Modeling of process correlation using computational intelligence techniques

• Main Author: Yuan, Shuo.
• Other Authors: Er Meng Joo
• Format: Final Year Project
• Language: English
• Published: 2012
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Control and instrumentation
• Online Access: http://hdl.handle.net/10356/50107
• Institution: Nanyang Technological University

Roll-to-Roll UV Embossing System utilizes pneumatic pressure generator and slot dies to spray and smooth liquid resin on a roll of flexible plastic, and UV embossing technology to cure and harden the liquid resin. This final year project intents to model the roll-to-roll UV embossing process and achieve tentative prediction of certain quality factors. The author looks into pneumatic pressure applied on liquid photosensitive resin, web running speed, temperature sensor data and UV intensity to provide an indication on the condition of the machine and/or the process, to identify the patterns in thickness measurements and to construct the model between these parameters. Linear Regression and Back Propagation Neural Networks (BPNN) are the two static modeling techniques performed based on the data collected during experiments. Moreover, to investigate system dynamic model, an FOPTD model is tentatively developed to study on the system dynamics. Some in-depth physical explanations and justifications are elaborated and discussed as well.