Monitoring the dressing operation of conventional aluminum oxide grinding wheels through damage index, power spectral density, and piezoelectric sensors

Monitoring the dressing operation of grinding wheels is crucial for optimizing the grinding process and ensuring quality outcomes. This study presents a novel data-driven method utilizing piezoelectric diaphragm signals, combined with the root mean square deviation index (RMSD) and power spectral density (PSD), to determine the optimal moment for interrupting the dressing operation of conventional aluminum oxide grinding wheels. By addressing the existing gaps in transition methods between dressed and undressed grinding wheels, as well as exploring untested metrics in digital signal processing, this research expands the use of alternative piezoelectric transducers for monitoring dressing. The proposed methodology utilizes a commercial acoustic emission (AE) sensor as a reference and employs experimental dressing tests to validate its effectiveness. The signals from both the AE sensor and the piezoelectric diaphragm are collected and subjected to digital processing to extract relevant features based on the proposed approach. Results demonstrate that the RMSD index successfully extracts information about the cutting surface conditions of the grinding wheel from signals obtained by both AE sensors and piezoelectric diaphragms. Furthermore, by selecting frequency bands that exhibit strong correlations with the grinding wheel’s cutting surface conditions, a threshold is defined, enabling timely interruption of the dressing operation, thereby ensuring the restoration of the grinding wheel for continued use in grinding applications. Ultimately, this study showcases the feasibility of a non-invasive method for monitoring the dressing operation of conventional grinding wheels, contributing significantly to the optimization of the grinding process.