Minimization of Force Model Uncertainty for CNC Milling Process Improvement

University Of New Hampshire, Durham NH

Investigators

Abstract

The objective of this research is to reduce the uncertainty of machining models through the use of wireless sensors embedded in the cutting tool. The approach is to investigate methods for in-situ model calibration, performed with sufficient frequency, such that model accuracy is maintained within acceptable error bounds. Methods based on cutting power, time domain force and torque signals, and frequency domain signals will be explored. New sensor technology using wireless Bluetooth transmission of tool mounted sensors will be investigated to measure cutting torque and force. Linearity, signal to noise ratios, and cross talk will be evaluated. Time synchronization of the wireless sensor data with the machine process data will be necessary to perform accurate model calibrations. If successful, this research will greatly improve the reliability of machine tools by giving them a self-knowledge of their process capabilities. Methods for decreasing model uncertainty will improve the utility of physics based models for computer numerical control process planning leading to greater part quality and higher machining efficiency. Machining strategies and cutting conditions (spindle speeds and feeds) can be chosen to appropriately reflect the current capabilities of the machine tool. Wireless sensors will be developed which will be broadly used in both industry and academic research. This project is consistent with the goals of the Manufacturing Engineering Laboratory at the National Institute of Standards and Technology and the Integrated Manufacturing Initiative, a public/private consortium of industry, academic, and government partners designed to strengthen the nation's manufacturing community.