|
Modeling and critical analysis of material removal rate in WEDM of Oil Hardening Non Shrinking Die Steel (OHNS) |
|---|---|
| รหัสดีโอไอ | |
| Creator | 1. Mangesh Phate 2. Shraddha Toney 3. Vikas Phate |
| Title | Modeling and critical analysis of material removal rate in WEDM of Oil Hardening Non Shrinking Die Steel (OHNS) |
| Publisher | Faculty of Engineering, Khon Kaen University |
| Publication Year | 2563 |
| Journal Title | Engineering and Applied Science Research |
| Journal Vol. | 47 |
| Journal No. | 3 |
| Page no. | 264-274 |
| Keyword | Wire EDM, Buckingham's Pi theorem, Dimensional analysis, Response surface method, Artificial neural network, Oil hardening non-shrinking die steel, Support vector regression |
| URL Website | https://www.tci-thaijo.org/index.php/easr/index |
| Website title | Engineering and Applied Science Research |
| ISSN | 2539-6161 |
| Abstract | Wire EDM is a complicated machining process thatis used for producing complex 2D and 3D shapes. In this work, the process parameters associated with the wire electrical discharge machining (WEDM) of oil hardening non-shrinkage (OHNS) die steel material wereinvestigated through response surface method (RSM) and an artificial neural network (ANN). Aquadratic model developed through RSM wasused to predict material removal rate (MRR) with appreciable precision. The various input variables, viz.pulse on time (PON), pulse off time (POFF), wire feed rate (WFR) and input current (I),have been considered. Acomparison between the predicted and measured values of MRR wasperformed for each experiment. It wasnoted that the RSM predicted values are in a 95%confidence interval. Statistical analysis shows the capabilities of the developed models to predict the MRR more accurately. Also, ANN model estimates MRR with high precision compared using theRSM model. Support vector regression (SVR) is also usedto analyze the impact of various process parameters. The resultsshow that all approaches are strongly capable of predictingthe response.Analysis the WEDM is a very effective.Of the three approaches ANN issuperior. |
