Online in-process monitoring ensures product quality and manufacturing productivity, providing fundamentals for promoting intelligent manufacturing. Motor current signature analysis (MCSA) allows for non-intrusive and cost-effective information perception and is widely accepted as a powerful tool for in-process monitoring. Signal amplitude-based methods from spectrum analysis provide good detection results but are not effective in diagnosing fine machining which causes small changes induced in motor current signals. This paper presents a Modulation Signal Bispectrum (MSB) based approach to analyzing current signals for accurate in-process monitoring. MSB can reliably utilize both amplitude and phase information and thereby achieve fine machine monitoring. It first presents the MSB analysis background analytically and then investigates the characteristics of modulation phases of motor signals numerically. Subsequently, an experimental verification was carried out based on a turning process of shaft-like workpieces with different cut parameters. It confirms that MSB phases allow for more accurate monitoring of the turning when the cut's depth is 0.5 mm. This is due to that MSB phases are relating to both static and dynamic cut loads. Comparatively, MSB amplitude shows high fluctuations as it correlates with dynamic loads, which can be attenuated significantly by long transmission paths in the turning process.
|Title of host publication
|Proceedings - 2020 3rd World Conference on Mechanical Engineering and Intelligent Manufacturing, WCMEIM 2020
|Institute of Electrical and Electronics Engineers Inc.
|Number of pages
|Published - 4 Dec 2020
|3rd World Conference on Mechanical Engineering and Intelligent Manufacturing - Virtual, Shanghai, China
Duration: 4 Dec 2020 → 6 Dec 2020
Conference number: 3
|3rd World Conference on Mechanical Engineering and Intelligent Manufacturing
|4/12/20 → 6/12/20