Predictive Maintenance for Utlrasonic Welding

This was my master’s research work work at Univeristy of Illinois Urbana-Champaign. Submited my master’s thesis and published a journal paper from this research work.

Objective

The objective of MS thesis was to develop an online tool condition moitoring system for Ultrasonic Welding aiming to optimize tool life and reduce production downtime cost.

Methodology

• Installation of an automated data acquisition system using IoT sensors and NI DAQ device
• Development of a GUI for data acquisition
• Design of experiments and data collection for research problem
• Data processing, feature engineering, and feature selection
• Machine Learning model development and hyper-parameter tuning

Tools & Technologies

DAQ, MATLAB, Feature Engineering, Machine Learning