture, finding the proper parameters can be time-consuming and the hardware burdensome. On the contrary, for the flow curve description the almost unused Radial Basis network offers a very easy training procedure and significantly shorter computing time under acceptable accuracy. The results of the submitted research should then serve as a background for the selection and following application of a suitable network architecture in the process of solving future flow curve description tasks.http://www.sciencedirect.com/science/article/pii/S2238785421007638Hot deformation behaviorHot flow curve descriptionMulti-layer feed-forward networkMulti-layer cascade-forward networkRadial basis networkGeneralized regression network |
| spellingShingle | Petr Opěla Ivo Schindler Petr Kawulok Rostislav Kawulok Stanislav Rusz Horymír Navrátil On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description Journal of Materials Research and Technology Hot deformation behavior Hot flow curve description Multi-layer feed-forward network Multi-layer cascade-forward network Radial basis network Generalized regression network |
| title | On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description |
| title_full | On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description |
| title_fullStr | On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description |
| title_full_unstemmed | On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description |
| title_short | On various multi-layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description |
| title_sort | on various multi layer perceptron and radial basis function based artificial neural networks in the process of a hot flow curve description |
| topic | Hot deformation behavior Hot flow curve description Multi-layer feed-forward network Multi-layer cascade-forward network Radial basis network Generalized regression network |
| url | http://www.sciencedirect.com/science/article/pii/S2238785421007638 |
| work_keys_str_mv | AT petropela onvariousmultilayerperceptronandradialbasisfunctionbasedartificialneuralnetworksintheprocessofahotflowcurvedescription AT ivoschindler onvariousmultilayerperceptronandradialbasisfunctionbasedartificialneuralnetworksintheprocessofahotflowcurvedescription AT petrkawulok onvariousmultilayerperceptronandradialbasisfunctionbasedartificialneuralnetworksintheprocessofahotflowcurvedescription AT rostislavkawulok onvariousmultilayerperceptronandradialbasisfunctionbasedartificialneuralnetworksintheprocessofahotflowcurvedescription AT stanislavrusz onvariousmultilayerperceptronandradialbasisfunctionbasedartificialneuralnetworksintheprocessofahotflowcurvedescription AT horymirnavratil onvariousmultilayerperceptronandradialbasisfunctionbasedartificialneuralnetworksintheprocessofahotflowcurvedescription |