College of Computing, Engineering & Construction
Master of Science in Electrical Engineering (MSEE)
NACO controlled Corporate Body
University of North Florida. School of Engineering
Dr. O. Patrick Kreidl
Dr. Alan Harris
Dr. John P. Nuszkowski
Most products sold today are packaged in a protective shell that involves the design of a box or wrapper. A subset of such products also adds a second layer of protection via sterilization. For both sterilized and non-sterilized products, a procedure referred to as the heat seal process creates the protective barrier from outside influence. For sterilized products, the American Society for Testing and Materials provides standards to test or verify seal strength, and this verification is normally accomplished by using a process called a Design of Experiments (DOE). The DOE method makes systematic use of powerful data collection and analysis tools, however, it also takes considerable time, capital, and resources to implement and verify. Moreover, when changes to the system or materials are necessary, the needed re-verification of the process compounds the effort needed to complete a subsequent DOE analysis. The objective of this thesis is to demonstrate the use of control-theoretic modeling and prediction algorithms to reduce the burden of DOE methods for heat-seal processes. Specifically, assuming the DOE analysis can collect data with sufficient instrumentation, we illustrate a two-pronged approach that employs (i) model identification from data to discern between success/failure of a heat-seal process and (ii) model-based feedback control to determine process reconfigurations towards failure recovery. Simulation experiments are presented that mimic the advent of heat-seal failures due to a new foil material and employ our approach to recover successful seals through minimal adjustments to the heater’s temperature profile. The extent to which the approach can apply to other process failure scenarios, different configurable inputs (e.g., seal pressure) or under non-ideal instrumentation assumptions is cited as future work.
Albanese, Charles, "A Predictive Modeling Approach to Counter Failures in Heat Seal Process Verification Methods" (2022). UNF Graduate Theses and Dissertations. 1142.