Aging Prediction Method for Laser Source and Lens of Laser Cutting Machine

• Main Authors: JHANG,CUN-WEI, 張存維
• Other Authors: YANG, HAO-CING
• Format: Others
• Language: zh-TW
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/huvnze

碩士 === 國立高雄第一科技大學 === 電機工程研究所碩士在職專班 === 106 === A laser cutting machine is pupularly used to slot wafer in the semiconductor packaging industry. The profile of a well cutted die should be evenness; however, it could appear crack or reamining metal when cutting quality getting worse such as insufficient laser Power output. A key for maintaining high availability is how to online estimate outputs of a cutting machine to timely change an aging component. This research develops a laser output prediction system, which can monitor and estimate the output Powers of the laser source and lens. In slotting wafer, the Powers of laser source with UV Power and the lens with PP Power are collected by frequencies 200 K, 100 K, and 40 KHz. This system supplies a web interface for indicating output Powers of the laser source and lens. In addition, the times to failure of the components are predicted based on the linear regression method. In results of three laser cutting machines, the failure durions of the target components can be estimtaed before their lead times after setting the corresponding lower bounds. Hence, the proposed system is promosing to monitor conponents and predcit thier times to failure and leading to change component for achieving the pourpose of predictive maintenance.. Keywords: Laser Slotting, Regression Prediction, Predictive Maintenance.