Neuro-fuzzy techniques for modeling process parameters in pulsed electrochemical machining of steel
Pulsed Electrochemical Machining (PECM) is an advanced manufacturing method used in the metalworking industry to manufacture metal tools or components utilizing the principle of electrolysis. As a result, chemical and electrical phenomena of different natures are involved, making it difficult to de...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universidad De La Salle Bajío
2025-07-01
|
| Series: | Nova Scientia |
| Subjects: | |
| Online Access: | https://novascientia.lasallebajio.edu.mx/ojs/index.php/novascientia/article/view/3629 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Pulsed Electrochemical Machining (PECM) is an advanced manufacturing method used in the metalworking industry to manufacture metal tools or components utilizing the principle of electrolysis. As a result, chemical and electrical phenomena of different natures are involved, making it difficult to describe the behavior that governs material wear accurately. An approach known as ANFIS (Adaptive Neuro Fuzzy Inference System) was employed to address this issue, integrating two artificial intelligence techniques: artificial neural networks and fuzzy logic. This algorithm performs inference-based mapping using fuzzy logic while adaptively adjusting parameters through a supervised criterion, as in neural networks. For this purpose, experimental data from electrochemical machining divided into subsets were used to train and test the model. In this way, a root mean square error in the prediction of the material removal rate of 0.394 and a coefficient of determination of 0.845 were obtained. Finally, the analysis of the residuals allowed to exclude the presence of heteroscedasticity, confirming an adequate predictive performance without bias and with constant variance according to the Breusch-Pagan statistical test criterion
|
|---|---|
| ISSN: | 2007-0705 |