1. Process monitoring of the AISI D6 steel turning using artificial neural networks.
- Author
-
Rubin, Victor Hugo Serafim, da Silva, Leonardo Rosa Ribeiro, Okada, Kenji Fabiano Ávila, de Souza, Felipe Chagas Rodrigues, Pimenov, Danil Yurievich, Gravalos, Marcio Tadeu, and Machado, Alisson Rocha
- Subjects
- *
ARTIFICIAL neural networks , *CUTTING machines , *STEEL , *COLD working of steel , *ELECTRIC currents , *BORON nitride , *ACOUSTIC emission - Abstract
This work implements artificial neural networks (ANNs) to monitor tool wear during turning of AISI D6 steel. The goal is to detect wear, prevent tool breakage, and enhance productivity. Experimental tests involve using polycrystalline cubic boron nitride tools to turn AISI D6 steel, measuring wear and surface roughness. ANNs are trained with machining forces, electric current, and acoustic emission data. Findings showed that cutting and machining forces, along with motor current, increased as the wear progresses. The dominant frequencies of the acoustic emission signal range from 30 to 40 kHz, with decreasing RMS as wear increases. The ANN achieved satisfactory results, with wear estimation deviations below 0.09 mm. However, its predictive capability beyond the training range is limited. Workpiece surface roughness estimation showed around 8% error compared to the experimental Ra values. Overall, the ANN-based approach enables real-time tool wear monitoring, reducing the risk of damage and optimizing productivity. [ABSTRACT FROM AUTHOR]
- Published
- 2023
- Full Text
- View/download PDF