Process monitoring and analysis of laser welding and of milling
Abstract (Summary)Defects during fabrication of a part can implicate large costs for the manufacturer. In industry they use several types of measurement platforms/sensors to try to make the process more stable and detect when a defect is happening or has happened. The systems used mainly focuses on setting an upper and lower level where the signal from the sensor should be held within to withhold a satisfactory product quality. This thesis investigates two different processes, laser welding and milling and their accompanying defects. Although they are not equal process wise (mechanical and thermal) they share some similarities when using sensors for the detection/diagnostic of a defect. In some cases even the same sensor can be used for both processes. A platform for monitoring and control of milling was implemented, that was also done with the laser welding process. Both processes are described and explained. The milling process has been simulated with the crucial points laid on the vibration and force components acting on the work-piece and tool giving data on how to run the machine-tool. The process was monitored using piezoelectric force transducers, accelerometers and a Laser Doppler Vibrometer (LDV). The data was analyzed using both in-house developed and commercial software. The laser welding process has been simulated with respect to the melt pool surface radiation and the oscillations of the keyhole and melt-pool edges. It was observed with a high-speed camera, monitored with photodiodes and analyzed with commercial hardware/software systems. The aim of this work is to gain knowledge in why and how a defect arises, how it is represented in a sensor signal and if this knowledge can help industry in becoming more self-confident in using commercially available monitoring systems for defect analysis and monitoring. The work as a whole, over the last five years, consists of sixteen papers where seven have been chosen to be included in the thesis.
School:Luleå tekniska universitet
Source Type:Doctoral Dissertation
Date of Publication:01/01/2008