Enabling efficient use and recycling of cutting tools using multifidelity optimization (CutMore)

Enabling efficient use and recycling of cutting tools using multifidelity optimization (CutMore).

Cutting tools are the most costly elements of any manufacturing process. In the absence of a robust technique to estimate a tool’s remaining useful life (RUL), operators monitor tool deterioration empirically by tracking the level of sound/vibration in the machine. Recent studies have shown that models derived based on data of reasonable accuracy cannot predict a tool’s RUL in real time. In this project, we aim to enable the indirect monitoring and prediction of tools’ RUL via a data driven reduced-order model (ROM) based on a multi-fidelity optimization approach informed by both real-time data and computational estimates. Thereby, we intent to furnish a game-changing method for the manufacturing sector that will drastically reduce energy and material waste while increasing recycling potential. The ROM we will develop shall constitute a crucial first step towards the long-term vision of green manufacturing.